<%BANNER%>

Synthesis and Characterization of Core-Shell Nanoparticles, and a Study of their Use as Encapsulation Agents

Permanent Link: http://ufdc.ufl.edu/UFE0021742/00001

Material Information

Title: Synthesis and Characterization of Core-Shell Nanoparticles, and a Study of their Use as Encapsulation Agents
Physical Description: 1 online resource (174 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: characterization, encapsulation, fluorescence, fret, nanoparticles, sequestration, solvation, uptake
Chemistry -- Dissertations, Academic -- UF
Genre: Chemistry thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: A mixed-surfactant system was used for the preparation of templates for the synthesis of core-shell particles. One of the amphiphiles used contained silanol groups that were condensed to fortify the templates. Subsequently, a silica coating was chemically attached, obtaining core-shell particles. It was observed by different techniques that the templates and core-shell particles were spherical and their diameter could be tuned by varying component composition. Atomic force microscopy studies after deposition on a hydrophilic substrate showed that these particles flattened dramatically, due to the formation of a flexible matrix in their interior. Formation of the shell increased their rigidity, preventing flattening to a certain degree. An experimental approach based on the use of hydrophobic fluorescent dyes was designed to study the polarity and rigidity of the templates and particles. By using a set of two different probes we found that the polarity of the interior of the particles decreased while its rigidity increased after coating. We used the fluorescence response of these dyes to study the uptake of hydrophobic compounds by the particles. It was determined that, based on their chemical structures, the sequestered probes were placed in different parts of the particles, sensing different environments. We used this information to study the compartmentalization of the interior of the templates and particles by using probes that could undergo resonance energy transfer if they were placed in close environments. These studies suggested that the sequestered probes could interact with each other producing a dynamic quenching of the donor. Finally, we covalently attached another hydrophobic fluorescent probe to the particles to study the level of interaction with the solvent, observing that the probe preferred to react in the interior pores of the particles to reduce interaction with the polar media.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Duran, Randolph.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021742:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021742/00001

Material Information

Title: Synthesis and Characterization of Core-Shell Nanoparticles, and a Study of their Use as Encapsulation Agents
Physical Description: 1 online resource (174 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: characterization, encapsulation, fluorescence, fret, nanoparticles, sequestration, solvation, uptake
Chemistry -- Dissertations, Academic -- UF
Genre: Chemistry thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: A mixed-surfactant system was used for the preparation of templates for the synthesis of core-shell particles. One of the amphiphiles used contained silanol groups that were condensed to fortify the templates. Subsequently, a silica coating was chemically attached, obtaining core-shell particles. It was observed by different techniques that the templates and core-shell particles were spherical and their diameter could be tuned by varying component composition. Atomic force microscopy studies after deposition on a hydrophilic substrate showed that these particles flattened dramatically, due to the formation of a flexible matrix in their interior. Formation of the shell increased their rigidity, preventing flattening to a certain degree. An experimental approach based on the use of hydrophobic fluorescent dyes was designed to study the polarity and rigidity of the templates and particles. By using a set of two different probes we found that the polarity of the interior of the particles decreased while its rigidity increased after coating. We used the fluorescence response of these dyes to study the uptake of hydrophobic compounds by the particles. It was determined that, based on their chemical structures, the sequestered probes were placed in different parts of the particles, sensing different environments. We used this information to study the compartmentalization of the interior of the templates and particles by using probes that could undergo resonance energy transfer if they were placed in close environments. These studies suggested that the sequestered probes could interact with each other producing a dynamic quenching of the donor. Finally, we covalently attached another hydrophobic fluorescent probe to the particles to study the level of interaction with the solvent, observing that the probe preferred to react in the interior pores of the particles to reduce interaction with the polar media.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Duran, Randolph.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021742:00001


This item has the following downloads:


Full Text





SYNTHESIS AND CHARACTERIZATION OF CORE-SHELL NANOPARTICLES, AND A
STUDY OF THEIR USE AS ENCAPSULATION AGENTS





















By

JORGE L. CHAVEZ BENAVIDES


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2008

































2008 Jorge L. Chavez Benavides




























To Gladys, Sarah and Victoria









ACKNOWLEDGMENTS

Many people helped me through the years this work was performed. The most important

people supporting me, despite the distance, have always been my parents and brother. Their

advice and faith have been crucial to finish this work without losing my mind.

Working in the Department of Chemistry has been an experience that changed my life in

different levels. I grew as a scientist and more importantly, as a person. In one of those long

days, when the pleasure of working was fading, I had the opportunity to meet the perfect match

for my unconventional soul. The lady that became my wife and, later, the mother of my

wonderful daughter, has been a source of strength and inspiration.

I had the chance to meet wonderful people in the Duran group, to all of them my respect

for their work and gratitude for their help with my research and the time shared, especially those

long hours in front of an instrument or a computer that, without them, would have been tedious

and difficult to stand.

All these years in Gainesville have been a great opportunity to learn about how human

beings from different parts of the world are so similar in some aspects and so different in others.

I have been fortunate to meet very interesting people, who have opened their hearts and minds to

me, without asking for anything. I hope my friendship has been for them as important as theirs to

me.

I thank Dr. Randy Duran for giving the chance to join his group and work in a project that

I enjoyed very much. I thank the staff in the Department of Chemistry at UF, especially in the

Butler Polymer Laboratories and the Graduate Program Office, for their help in so many things.

A significant part of the work presented here was carried out in the Materials Chemistry

Characterization Laboratory. I thank Dr. John R. Reynolds and Dr. Kirk Schanze for allow me to









use their facilities. Finally, I thank my advisory committee for taking the time to review this

contribution and for their comments and suggestions.









TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ..............................................................................................................4

L IST O F T A B L E S ......... .... .............. ..................................... ...........................10

LIST OF FIGURES .................................. .. .... ..... ................. 12

ABSTRAC T ................................................... ............... 16

L IST O F A B B R E V IA T IO N S ........................................................... ....................................... 18

CHAPTER

1 REVIEW OF SOME OF THE MOST IMPORTANT STRATEGIES OF
ENCAPSULATION IN NANOPARTICLES ............................................. ............... 20

1.1 Introduction ..... .................. ............................................. 20
1.2 Different Strategies for Encapsulation..................................... ...............21
1.2.1 Dendritic Architectures........................................................ 21
1.2.2 C ore-Shell and H follow Particles ........................................... .....................25
1.2.3 M icroem ulsions ............................ ....... ........... ................. 31
1.3 C conclusions ............................................................................................. ............... 36

2. METHODS AND EXPERIMENTAL PROCEDURES............... .... .......... ............... 39

2 .1 In tro d u ctio n ............................................................................................................ 3 9
2 .2 T techniques and M methods ..................................................................... ..................39
2.2.1 D ynam ic Light Scattering........................................... .......................... 39
2 .2 .1.1 T h eory ................................................................... ............ 39
2.2.1.2 Experim ental settings ........................................ .......................... 44
2.2.2 Transmission Electron Microscopy .............. .............................................45
2 .2 .2 .1 T h eory ................................................................... ............ 4 5
2.2.2.2 Experim ental settings ........................................ .......................... 46
2.2.3 A tom ic Force M icroscopy ...................................................... ..... .......... 47
2 .2 .3 .1 T h eory ................................................................... ............ 4 7
2.2.3.2 Experim ental settings ........................................ .......................... 48
2 .2 .4 p-P o te n tia l ................................................................................................... 5 0
2 .2 .4 .1 T h eory ................................................................... ............ 50
2.2.4.2 E xperim ental settings ........................ .... ................ ............... .... 52
2.2.5 Study of the Polarity and Viscosity of a Matrix by Steady State
Fluorescence Spectroscopy............................................................ .... ........ 53
2.2.5.1 Use of pyrene as polarity and rigidity probe ...........................................53
2.2.5.2 Use of C153 as a polarity probe .... ........... ..................................... 54
2.2.6 Time-Domain Lifetime M easurements .................................... ............... 55



6









2.2.6.1 Fluorescence resonance energy transfer studied by time-resolved
fl u o rescen ce .......................... ... ... .............. ..........................5 6
2.2.6.2 Time-correlated single photon counting (TCSPC).................................59
2.2.6.3 Determination of the degree of solvation of dansyl chloride by
fluorescence lifetime measurements ........... ................... ................60
2.2.6.4 Experim ental settings ........................................ .......................... 61
2 .3 E xperim mental P procedures ..................................................................... ..................6 1
2.3.1 Preparation of Tem plates............................................................. ............... 61
2.3.2 Synthesis of Core-Shell Nanoparticles.................................. ............... 62
2.3.3 Coating of the Particles with 3-aminopropylsilane ................................ ....63
2.3.4 Synthesis of the Complex between 3-aminopropylsilane and Dansyl
C h lo rid e ................ ....................... ............. ..................... ............... 6 3
2.3.5 Coating of the Templates and Core-Shell Particles with APSDC1
C o m p lex ................................................ ............ ..... .......... .... ....... 6 3
2.3.6 Loading the Dyes to Templates and Core-Shell Particles .............................64
2.3.7 Fluorescence Im ages................................................ ............................. 64
2.3.8 Steady State Fluorescence Spectroscopy .............. .. ... .............. ...................64

3 SYNTHESIS AND CHARACTERIZATION OF SURFACTANT-BASED
TEMPLATES FOR THE FORMATION OF CORE-SHELL PARTICLES .........................68

3.1 Preparation of Surfactant-Based Templates...................................... ............... 68
3.1.1 Strategy ..................................... ................................ ..........68
3 .1.2 R esu lts ................................................................... 6 8
3 .1.2 .1 S ize an aly sis ............................................................7 0
3.1.2 .2 T E M analy sis........... ......................................................... .... .... .... .. 74
3.1.2 .3 A F M analy sis............ .......................................................... .... .... .... 76
3.2 Preparation of D oped Tem plates ....................................................... ...................78
3.2.1 R results .......................................................... ... .....................79
3.2.1.1 Size analysis of ethyl butyrate-doped templates................... ......... 79
3.2.1.2 AFM analysis of ethyl butyrate-doped templates................ ......... 79
3.2.1.3 Size analysis of hexadecane-doped templates.................. .. ............. 80
3.2.1.4 AFM analysis of hexadecane-doped templates ............. ..................81
3.3 Other System s Studied ........... .... .. .... ... .. ..... ..... ... ................ 81
3.3.1 Use of a Cationic Surfactant in the Synthesis of the templates........................81
3.4 Preliminary Studies of the Strength of the Templates ...........................................82
3 .4 .1 S trate g y ....................................................................................................... 8 2
3.4.2 Results ........................... .... .... .. ....................82
3.4.2.1 Calcination of dopant-free templates............... ..... .. ............... 82
3.4.2.2 Calcination of ethyl butyrate-doped templates.......................................83
3 .5 C onclu sion s ................................................................................................ .... 83

4 TEMPLATE-BASED SYNTHESIS OF CORE-SHELL PARTICLES..............................97

4.1 Synthesis of C ore-Shell P articles...................................................... .....................97
4.1.1 Strategy ...................................... ................................ ..........97
4 .1 .2 R e su lts ...................................................................... 10 0









4.1.2.1 Size analysis of the core-shell particles.............................. .............. 101
4.1.2.2 TEM analysis of the core-shell particles .............. .................101
4.1.2.3 AFM analysis of core-shell particles..................................................... 102
4.1.2.4 Analysis of the dimensions of the templates and particles in
suspension and after drying ........................................ ............... 103
4.2 Derivatization of the Surface of the Nanoparticles............................105
4.2.1 Derivatization with 3-aminopropyltrimethoxysilane..............................105
4.2.2 Derivatization with dansyl chloride-APS complex ............ ... .................106
4.3 Preliminary Results on Other Systems Studied ..................................... ............... 108
4.3.1 Preparation of Large Templates and Core-Shell Particles for Calcination
S tu d ie s .................................................................... 1 0 8
4.4 Conclusions ..................................... ................................. ......... 110

5 SEQUESTRATION AND ENCAPSULATION OF FLUORESCENT DYES IN
TEMPLATES AND CORE-SHELL PARTICLES.............................................................120

5.1 Sequestration of Dyes by the Templates.................. ............................................. 120
5 .1.1 Strategy .......................................................................................... ..... 12 0
5.1.2 Use of Pyrene as a Polarity and Rigidity Probe .............................................122
5 .1 .3 R e su lts ................ ...... ............. ..... ........... ........ .. ...... ............... 12 3
5.1.3.1 Determination of the optimal conditions for the study of the
fluorescence of sequestered pyrene ....................................... ............ 123
5.1.3.2 Sequestration of pyrene by ethyl butyrate-doped templates....................124
5.1.3.3 Sequestration of pyrene by hexadecane-doped templates .......................125
5.1.3.4 Study on the retention of the dye after solvent exchange.......................126
5.2 Sequestration of Dyes by the Dopant-Free Core-Shell Particles............................126
5.2.1 Sequestration of Pyrene by Dopant-Free Core-Shell Particles.......................127
5.2.2 Sequestration of C153 by Dopant-Free Core-Shell Particles .........................127
5 .2 .2 .1 S trateg y .............................................................. ................ 12 7
5 .2 .2 .2 R e su lts .................................................................... .. 12 8
5.3 Encapsulation of Dyes by the Templates and Core-Shell Particles.........................129
5.3.1 Strategy ............... ......... ....... ......... 129
5.3 .2 R results ....................................... ..... ............................129
5.3.2.1 Encapsulation of pyrene in the templates and core-shell particles..........129
5.3.2.2 Encapsulation ofC153 in the templates and core-shell particles ............130
5.4 Conclusions .............. ...... .. ........... ........... ...... ................. 131

6 STUDY OF THE COMPARTMENTALIZATION OF THE TEMPLATES AND
CORE-SHELL PARTICLES THROUGH ENERGY TRANSFER BETWEEN DYES.....137

6.1 Energy Transfer in the Tem plates.................................................................. 137
6.1.1 Strategy ............. ........................................ ..... .. ........ 137
6.1.2 R results .................... .................... .......... 139
6.2 Energy Transfer in the Core-Shell Particles ................................... ............... 141
6 .2 .1 S trateg y ......... ............... .................................... ..................................... 14 1
6 .2 .2 R esu lts ...................................................... 14 2
6.2.2.1 Pyrene-C153 pair ......... ......................... 142









6.2.2.2 C 153-D CM pair......... .......................... .......................... 144
6.3 Energy Transfer in the Core-Shell Particles Studied by Fluorescence Lifetime
M easurem ents ....................................................... 145
6.3.1 Strategy ............... .... ......... .................. 145
6.3.2 Results .............................................146
6.4 Use of the Fluorescence Lifetime of a Probe Chemically Attached to Study
Microenvironments Inside the Core-Shell Particles ......................................... 147
6.4.1 Study of the Fluorescence of Dansyl Chloride Attached to the Periphery
of th e P articles.............. ............ ...................................... ............. ..... 14 7
6.4 .1.1 Strategy .............. ........................................................... .......... ... ... 147
6.4 .1.2 R results .............. .............. ...... ..... ..... ..................... ...............147
6.4.2 Preliminary Studies on the Energy Transfer between Dansyl Chloride
Attached to the Periphery of the Particles and Sequestered DCM ...................149
6 .4 .2 .1 Strategy ............... ........................................................... .......... ... ... 149
6 .4 .2 .2 R esu lts ...............................................................14 9
6.5 Conclusions ......... ...... ....... ........................... .................. 151

7 CONCLUSIONS AND SUGGESTIONS FOR FUTURE WORK................................158

7 .1 C o n c lu sio n s ............... ......................... ................................................................ 1 5 8
7.2 Suggestions for Future W ork .......................... .. .............................. ............... 162

APPENDIX

A APSD C1 COM PLEX NM R ....................................................... ........ .......164

B MODIFIED STERN-VOLMER PLOTS FOR EB AND HX TEMPLATES ......................165

C FLUORESCENCE LIFETIME FITS FOR C153-DCM FRET ..........................................166

L IST O F R E F E R E N C E S ......... ................. ............................................................................. 167

B IO G R A PH IC A L SK ETCH ........................................................................................ 174



















9









LIST OF TABLES


Table page

3-1 Formulations used for the preparation of templates and characterization results
obtained by D L S and TEM ....................................................................... ...................85

3-2 Size analysis results obtained by DLS and AFM for templates D2 and D3............... 85

3-3 Formulations used for the formation of ethyl butyrate-doped templates, and size
analy sis results obtained by D L S ........................................ ........................................ 85

3-4 Topographic information for ethyl butyrate-doped templates obtained by AFM. DLS
data is show n for com prison. ................................................ ............................... 86

3-5 Formulations used for the formation of hexadecane-filled templates and size analysis
results obtained by DLS and AFM .............................................................................. 86

4-1 Formulations used for the synthesis of templates and corresponding core-shell
p articles ........................................................... ....... ......................................... 1 1 1

4-2 DLS characterization results for templates and the corresponding core-shell particles.. 112

4-3 AFM characterization results for templates (1) and core-shell particles (2) prepared
as indicated in Table 4-1 .................. ................................. .. .. ........ .... 112

4-4 Volume of templates and coated particles based on DLS and AFM data .......................112

4-5 Formulation and size characterization results for APS-coated nanoparticles (NPAPS)..112

4-6 Formulations used for coating of the particles with APS-DC1 and DLS
characterization results ..... ............. .............. ...... ...... ...... .... .. .......... .. 112

5-1 13/I1 values for pyrene in ethyl butyrate (EB) and hexadecane (Hx) and EB and Hx-
doped tem plates. ........................................................................ 132

5-2 Formulation used in the synthesis of core-shell particles for sequestration studies and
size characterization results obtained by DLS. ............. ........................................132

5-3 Fluorescence data for pyrene and C153 sequestered by dopant-free core-shell
particles. ...................... ...................................... 133

5-4 Steady State Fluorescence data for templates and core-shell particles used to
encapsulate pyrene (NPY) and C153 (NCU) .... ........... ....................................... 133

6.1 Changes in the intensity of the first peak of sequestered pyrene emission (371 nm)
after addition of C153 ............. ................. .......... ........................ 152









6.2 Emission maximum of sequestered C153 in pyrene-doped particles and comparison
of the intensity of its emission obtained when excited at 339 and 420 nm .....................152

6-3 Steady state fluorescence data for the pyrene-C153 encapsulated pair (Xex= 339 nm)....153

6-4 Steady state and fluorescence lifetime measurements results for the energy transfer
betw een C 153 and D CM .................. ........................................................... .. .. 153

6-5 Steady state and fluorescence lifetime measurements results for templates and
particles coated w ith dansyl chloride........................................ ........................... 153

C-1 Fluorescence lifetime fits for C153-doped templates used for FRET with DCM..........166









LIST OF FIGURES


Figure pe

1-1 Synthesis of dendrimers for encapsulation. Divergent (A, D) and convergent (B, C)
approaches to the synthesis of dendrimers for the encapsulation of active cores or
active e m o lecu les ..................................................... ................ 3 7

1-2 General approach of the synthesis of hollow particles using a removable core for the
encapsulation of active species. ............................................... .............................. 37

1-3 Synthesis of hollow particles based on the use of lipids. These particles can
encapsulate active molecules in the water pool in their interior .....................................38

1-4 Particle synthesis by microemulsion polymerization. ............................ ...................38

2-1 Possibilities for the interaction of a laser beam with a liquid sample. In a
homogeneous medium, waves of the scattered light interact destructively, producing
no scattering (A); while in a heterogeneous medium scattering is produced (B)..............65

2-2 Comparison of the correlation function of two set of templates with different
diameters as determined by DLS. The diameters obtained were 47 nm (solid line)
and 72 nm (em pty circles).............................................. ................... ............... 66

2-3 A simple representation of the basic concept of a transmission electron microscope
operating in the bright field m ode........................................................ ............... 66

2-4 A simple representation of the basic setup of the atomic force microscope....................67

2-5 Data analysis of a Z-axis calibration grid performed in contact mode............................67

3-1 Schematic representation of the synthesis of core-shell particles................................86

3-2 Chemical structures for surfactants, amphiphiles and dopants used for this study. ..........87

3-3 Ideal schematic for the hydrolysis and condensation of the methoxy groups in OTMS
at basic conditions ............................ ................ ......... 88

3-4 Size distribution obtained by DLS for a typical set of templates A) before and B)
after p u rificatio n ..................................................... ................ 8 8

3-5 (-potential variation as a function of pH for non pure (circles) and purified (squares)
tem plates ........................................................ .................................89

3-6 Variation of the normalized correlation function obtained by diluting a suspension of
particles to reach different light scattering intensities. A) templates, intensity: 3.5 x
105 kcps-purple line, 2.5 x05 kcps-green line, 1.1 x 105 kcps-red line, 0.5 x 105









kcps-blue line, and B) SO/t80 micelles, intensity: 3.0 x 105 kcps-purple line, 2.0 x
10 kcps-green line, 1.0 x 105 kcps-red line, 0.5 x 105 kcps-blue line............................89

3-7 TEM images for templates: a) B2, b) C2, c) D2. Scale bars 200 nm. ...........................90

3-8 TEM images showing what are believed to be templates deposited with their longest
axis perpendicular to the surface of the substrate (indicated by the red boxes). Scale
b ars 5 0 0 n m ........................................................................... 9 0

3-9 AFM tapping mode images for samples A) D2 and B) D3. On the bottom part of
each im age the section analysis is shown. .............................................. ............... 91

3-10 AFM images and section analysis for templates A) EBB, B) EBC, C) EBE and D)
EBF. The arrows indicate what are believed to be folded templates..............................92

3-11 AFM images and section analysis for templates A) HxB, B) HxC, C) HxE and D)
HxF. The arrows indicate what are believed to be folded templates...............................93

3-12 AFM image of two set of templates prepared with DTAB replacing SO ..........................94

3-13 TEM images of dopant-free templates before (A) and after (B, C, D) calcination.
Scale bars 200 nm ........................................... ............................ 94

3-14 AFM images of dopant-free templates before (A) and after (B, C, D) calcination. .........95

3-15 TEM images of ethyl butyrate-doped templates before (A) and after (B, C, D)
calcination. Scale bars 200 nm ................................................ ............................... 95

3-16 AFM images of dopant-free templates before (A) and after (B) calcination.....................96

4-1 Reaction scheme for the hydrolysis of TMOS, shown for the first A) and second B)
hydrolysis products. The reaction of these species with the silanol groups on the
surface of the templates is shown, as well as the final product after complete
hydrolysis at neutral pH. ........... .............................. ......... ...... ......... 113

4-2 Crosslinking resulted when templates were not diluted before the addition of TMOS.
Scale bar 500 nm .............. ............. ....... .......................... ........ 114

4-3 TEM and AFM height images of NEB A) templates and B) core-shell particles.
Scale bars in all im ages 500 nm ....................................................................... 114

4-4 TEM and AFM height images of NCU A) templates and B) core-shell particles.
Scale bars in all im ages 300 nm ....................................................................... 115

4-5 TEM and AFM height images of NPY A) templates and B) core-shell particles.
Scale bars in all im ages 500 nm ....................................................................... 116

4-6 Attachment of APS to core-shell nanoparticles. ............. ............................................ 116









4-7 AFM images for the templates and APS-coated nanoparticles reported on Table 4-4....117

4-8 Comparison of (-potential results before (circles) and after (squares) derivatization
with APS: A) NPAPS; and APS-DC1: B) NPDCl-1A, C) NPDCl-1B and D) NPDC1-
2A (squares) and N PD C12B (triangles). ..........................................................................117

4-9 Coupling between dansyl chloride and APS..................................................118

4-10 TEM and AFM images ofN-benzyl maleimide-doped templates (A and C) and core-
shell particles (B and D). Scale bars 500 nm. .......... .......... ......... ......................... 118

4-11 TEM images of N-benzyl maleimide-doped templates (A and B) and core-shell
particles (C and D) after calcination at 600 OC. Scale bars 500 nm.............................119

5-1 Fluorescence response detected after excitation of a dye-free nanocapsule
suspension. The peak observed is due to scattering of light by the particles, excitation
w avelength 270 nm ....................................................................... ........ .. 33

5-2 Chemical structures of the dyes used in this study. ........................................................ 133

5-3 Emission of a 1 [tM pyrene solution in A) hexadecane and B) ethyl butyrate. ...............134

5-4 Determination of conditions to study the polarity of templates with pyrene. A)
Pyrene titration for hexadecane-doped templates, numbers in the box indicate pyrene
concentration, B) plot of the ratio of the intensities of the third and first emission
peaks of pyrene (13/11) at different dye concentrations. ...................................................134

5-5 Representation of the removal of dopant molecules during purification of the
templates. The positions preferred by pyrene molecules when sequestered by a
suspension of templates, is shown as well. ........................................ ...............135

5-6 Emission of pyrene loaded in A) dopant-free templates and B) EB-doped templates,
before (black line) and after (grey line) dialysis ............................. ... ......... ....... 135

5-7 Representation of the positions preferred by pyrene and C153 molecules when
sequestered by core-shell particles........................................................ ............... 135

5-8 Normalized emission of pyrene encapsulated in templates and in the corresponding
core-shell particles (blue) for the NPY system ..................................... ............... 136

5-9 Fluorescence microscopy images for C153 encapsulated in core-shell particles, scale
bar 10 tm. Image was obtained in black and white, color was added for clarity B)
Normalized emission of C153 encapsulated in templates (red line) and in the
corresponding core-shell particles (blue line)................................. ............. ........... 136

6-1 Energy transfer between pyrene and C153 in templates. A) Overlap of the emission
of sequestered pyrene and excitation of sequestered C153, B) changes in the
fluorescence spectrum of pyrene (1 aM) after addition of C153 C) typical excitation









spectra of sequestered C153 (0.6 pM, x,,e 520 nm) in pyrene-loaded templates (1
pM), D) efficiency of the energy transfer observed for templates EBB (diamonds),
EBC (squares), HxB (triangles), HxC (circles). ........................................ ..................154

6-2 Energy transfer between pyrene and C153 in core-shell particles. A) Emission
spectra of encapsulated pyrene before (solid line), and after sequestration of to
different concentrations of C153 (0.14 pM-empty circles and 0.7 pM-empty
triangles); B) comparison between the emission of the doubled-doped nanoparticles
excited at the excitation maxima of pyrene (solid line) and C153 (empty circles). ........155

6-3 Schematic representation of the energy transfer between probes based on their
placement inside the nanoparticles. The particles are divided in zones with different
polarities based on the penetration of the solvent....................................................... 155

6-4 Energy transfer between C153 and DCM. A) Overlap between C153 emission (solid
line) and DCM excitation (empty circles); B) emission spectra of nanoparticles
doped with sequestered C153 (0.7 pM, empty circles), and after sequestering DCM
(0.14 pM-empty triangles and 0.7 pM-solid line); C) excitation spectra for
sequestered C153 (0.7 pM, empty circles, emission 539 nm), sequestered DCM (0.7
pM, solid line, emission 615 nm) and DCM in nanoparticles with both dyes
sequestered (0.7 pM each, empty triangles, emission 610 nm); D) comparison
between the emission of the doubled-doped nanoparticles (0.7 pM each dye) excited
at the excitation maximum of C153 (420 nm, solid line) and DCM (475 nm, empty
circles). 156

6-5 Energy transfer between DC1 and DCM. A) Nanoparticles system APSDC1-2A, X.ex
370 nm (blue line), ,xc 450 nm (black line); solid DCM added: ,xc 370 nm (green
line), Xexc 450 nm (red line). B) Nanoparticles system APSDC1 2B: Xexc 340 nm
(black line); after sequestration of DCM: 0.5 iM (red line), 2.5 iM (green line) and
2 .5 M ( 4 50 n m ) .................................................................................... .... 157

A-i Chemical structure of the APS-DCl complex......... ...... ...................... ........ ........... 164

A-2 NMR spectrum of the APS-DC1 complex. ........... .. ............................ ................. 164

B-1 Modified Stern-Volmer plots for hexadecane-doped templates .....................................165









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

SYNTHESIS AND CHARACTERIZATION OF CORE-SHELL NANOPARTICLES, AND A
STUDY OF THEIR USE AS ENCAPSULATION AGENTS

By

Jorge L. Chavez Benavides

May 2008

Chair: Randolph S. Duran
Major: Chemistry

A mixed-surfactant system was used for the preparation of templates for the synthesis of

core-shell particles. One of the amphiphiles used contained silanol groups that were condensed to

fortify the templates. Subsequently, a silica coating was chemically attached, obtaining core-shell

particles. It was observed by different techniques that the templates and core-shell particles were

spherical and their diameter could be tuned by varying component composition. Atomic force

microscopy studies after deposition on a hydrophilic substrate showed that these particles

flattened dramatically, due to the formation of a flexible matrix in their interior. Formation of the

shell increased their rigidity, preventing flattening to a certain degree. An experimental approach

based on the use of hydrophobic fluorescent dyes was designed to study the polarity and rigidity

of the templates and particles. By using a set of two different probes we found that the polarity of

the interior of the particles decreased while its rigidity increased after coating. We used the

fluorescence response of these dyes to study the uptake of hydrophobic compounds by the

particles. It was determined that, based on their chemical structures, the sequestered probes were

placed in different parts of the particles, sensing different environments. We used this

information to study the compartmentalization of the interior of the templates and particles by

using probes that could undergo resonance energy transfer if they were placed in close









environments. These studies suggested that the sequestered probes could interact with each other

producing a dynamic quenching of the donor. Finally, we covalently attached another

hydrophobic fluorescent probe to the particles to study the level of interaction with the solvent,

observing that the probe preferred to react in the interior pores of the particles to reduce

interaction with the polar media.









LIST OF ABBREVIATIONS


1-d 1-dodecene

AFM Atomic force microscopy

APS 3-aminopropylsilane

C153 Coumarin 153

d Diameter

DC1 Dansyl chloride

DCM 4(dicyanomethylene)-2-methyl-6-(p-dimethylaminostyryl)-4H-pyran

DHR Diameter to height ratio

DLS Dynamic light scattering

DTAB Dodecyltrimethylammonium bromide

EB Ethyl butyrate

FRET Fluorescence resonance energy transfer

h Height

kcps Kilocounts per second

NCU Coumarin 153-doped particles

NEB Ethyl butyrate-doped particles

np Nanoparticles

NPY Pyrene-doped particles

OTMS Octadecyltrimethoxysilane

RT Room temperature

rpm Revolutions per minute

SO Sodium octanoate

ST Shell thickness

t80 Poly(oxyethylene)[20]-sorbitan monooleate









TCSPC Time-correlated single photon counting technique

TEM Transmission electron microscopy

TMOS Tetramethoxysilane

UA Uranyl acetate









CHAPTER 1
REVIEW OF SOME OF THE MOST IMPORTANT STRATEGIES OF ENCAPSULATION IN
NANOPARTICLES

1.1 Introduction

Encapsulation of a variety of species has been the focus of attention of many research

groups in the last several years.13 Reaction vessels,4-8 drug delivery9-11 and catalysis12 are three

examples of many potential applications that encapsulation systems offer. Different

methodologies for the synthesis of such systems have been presented in the literature, including

microemulsions,13-16 polymers,17-20 and dendrimers.21-27 Especially important in this field is the

study of core-shell particles. These systems offer the possibility of combining two different

environments in the same particle, which enhances the solubility of the encapsulated molecule if

the polarity of the core is the opposite to that of the shell. Moreover, protection of the

encapsulated agent from environmental changes that it could experience, like pH, ionic strength,

and the like is another important property that can be accomplished by these architectures.

When such encapsulating agents fit into the nanometer scale, new and exciting options for

their use appear. This is especially important in applications related to medicine, since these

devices can be injected in the blood system without any risk of embolization. Moreover, since

the encapsulated molecule is not in direct contact with the environment, irritation at the site of

administration will be reduced.28

In this chapter, our goal is to review the most relevant contributions of different groups in

this field in the last years. We have focused our attention on contributions based on several key

synthetic approaches to sub-micron encapsulation, a brief description of the synthetic

methodology will be presented, and results that show their potential toward novel applications

will be discussed.









1.2 Different Strategies for Encapsulation

1.2.1 Dendritic Architectures

Since their introduction in 198529 dendrimers have become a powerful tool for the

development of new technologies. Figurel-1 shows two typical approaches to the encapsulation

in dendrimeric structures. Based on their synthesis, the dendrimers can be prepared by a

convergent or divergent approach. The encapsulated molecule can be used as the core of the

dendrimers or it can be incorporated after the macromolecule has been synthesized. In recent

years, both strategies have been reported in the preparation of molecular capsules. In this section

we will review the most important contributions in this area.

In a convergent approach, Morgan et al.30 have synthesized different generations of

poly(glycerol succinic acid) and tested them for encapsulation of a dye and drug delivery. The

synthesis of these polymers was accomplished by first reacting succinic acid with cis 1,3-0-

benzylidenglycerol (Bdg). Deprotection gave as a product a tetrahydroxy core. The synthesis of

the dendrons was performed reacting Bdg with succinic anhydride in pyridine. Coupling of these

two species gave the first generation dendrimer. Higher generation dendrimers were synthesized

in a similar way. Encapsulation of Reichardt's dye (2,6-diphenyl-4-(2,4,6-

tryphenylpyridinio)phenolate) was achieved by adding the dye to a methanolic solution of the

dendrimer. It was found that the dye could be encapsulated only with third and fourth generation

dendrimers. The authors used UV-vis spectroscopy to show that the polarity of the core was

similar to that of glycerol. NMR was used to elucidate the interaction of the dye with the

dendrimer, showing that the dye molecules experience restrictions in their motion. These

macrostructures were tested as delivery agents for 10-hydroxycamptothecin (10HCPT), a

topoisomerase I inhibitor. One important feature is that the physical properties of this molecule

are different than Reichardt's dye. This shows the versatility of the systems used. 1H-NMR NOE









was used to prove the internalization of 10HCPT within the dendrimer. The anticancer activity of

the encapsulated inhibitor in human breast cancer cells was evaluated after being released from

the capsules. In a more elegant, divergent approach, Furuta et al.31 have achieved the site

isolation of two dyes, Coumarin 343 and pentatiophene, in the core of dendrimers. The goal of

this work was to prevent quenching of the emission of the dyes when they were used in close

proximity in the same device. This is needed in order to avoid energy transfer between dyes that

leads to emission of solely the dye with the lowest HOMO-LUMO band gap. This group

performed the synthesis by placing the dye molecules at the center of a Frechet-type dendrimer.32

Solubilization of the higher generation dendrimers was enhanced by alkyl ether functionalization

of the periphery of the dendrimer. The surface of these molecules was further functionalized with

hole-transporting triarylamine groups whose emission overlaps with the absorption bands of the

dyes investigated. Films of the dendrimer were prepared with different thicknesses and showed

to be uniform and aggregate-free. Thin film photoluminescence (PL) experiments demonstrated

that the isolation of the dye was more effective as the size of the dendrons surrounding the cores

increased. PL studies of the encapsulated dyes incorporated in a glassy polystyrene matrix

showed that the emission of both dyes could be detected without significant quenching. This

confirmed that the isolation of the dye by encapsulation in the dendrimer reduced the extent of

the energy transfer process.

In a divergent approach, Sunder et al.33 synthesized hyperbranched polyglycerols based on

the anionic ring-opening multibranched polymerization (ROMBP) of glycidol, obtaining

polymers with narrow polydispersities. Further derivatization of the free hydroxyl groups with

different fatty acid chlorides resulted in dendrimers composed of hydrophilic cores and

hydrophobic chains on the surface. Congo red, an anionic-water soluble dye, was chosen for









encapsulation studies. After mixing an aqueous solution of the dye with a chloroform polymer

solution, the probe was extracted quantitatively below the saturation concentration in the

capsules. This behavior was shown to be general, as observed for other dye-solvent

combinations. The response of the encapsulated dye to environmental changes was studied by

changing the pH of the water phase in contact with the organic solution. It was shown that the

dye, despite being protected by a hydrophobic shell, was still able to respond to external

stimulus. In a more elaborated process, Schappacher and Deffieux34 synthesized dendrimers by

polymerizing a-diethylacetal polystyrillithium (DEAPS) onto poly(chloroethylvinyl ether)

(pCEVE) to form a comb-polymer core. After purification, the polymer chains were extended by

cationic polymerization of CEVE. A second chain extension process was performed

polymerizing DEAPS. The hydrophilic extension was performed by polymerizing, in sequence,

2-((tert-butyldimethylsilyl)-oxy)ethyl vinyl ether and di-O-isopropylidene-6-O-(2-

vinyloxyethyl)-D-galactopyranose. The final step involves the deprotection of the poly(vinyl

ether) blocks. As a result, a hyperbranched core-shell architecture is achieved when dissolved in

an aqueous solution. Dynamic Light Scattering (DLS) measurement in water and a mixture of

solvents showed that the dendrimer had a diameter of approximately 100 nm. Zinc, cobalt and

iron metallo-porphyrins were encapsulated by pouring a solution of the porphyrins in DMF into a

large volume of a water/DMF solution containing the dendrigraft. Size exclusion

chromatography was used to prove the encapsulation of the metallo-porphyrins into the

polymers. UV-vis absorption of the metallo dimer complex was used as the detection technique.

The dendrimers synthesized by this group showed high encapsulation capacities. The authors

observed that encapsulation produced significant changes in the size of the capsules, and, as a

consequence, in their volume. The dendrigraft capacity for porphyrin complexation was shown









to be dependent on the metal and the structure of the porphyrin used. The shift in the soret band

of the porphyrins showed that the encapsulated species did not present significant conformational

restrictions. This is in agreement with the DLS results, which indicated that the porphyrin

molecules occupy a high volume in the dendrimers. To demonstrate the polarity of the

environment in which the metallo-porphyrin complexes were located, UV absorption was used,

determining that it was a poly(styrene)-like environment.

Chen and Guan35 have accomplished the synthesis of dendrimeric micelles by

copolymerizing ethylene and its poly(ethylene glycol) derivative. Using the Brookhart

palladium-ac-diimine chain walking catalyst, they placed the poly(ethylene glycol) moiety at the

end of the PE chain. In solution, these dendrimers produce water-soluble unimolecular micelles.

These macromolecules were tested as encapsulating agents by dissolving nile red, a hydrophobic

dye, in the core of these dendrimers. UV absorption of the encapsulated dye was compared to

that of the probe in an aqueous suspension of sodium dodecyl sulfate (SDS), observing that the

core of their capsules was more hydrophobic than the SDS micelles. They also showed that the

capacity of solubilizing nonpolar dyes was related to the molecular weight of the hydrophobic

core.

Niu and Crooks36 derivatized dendrimers with hydrophobic groups on the exterior of the

molecules and encapsulated metals nanoparticles in the core. The authors functionalized

poly(propylene imine) 37 dendrimers by reacting the amine terminal groups with palmitoyl and

hexanoyl functionalities. The derivatization was performed by dissolving the polymer in dry

CH2C2 and mixing it with triethylamine under Ar. The final step involved the addition of

palmitoyl chloride. A similar procedure was used for the hexanoyl derivative. The dendrimer

encapsulated metal nanoparticles (DEMNs) were synthesized by dissolving the polymer in a









mixture of CHC13/MeOH. CuC12 was dissolved in the same solvent mixture and this was added

to the previously prepared dendrimer solution. Subsequently, the metal ion was reduced with

NaBH4. The encapsulation of the metal ions was followed by UV; observing that the loaded

capsules showed a different spectrum than the dendrimer or ion alone. After the reduction of the

metal ion, precipitation was not observed, which showed the stabilization due to the

encapsulation in the dendrimers. Moreover, the absorption properties of the complex changed.

Since the encapsulation agents studied by this group did not contain any peripheral amine groups

and the interior amine groups did not seem to coordinate strongly with metal ions, the

encapsulation process is not driven by coordination to the metal ion. Based on these

observations, the authors concluded that the encapsulation process is more likely to occur based

on solubility differences.

1.2.2 Core-Shell and Hollow Particles

Figure 1-2 shows a general approach to the synthesis of core-shell and hollow particles for

the encapsulation of different species, based on a removable core-templating strategy. This can

be done in the micro and nano scale, which offers the possibility of their use in different

applications, as discussed below.

Skirtach et al.38 used two different methods to synthesize capsules containing a core in the

micrometer range (4-15 [tm) coated electrostatically. In the first case, multiple layers of

alternating polyelectrolytes, poly(sodium 4-styrenesulfonate) (PSS) and poly(allyl-amine

hydrochloride) (PAH) were deposited and doped with Ag nanoparticles. In the second case, a

cationic polyelectrolyte and an ionic dye (IR-806) were alternatively deposited. After deposition

of the layers, the cores were dissolved to result in hollow microcapsules with light absorption

enhancers. When these systems were irradiated with a laser source, light in the IR region (830









nm) was absorbed by the dye or the Ag nanoparticles. The authors showed that irradiation of the

doped capsules produced distortion of the shell, and depending on the size of the particle and

intensity of the laser, this effect could be perforation, deformation, or an ablative process. One

important aspect of this work is that the wavelength of the laser used corresponded to the

biologically friendly window (where the absorption of tissue, biomedical substances, and water

is minimal). Using the same system, Peyratout et al.39 promote the crystallization of the cationic

dye, 1,1'-diethyl-2,2'-cyanine (PIC) in the interior of the hollow particles. Monomeric PIC does

not show any significant fluorescence, but at high concentration it forms the fluorescent so-called

J-aggregates. Since the formation of these fluorescence aggregates is promoted by the presence

of negatively charged polyelectrolytes, PSS was synthesized in the interior of the capsules and

subsequently PIC was added. In order to precipitate the dye forming the PIC-PSS complex,

tetraphenyl borate (TPB) was added to the interior of the capsules. Changes in the absorption and

emission spectra showed that a PIC-TPB complex was formed, sequestering the dye from the

PSS matrix. Confocal laser scanning microscopy (CLSM) and atomic force microscopy (AFM)

were used to characterize the species formed and their location in the hollow capsules. In a

combination of the use of the layer by layer deposition technique and the use of dendrimers as

potential encapsulating agents, Khopade et al.40 have deposited PSS and a positively charge

fourth generation poly(amidoamine) dendrimer on spherical latex colloids. (-potential was used

to confirm the alternate deposition of the charged polymers on the surface of the growing

particles. Carboxyfluorescein was loaded in the capsules by exposure of the colloids to an

aqueous solution of the fluorescent dye. The internalization of the dye was studied by CLSM,

showing that the dye placed itself in the shell of the dendrimer. The release of the dye in a NaCl

aqueous solution was evaluated as a function of time. The authors observed a Fickian type









release. Finally, hollow capsules were obtained by a similar approach, by using MF as the core

and fortifying the final product with a PAH/PSS outer coating.

Sun and Chiu41 prepared Ag nanoparticles with sizes between 25 and 200 nm by reducing

silver nitrate in an ethylene glycol/poly(vinyilpyrrolidone) solution. Subsequently, these particles

were coated with a silica shell and the silver core was dissolved in an ammonia solution. Two

techniques were used for encapsulation. The first method consisted of mixing a nanocapsule

aqueous suspension with a fluorescein solution in 2-propanol/chloroform. The second method

consisted of drying the nanocapsule suspension, mixing with a 2-propanol/chloroform

fluorescein solution and resuspension in borate buffer. They proved that these systems could be

used for delivering small molecules using larger capsules, commercially available polystyrene-

acrylic-based hollow microcapsules coated with a silica shell. This process was followed by

widefield fluorescence imaging and confocal point detection. The authors also investigated the

delivery of carbachol from the nanocapsules. They trapped cells loaded with muscarinic

acethylcholine receptor type I with a laser. Subsequently, a small volume of a suspension of the

particles was added near the trapped cells. Since carbachol is an agonist of this receptor, the

calcium signalling in the cells transfected with these receptors was observed as an indication of

the delivery of the carbachol molecules.

Taking advantage of the core-shell architecture of the precursor used in this technique,

encapsulation in the pores of the core material has been proposed as a new vehicle for

encapsulation. Wang and Caruso42 have encapsulated enzymes in mesoporous silica particles and

deposited polyelectrolyte layers on their surface to protect the encapsulated agent and avoid

leaking. The authors studied silica particles with two different pore size distributions. Bimodal

mesoporous silica particles (BMS) with 2-3 nm and 10-40 nm pores sizes and 2-4 |tm diameter









were used for these studies. BMS showed a high enzyme-loading capacity due to their large pore

size, although the amount of enzyme loaded depended on the molecular weight (MW) of the

enzyme. CLSM was used to study the loading ability of the particles after derivatization with

fluorescein isothiocyanate-labelled peroxydase. It was shown that the enzyme adsorbed in BMS

could be distributed on the surface and inside the particle. The coated capsules were obtained by

alternate assembly of three layer pairs of poly(diallylammonium chloride) and 21 nm silica

nanoparticles. The authors proved the protective ability of the coating layers towards catalase (C-

100) by comparing its behavior against non-coated particles. BMS spheres showed that 25% of

the enzyme had leaked after 6 h of being exposed to a PBS solution, while no leaching was

detected for the coated particles. Similarly, the encapsulated enzyme presented an enhanced

stability against changes in pH and H202 degradation. When the stability against proteolysis was

studied, it was observed that the as-prepared capsules retained 75% of the enzyme activity and

total protection was obtained when eight layers of PAH-PSS were deposited. Balabushevich et

al.43 studied the release of the enzyme encapsulated by different methods in these particles. The

one consisted of the formation of a complex between insulin and dextran sulfate (DS), while the

second was based on the salting out of insulin microaggregates coated with DS and/or protein.

Both systems showed that the protein could be release at pH values higher than 5, with a faster

release when the pH was further increased. At the same time, when more polyelectrolyte layers

were deposited, the release of the enzymes at the same pH was lowered.

A different technique for the synthesis of hollow capsules is based on a core-templated-

free approach, where the use of a soluble core is avoided. This is preferentially used when the

synthesis of the capsules involves a polymerization process. Turner and Wooley44 have

synthesized a diblock copolymer formed with poly(tert-butyl acrylate) and polyisoprene and









used them as templates for the formation of nanocage structures. Acidolisys of the tert-butyl

acrylate groups with a methanesulfonic acid catalyst transformed this macromolecule into

amphiphilic structures that self-assembled in mixed solvent environments. Crosslinking of the

shell formed and subsequent ozonolysis of the core resulted in hollow structures. Reduction of

the intermediate ozonides was performed with sodium sulphate. This gave as a result aldehyde

and ketone groups in the interior of the hollow capsules. Among other studies, the authors

compared the capabilities of the core-shell structures and the hollow particles for the

encapsulation of BODIPY, a fluorescent dye. BODIPY was dispersed from methanol into an

aqueous solution of the polymer. After ozonolysis and reduction of the inner functional groups,

the interior of the particles was less hydrophobic than the interior of the core-shell precursor.

This was confirmed by the lack of encapsulating ability in the hollow particles. In another

polymerizable system, Hu et al.45 synthesized hollow polymeric nanospheres with biocompatible

and biodegradable polymers for the controlled release of an antitumor drug. The biopolymer

chitosan (CS) with 90% deacetylation was dissolved in an acrylic acid (AA) aqueous solution.

Due to their different charges, the AA monomers are attracted to the cationic biopolymers, being

trapped in the CS micelles. AA was polymerized within the micelles by initiation with K2S208,

and CS was crosslinked with glutaraldehyde. DLS and (-potential measurements were used to

study the formation of the hollow capsules. After polymerization of AA and crosslinking of CS,

the micelles showed a dramatic decrease in their radio and (-potential values. Transmission

electron microscopy (TEM) showed the different species present at the different stages of the

synthetic process, confirming that hollow structures were formed at the end. In order to

encapsulate doxorubicine, the drug was added to the nanospheres prior to the CS crosslinking.









After purification, the release of doxorubicine was followed by steady state fluorescence,

observing that the drug could be delivery for up to 15 days.

In a simple and biocompatible design, lipids are used in solution to form vesicles to be

used as encapsulating agents, as shown in Figure 1-3. In this case, the encapsulation is not used

to enhance solubility, but to separate reactive species from the solution. The walls of the hollow

capsules are used as a selective membrane for the intake of reactive compounds dissolved in the

surroundings of the capsules. Graff et al.46 used a liposome system to form vesicles that were

fortified by polymerizing a hydrophobic monomer inside the walls of the capsules. Membrane

channels were added to the lipid bilayer in order to enhance the permeability of the capsules.

Preparation of the hollow particles involved dissolving the lipid, 1-palmitoyl-2-oleoyl-sn-

glycero-3-phosphocoline (POPC) in chloroform. After evaporating the solvent, the membrane

channel, OmpF, dissolved in an emulsion was added and vortexed. A film was obtained by

drying the mixture and the enzyme p-lactamase was added at pH 7.4. Purification and

subsequent homogenization of the sample produced hollow capsules with diameters of

approximately 300 nm. A suspension of the capsules was mixed with an n-butyl methacrylate

solution and ethylene glycol dimethacrylate (the initiator). Polymerization was performed by

irradiation of the sample with UV light (254 nm). The activity of the enzyme in the hydrolysis of

ampicillin was analyzed for the capsules with and without the OmpF channels. The authors

observed that there was no enzymatic activity if the channels were not present to enhance the

permeability of the walls of the capsule towards ampicillin. Moreover, the activity of the

encapsulated enzyme was comparable to that of the free enzyme in high substrate concentration.

The effect of the presence of the monomer on the enzyme activity was studied as well. It was

shown that, after the polymerization, the activity of the functionalized vesicles decreased









considerably. The authors interpreted this as a consequence of the closure or expulsion of some

of the channels in the liposomes. Similarly, Lawson et al.47 have used derivatized phospholipids

for the preparation of vesicles to be used as encapsulation agents. The authors place 3,5-divinyl

benzoyl, a polymerizable group, in the polar heads of two lipids 1,2-dipalmitoyl-sn-glycero-3-

phosphoethanolamine and 1,2-dilauryl-sn-glycero-3-phosphoethanolamine, in order to

polymerize them. The authors observed by TEM that the vesicles have the same diameter before

and after the polymerization, approximately 70 nm. The same technique was used to prove that

the presence of the polymerized units in the surface stabilized the capsules compared to the non-

polymerized ones. A mixture of derivatized phospholipids was used to prepare hollow capsules

loaded with wild-type phosphotriesterase. The ability of the encapsulated enzyme to hydrolyze

methyl paration (MPT) was studied, showing that the encapsulated enzyme could hydrolyze

MPT. In a more sophisticated approach, Bolinger et al.48 have encapsulated small vesicles (SV)

in large vesicles (LV), in order to control the release of molecules from SVs into LVs. Addition

of an anionic lipid in both types of vesicles (LV and SV) helped maintaining SVs suspended

inside LVs. After encapsulation of a fluorescent dye at a self-quenching concentration in the

interior of the SVs, the thermotropic phase transition-induced release of the dye was analyzed by

fluorescence confocal microscopy. The authors observed an increase in the intensity of the

fluorescence inside the LVs after changing the temperature of the system from 25 to 45 C.

1.2.3 Microemulsions

Microemulsion systems have been used in the past due to their ability to enhance solubility

of active agents in aqueous solutions, especially in biomedical applications. In the synthesis of

encapsulating systems, organic, non-polar compounds are suspended in the presence of

surfactants in order to react or crystallize, as shown in Figure -4 for a suspension of a monomer.









If polar reactive groups are placed on the surface, they can be used to grow a coating to fortify

the particles.

Yang and Zhang49 synthesized CdSe quantum dots (QD) in the micro- and nanometer

range and encapsulated them with a polymer shell. Encapsulation was achieved by incorporating

a toluene solution of QD into a CTAB emulsion. A mixture of monomers, styrene,

divinylbenzene and methacrylic acid, including the initiator (AIBN) was added. After 20 h of

polymerization at 70 oC, the polystyrene particles in the emulsion were precipitated and washed

with water and ethanol to remove the surfactant. After the addition of NaOH, the negatively

charged particles were coated with a silica shell by mixing an acidic solution of tetraethoxysilane

(labelled with a small amount of rhodamine) with an aqueous solution of the polymer-coated

particles. Control on the size of the particles was achieved by changing the oil/water ratio and the

amount of surfactant. The particles were characterized by QD and rhodamine fluorescence. It

was shown that the emission of the QD was not quenched by coating.

Wang et al.50 used a surfactant solution as a template for the synthesis of core-shell and

hollow nanoparticles. The core was synthesized in an aqueous basic solution of benzethonium

chloride by polymerizing dimethoxydimethylsilane (D). This process was followed by

endcapping of the free hydroxyl groups with methoxytrimethylsilane (T). In order to synthesize

core-shell particles, a mixture of D and T was added to the previously formed core. Free

hydroxyl groups on the shell were encapped by mixing with T and hexamethyldisilazane. As a

result, core-shell microparticles were obtained. Hollow nanocapsules were formed by dissolving

the capsules in THF and subsequent extraction of the core. TEM showed that the architectures

obtained were strong enough to maintain their shape after the extraction. The average size of the

core-shell particles was approximately 300 nm. After the removal of the core the average sizes of









the hollow particles shrunk to approximately 200 nm. TEM images showed that the shell

thickness was approximately 70 nm. AFM was used to confirm the hollow features of the

capsules. Indomethacin (Idm) was encapsulated by dispersing the capsules in an aqueous

bezethonium chloride solution swollen with an indomethacin/CH2Cl2 emulsion. TEM

micrographs showed that Idm could diffuse into the capsules. Jungmann et al.51 used similar

core-shell and hollow nanocapsules for the encapsulation of dyes with different polarities and

charges. These capsules contained a cationic additive in the core of the core-shell architectures or

in the inner part of the hollow capsules. The synthesis is based on the polymerization of T,

(chloromethylphenyl)trymethoxysilane (ClBzT) and D using benzehtonium chloride as a

surfactant.52 The core-shell particles become hydrophilic by quaternarization of ClBzT. In order

to prepare hollow nanocapsules, the core was not covalently attached to the growing shell, so it

could be removed after the shell was synthesized. Solid-liquid and liquid-liquid phase transfer

techniques were used to load dye molecules in the capsules. In the former, a solid dye is added to

the dispersion of the capsules in toluene, in the latter, an aqueous solution of the dye is covered

with a layer of the nanocapsules dispersion. Both techniques were showed to produce

encapsulation. This process seemed to be dependent on the presence of the quaternary amino

groups. This is especially important since it was observed that cationic dyes were able to be

encapsulated (to a lower extent), despite electrostatic repulsion could be thought as an

unfavorable interaction. The mesh width of the nanocapsules was investigated. The lack of

uptake of a sterically demanded dye showed that this is an important parameter that allows

control over the uptake process.

Kumar et al.53 have achieved the enzymatic synthesis of copolymers in order to form

nanomicelles for the encapsulation and delivery of drugs. The synthesis of the copolymers was









performed by mixing PEG (Mn 600) and dimethyl 5-hydroxyisophtalate with the enzyme

Novozyme-435 (Candida Antarctica Lipase B), at 90 OC for 48 h. The macromolecule obtained

was reacted with different alkyl bromides in acetone, to functionalize the polyester with different

pendant groups. Due to the amphiphilic character of the copolymers, they formed micelles in

hydrophilic and hydrophobic solvents. In order to encapsulate aspirin, the copolymer and the

drug were dissolved in chloroform. After evaporation of the solvent, the mixture of both species

was dissolved in water, resulting in drug loaded nanomicelles. 1H NMR longitudinal relaxation

time studies were performed to study the micellization process. DLS and Static Light Scattering

(SLS) were used to determine the size and shape of the micelles formed. The particles obtained

showed sizes between 15-20 nm for the different polyesters derivatives and their shapes varied

from hollow to star-like spheres. SLS and UV absorption were used to confirm the encapsulation

of aspirin and napronex. A slight increase in the capsules Rg compared to the hollow capsules

was observed and the UV absorption of the loaded capsules showed the peaks of both, the

nanospheres and the drug. The release of both active molecules was tested in animals, showing a

significant efficiency in the reduction of inflammation compare to a similar dose of the

commercially available drugs.

Panyam et al.54 used poly(D,L-lactide-co-glycolide) (PLGA) and polylactide (PLA), in an

emulsion-solvent evaporation technique in order to prepare micro-and nanocapsules loaded with

drugs. PLGA and PLA of different molecular weights were mixed in different ratios with the

drug in order to study its influence in the capsules synthesis. After this, the loading capacity of

the particles and the in vitro release of the drugs was studied. Tritiated-labeled dexamethasone

was dissolved in methanol and mixed with a polymer solution in chloroform. Polyvinyl alcohol

was added as an emulsifier and the mixture was sonicated first and then stirred for 18 h to









evaporate the organic solvent. In the microparticles formation sonication was replaced by

vortexing. Purification was performed to wash the emulsifier out of solution and separate the

non-encapsulated drug. DLS results showed that the size of the nanoparticles varied between 240

and 270 nm. When low MW polymers were used, the size of the particles increased up to 740

nm. The authors found that polymer combinations with lower amount of polar groups produced

capsules with higher drug loading. X-ray diffraction and differential scanning calorimetry were

used to examine the physical state of the drug, observing that there was no crystalline drug

present in most of the formulations, except when low MW polymers were used (and larger sizes

were obtained). When studying the in vitro release of the drug, it was observed that capsules with

a higher loading capacity had lower release rates. This was interpreted as a consequence of the

lower partitioning of the drug from the polymer to the external aqueous phase. Moreover, the

release was found to be faster when the difference between the solubility parameters of the

matrix and the drug was higher. In a similar approach, Bae et al.55 have synthesized polymeric

micelles based on poly(ethylene glycol)-poly(aspartate-hydrazone-adriamycin) (PEG-PAHA).

The synthesis was based on the ring opening polymerization of P-benzyl-L-aspartate N- carboxy-

anhydride with ca-methoxy-co-amino poly(ethylene glycol) in dimethylfomamide. Extension of

the polymer chain ends was accomplished by adding hydrazide groups that were further

derivatized with adriamycin (ADR, a fluorescent, anticancer agent) through an acid labile

hydrazone bond. The amphiphilic polymer formed micelles in a polar solvent, placing the drug

molecules in the core. DLS analysis showed that the diameter of the micelles was approximately

65 nm. The influence of the pH on the release of ADR from the polymeric matrix was studied by

reversed-phase liquid chromatography. It was observed that, at pH below 5.5, the active

molecule-polymer bond broke, delivering ADR. The loaded drug micelles and the drug by itself









were incubated in a cell culture medium. CLSM was used to identify the intracellular

localization of the micelles. Since the fluorescence of the drug is only possible when is outside

the micelles (and not inside due to quenching for the high concentration when encapsulated), the

authors showed that after 24 h of incubation the micelles were located in the cytoplasm and that

ADR was released.

1.3 Conclusions

A variety of synthetic techniques have emerged in recent years that greatly enhance the

ability to achieve controlled encapsulation of molecular species. In this Chapter, recent advances

in several sub-micron encapsulation strategies have been discussed. In particular, dendritic

architectures, hollow particles, microemulsion-based systems, and core-shell nanocapsules have

been highlighted. Together, these techniques show promise for the encapsulation of a variety of

species in a diverse range of conditions, which makes them a valuable tool for the design of

nanocontainers, fluorescent tags and delivery systems.

In the work presented in the following chapters, we offer a different approach to the

synthesis of nanoparticles. Our goal was to design a simple, fast and relatively cheap

methodology for the synthesis of core-shell particles for different applications. We decided to

use a combination of the most practical aspects of the approaches reviewed here. We used a

templating system based on the use of surfactants. This approach takes advantage of the self-

assembly of amphiphiles in aqueous suspensions, producing species with a hydrophobic core and

a hydrophilic periphery. We then used sol-gel chemistry to chemically connect the polar groups

of one of the surfactants and used the free groups on the surface of the particles to attached

different species. In order to form the core-shell particles, a silica shell was grown on the

periphery of the particles by a sol-gel process.








A B


/A
vI^


NW


VO Dendron
Active Core
S Non Active Core
Active Molecule


I


Figure 1-1. Synthesis of dendrimers for encapsulation. Divergent (A, D) and convergent (B, C)
approaches to the synthesis of dendrimers for the encapsulation of active cores or
active molecules


Removal

SAnionic Cationic
*Soluble core p= Polyelectrolyte Polyelectrolyte


Active
*" Molecule


Figure 1-2.General approach of the synthesis of hollow particles using a removable core for the
encapsulation of active species.


c .


T

I


D

N'


I


















= Lipid


0 Active Molecule


Figure 1-3.Synthesis of hollow particles based on the use of lipids. These particles can
encapsulate active molecules in the water pool in their interior.


T+0


sh&W
Formaon


Sufactant
Removal


S- Surfactant


N-Monomer


Figure 1-4.Particle synthesis by microemulsion polymerization.









CHAPTER 2
METHODS AND EXPERIMENTAL PROCEDURES

2.1 Introduction

The templates and core-shell particles were characterized by dynamic light scattering,

transmission electron microscopy, atomic force microscopy and (-potential measurements. It was

crucial for us to characterize the particles in detail, in order to learn how to control their size.

This is especially important if we keep in mind that different applications may require the

synthesis of particles with different sizes. In addition, to elucidate the polarity of the interior of

the particles, we used steady state fluorescence spectroscopy and fluorescence lifetime

measurements to study the response of different probes internalized by the templates and core-

shell particles. Below we describe briefly the basic principles of the techniques used for the

characterization studies, with emphasis on their application to study particulate systems. At the

end of each section we include the details of the experiments performed in this study.

2.2 Techniques and Methods

2.2.1 Dynamic Light Scattering

2.2.1.1 Theory

The principle of the dynamic light scattering technique is that the scattering of light can be

viewed as a result of microscopic heterogeneities within the illuminated volume. When a volume

of a homogeneous sample is illuminated with a beam of light, the scattered waves will have the

same amplitude and interfere destructively in all directions, except in the direction of the incident

beam. If a heterogeneous sample is being analyzed, the index of refraction would differ from the

average value at some location, as a consequence, the wave that is scattered at this location

would not be compensated for and some light would be observed in directions other than the

incident light and light scattering occurs, as shown in Figure 2-1.56









There are two ways to approach the phenomenon of light scattered by particulate matter in

solution. The one is to consider the suspension as a homogeneous medium and ascribe light

scattering to the spatial fluctuations of the solute. This is appropriate for solutions of small

molecules in which the average distance between the center of the scatterers is small compared

to the wavelength of light. The second is to consider each individual solute particle as a

heterogeneity and therefore as a source of light scattering. The latter is more appropriate for

solutions of large macromolecules and colloids, where the average distance between particles

centers is comparable to the wavelength of light.57

In cases in which the size of the scatterers is not small compared to the wavelength of

light, the interference of the electromagnetic waves scattered by the solute is not all constructive

and the phases of these waves must be taken into account. If the phase of a wave scattered at the

origin is used as a reference, the phase of a wave scattered at a point with radius vector r is q.r.

The vector q is called the "scattering vector," which is a fundamental characteristic of any

scattering process. The length of the vector is indicated in Equation 2-1.56


q = q = 4--sin where rI is the refractive index of the medium (2-1)
A 2 X is the wavelength of light
0 is the scattering angle

The essence of the DLS technique is to measure the temporal correlations in the

fluctuations in the intensity of the scattered light and to reconstruct from these data the physical

characteristics of the scatterers. In a suspension of particles, the scatterers are randomly

distributed within the scattering volume. Since the size of the scattering volume is much larger

than q -1 (with the exception of nearly forward scattering, where q- 0 = 0), the phases of the

waves scattered by different particles vary dramatically. As a result, the average amplitude of the

scattered light is proportional to N1/2, where N is the number of scatterers, and the average








intensity is simply N times the intensity scattered by an individual particle. The local intensity,

however, fluctuates from one point to another around its average value. As the scattering

particles move, the interference pattern changes in time resulting in temporal fluctuations in the

intensity of light detected at the observation point.5

In a DLS measurement, the instrument detects a random signal. The information is

contained in the correlation function of this signal i(t), which in the case of DLS is the

photocurrent, defined as in Equation 2-2

G 2()= (i(t) i(t + ) ) (2-2)

The notation G2(r) is introduced to distinguish the correlation function of the photocurrent

from the correlation function of the electromagnetic field G'(r) (which is the Fourier transform

of the light spectrum)

G(r) = (E(t) .E*(t + T) ) (2-3)

The angular brackets denote an average over time t. This time averaging, an inherent

feature of the DLS method, is necessary to extract information from the random fluctuations in

the intensity of the scattered light.

In the majority of practical cases, the scattered light is a sum of waves scattered by many

independent particles and therefore displays Gaussian statistics. This being the case, there is a

relation between the intensity correlation function G2(r) and the electromagnetic field correlation

function G' (r):

G2(r) = I2 (+ yg(r_2) (2-4)

Here gl(r)= G1(z)/ G1(0) is the normalized field correlation function, Io is the average

intensity of the scattered light and y is the intensity factor.5









Temporal fluctuations of the intensity of the scattered light are caused by the Brownian

motion of the scatterers. The speed of the particles is related to their size, small particles move

faster than large particles. Though each particle moves randomly; in a unit time more particles

leave regions of high concentration than region of low concentration. This results in a net flux of

particles along the concentration gradient. Brownian motion is then responsible for the diffusion

of the solute, and is quantitatively characterize by the diffusion coefficient, D. Rigorous

mathematical analysis of the process of light scattering by Brownian particles leads to the

following expression for the correlation function of the scattered light:

g () = exp(-Dq 2) (2-5)

The diffusion coefficient in an infinitely dilute solution is determined by particle geometry.

For spherical particles, the relation between the radius R and its diffusion coefficient D is given

by the Stokes-Einstein Equation:

k T where: kb is the Boltzman constant
D = b- r is the viscosity of the solution (2-6)
67rrR Tis the absolute temperature

Hence, from Equation 2-5 the diffusion coefficient can be obtained from the correlation

function g'(r). Assuming that the scatterers are spherical, Equation 2-6 can be used to obtain the

hydrodynamic radius of the particles. We used the imaging techniques to investigate whether the

shape of the particles matched this requirement.

In our studies, the suspensions obtained where not monodisperse but rather, made of

particles a range of sizes. This made the analysis of the data obtained more complicated. In the

case of polydispersed samples, a different analysis of the correlation function is required. For a

continuous distribution of scattering particle size, the correlation function is obtained from the

following equation










g(1'r)= I(D)exp(-Dq2r) dT (2-7)

where I(D)dD= N(D)Io(D)dD is the intensity of light scattered by particles having their

diffusion coefficient in the interval [D,D+dD], [N(D)dD] is the number of these particles in the

scattering volume, Io(D) is the intensity of light scattered by each of them.58

The main goal of the software is to reconstruct as precisely as possible the distribution

function I(D) from the experimentally measured function G2exp(r).The main problem encountered

is that dramatically different distributions I(D) lead to nearly identical correlation functions of

the scattered light and therefore are equally acceptable fits of the experimental data. There are

three typical approaches to solve this problem:

* Direct fit method: In which a functional form of I(D) is assumed apriori and the parameters
that lead to the best fit of G2theor(r) to G2exp(r) are determined.
* Method of cumulants: This approach focus on the "stable" characteristics of the
distribution, which are the moments of the distribution, or closely related quantities called
cumulants. The first cumulant of the distribution that gives the average diffusion coefficient
D*, can be determined from the initial slope of the field correlation function, as shown in
Equation 2-8. The second cumulant, the width of the distribution, is obtained from the
curvature of the initial part of the correlation function.
-d1
-dIng(1)(j = i I(D)Dq2dD = Dq2 (2-8)
dr 19 1O
* Regularization: Is a combination of the first previous methods mentioned, it assumes that
the distribution I(D) is a smooth function and seeks a non negative distribution producing the
best fit to the experimental data. This method is used in different approaches used to
reconstruct the scattering particle distribution from DLS data. The key point is the selection
of the smoothness of the distribution. If the smoothing is too strong, the distribution will be
stable but will lack details. If the smoothing is too weak, false spikes will appear in the
distribution. The moments of the distribution reconstructed by the regularization procedure
gives unbiased (apart from smoothing) information on the shape of the distribution.58


The primary technique used to follow the effectiveness of the synthesis of the templates

and particles was DLS, due to its ease of operation and the small amount of sample needed. As

shown in Figure 2-2, the difference in size between different species can be visualized qualitative

by a comparison of the normalized correlation function obtained. Similarly, a quantitative









analysis can be performed in the case of coating particles with a layer of other material by

determining the diameter of the particles before and after the coating process. In our case, from

the difference in sizes between the templates and core-shell particles, a shell thickness can be

calculated using Equation 2-7. The size determination was performed after purification of the

species obtained, to minimize the influence of the presence of small micelles in the

determination of the correlation function. In addition, the suspensions had to be filtered to

eliminate dust or aggregated particles that could not be separated by sonication. This is

especially important due to the dependence of the intensity of the light scattered by a particle to

the sixth power of its radius, which would result on the signal of the particles being obscured by

the aggregates, producing an overestimation of their diameter.

ST = (dz d,)/2 where: di is the diameter of the templates (2-9)
d2 is the diameter of the core-shell particles

2.2.1.2 Experimental settings

Throughout this work two instruments were used, a PDDLS/CoolBatch 90T detector with

a PD2000DLSplus correlator, manufactured by Precision Detectors Inc, and a 90 Plus/BI-MAS

detector with a BI 9000AT digital correlator manufactured by Brookhaven Inc. In both cases the

detector was placed at 900 from the incident beam. Each instrument works with its own software

for size determination, the Precision Deconvolve 32 in the first case, and the BI-ISDAW

advanced size distribution software, in the second case.

The templates and core shell particles were sonicated for approximately 30 minutes and

filtered before DLS analysis. Sonication was intended to break any aggregates formed, which

was observed after storing the samples for months, especially after purification, while filtering

was used to remove aggregates that could not be broken up. The samples needed to be diluted to

avoid multiple scattering. The extent of dilution differed from templates to core-shell particles









due to the differences in sizes and, as a consequence, on the intensity of light they scattered. On

average, the concentration of templates and core-shell particles used was below 0.1 mg/mL. The

appropriate concentration was determined on a sample by sample basis by examination of the

intensity of the light scattered through the instrument. As indicated in the manuals, a total

intensity of the scattered light between 100 to 300 kilocounts per second (kcps) was used; being

the concentration of scatterers adjusted until these values were reached. Subsequently, the

sample time was adjusted after a quick examination of the correlation function. The sample time

is directly related to the size of the particles, so it can be adjusted to obtain the proper fit of the

correlation function. After these settings were adjusted, the measurements were performed in 10

minutes runs. This time was found to be optimal to obtain a good signal to noise ratio, resulting

in stable and reproducible results. The final results reported were obtained by the average of five

measurements.

2.2.2 Transmission Electron Microscopy

2.2.2.1 Theory

The need to use the electron microscope is due to the size of the particles to be imaged.

Since their diameter is similar to the wavelength of the light used in common light microscopes,

they cannot be characterized by this technique. In such cases it is necessary to use the

transmission electron microscope, in which the use of light has been replaced with a beam of

electrons. This results in better resolution, due to the shorter wavelength of the electrons,

compared to photons. In the electron microscope, electromagnetic lenses are used to focus the

beam of electrons into a tight coherent beam, which is then focused on the sample. Data is

collected after the beam has passed through the sample. Our imaging has been performed in the

bright field mode. This means that after the beam of electrons has passed the sample deposited

on the substrate, an objective aperture has been inserted. This aperture allows the electrons in the









transmitted field to pass and contribute to the resulting bright field image, rejecting the electrons

that had been scattered by the sample,59 as shown in Figure 2-3.

One of the problems encountered with our templates and core-shell particles is their lack of

electron density. This could be due to the formation of a porous matrix in the interior of the

particles or the formation of disk-like species, as will be discussed later. To overcome this

problem different staining techniques were tested, including the use of OsO4 to stain

unsaturations and uranyl acetate to stain the hydrophilic part of the particles (mostly their

periphery). This improved the quality of the images obtained, especially by the use of uranyl

acetate.

2.2.2.2 Experimental settings

The instrument used in this study was a Hitachi H-7000, with a maximum resolution of 0.2

nm and a maximum acceleration voltage of 125 kV. Imaging the particles synthesized in this

work was not an easy task due to their low electron density, as explained above. Moreover, due

to their size, especially the templates, getting a decent focus that could be used to measure the

size of the particles was difficult. Even trying higher acceleration of the electron beam through

higher voltages failed, as a consequence, our images were taken with a voltage of 75 kV, the

usual setting of the instrument. Considering that the instrument was not available for use at our

will and the fee for its use, we decided to use this technique to image our particles and confirm

the geometries observed by AFM, technique then selected for the size characterization of the

particles, as explained below.

Similarly to what was observed with DLS, the particles needed to be diluted before loading

them on the grids. If the concentration was too high, the particles were deposited as multilayer,

making impossible their visualization. It was determine that a concentration of about 0.1 mg/mL

was optimal in most of the cases to load the grids. At the beginning of these studies, particles to









be imaged were made with 1-dodecene to stain its unsaturation with Os04 and visualize the core

and shell of the particles. After deposition of the suspensions on the grid and evaporation of the

solvent, the grids were placed in a small glass chamber where they were exposed to Os04 vapors,

for approximately 30 minutes. Using longer times made the images too dark and oxidization of

the grids was observed in some cases. Before insertion of the samples containing the grids in the

microscope, the samples were stained with a methanolic solution of uranyl acetate (UA) for 1

minute. This is a common staining agent for biological samples, used primary to stain

hydrophilic regions. Images were taken at different spots of the same grid to confirm that the

species observed were present in most of the suspension and that they were not an unusual

finding. Size analysis was performed by comparison of the diameter of the particles and the scale

bar in the image. This was done in the Adobe photoshop software and later with the Image J

software.

2.2.3 Atomic Force Microscopy

2.2.3.1 Theory

Atomic force microscopy, since its introduction in 1986, has been used in many different

applications, especially because of the improved resolution compared to other microscopy

techniques. In our case, the main reason to use this technique is the ability of AFM to obtain

information on three dimensions without the need of any coatings or stains, an advantage over

TEM. Especially important in our case is the determination of the particles height after

deposition on a substrate, which we used as an indication of the rigidity of the particles, and how

the coating of the templates affects this.

The technique is based on the interaction of the tip of a cantilever with the sample

deposited on a substrate. The instrument measures the forces between the tip of the flexible

cantilever and the sample. The basic idea is that the local or attractive forces between the tip and









the sample are converted into a bending, or deflection, of the cantilever. The key feature is that

the force between the probe and the sample is maintained constant while the probe is raster

scanned across the surface. In order to detect the probe bending, a laser beam is reflected from

the back of the cantilever onto a detector, in such a way that a small change in the bending angle

of the cantilever is converted to a measurable large deflection in the position of the reflected

spots. The deflection observed then is converted into an electrical signal, to produce an image of

the surface. A simple representation of the instrument setup is shown in Figure 2-4. In order to

avoid any non desired interaction between the probe and the particles, the instrument was used in

tapping mode. In this mode, the cantilever is oscillated close to its resonance frequency and the

tip taps the surface only periodically, reducing significantly the lateral force. This means that the

probe is free to oscillate up and down at its resonance frequency as a consequence of the

interaction with the substrate when it comes extremely close to it. In tapping mode, the image is

obtained by imaging the force of the oscillating contacts of the tip with the sample surface.60'61

2.2.3.2 Experimental settings

The instrument used in this study was a Nanoscope III, manufactured by Digital

Instruments, Inc. Imaging was performed in tapping mode, using silicon probes (Nanosensors,

with dimensions: T=3.8-4.5 im, W= 26-27 im, L= 128 im). The z-calibration was performed

with a silicon grating (TGZ01, Mikromash), with a step height of 20 nm (accuracy 1 nm).

Images were analyzed with the Nanoscope III software provided by the manufacturer.

To determine how accurate the z-measurements were in the instrument, we analyzed a

silicon grating calibration grid with a step height of 19.66 +/- 1 nm, as reported by the

manufacturer. The instrument was used in contact mode. The area was scanned at 2.44 Hz and at

256 samples per line. The image was flattened with a first order fit, as suggested by the

instrument manual. Figure 2-5 shows the bearing analysis, in which the software determines the









distribution of data points in the z-axis. The two spikes in the histogram correspond to two

elevations on the surface: the bottom and top surface. The red box (Hist rel depth) indicates the

distance between the two cursors on the histogram (19.93 nm), which represent the height

determined by the instrument. The result showed that the height was measured with an accuracy

of 1.4 %, which confirmed that the measurements performed were accurate.

The samples were deposited on mica by adding a couple of drops of the suspension and the

solvent left to evaporate at room temperature. No other treatment was performed before analysis.

The images were obtained with the scanner E, which offers a maximum imaging area of 15 x 15

im. The exact conditions for the different parameters used varied from sample to sample,

although average values or ranges can be reported. To image the samples, first, a fast scan was

performed to identify areas were the sample was deposited, after which the tip was taken to one

of this spots to zoom in and acquire an image of proper quality. Typically, we used 512 samples

per line, the best quality possible with our instrument. Next, the amplitude setpoint was

decreased slightly to optimize the force use to scan the surface. This was stopped when the trace

and retrace profiles in the section analysis looked similar. The scan rate was adjusted based on

the dimension of the image, for areas between 2 x 2 2.5 x 2.5 im, a typical scan rate between 1-

1.5 Hz was used, and this value had to be increased for smaller areas scanned, typically to 2-2.5

Hz. This was done to maintain good track of the tip on the z-axis, which allowed the

visualization of the changes in the height profile. Finally, the integral and proportional gains

were adjusted to improve the quality of the images, with typical values of 0.35-0.45 and 0.4-0.6,

respectively.

In general, the templates were easier to image, since they tended to be deposited as layers

of relatively homogeneous height, which made it easy for the tip to identify the topographic









features of the species deposited. The opposite was observed for the core-shell particles, which

were deposited as agglomerates of different heights and sizes, as a consequence, images of good

quality were more difficult to obtain, but usually it was a matter of finding a spot with a low

concentration of particles. In the first studies, a flattening of the images was performed through

the software, but since we became interested in the topographic features, this was not used in the

study presented here. This was done to avoid any kind of artifacts that could affect the

determination of the height of the particles, as a consequence of the leveling of the surface done

through the flattening option. The section analysis was performed then with the images without

any post-treatment. The particles used to measure the diameter and height were selected from

each image taken after zooming in the area of interest and confirmation that both dimensions

were possible to determine without ambiguities. Unless otherwise stated, the height images were

used for the characterization of the particles, although in many cases it was easy to visualize the

shape of the particles in the phase image, but, due to the lack of understanding on the meaning of

these images in the literature, they were not used in our studies.

2.2.4 4-Potential

2.2.4.1 Theory

We used this technique to characterize changes in the charge of the surface of the particles,

after different treatments given to the templates and core-shell particles. The electrostatic

potential at the surface of shear of a kinetic unit is the (-potential. The kinetic unit consists of the

particle, ions adsorbed onto the surface, counter ions contained within the surface of shear, plus

solvent molecules strongly attached to the surface ions and counter ions in the double layer. This

kinetic unit is also called the hydrodynamic radius, obtained experimentally by DLS as explained

in section 2.2.1. The magnitude of the electrostatic repulsive forces between particles is

determined from (-potential.62









The technique works based on the detection of the mobility of the particles under the

influence of an electric field. A laser beam passes through the sample, the light scattered detected

is Doppler shifted because of the motion of particles under the influence of the electric field. A

portion of the beam directed to the samples is split off and then recombined with the scattered

beam and modulated at 250Hz. With this, the Doppler shift of the signal, which is proportional to

the velocity of the particles, is easy to detect as a shift in the frequency of the recombined beam.

Two factors affect the motion of the particles, the sign of their charge, which determines the

direction they move, and the value of this charge, which determines the speed they move at.

Experimentally, it is the velocity of the charged particles that is measured, from which the

mobility and (-potential are calculated. The calculation of mobility is straight forward, contrary

to the Q-potential.62'63

The relationship between (-potential and mobility depends on the theoretical model

chosen. There are two classical models that results in two classical limits: the Smoluchowski and

the Huckel equations. They each applied in two opposite limits. Simply stated, these limits have

a common root: the magnitude of the dimensionless product Ka, where K is the inverse of the

double layer thickness and a the radius of the kinetic unit. In general, mobility is related to the Q-

potential as shown in Equation 2-10


e 2e (2-10)
(3h)f(Ka,)

wheref(ka, ) is a model dependent function. In the simplest model, for 1:1 electrolytes

f(Ia,;))= f(a)= 3/2-4.5/(Ka)+37.5/(Ka)2-330(Ku)3 (2-11)

The Hickel limit is satisfied when Ka << 1, typical for ions. In this limit (-potential and

mobility are related as shown in Equation 2-11









Pe 2e,; where: q is the viscosity of the suspending liquid (2-12)
3r

On the other hand, the Smoluchowski limit is satisfied when Ka >>1, obtaining Equation 2-

13


Pe e (2-13)


For most colloidal particles it is impossible to satisfy the Hickel limit. As a consequence,

in our studies, the Smoluchowski model was used. To assure the limits of this approximation

were satisfied to the best of our possibilities, the suspensions were analyzed suspended in a 10

mM KCl solution, to reduce the thickness of the double layer and increase the product Ka.62 63

The detection system is the same as the one explained in the dynamic light scattering

section. The instrument used in this study was the Brookhaven instrument described in the DLS

studies. The difference in this case was that the suspensions were placed in a cuvette to which an

electrode was attached, and the detector placed at 150 from the original laser beam.

2.2.4.2 Experimental settings

The suspensions of the particles were treated previous to the measurements similarly to the

case for DLS studies. The samples were sonicated and filtered. Before the measurements, the

particles were analyzed by DLS to confirm the size of the particles, assure the absence of

aggregates and adjust the concentration to modulate the intensity of the signal reaching the

detector. Before each measurement, the electrodes had to be preconditioned. This was done as

indicated by the manufacturer by running a NaCl solution in 3 repetitions of 10 cycles consisting

of 10 runs each. With this, the electrodes became black, producing a dark brown solid that was

discarded. The electrodes were rinsed with Millipore water before the measurements. This was









found to be a very critical step in the setting of the conditions for the measurement. If this

preconditioning was not done properly, the values obtained were difficult to reproduce.

We usually started our measurements at the pH the samples were stored, approximately, 7.

We then increased the pH by addition of the proper volume of NaOH 0.5 M. After reaching a pH

of around 10, the measurement at the first pH value tried was repeated. If the values were off, the

preconditioning was considered improperly done, the suspension discarded and the measurement

redone. Lower pH values were obtained by adding HC1 0.5 M. With the templates, the pH of the

suspension could be taken to a lower value compared to the core-shell particles. The ones

derivatized with amino groups, tended to precipitate when the pH was below 3. In all the cases,

below pH 3, addition of larger volumes of the acid was necessary to achieve changes in the pH

of the suspensions, since protonation of the groups on the surface takes placed preferentially

under these conditions. The measurement at each pH was repeated five times and the mean value

was used in the plots presented through the text.

2.2.5 Study of the Polarity and Viscosity of a Matrix by Steady State Fluorescence
Spectroscopy

2.2.5.1 Use of pyrene as polarity and rigidity probe

Pyrene has been used extensively as a probe for the study of polarity of different matrices.

It has been established that the intensities of its various vibronic bands show a strong dependence

on the solvent environment. In the presence of polar solvents there is an enhancement of the 0-0

band at the expense of others. The polarity index, defined as the ratio of the third and first peak

intensities (I3/11) has been qualitatively related to changes in polarity in the environment where

the dye is placed.64'65 This is a consequence of the perturbations of the intensities of the

vibrational fine structures in its emission spectrum, due to the extent of interaction between the

solvent dipoles and the excited singlet states of pyrene. When there is minimal coupling, the









polarity index is 2.0 (as in perfluoromethylcyclohexane). However, efficient dipolar coupling,

with solvents such as acetonitrile or dimethyl sulfoxide, causes a drop in the I3/11 value to as low

as 0.5.

Pyrene has been used as a probe for rigidity as well. The photoexcited molecule, during its

lifetime, may approach an unexcited dye molecule to form a collisional complex called excimer.

Since this process is dominated by translational diffusion, it has been related to the viscosity of

the media where the dye resides.66 The excimer emission appears as a broad peak centered

around 480 nm. Hence, the ratio of the intensities of the emission of the monomer and excimer

(I1/Iexc) is related to the ability of the dye molecules to form dimers, and can be used as an

indicator of changes in the viscosity of the media.65 Due to its very low solubility in water,

pyrene has been used as a probe in studies of micellar systems as well, observing that it is

preferentially solubilized in the most external hydrophobic regions of these aggregates, in

contact with water molecules to a certain degree.67 Hence this probe has been used for

determination of CMC values. Pyrene would be dissolved in water if the surfactant concentration

is low enough to prevent the formation of micelles. Once the surfactant molecules start

aggregating, the probe will be incorporated in the micelles due its hydrophobicity, which is

observed as a dramatic shift in the emission of the dye and on the intensity of its emission.

2.2.5.2 Use of C153 as a polarity probe

C153 has been used extensively in different systems as a probe for polarity due to its large

change in the dipole moment when excited.68-70 After excitation, the dipole moment of the

molecule increases from 6.6 in the ground state to a value between 14.2 and 16, depending on the

solvent.7 When the probe is excited, reorganization of the solvent molecules leads to a relaxed

state of lower energy. The relaxation is larger when the solvent polarity is larger. Only one single

excited state (Si state) of intramolecular charge transfer is responsible for the fluorescence of









C153.69 Thus, complications arising due to the participation of multiple fluorescence states are

avoided. As a result, the dye emits with a maximum between 450 nm to 550 nm. This excitation

and emission wavelengths are very similar to dyes commonly used in protein labeling which

makes it useful for fluorescence particles intended to label biologically relevant species.

2.2.6 Time-Domain Lifetime Measurements

Time-resolved fluorescence data usually contain more information than is available from

the steady state studies. For instance, it can give information on the presence of one fluorophore

in different environments, by reporting different decay times. Another important use of this

technique deals with the differentiation between static and dynamic quenching. Two methods of

measuring time-resolved fluorescence are common, the time-domain and the frequency-domain.

In our case we worked with the former, as a consequence, we will focus our discussion on this

technique. In time-domain, the sample is excited with a pulse of light, made as short as possible,

preferentially much shorter than the decay time of the sample to be examined. The time-

dependent intensity is measured following the excitation pulse, and the decay time T is calculated

from the slope of a plot of logI(t) versus t, or from the time at which the intensity decreases to

1/e of the value at t=0.65

In general, the inverse of the lifetime is the sum of the rates which depopulate the excited

state. The lifetime of a fluorophore is also interpreted as the average amount of time a probe

remains in the excited stated following excitation. This lifetime is a statistical average, and

fluorophores emit randomly throughout the decay. For a large number of fluorophores, some

emit quickly following the excitation, and some will emit at times longer than the lifetime. This

time distribution of emitted photons is the intensity decay.65









2.2.6.1 Fluorescence resonance energy transfer studied by time-resolved fluorescence

Fluorescence resonance energy transfer (FRET) is transfer of the excited-state energy from

the initially excited donor (D) to an acceptor (A). The rate of energy transfer depends upon the

extent of spectral overlap of the emission spectrum of the donor with the absorption spectrum of

the acceptor, the quantum yield of the donor, the relative orientation of the donor and acceptor

transition dipoles, and the distance between the donor and acceptor molecules. The extent of

energy transfer is also influenced by the presence of the donor-to-acceptor diffusion during the

donor lifetime, which is usually studied by time-resolved measurements.

Resonance energy transfer (RET) is a through space interaction which is mostly

independent of the intervening solvent. The term RET is preferred over FRET since this process

does not involve emission and reabsorption of photons. The theory of RET is based on the

concept of a fluorophore as an oscillating dipole, which can exchange energy with another dipole

with a similar resonance frequency. In principle, the orientation of the donor and acceptor can

prevent energy transfer between a closely space D-A pair, but such a result is rare. Hence one

can assume that RET will occur if the spectral properties are suitable and the D-A distance is

short enough.65

In the presence of an acceptor, the donor decay becomes significantly nonexponential,

especially at higher concentrations of the acceptor. The origin of the nonexponential decay is the

time-dependent distribution of acceptors around the excited donors. At short times, following

excitation, there exist more donors with nearby acceptors. These decay more rapidly because of

the distance dependence of energy transfer. At later times, the donors with nearby acceptors have

decayed, and emission results preferentially from donor without nearby acceptors. The decay of

these donors is longer, owing to a slower rate of RET to more distant acceptors. The distance at

which RET is 50% efficient, called the Forster distance, is typically in the range 20-60 k. The









rate of energy transfer from donor to acceptor (kT) is given by Equation 2-14, where D is the

decay time of the donor in the absence of acceptor, r is the donor-to-acceptor distance, and Ro is

the Forster distance


kT I ) (2-14)
TD" r

The Forster distance is defined as in Equation 2-15, where Kc2 is the orientation factor, r is

the refractive index of the medium and QD is the quantum yield of the donor in the absence of the

acceptor

R = 0.211[K 2r-4 QDJ() /6 (2-15)

The normalized fluorescence intensity (FD) of the donor in the absence of the acceptor and

the extinction coefficient of the acceptor (eA) are related to J(A) as shown in Equation 2-15


FD (Z)EA (A)A4dA
J(A)= 0 (2-16)
J FD (A)dA
0

The transfer efficiency is typically measured using the relative fluorescence intensity of the

donor, in the absence (FD) and presence (FDA) of acceptor. The transfer efficiency can also be

calculated from their lifetimes (TD and TDA), as shown in Equations 2-17 and 2-18

F
E= 1 DA (2-17)
FD


E =1 DA (2-18)
TD

In Equations 2-17 and 2-18, it was assumed that the lifetime of the donor was not altered

by binding of the acceptor, other than by the rate of energy transfer. Allosteric interactions

between the donor and acceptor sites could alter the donor lifetime by enhancing other decay









processes. Under these circumstances, more complex analysis of the apparent transfer efficiency

is required, typically by a comparison of the apparent efficiency by donor quenching and

enhanced acceptor emission.65 72 73

Quenching of the fluorescence lifetime of the donor can be used to study RET, considering

a few assumptions, including the following:

* Formation of static ground state complexes does not decrease the decay times of the
uncomplexed fluorophores, because only the unquenched probes are observed in a
fluorescence lifetime experiment.
* Dynamic quenching is a rate process acting on the entire excited-state population and thus
decreases the mean decay time of the excited-state population.
* A fluorophore usually displays the same radiative decay in different environments.


There are two cases for quenching of the lifetime of a fluorophore, static and dynamic

quenching. Static quenching is based on the formation of a nonfluorescent complex between the

donor and the quencher. When this complex absorbs light, it immediately returns to the ground

state without emission of a photon. For static quenching, there are no changes in the fluorescence

lifetime of the donor (D/ TDA =1), since the uncomplexed fraction is unperturbed. In a collisional

quenching process, a decrease in the lifetime of the donor occurs because quenching is an

additional rate process that depopulates the excited state. At the same time, there is a decrease in

the yield of the fluorescence because quenching depopulates the excited state without emission.

As a consequence, the decrease in the intensity is proportional to the change in the fluorescence

lifetime (rD/ TDA= FFDA). Collisional quenching of fluorescence is described by the Stern-

Volmer (S-V) Equation

S= 1 + kqr[Q]= 1 +KD[Q] (2-19)
F

In Equation 2-19, Fo and F are the fluorescence intensities in the absence and presence of

the quencher, respectively, kq is the bimolecular quenching constant, to is the lifetime of the









fluorophore in the absence of quencher, and [Q] is the concentration of the quencher. If the

quenching is known to be dynamic, KD is used. A linear S-V plot is indicative of a single class of

fluorophores, all equally accessible to the quencher. If two fluorophore populations are present,

and one is not accessible to the quencher, then the S-V plots deviate from linearity towards the x-

axis. Observation of a linear S-V plot does not prove a collisional quenching, since, under some

circumstances, static quenching also results in linear S-V. The best way to distinguish them is by

lifetime measurements.65

2.2.6.2 Time-correlated single photon counting (TCSPC)

In this technique, the sample is repeatedly excited using a pulse light source. The time

measurement clock starts when the sample is excited. The start signal is used to trigger the

voltage ramp of the time-to-amplitude converter (TAC) and it is stopped when the first

fluorescence photon of the sample is detected. The TAC provides an output pulse whose voltage

is proportional to the time between the start and stop signals. Subsequently, a multichannel

analyzer converts this voltage to a time channel using an analog to digital converter. The

experiment is continued until more than 10 000 counts are collected in the peak channel, to

obtain an optimal signal to noise ratio. The instrument used in this study is a home-made single-

photon-counting system. For data acquisition, the hardware (control electronics and the

multichannel buffer (MCB)) is from PRA and the software used was Maestro-32 (version 6.05)

from Ortec.

For data analysis (curve-fitting), the software used was DAS6 from IBH, which is based on

the Nonlinear Least-Squares analysis (NLLS). Its goal is to obtain estimates of parameter values

which have the highest probability of being correct. These values would provide the best match

between the data N(to and the calculated decay Nc(t). This is accomplished by minimizing the









goodness-of-fit-parameter X2, where n is number of channels, and o, is the standard deviation of

each data point

22 [N(t N([N(t) )- NC(t,)]2
X2= 1 [N(t)- (t) (2-20)
=c12 C C=1 N(tQ )

A multiexponential model was used to fit the data, assuming that the decay is the sum of

individual single-exponential decays, as shown in Equation 2-21, where a, is the amplitude of the

components at t= 0, Ti is the decay times and n is the number of decay times


I(t)= a, exp(-t/r,) (2-21)
1-1

This is the most common used model, but the meaning of the parameters (a, and r,)

depends on the system studied. The most obvious case is to a mixture of fluorophores, each

displaying one of the decay times, r,. If the probe can exist in two environments, such as exposed

and shielded from the solvent, then a decay time can be assigned to each of these states. In this

case, which is the one we worked with, it is generally safe to assume that the fluorophore has the

same radiative decay rate in each environment and the a, values represent the fraction of the

molecule in each conformation at t=-, which corresponds to the ground state equilibrium.65

2.2.6.3 Determination of the degree of solvation of dansyl chloride by fluorescence lifetime
measurements

Dansyl chloride has been extensively used as a fluorescence probe after its introduction by

Weber.74 The family of dansyl probes shows characteristic changes in its fluorescence maxima

and intensity, which have permitted their use to determine "polar" and "nonpolar" sites in

macromolecules.75 The fluorescence band of dansyl chloride is associated with a transition

involving an excited state having a considerable charge-transfer character, due to the promotion

of a lone-pair electron of the amino group into an antibonding orbital of the naphthalene ring,









called twisted intramolecular charge transfer (TICT). TICT phenomenon occurred in some

especial fluorophores, which may have two parts in their structure, a donor and an acceptor part.

When excited, the donor part may rotate around a single bond and transfer an electron to the

acceptor part. Therefore there exists two excited states, one is the TICT and the other is the non

charge transfer state.76 Because of its nature, the energy of the TICT state is very sensitive to the

polarity of the solvent.7 It has been identified that changes in polarity of the media where the

dyes resides can be correlated to variations in the emission maximum of the dye, which has been

used extensively in the literature.6 Similarly, dansyl chloride can be used to monitor the local

viscosity of different media, as demonstrated by Leezenberg and Frank.78 If a dansyl probe is

locked in a rigid environment, solvent dipoles may not relax around the excited molecules and

lower its emission. Moreover, a rigid environment hinders the rearrangement of the naphthalene

group, increasing the energy of the excited state, observed as a blue shift in its emission.79

2.2.6.4 Experimental settings

The samples were characterized by steady-state fluorescence spectroscopy prior to the

lifetime measurements, to obtain the maximum in the excitation and emission spectra of the

encapsulated probe. In some cases, the intensity of the doped particles was slightly low, which

required performing the experiments for longer than 30 minutes to obtain a good signal to noise

response. The data obtained was analyzed with different exponentials, taking the x2 value as an

indication of the good fit between the experimental and obtained data, being the values closer to

one considered the best.

2.3 Experimental Procedures

2.3.1 Preparation of Templates

Poly(ethylene[20]-sorbitan monooleate) (t80) and sodium octaniate (SO) were mixed with

13.4 mL of saline (9 g NaCl in 1 L deionized water) at room temperature (RT) in an orbital









shaker (J-Kem Scientific, model BTS 1500) at 300 rpm. After 4 h, octadecyltrimethoxysilane

(OTMS) was added. The mixtures were placed in an orbital shaker at 300 rpm, and heated to 75

0.5 C overnight to promote the inclusion of OTMS in the system. After cooling down to RT,

the pH of the solution was increased to approximately 10 using a 0.5 M NaOH aqueous solution.

After 30 min, the pH was decreased to 7.4 using a 0.5 M HC1 aqueous solution. Prior to any

characterization or fluorescence studies, the templates were purified through 25 nm pore size

membranes (Millipore). Finally, the concentrated suspensions were diluted to their original

volume, sonicated for 15 min and filtered with 450 nm pore size filters (Fisherbrand). For doped-

templates, the surfactants were mixed with the dopant and heated until a homogenous suspension

was obtained. Subsequently, OTMS was added and the synthesis continued as explained for the

dopant-free templates.

2.3.2 Synthesis of Core-Shell Nanoparticles

The core-shell particle synthesis started with the templates formation as explained in the

previous section. For dye-loaded particles, the dye, pyrene (system NPY) or coumarin 153

(system NCU), was mixed with the surfactants at room temperature in the first step of the

synthesis. For the NEB system, ethyl butyrate (EB) and 1-dodecene were mixed with the

surfactants at 75 0.5 C for 4 h. This was followed by OTMS addition as explained for the

synthesis of the templates. Once the templates were obtained, a 5-fold dilution of the original

templates suspension was necessary before adding tetramethoxysilane (TMOS), in order to avoid

crosslinking. TMOS was added in small portions (approximately 20 mg every 24 h) and the

solution was allowed to stir until the size reached the desired range, as observed by DLS. Unlike

the templates, the core-shell particles could be centrifuged. Based on this, the nanocapsules were

centrifugated in a Sorvall RC 5B centrifuge (GMI, Inc) at 2990xg for 80 minutes to remove

excess reagents. The nanocapsules were recovered as a pellet and resuspended by sonication for









approximately 1 h. This process was repeated twice. After resuspension, the samples were

filtered with 450 nm pore size filters.

2.3.3 Coating of the Particles with 3-aminopropylsilane

The particles were not purified before reacting with 3-aminopropylsilane (APS). The

suspensions were diluted 5 times with Millipore water and the pH was adjusted to approximately

9.5 with a NaOH 0.5 M solution. The suspension was placed in an oil bath and the temperature

was increased to 700 C. APS was added in small portion of approximately 15 mg every 4 h until

the desired amount to add was reached. After this, the suspensions were purified as explained

earlier by centrifugation.

2.3.4 Synthesis of the Complex between 3-aminopropylsilane and Dansyl Chloride

Dansyl chloride (DC1, 1.00 g, 3.71x10-3 mol) was dissolved in 80 mL of CH2C2.

Separately; APS (0.8 mL, 4.6x10-3 mol) was dissolved in 20 ml CH2C2. The two solutions were

combined in a 250 mL round bottom flask under Ar. Finally, triethylamine (0.6 mL, 2.5x10-3

mol) was added to neutralize the acid formed. The mixture was stirred at RT for 3 h, after which

the solvent and excess triethylamine were evaporated under vacuum. The solid left was dissolved

in THF and the insoluble hydrochloride salt was removed by filtration. The product (APSDC1

complex) was obtained as a yellow viscous liquid. The excess DC1 that did not react was not

separated, since due to the silanol groups attached to the probe, column chromatography was not

possible to use. After reaction with the particles, we expected the non-derivatized dye to be

removed from the suspensions during the dialysis, as observed by UV.

2.3.5 Coating of the Templates and Core-Shell Particles with APSDC1 Complex.

A suspension of the particles or templates before purification was mixed with excess of the

APSDC1 complex at pH 9.5. The suspensions were left stirring in the dark for three days, after

which the excess solid dye was removed by filtration. The nonreacted probe and nonderivatized









dye were removed by dialysis with membranes with a 6-8000 Da MWCO (Spectrum Labs).

After this, the core-shell particles were centrifugated and the templates were purified with the 25

nm pore size membranes. Subsequently, the samples were resuspended and diluted to their

original volume.

2.3.6 Loading the Dyes to Templates and Core-Shell Particles

A volume of the nanocapsule or templates suspensions (purified and resuspended to its

original volume) was mixed with the proper volume of a stock solution of the dye dissolved in

methanol, to obtain the desired final concentration of the dye. The samples were vortexed and

then stirred for 1 h. In the case of energy transfer studies, the donor was added to the suspension

and the system vortexed and stirred for 30 min. Subsequently, an aliquot of the acceptor was

added to reach the desired concentration, stirred for 15 minutes and the fluorescence recorded.

This was repeated until the maximum concentration was reached.

2.3.7 Fluorescence Images

A drop of the nanocapsule suspension was placed on a clean glass slide (Corning Inc.) and

dried at room temperature. Imaging was performed on a fluorescence microscope (Olympus,

model IX70) with excitation maximum at 360 nm (+/- 20 nm) and emission maximum at 525 nm

(+/- 25 nm).

2.3.8 Steady State Fluorescence Spectroscopy

Steady state fluorescence spectroscopy measurements were carried out in a

FluorologMax-3 (Horiba Jobin Yvon) with the following set up: 2 mm slits and 1 s integration

time. The dye-doped particles were dialyzed until no more dye could be removed. After this,

they were stored at room temperature. Before the measurements, they were diluted five times to

reduce scattering of light due to the size of the particles. It was not possible to avoid the presence

of the scattering peaks, as observed in the emission spectra of C153, but these appeared far from








the maximum. As a consequence, no complications in the determination of the maximum in the

emission of the dye were encountered. The suspensions were stirred for one hour before the

measurement to homogenize them. The dye-loaded particles were diluted and homogenized

similarly to the doped particles. After addition of the dye, the samples were vortexed and stirred

for one hour before measuring their emission.


incident light


no scattering


homogeneous medium


B


scattering



i


heterogeneous medium


Figure 2-1.Possibilities for the interaction of a laser beam with a liquid sample. In a
homogeneous medium, waves of the scattered light interact destructively, producing
no scattering (A); while in a heterogeneous medium scattering is produced (B).


I



















05 10 15 20 25 30
log t



Figure 2-2.Comparison of the correlation function of two set of templates with different
diameters as determined by DLS. The diameters obtained were 47 nm (solid line) and
72 nm (empty circles).


Z incident beam



specimen


objective lens


transmitted beam


diffracted beam


Figure 2-3.A simple representation of the basic concept of a transmission electron microscope
operating in the bright field mode.


1\ \
















laser source


detector


cantilever

sample t interaction




substrate




Figure 2-4.A simple representation of the basic setup of the atomic force microscope.


Bearing


Box area 12.689 pm'
Center line av 3.015 nm
Bearing area 6.864 pm'
Bearing area % 54.092
Bearing depth 12.059 nm
Bearing volume 6.9456e-2 pm'
Hist area 691674 nm2
Hist % 5.451
Hist rel depth 19.926 nm

Bearing area % Depth
BA% 1 0.011 0.000094
BA% 2 0.011 0.000094
BA% 3 0.011 0.000094
BA% 4 0.011 0.000094
BA% 5 0.011 0.000094
BA% 6 0.011 0.000094

calgridtgz01.007


0 2.50
Hist. %


5.00 0 45
Bearing area %


Figure 2-5.Data analysis of a Z-axis calibration grid performed in contact mode.


BII~WW ~8~ a~a ahorll









CHAPTER 3
SYNTHESIS AND CHARACTERIZATION OF SURFACTANT-BASED TEMPLATES FOR
THE FORMATION OF CORE-SHELL PARTICLES

3.1 Preparation of Surfactant-Based Templates

3.1.1 Strategy

The synthesis of core-shell particles proposed in this work is based on the formation of

templates with reactive groups on their surface that can be used to grow a shell, as shown in

Figure 3-1. The first step of the synthesis involves the formation of spherical micelles, to which a

reactive amphiphile is added. Since micelles can be broken down by dilution, the reactive groups

in the micelles will be chemically connected in subsequent steps to fortify them. The unreacted

groups left after this process can act as anchoring points to grow a shell that is chemically

connected to the core, fixing the shape and size of the species obtained. This would produce

more robust particles which would not be affected by changes in concentration, ionic strength,

pH or the like. We decided to use OTMS as the reactive amphiphile due to its well-known

chemistry (see Figure 3-2 for its chemical structure). The chemical properties of OTMS have

been extensively studied, especially at the air-water interface, with emphasis in the influence of

the pH in the hydrolysis and condensation of its methoxy head group.80-82 It has been established

at the air-water interface that low pH (3.9-4.8) and high pH (10.3-11.7) promoted the hydrolysis

and condensation of its methoxy groups, although totally hydrolyzed or totally condensed

species are not obtained unless the pH is lower than 3.9 or higher than 11.7.

3.1.2 Results

Based on the information about the reactivity of OTMS discussed above, we decided to use

this molecule to fix the structure of the templates obtained through its hydrolysis-condensation

chemistry. In previous work in our group,83 t80 was used to produce spherical micelles with

diameters of approximately 7 nm, as observed by DLS.84 Subsequently, OTMS was added to the









suspension to produce the templates. The first problem encountered was the low miscibility of

t80 and OTMS. To overcome this, the system was heated to 75 OC after the addition of the

chemically active amphiphile. At this temperature, the ethylene glycol chains in t80 become

more hydrophobic which promoted the incorporation of OTMS into the micelles already

formed.85 This approach formed the desired templates but the yield was low and excess

surfactant was left in the suspension, which made the purification process tedious. Preliminary

studies showed that using increasing amounts of OTMS in the suspension did not produce more

templates, but instead resulted in phase separation. To overcome this problem, a combination of

t80 with sodium octanoate was selected. This small surfactant is well known for its extreme self

assembly behavior, forming small aggregates and having a high critical micelle concentration

(CMC).86 Based on this, it has been used in different biomedical applications, especially to

promote mixing of the different components of microemulsions.87 In the study presented here,

the concentration of t80 was kept above its CMC, to assure that spherical micelles were present

in the suspension, while the concentration of SO was maintained below its CMC to promote its

addition to the already formed t80 micelles. After addition of OTMS and heating, the suspension

changed from transparent to white-blue, which is a clear indication of the increase in the size of

the species obtained. Subsequently, the methoxy groups in OTMS were hydrolyzed and

condensed by increasing the pH of the suspension to approximately 10.5 for 30 minutes, as

shown in Figure 3-3. We chose to work under basic conditions due to the presence of sodium

octanoate in the formulation, which would precipitate at low pH. As explain earlier, at this pH

the condensation process would not produce a totally condensed network, but we decided not to

use a higher pH to avoid harsh conditions that could affect the other components of the system.

After this reaction, a lose network had been formed, producing the templates. The particles









obtained are strong enough that they could be diluted without disruption of the system, as

observed by DLS. In any case, the condensation process would continue at a lower pace,

producing a tighter network at pH 7.4, conditions at which the templates are stored.

3.1.2.1 Size analysis

Different SO/t80/OTMS mixtures were prepared to study the influence of each component

on the size and shape of the templates obtained. Table I shows the formulations used and the size

characterization results. Samples A, B and C were prepared with different SO/t80 mol ratios, and

each one was mixed with three different amounts of OTMS. DLS was used for size analysis

before the addition of OTMS to the three original samples, showing low intensity of the scattered

light and diameter values between 6-8 nm in all cases, corresponding to t80 micelles.84 This

indicates that the inclusion of SO in the system did not perturb the size of the micelles, probably

due to the small size of the anionic amphiphile, which allows its packing in between the

poly(ethylene glycol) (PEG) chains of the t80 polar head. After addition of the reactive

surfactant, the transparent suspensions became white to various degrees, depending on the

amount of OTMS added. The routine size analysis of the templates and core-shell particles was

performed by DLS. This is a more convenient way to characterize the particles compared to the

microscopy techniques, since DLS characterization is performed in solution, with conditions

similar to what we expect the particles to be used in different applications. Before analysis, the

suspensions were purified by passing them through a cellulose membrane with a size cut-off of

25 nm. Subsequently, the suspensions were diluted to their original volume and sonicated to

separate any aggregates formed. This process removed the excess surfactants used for the

synthesis of the templates and the ions present in the original mixture, as observed by DLS in

Figure 3-4 and (-potential, in Figure 3-5. It was observed that if the particles were not

completely dried during the purification, they were easily resuspended after 15 minutes of









sonication. If all the solvent was removed, the templates formed a paste which required extensive

sonication to be resuspended, and a significant mass of particles was lost when the suspensions

were filtered.

The DLS graphs show that after purification of the templates the signal for the micelles

was lost. This indicated that the surfactant molecules were removed of the system, which was

immediately noticeable when the suspensions were shaken, as no bubbles were formed. This was

also confirmed by analyzing the electrophoretic mobility of the samples, as shown in Figures 3-

5. In a typical (-potential measurement, the particles are suspended in a solution of an electrolyte

and the velocity at which they moved when an electric field is applied is measured. From this,

the electrophoretic mobility of the particles is directly calculated and this value is used to

calculate the (-potential of the particles. The sign of its value corresponds to the charge of the

particles and its absolute value is an indication of the stability of the particles in suspension.

Large (-potential values indicate a low tendency for aggregation. In this study, it was observed

that the (-potential of the templates varied differently when the pH of the suspensions was

changed before and after purification. The pure templates showed more negative (less positive)

values than the nonpure ones. The (-potential of the pure templates was similar to the ones

obtained by Katagiri et al. for colloids coated with a lipid bearing a triethoxysilil moiety as the

polar head,88 and more negative that the ones obtained by Bourgeat-Lami et al. for SiOH

functionalized polymer latexes.89 The difference after purifying the templates is believed to be a

consequence of the presence of the non ionic surfactant (t80) adsorbed on the surface of the

templates before purification, which hides the charges of the silanol groups, due to its

poly(ethylene glycol) polar head.90 This resulted in the diffusion of the templates being less









affected by changes in the external field applied to the suspensions, producing (-potential values

closer to zero.

Size analysis by DLS showed that the samples labeled as 1 produced a small amount of

templates, easily recognizable due to the pale white color of the suspensions and confirmed by

the low intensity of the scattered light detected. Moreover, the size distribution was largely

dominated by the micelles. Considering that the intensity of the light scattered by a particle is

related to its radius to the sixth power,5 and the large difference in size between the micelles and

templates, the high intensity of the signal observed for the micelles indicated that they

outnumber the templates by a large factor. Samples labeled as 2 and 3, showed a white-blue

color as well as high intensity of scattered light, indicating that the number of aggregates, as well

as their sizes, were higher than the ones obtained in the sample set labeled as 1. In this case, the

size distribution is largely dominated by the templates, with just a small signal for the micelles,

as shown in Figure 3-4 A. These results are an indication of the incorporation of OTMS into the

aggregates, which suggests that SO is promoting this process, increasing the yield of templates

obtained. Samples A2 and A3 still included a high percentage oft80 in the formulation, but since

one of our goals was to reduce the mass fraction of this surfactant, they were not further

analyzed.

As explained in the previous chapter, in the DLS technique, the diffusion coefficient of the

scatterers is obtained from the variation of the intensity of the light scattered by the particles due

to Brownian motion. From this information a correlation function is obtained, and the diffusion

coefficient of the particles determined. This value is used in the Stokes-Einstein Equation to

calculate the diameter of the particles, assuming that the particles are spherical. The values

obtained by this technique are considered hydrodynamicc", meaning that the size estimated









includes any species adsorbed on the surface of the particles, like ions, surfactant molecules and

the like, which influence their motion in solution.57 As can be observed in the DLS results

reported in Table I, the incorporation of OTMS in the system produced an increase in the size of

the aggregates obtained of approximately ten times, compared to the SO/t80 micelles. This is

clearly an indication of the incorporation of the reactive amphiphile in the micelles, although

such dramatic effect on their diameter is not completely understood. It is known that the

aggregation number in t80 micelles increases with the concentration of the surfactant by

association of four primary micelles.91 The incorporation of OTMS in the system may be

promoting this association, producing larger aggregates. The results for the samples B and C

indicated that, for a fixed SO/t80 combination, larger aggregates were obtained by the addition of

higher amounts of OTMS. In both cases, by increasing the OTMS mol fraction four times an

increment of approximately 40% in the diameter of the templates was observed. This suggested a

mean to control the size of the particles.

The stability of the templates was compared to that of the SO/t80 micelles by analyzing

their correlation function in the DLS instrument. The suspension of the micelles without OTMS

was used "as is" for the first measurement, detecting an intensity of scattered light of

approximately 3.0 x 105 kcps. The suspensions of the templates was diluted to obtain a similar

intensity of scattered light and the correlation function of each of these was obtained, as shown

in Figure 3-6. Both samples were diluted until the signal was weak but still strong enough to

obtain a good fit for the correlation function, which was observed to be 0.5 x 105 kcps. The data

for the templates (Figure 3-6 A) shows that, despite changes in the background signal for the

correlation function, the results for the templates did not change, an indication that the particles

did not break after dilution. Different results were observed for the mixed micelles, as shown in









Figure 3-6 B. In this case, the typical signal for the micelles was obtained at 3.0 x 105 kcps, a

distortion of the normalized correlation function was observed upon dilution. The correlation

function showed that the micelles broke down and bigger species were formed. These results

showed the effect of the fortification of the particles after the condensation of OTMS which was

one of the goals of this work.

To validate the results obtained by DLS we imaged the particles in the electron

microscope, the results of which will be discussed in more detail in the next section. In all cases,

spherical particles with low polydispersity were observed, independently of the mol ratio of

surfactants used, as shown for samples B2, C2 and D2 in Figure 3-7. This confirmed that the

particles were spherical; and that their diameter can be obtained with the Stokes-Einstein

Equation with the DLS software. As can be observed in Table I, a 1:1 mass ratio of SO/t80

produced smaller templates after the addition of OTMS; as a consequence, this combination was

selected to study the formation of core-shell nanoparticles. The sample set D was prepared based

on this combination but with a lower total mass of surfactant in order to avoid complications

when purifying the nanocapsules.

3.1.2.2 TEM analysis

To investigate the geometry of the templates we used the electron microscope. Samples for

imaging were prepared by placing a drop of a dilute suspension of the templates on a formvar

coated carbon grid and the solvent evaporated. Figure 3-7 presents typical images of the

templates. The size of the particles was determined from the images using Adobe Photoshop

software and Image-J software, by comparison of the diameter of the particles and the scale bar

(Table 3-1). The values obtained were similar to the ones observed in the DLS analysis, although

a slight overestimation of the diameter was observed by TEM, especially for samples D. This is

interesting since DLS data is biased to the larger particles, while TEM gives a number average.









Based on this, larger values would be expected in the DLS results, the opposite trend observed

may indicate that the particles expanded after being deposited on the substrate. One important

detail for these images is the low electron density of the templates, which gave a low contrast

when imaging. This means that the electron beam did not encounter enough electron density

when passing through the particles, which indicates that the templates are porous. In the images

showed in Figure 3-7 and 3-8, uranyl acetate was used to stain the templates prior to imaging.

This staining agent is used to interact with hydrophilic regions in biological systems, increasing

the electron density of these areas.92 It was observed that the staining improved the ease of

imaging by making the periphery of the particles more defined; hence, size analysis was easier to

perform. Figure 3-8 shows evidence that the particles flattened after the solvent is evaporated.

The red boxes showed what are believed to be particles that were deposited with their longer axis

perpendicular to the surface of the grid, showing a side view of the templates. It is clear in these

images that the particles show a disk-like geometry when dried. Moreover, this does not seem to

depend on the use of a dopant in the synthesis of the templates, since the sample on Figure 3-8 B

was prepared with ethyl butyrate as a dopant, as explained in the next sections. Interestingly, the

longer axis of these particles is similar on average to the spherical ones, confirming that they are

the same kind of species. It is important to mention that this side view of the templates was not

commonly found when imaging, and they seemed to occur when the concentration of the

templates in the suspension used to load the TEM grid was high. Another significant detail

comes from the fact that these images are different from the ones typically obtained for hollow

particles, in which a thin skin, that usually collapses, is visualized.93 This may indicate that the

OTMS condensation is not restricted to the surface of the templates but it occurred in their

interior as well.









We performed a simple calculation to have an idea of the extent the particles expanded on

a substrate. We calculated the area of the templates D2, considering them as spheres, using the

formula 47nr2, where r is the hydrodynamic radius of the templates (in this case, 32 nm). We then

obtained the radius of the maximum area possible to obtain after flattening, considering the

formation of two circular sheets, with total area of 27r2. The radius of the totally flattened

particles was determined to be 45 nm. Comparison to the value obtained by TEM (41 nm)

showed that, after drying on the substrate; the diameter of the templates reached 91% of the

maximum possible, indicating a significant expansion.

3.1.2.3 AFM analysis

Atomic force microscopy was found to provide more information on the shape of the

particles due to its ability to construct a 3-D image of the sample. Analysis of the images

obtained by TEM suggested that the particles flattened after deposition on the substrate; hence,

we decided to study the morphology of the templates into more detail. One of the advantages of

AFM over TEM is that no staining is necessary to image the particles, so the probability of the

images containing artifacts is reduced. Moreover, TEM imaging is performed under vacuum,

while AFM imaging was performed under standard conditions. One significant difference in this

analysis was the use of mica, a hydrophilic material, as a substrate for AFM imaging, contrary to

TEM analysis, in which hydrophobic grids were used. As a consequence, the interaction between

the particles and substrate is expected to differ in each technique.61 Imaging was performed in

tapping mode to avoid direct interactions between the tip and the particles that can cause damage

to the sample (or the tip). In this mode, the tip of the cantilever is set to interact with the sample

through a constant force, without touching the surface but interacting with it probably through

the layer of water adsorbed on the surface.94









Typical images showed that the particles formed aggregates of a few micrometers in size

after drying, similarly to what Jungmann et al. observed in the characterization of nanospheres

made of poly(organosiloxanes).52 We believe these are formed during the solvent evaporation,

since these same suspensions were analyzed by DLS and no aggregates were detected. The only

difficulty encountered during imaging was related to the sample concentration, if this was too

high the particles were deposited as multilayers. This produced large differences in the height of

different parts of the sample, which was difficult for the tip of the cantilever to track, resulting in

images with poor resolution. This was avoided by using very dilute suspensions. Similar

problems arose when templates and core-shell particles with large diameters were imaged, since

the differences in the features on the z-axis became difficult for the instrument to track. Both

problems were more frequent with the core-shell particles than with the templates.

Templates D2 and D3 were selected to study their topographic features by AFM, as shown

in Figure 3-9, to complement the results obtained with TEM and DLS. The images were

analyzed with the Nanoscope 5.3r software, observing spherical particles when a top view was

used. The section analysis option in the software allows an accurate estimation of the diameter

and height of the particles. We used this feature to study the flattening of the particles after

deposition on the substrate suggested by the TEM images. As expected, it was observed that the

height of the templates was considerably lower than their diameter and the latter was larger than

the ones obtained in the DLS studies. Confirming what the TEM images suggested, the templates

did not look like hollow particles made of a thin skin, which usually appear as a thin film of the

material that collapses.95 Interestingly, when the diameter observed for D2 and D3 were

compared, a larger diameter for D3 was matched by a larger height value, as shown in Table 3-2.

We believe this is a consequence of the higher amount of OTMS included in the formulation









used to prepare the templates labeled as D3. Higher loadings of OTMS offered more head groups

to condense which would produce a more rigid system, preventing the templates from spreading

as much on the substrate. We used this flattening of the templates as an indication of their

flexibility, expressed as the diameter to height ratio (DHR). This value will be compared later to

the corresponding core-shell particles to study the effect of the coating of the particles on their

flexibility. The DHR values are 10.4 for D2 and 8.9 for D3, showing an increase in the rigidity

(lower DHR) as a consequence of the increase in the OTMS loading in the templates.

Comparison of these values to the ones obtained by Wang et al., who obtained DHR values

above 20 for hollow particles, indicated that, despite being flexible, these templates are not

hollow.50

3.2 Preparation of Doped Templates

The templates studied above can have different potential applications, including their use

as nanoreactors, drug delivery systems, agents for sequestration and encapsulation, and the like.

It is desirable to have control over the size of the particles to meet the requirements of the

application in mind. It was observed on the results showed in the previous sections that changing

the mol fraction of OTMS offered some degree of control over the diameter of the templates, but

this was limited by phase separation if the fraction of OTMS was high. As an alternative, the

templates could be prepared including a dopant. Molecules of different polarities and sizes could

be used to fill the interior of the droplets formed before the addition of OTMS, which would

allow control over the diameter of the precursors of the templates. After purification, these

molecules, depending on their polarity, could be removed by dialysis or other means, leaving

pores in the interior of the templates that could be used to trap other species, especially if they

are nonpolar. To explore these ideas we prepared templates with two different dopants, one









moderately polar and small, ethyl butyrate, and one nonpolar and larger, hexadecane (see Figure

3-2 for structures), as shown in Table 3-3.

3.2.1 Results

3.2.1.1 Size analysis of ethyl butyrate-doped templates

The formulations used to prepare the ethyl butyrate-doped templates (EB) are shown in

Table 3-3, as well as the size characterization results obtained by DLS. The EB templates were

prepared similarly to the templates labeled as D2, but the dopant was mixed with the surfactants

in the first step of the synthesis. It was observed that the suspension needed to be heated up to

obtain a homogeneous suspension and that the time needed depended on the dopant used, in this

case 2.5 h. After this, OTMS was added and the mixture heated overnight as explained above.

Samples G, H and I were not considered successful systems since a significant amount of

material was lost when the templates were passed through filters with 450 nm pore size. As a

consequence, they were analyzed by DLS only. The size analysis reported in Table 3-3 showed

that the size of the templates could be varied by changing the concentration of the dopant. An

increase in size of approximately 30% was observed when the concentration of the filler

molecule was doubled, and 50% when it was tripled. Differently to the templates studied above,

the concentration of OTMS did not seem to have a significant effect on the diameter of the

templates, but on the amount that were formed. This was clearly observed by the scattering of the

samples after the formation of the templates was finished. We then used the atomic force

microscope to study the topographic features of these templates.

3.2.1.2 AFM analysis of ethyl butyrate-doped templates

The ethyl butyrate-doped templates were imaged with the atomic force microscope, and

analyzed similarly to the set D, as shown in Table 3-4 and Figure 3-10. The size increase

detected by DLS was observed in a similar way in the AFM images. The templates showed, in









all cases, spherical particles and a higher polydispersity compared to D2 and D3. The data on

Table 3-4 showed that the increase in the diameter of the particles matched an increase in their

height after deposition on the substrate. Contrary to the templates studied before, a significant

number of "hollow" particles were identified in the images, as indicated with black arrows in

Figure 3-10. We believe these are templates that are squeezed in between other ones. This is an

indication of the influence of the dopant in the final properties of the templates, producing

particles that retained their geometry after purification, but are flexible enough so that they can

be folded without breaking. This opposes the tendency observed with the DHR values, which are

lower than the D set, indicating a more rigid system. Nonetheless, in each pair when more

OTMS is used in the synthesis of the templates, a more rigid system is obtained, as explained for

the D templates. This disagreement may be a consequence of the larger diameter of the templates

obtained when dopants are loaded in the templates.

3.2.1.3 Size analysis of hexadecane-doped templates

The hexadecane system showed a similar trend that the EB templates (Table 3-5). In this

case, the amount of the dopant used was lower, since an increment of its concentration did not

produced a homogeneous suspension, even over longer periods of heating. Hence, the

concentration of hexadecane was kept at approximately one order of magnitude lower than ethyl

butyrate, despite using the same concentration of surfactants. In this case, the samples were

heated for 3.5 h. DLS analysis showed that the size range obtained was similar to the ones

prepared with ethyl butyrate. Moreover, when the hexadecane concentration was doubled, a 30%

increment in the diameter was observed. This shows that the diameter of the templates can be

tuned by varying the concentration of the dopant.









3.2.1.4 AFM analysis of hexadecane-doped templates

Images obtain with AFM and the corresponding section analysis for the hexadecane-doped

templates are shown in Figure 3-11. These templates looked similar to the ethyl butyrate system,

with a significant number of"hollow" particles observed in all the cases. This indicates that this

is a trend for the templates obtained when a dopant is used in their synthesis. Comparison of the

results for the HxE and HxF templates suggested that, at a higher hexadecane concentration, the

amount of OTMS plays more control over the flexibility of the particles. Despite having the

same diameters in solution, as observed by DLS, their dimensions are considerably different

when imaged by AFM. Templates HxE showed a larger diameter and height, which seemed to be

due to the lower OTMS content.

3.3 Other Systems Studied

3.3.1 Use of a Cationic Surfactant in the Synthesis of the Templates

In an attempt to develop a robust approach to the synthesis of the templates we tried to

extend the choices of surfactants that could be used in this methodology. We believe that the

encapsulation of anionic species would be promoted by the use of a cationic surfactant. As a

consequence, we explored the use of dodecyltrimethylammonium bromide (DTAB) for the

formation of the precursors of the templates instead of sodium octanoate. Figure 3-12 shows

AFM images of two sets of templates obtained with different surfactant loadings, and diameters

of 53 nm (A) and 34 nm (B), as obtained by DLS. These templates looked very similar to the

ones obtained with SO without dopants. Work is needed to determine if these templates could be

used for the synthesis of core-shell particles.









3.4 Preliminary Studies of the Strength of the Templates


3.4.1 Strategy

The templates synthesized in this work were formed by the alkyl chain of the OTMS

molecules and their chemically attached head groups. We believed, based on the DLS and i-

potential data (Chapter 2), that most of the t80 and SO molecules are separated from the

suspensions during the purification steps. The particles flattened but did not collapse when

deposited on a substrate, which indicates that they are made of a thick skin of inorganic material,

probably with pores in their interior where the tails of OTMS could be accommodated. These

pores were probably formed by aggregation of the surfactant and dopant molecules during the

condensation of OTMS. To investigate the strength of the structures formed, they were heated up

slowly to temperatures at which the organic tails of the amphiphile would be removed. We

investigate what kind of species were obtained after this process by TEM and AFM.

3.4.2 Results

3.4.2.1 Calcination of dopant-free templates

Figure 3-13 shows TEM images of dopant-free templates with diameter of 60 nm, as

obtained by DLS. Figure A shows the typical image of templates as reported in the previous

sections. Figures B, C and D showed what was found in the crucible after the calcination was

done. The solid left was scratched although not all of it could be removed, which means than

some other species could be present in the final mixture. In any case, most of what was observed

looked like fibers (Figure 3-13 B and C), although a few dark, spherical species with different

sizes were observed as well (Figure 3-13 D). In general, it looked like the templates were

destroyed and what was left was a fiber-like material, probably made of what originally was the

siloxane network formed by the polar heads of OTMS, since the hydrocarbon part of the particles

was removed. Similarly to the templates, these fibers showed a very low electron density.









Interestingly, in the AFM images (Figure 3-14), more species similar to the original

templates were observed. Although they are not completely spherical, there seems to be a variety

of particles with different sizes, some of which have the same dimension of the templates. This

may indicate that the templates broke into smaller pieces during calcination. It is not clear why

these species were observed more frequently on AFM compared to TEM. It is possible that the

fibers observed in the electron microscopes were very thin (as indicated by their low electron

density) and thus, difficult for the tip of the cantilever to detect them, although Figure 3-14 D

look similar to the images obtained by TEM.

3.4.2.2 Calcination of ethyl butyrate-doped templates

Similar results to the ones obtained with dopant-free templates were observed when using

ethyl butyrate as a dopant. Figure 3-15 A shows the original spherical templates, which were not

found after the calcination, as shown in Figures B, C and D. Small particles with different sizes

as well as fibers were observed. Images obtained by AFM (Figure 3-16) showed particles of

different sizes and shapes, similar to the previous case, confirming that the particles tend to break

apart at high temperatures. In the Chapter 4 results for similar studies for the core-shell particles

will be examined.

3.5 Conclusions

This chapter focused on the synthesis and characterization of templates for the preparation

of core-shell particles. Our goal was to obtain spherical particles with chemical groups on the

surface that could be reacted in subsequent steps with a coating agent to build a shell on their

periphery. A ternary amphiphile system was used, including octadecyltrimethoxysilane as a

chemically active surfactant. The heads groups of OTMS were covalently connected to fix the

shape of the structures obtained. Dynamic light scattering showed that templates of sizes in

between 60 to 80 nm were obtained. Transmission electron microscopy images were used to









confirm that the particles were spherical. These images suggested that after deposition on a

substrate and solvent evaporation, the particles flattened, giving disk-like particles. This was

confirmed using the atomic force microscope, which let us obtained more details on the

geometry of the particles through the section analysis option in its software. It was observed that

the size of the particles could be varied by the amount of OTMS loaded in the formulation used

to synthesize the particles, although this was limited by phase separation. To have more control

on the size of the particles obtained, dopant molecules of different polarities were used in the

synthesis of the templates. Analysis by DLS showed that this produced an increase in the

diameter of the templates, as well as in their polydispersity. It was observed that the polarity of

the dopant used affected its maximum loading in the formulations, obtaining, in average,

particles with hydrodynamic diameters in between 150 to 300 nm. Imaging with the atomic force

microscope showed that the doped-templates were more flexible than the dopant-free ones. We

were able to observe "folded" particles which suggested that the dopant molecules were lost after

purification leaving a porous and flexible material. In this case, the amount of doped-templates

obtained seemed to be influenced by the amount of OTMS loaded in the system. We started

studies on the used of different surfactants in the synthesis of the core-shell particles, showing

that DTAB could be used instead of SO. This may facilitate the use of anionic dopants through

ionic interactions although this needs more work to be confirmed. In general, the templates can

be diluted without affecting their size or shape, as observed by DLS and AFM, suggesting they

are robust species. To test this, we calcinated templates prepared with and without dopants.

Preliminary studies based on AFM and TEM images showed that the particles break apart after

heating them up to 600 OC. We believe the results summarized here indicate that the templates

are composed of a thick skin of a siloxane network formed by the hydrolysis and condensation of









the OTMS polar groups. This skin retains its spherical shape in solution but deflates after

removal of the dopant, just like a soccer ball after removing air from its interior, producing

flexible structures. We believe these particles are strong enough to be used in solution in

different applications. In the next chapter, we explore the coating of the templates with a silicate

precursor and study how this affects their flexibility.


Table 3-1 Formulations used for the preparation of templates and characterization results
obtained by DLS and TEM.
Al A2 A3 B1 B2 B3 Cl C2 C3 D1 D2 D3
tween80(g) 1.0 1.0 1.0 0.7 0.7 0.7 0.4 0.4 0.4 0.5 0.5 0.5
SO (g) 0.4 0.4 0.4 0.7 0.7 0.7 1.0 1.0 1.0 0.5 0.5 0.5
tween80/SO (mol) 1/3 1/8 1/19 1/8
OTMS(mg) 11 20 42 10 22 43 7 22 45 11 31 45
tween80/OTMS (mol) 26 14 7 20 9 5 16 5 3 13 5 3
dDLS (nm) 81 65 47 54 64 63 75 88 50 64 72
dTEM (nm) 62 69 68 84 82 96


Table 3-2 Size analysis results obtained by DLS and AFM for templates D2 and D3.
dDLS dAFM hAFM DR
D2 64 80 8 10.4
D3 76 86 10 8.9
All values in nm, d: diameter, h: height, DHR: diameter to height ratio.


Table 3-3 Formulations used for the formation of ethyl butyrate-doped templates, and size
analysis results obtained by DLS.
EBA EBB EBC EBD EBE EBF EBG EBH EBI
SO 125 126 126 126 126 128 124 128 124
t80 125 123 128 123 128 129 126 124 126
EB 13 13 13 26 26 26 40 40 40
OTMS 13 26 44 13 26 44 13 26 44
dDLS 180 158 176 243 205 232 251 272 261
All quantities in mg, except diameter (d) in nm.







Table 3-4 Topographic information for ethyl butyrate-doped templates obtained by AFM. DLS
data is shown for comparison.
EBB EBC EBE EBF
dDLS 158 176 205 232
dAFM 184 232 245 260
hAFM 23 43 35 40
DHR 8.0 5.4 7.0 6.5

Table 3-5 Formulations used for the formation of hexadecane-filled templates and size analysis
results obtained by DLS and AFM.
SO t80 Hx OTMS dDLS dAFM hAFM DHR
HXB 126 123 6 26 141 140 15 9.3
HXC 124 128 6 44 160 142 17 8.4
HXE 126 128 12 26 204 178 44 4.0
HXF 128 129 12 44 203 146 18 8.1
All quantities in mg, except diameter (d) and height (h) values, in nm.


9 0Y


S0


= tween 80 ?= sodium octanoate


TotranthoxyIllane


. Octadecyltrimethoxysilane


Figure 3-1. Schematic representation of the synthesis of core-shell particles.












0 0


Na
sodium octanoat thyl utyratc
HCO
C_,PC


odlmecybllrimlhwxys3nii \-

H OHCH^- S






Tworn RO H,O -CHCH0 ,O S

x-vyw+7=20 N-brzmy]






hexadecane



DTAB
Br





Figure 3-2. Chemical structures for surfactants, amphiphiles and dopants used for this study.











MeO OMe
SI
3 OH


-3 MeO


OH
HO OH
COH
Si H
OH


- OH
/OH
$1I,..OO


OH
HO OH
Si'


OH
o0 H
O-OH
Si

+ -


Figure 3-3.Ideal schematic for the hydrolysis and condensation of the methoxy groups in OTMS
at basic conditions.


relative
intensity
0.30

0.25


0 25 50 75
diameter (nm)


0.20

0.15

0.10

0.05

0.00 E-1
0 25 50 75
diameter (nm)


Figure 3-4. Size distribution obtained by DLS for a typical set of templates A) before and B) after
purification.


0


relative
intensity
0.30-

0.25-

0.20-
















-10- \ "


-20-


E -30


-40 \


-50

2 3 4 5 6 7 8 9 10 11
pH







Figure 3-5. -potential variation as a function of pH for non pure (circles) and purified (squares)
templates.


A 1.01E+00

- 9.90E-01

S 9.70E-01

S9.50E-01

9.30E-01

9.10E-01

8.90E-01
1.8 2.3 2.8


3.3 3.8 4.3 4.8

IoE(r)


Figure 3-6.Variation of the normalized correlation function obtained by diluting a suspension of
particles to reach different light scattering intensities. A) templates, intensity: 3.5 x
105 kcps-purple line, 2.5 x05 kcps-green line, 1.1 x 105 kcps-red line, 0.5 x 105 kcps-
blue line, and B) SO/t80 micelles, intensity: 3.0 x 105 kcps-purple line, 2.0 x 105
kcps-green line, 1.0 x 105 kcps-red line, 0.5 x 105 kcps-blue line.


S1.00E+00
3
c 9.80E-01
-9
E 9.60E-01

- 9.40E-01

E 9.20E-01
E
9.00E-01 -
0.8


log r)




















Figure 3-7.TEM images for templates: a) B2, b) C2, c) D2. Scale bars 200 nm.


Figure 3-8. TEM images showing what are believed to be templates deposited with their longest
axis perpendicular to the surface of the substrate (indicated by the red boxes). Scale
bars 500 nm.




























;-

0 0.25 0.;0 0.75 0 c 2, 0.50 ;Q5 1.500 n



Figure 3-9.AFM tapping mode images for samples A) D2 and B) D3. On the bottom part of each
image the section analysis is shown.


























.LJc-ry?


S I I ,2I (i D ?5 1 1 H ,
0.25 0.$0 0. ? 1.C.W p, a D.2 c.' o. :.0 i.. l '


a,:25 6.i O .~7 r a


Figure 3-10. AFM images and section analysis for templates A) EBB, B) EBC, C) EBE and D)
EBF. The arrows indicate what are believed to be folded templates.


~~-J~S3*"`?



















i;


"r. 0. V^5V'4


P 1A
"'-r t!'rwr


Figure 3-11. AFM images and section analysis for templates A) HxB, B) HxC, C) HxE and D)
HxF. The arrows indicate what are believed to be folded templates.


















7 A
-,~ t ,III r


Figure 3-12. AFM image of two set of templates prepared with DTAB replacing SO.


Figure 3-13. TEM images of dopant-free templates before (A) and after (B, C, D) calcination.
Scale bars 200 nm


r.b~v' d r.;l ;.;; ..ir I.b ,I.


:I u.;o


B r-r












I:I


Figure 3-14. AFM images of dopant-free templates before (A) and after (B, C, D) calcination.


Figure 3-15. TEM images of ethyl butyrate-doped templates before (A) and after (B, C, D)
calcination. Scale bars 200 nm.























Figure 3-16. AFM images of dopant-free templates before (A) and after (B) calcination.
Figure 3-16. AFM images of dopant-free templates before (A) and after (B) calcination.


C, hI









CHAPTER 4
TEMPLATE-BASED SYNTHESIS OF CORE-SHELL PARTICLES

4.1 Synthesis of Core-Shell Particles

4.1.1 Strategy

The templates studied on Chapter 3 were formed with the idea of having reactive groups

on their surface that could be used as anchoring points for the growth of a shell. In order to build

a hydrophilic coating that would produce more robust particles, we turned our attention to

alkoxysilanes. This would allow us to work with the same kind of chemistry used for the

synthesis of the templates. Our case would be similar to the synthesis of a silicate, but instead of

letting the monomers hydrolyze and condense to form particles, we needed to promote the

reaction of the hydrolyzed monomers with the silanol groups already present in the templates. As

a consequence, we based the design of our synthesis on the knowledge of the preparation of

silicate gels. In general, silicate gels are synthesized by hydrolyzing monomeric, tetrafunctional

alkoxide precursors, employing a mineral acid or base as a catalyst. Under most conditions,

condensation commences before hydrolysis is complete, as shown in Figure 4-1. Because water

and alkoxysilanes are immiscible, alcohols are used to homogenize the mixtures. However, gels

can be prepared without a homogenizer, since the alcohol produced as the by-product of the

hydrolysis reaction is sufficient to homogenize the system. The most common tetraalkoxysilanes

used in the sol-gel process are tetraethoxysilane and tetramethoxysilane. The steric bulk and the

electron donor characteristic of the alkoxide or the organic substituent attached to silicon will

largely determine the kinetics of the hydrolysis and condensation reactions. Therefore, the use of

specific precursors is often dictated by kinetics considerations or compatibility with precursor or

other network-forming elements in multicomponent silicate gel synthesis. Numerous

investigations have shown that variations in the conditions of the synthesis cause modifications









in the structure and properties of the polysilicate products, although a consistent trend is

apparent. Acid-catalyzed hydrolysis with low H20:Si ratios produced weakly branched

polymeric sols, whereas base catalyzed hydrolysis with large H20:Si ratios produced highly

condensed "particulate" sols. Intermediate conditions produce structures intermediate to these

extremes.96,97

In this study, we selected tetramethoxysilane as the silicate precursor to coat the templates.

This was selected over tetraethoxysilane due to the dependence of the rate of the hydrolysis of

the alkoxysilanes on the sterics of the substituents attached to the silicon center. We believe a

faster hydrolysis would produce a rapid homogenization of the system, promoting the reaction

between the silanol groups on the surface of the templates and the silicate precursor.97 The

hydrolysis product, salicylic acid is known to grow primarily by addition of monomers to highly

condensed particles, rather than by particle aggregation, at pH above 7. Moreover, as explained

earlier, the condensation of the alkoxysilanes is expected to start before the hydrolysis is

completed, as shown in Figure 4-1. TMOS, being smaller, would encounter less steric hindrance

to approach the surface of the templates and react with its silanol groups. This is important, since

a competition with the reaction between the alkoxysilane molecules is expected, which would

produce a gel and, possibly, smaller particles. We carried out the reaction between TMOS and

the hydrolyzed and condensed polar groups in OTMS on the surface of the templates at pH 7.4,

since at this and higher pH the reaction between larger, more highly condensed species, which

contain acidic silanols, and smaller, less weakly branched species is favored.96

In order to investigate the effectiveness of the coating of the templates to produce core-

shell particles, different nanoparticle systems were prepared to be characterized before and after

the shell growth by DLS, AFM and TEM. Based on the study of the formation of the templates









discussed in the previous chapter, a combination of equal masses of surfactants was selected for

the synthesis of core-shell particles (Table 4-1). The shell growth was determined by DLS,

comparing the diameter of the templates and the coated particles.98 99 An increase in the average

size of the particles was considered as an indication of a successful coating. Once this process

was finished, the nanocapsule suspensions were centrifuged at 1600xg twice, followed by

sonication for approximately 1 h in order to resuspend the particles. Similarly to the templates, it

was observed that after sonication and filtration, most of the original mass of particles was

recovered. The diameter of the templates and the particles obtained by DLS after purification

were compared to obtain the shell thickness (ST). Equation 4-1 was used assuming that one

template produced one core-shell particle, where di is the diameter of template and d2 is the

diameter of the core-shell particle.

ST = (d2- di)/2 (4-1)

We selected three representative systems for the physical characterization of the core-shell

particles obtained. The systems NCU and NPY were prepared with formulations similar to the

template set D. They included small amounts of hydrophobic dyes for fluorescence studies,

whose results will be presented in the next chapters. Due to the low electron density observed for

the templates studied in the previous chapter, we decided to try a different staining technique to

improve the quality of the images obtained. The system NEB was prepared with a combination

of ethyl butyrate and 1-dodecene. With this combination of dopants we intended to obtain large

templates for ease of imaging. 1-dodecene was selected due to the possibility of reacting its

unsaturation with OsO4 through an oxidative addition.100 A highly hydrophobic molecule as 1-

dodecene would be difficult to remove from the core-shell particles during the purification steps,

especially due to the expected protection of the interior of the particles from the solvent after









their coating. The reaction of the unsaturation of 1-dodecene would also fix the metal center in

the particles, producing an increase in the contrast of the images obtained. As a second step, the

three systems were stained with UA to visualize the periphery of the particles, as discussed for

the templates in the previous chapter.101 All the systems were analyzed by AFM in a similar

approach to the one used for the characterization of the templates presented in Chapter 3. In this

case, we focused our attention on the differences in the values of the diameter and height

obtained in the section analysis before and after the synthesis of the shell, as shown in Table 4-3.

4.1.2 Results

The templates were used for the formation of core-shell particles without any further

purification, since it was observed that the presence of the surfactants used in their synthesis was

necessary for the shell to grow. Interestingly, when the coating of the particles was performed

with the templates suspensions "as is", precipitation occurred, due to chemical crosslinking and

formation of a gel, as shown in Figure 4-2. To prevent this, a 5-fold dilution of the original

suspension was used and TMOS was added in small portions of approximately 20 mg every 24

h, under vigorous stirring. The dilution assured that the templates were separated enough in

solution to prevent their crosslinking during the growth of the shell. The addition of TMOS in

small volumes prevented the silicate precursor molecules from reacting in between them to

produce a gel. As mention above, the core-shell particles could be centrifuged, contrary to the

templates. This indicates that the templates are very light and that there is an increment in their

mass after the coating, as expected due to the growth of the silicate network. This gave us an

effective way to separate the core-shell particles from unreacted materials and templates that did

not form a shell. After centrifugation, the particles were resuspended by sonication for

approximately 1 h. It was observed by DLS that the size of the particles was not affected by

sonication, confirming that the core and shell of the particles were firmly connected.









4.1.2.1 Size analysis of the core-shell particles

The size analysis report here was performed similarly to the study of the templates

discussed in Chapter 3. The system NEB produced particles with diameter values twice the ones

of the dye-doped templates, as expected due to the large difference in the concentration of the

dopant. For the three systems shown in Table 4-1 there was an increase in the average size of the

particles after addition of TMOS, as shown in the characterization results on Table 4-2. The shell

thickness, calculated with Equation 4-1, was surprisingly higher for NCU than for NPY. Both

systems were prepared similarly, including equimolar amounts of dyes, which produced

templates with similar dimensions, although the NCU templates were slightly larger. The

difference in the shell thickness may indicate that pyrene, being a more hydrophobic molecule,

promoted the formation of a lower amount of templates compared to C153. Since a similar

amount of TMOS was added to both systems, the one with a lower number of templates will be

coated with a thicker shell. A thin shell was produced in the case of NEB, despite the use of

approximately three times the amount of TMOS used for NCU and NPY. This is expected due to

the larger radius of these particles which results in a larger area per particle to cover. An

interesting general observation is that the increment in size is not immediately observed after

TMOS addition. Despite observing that the suspensions became whiter, only after a certain

amount of TMOS was added the diameter of the particles changed. It was observed that the

initial mass of the silicate precursor necessary to produce this change was larger for bigger

templates.

4.1.2.2 TEM analysis of the core-shell particles

A number of authors have used TEM for the characterization of core-shell particles,

especially in cases where the shell was prepared with silicate precursors.35'44, 102 Figure 4-3, 4-4

and 4-5 (top) show TEM images of the templates and the corresponding core-shell particles. In









all cases spherical particles were observed, although no significant differences in the contrast

between the core and shell were found, as has been observed by other groups.97 The images of

the NEB templates looked similar to the ones obtained for the sample set D, although imaging

was easier due to their larger size. The effect of doping the particles with 1-dodecene was

observed in the higher quality of the images obtained. The particles looked homogeneous, which

indicates an even distribution of the dopant inside the templates. Moreover, a darker ring on the

outside can be observed on their periphery as a consequence of the use of UA. The core-shell

particles seemed to show some differences in the electron density of the different parts of the

particles, but no trend could be identified, since some of them showed a darker interior while

others showed a higher electron density in their external part. We believe this is not related to the

presence of the dopant but that it is an artifact obtained while manipulating the instrument.

Systems NCU and NPY did not showed any special features as expected, although the pyrene-

loaded core-shell particles looked darker on the microscope compared to the other particles

studied, probably as a consequence of the thicker shell obtained (Figure 4-5). At the same time,

small particles were observed on the images, which seem to be produced when a thicker shell

was formed. Due to the larger difference in size between the templates and core-shell particles it

was possible to observed a clear difference in their sizes in the TEM images, which was not the

case with the other two systems with thinner shells.

4.1.2.3 AFM analysis of core-shell particles

AFM was used to study the influence of the coating process in the rigidity of the particles,

by comparing the diameter to height ratio (DHR) values for each system before and after the

coating of the templates, as shown in Table 4-3. As observed in the previous chapter for the

doped-templates, the NEB templates showed "hollow-like" structures, with a DHR of 15.3,

indicating a very flexible system. After the shell growth, the images showed remarkable









differences in the section analysis, where no "hollow-like" structures were observed. The

average height of the particles increased, producing a lower DHR value (9.3), indicating a

significant increase in the rigidity of the system. Based on these results, we believe the shell built

on the periphery prevented the particles from deforming as much as the templates after

deposition on a substrate, producing a more robust system. Apparently, the shell being a

chemically connect network, is less flexible than the material the templates are made of. This is

confirmed by the results obtained from Figures 4-4 and 4-5 for the NCU and NPY systems,

respectively. In both cases, the same diameter and height values for the templates were obtained,

63 and 5 nm, with a DHR value of 12.6. This value is lower than the one for the NEB system,

meaning that these templates are more rigid, probably due to the small amount of dye in their

interior. After the shell growth, both, the diameter and height, increased considerably for both

dye-doped systems. As mentioned above, despite using similar amounts of TMOS, NPY grew a

thicker shell than NCU, observed as different diameter results by DLS (200 and 120 nm,

respectively). These values correlate to the diameters obtained in the AFM analysis (141 and 87

nm, respectively), and more importantly, to the height values, which increased to 41 and 21 nm,

resulting in a decrease in the DHR values to 3.4 for NPY and 4.1 for NCU. This is a more

dramatic change than the NEB templates to particles since the dye-doped particles grew a thicker

shell and thus, become more rigid.

4.1.2.4 Analysis of the dimensions of the templates and particles in suspension and after
drying

In order to understand the changes in the dimensions of the templates and particles in

solution and when dried, we proceeded to analyze the changes in the volume of the particles

based on the data obtained by DLS and AFM, respectively. We selected the systems NPY and

NCU due to the similar dimensions of their templates but different shell thicknesses. Since the









radius of the particles calculated by DLS is obtained by fitting the data to the presence of

spherical scatterers, we calculated their volume with the formula 4/3 x (xr3), with r being the

hydrodynamic radius. When the particles were deposited on mica and the solvent evaporated,

disc-like structures were observed, as a consequence, the volume was obtained by approximating

the particles to a cylinder, using the formula (7r2) x h, where r is half the diameter and h the

height obtained through the section analysis in AFM. Table 4-4 shows the results of these

calculations. It should be mentioned that the values obtained are approximate and used only for

qualitative comparisons. For both set of templates similar results were obtained, as expected

based on the similarities of their dimensions. There is a dramatic decrease for the volume of

both, the NCU and NPY templates, after drying (17 and 14-fold decrease, respectively). This

indicates that the templates contained a large amount of water molecules that were removed

during drying. Interestingly, after coating, the volume of the particles changed differently

depending on the thickness of the shell. The system NCU, made of particles with a thin shell,

shrunk less than the templates after evaporation of the solvent (7-fold decrease in their volume).

This may indicate that the coating is preventing solvent molecules to penetrate to the interior of

the particles and that only the shell is hydrated. This would result in less water molecules being

evaporated, producing a smaller change in volume. The system NPY with a thicker shell, shrunk

13 times, as expected due to the larger volume available in the shell to store solvent molecules.

These are interesting results that can be used for the design of carrier systems for different

species. For instance, species placed on the shell of the particles are expected to be solvated,

even if they are not in the periphery. On the other hand, hydrophobic species that are included in

the mixture used to synthesize the particles could be expected to be preferentially located in their









interior, avoiding contact with solvent molecules and, as a consequence, they would be difficult

to remove during the purification.

4.2 Derivatization of the Surface of the Nanoparticles

4.2.1 Derivatization with 3-aminopropyltrimethoxysilane

One of the advantages of coating the templates with a siloxane matrix is the different

options for further derivatization of this material. 3-aminopropylsilane (APS) is the common

choice for attaching amino groups in silica particles and surfaces coated with silicate precursors.

The advantage of using this molecule is related to the ability of the amino group to act in

subsequent reactions as a nucleophile, which provides a means to attach different groups on the

surface of the particles by relatively simple chemistry, as shown in Figure 4-6. Based on this, we

prepared core-shell particles and let them react with APS (system NPAPS). For the reaction to

occur, it was necessary to raise the pH to 9.5 and the temperature to 70 OC. Similarly to what was

observed for the reaction with TMOS, the templates or particles could not be purified for the

reaction to work and the suspensions had to be diluted five times. To avoid crosslinking, the

coating agent had to be added in small volumes under strong agitation. The products were

dialyzed after the reaction took placed to separate any unreacted material. Size analysis was used

as explained before to confirm the growth of the shell, as shown in Table 4-5. It was observed

that after coating of the particles with TMOS, the addition of APS did not produce a significant

change in the diameter of the particles, suggesting that only a thin layer of APS was deposited on

the surface of the particles.

Typical AFM images for the templates and corresponding APS-coated core-shell particles

are shown in Figure 4-7. As expected based on the results discussed in the previous sections, an

increment in the diameter and height of the particles was observed, correlated with the changes

reported by DLS, as shown in the section analysis results reported in Table 4-5. Interestingly, the









APS-coated particles showed a lower tendency to aggregate after deposition on mica, compared

to the templates and particles coated with TMOS.

To confirm the presence of amino groups on the surface of the particles, we studied the

changes in the overall charge of the particles at different pH conditions. Figure 4-8 A shows

typical (-potential plots for the system NPAPS before and after the reaction with APS. As

explained in the previous chapter, the sign of the (-potential corresponds to the charge of the

particles and its absolute value is an indication of the stability of the particles in suspension. In

general, it was observed that the total charge of the particles was dictated by the silanol groups

on the surface, which are deprotonated, producing negative (-potential values. The particles

seemed to be stable when the pH of the suspensions is above 4, as observed by the high absolute

value of their (-potential. The response of the APS-coated particles is slightly different than the

particles without the amino groups at the same pH, although these differences are not dramatic.

The less negative (-potential values after coating are produced by the reduction of the number of

silanol groups, which reduced the number of negative charges on the surface of the particles and

the incorporation of more basic amino groups. Apparently, the particles were coating with a thin

layer of APS, resulting in a low density of amino groups. This is in agreement with the DLS

data, which showed no differences in size after the coating.

4.2.2 Derivatization with Dansyl Chloride-APS Complex

Dansyl chloride, a hydrophobic fluorescent dye, was attached to the surface of the particles

to study its fluorescence, which will be discussed in the next chapters. In this case, a reaction

between APS and dansyl chloride was performed as shown in Figure 4-9 to create a silane

derivative of the dye (APSDC1 complex). The product obtained was a yellow thick paste, which

was kept under argon until used. Two set of templates were prepared for derivatization with the

APSDC1 complex. The system NPDC1-1 was obtained from templates prepared without a









dopant, while the system NPDC1-2 was doped with EB. The suspensions of the templates were

divided in two, to grow coatings of different thicknesses with TMOS, before the attachment of

the APSDC1 complex, as shown in Table 4-6. Once the desired size range was obtained, an

excess of the APSDC1 complex was added as a solid to the suspensions, and left stirring for a

week under strong agitation. After this time, the excess complex was separated through filtration

and any hydrolyzed APSDC1 that did not react with the surface was separated through dialysis,

until no more DC1 was detected by UV.

Figure 4-8 B and C show the (-potential values at different pH for particles prepared with

the same templates containing no dopants (systems NPDCl-1A and -1B). The system 1A showed

clear differences at low pH values. This is due to the presence of amino groups on the surface of

the particles, which were protonated at acidic conditions, resulting in more positive (-potential

values. These differences became more noticeable for the sample NPDCl-1B, which grew a

thicker shell. We believe this is a consequence of the larger surface area available for coating for

the bigger particles. This would result in more amino groups available for protonation at low pH.

Moreover, the differences are more significant than the ones for the system NPAPS since each

APS-DC1 complex contains two amino groups, compared to only one in APS. This doubles the

number of amino groups added to the surface of the particles per molecule of the dye complex

attached. In both cases, aggregation of the amino-coated particles was observed when the pH of

the suspensions was close to their isoelectric point. This made the accurate determination of the

charge of the APS-coated particles challenging, at pH values close to 2.

Some interesting results were obtained for the particle set NPDC12, as shown in Figure 4-6

D. Similarly to the previous case, both dye-coated particles aggregated when the pH was reduced

to 3 or lower values, a clear indication of the presence of amino groups on the surface. In this









case, the first comparison is made between the templates and particles coated with APSDC1 but

no TMOS. Surprisingly, the templates showed a more positive charge than the amino-coated

particles, which contradicts the fact that the dye-coated particles precipitated at low pH and the

templates did not. These results are not completely understood, although it could be explain by

the presence of excess surfactant in the templates that was not completely removed despite the

extensive purification performed. The particles coated with TMOS and the dye, on the other

hand, showed the expected results based on their larger size, which can accommodate more

APSDC1 molecules, resulting in more positive (-potential values.

4.3 Preliminary Results on Other Systems Studied

4.3.1 Preparation of Large Templates and Core-Shell Particles for Calcination Studies

We started to investigate dopants that can produce larger templates and core-shell particles

than the ones obtained with hexadecane and ethyl butyrate. N-benzyl maleimide was used during

the synthesis of the particles as a dopant with surprising results. It was difficult to disperse this

material in the original surfactant mixture, even after long periods of heating. Despite this, large

templates and core-shell particles were obtained with diameters of 355 nm and 410 nm,

respectively, as observed by DLS. Similarly to the previous systems, the polydispersity of both,

the templates and core-shell particles, was high. Due to the large size of the templates, the

coating of the particles was difficult. As observed previously for other large templates, the initial

addition of TMOS did not produce an increase in the average size of the particles. After this, a

first small increased in their diameter was obtained, after which no more changes were observed

with DLS, even after adding large amounts of TMOS. Apparently, when trying to coat large

particles, only a thin shell grows, after which small particles are formed. Surprisingly, despite

using the same staining techniques used for the other templates, the TEM images showed very

dark particles, as observed in Figure 4-10 A. This may indicate that the dopant was not removed









from the system during the purification steps, due to its very low solubility in water, increasing

the electron density of the particles. Moreover, in this case, due to the ease of imaging, the

coating is observed as a skin on the surface of the particles (Figure 4-10 B). On the other hand, it

was difficult to image the particles in the atomic microscope due to their large size, as observed

in Figure 4-10 C and D. The tip of the cantilever had problems tracking the topology of the

surface due to the large differences of the features on the Z-axis, which made the size analysis

difficult. Approximate values were obtained for the diameter and height of the templates (450

and 95 nm, respectively), and for core-shell particles (550 and 300 nm, respectively). Despite

these difficulties, spherical particles were observed with only a few "hollow particles" in both

cases, as expected due to the presence of the dopant in the interior of the core-shell particles.

Calcination of these particles was performed similarly to the templates studied in Chapter

3. The images obtained for the templates (Figure 4-11 A and B) show the presence of small

particles, as observed with the smaller templates studied before. Interestingly, ring structures

with sizes similar to the templates were observed. At the same time, darker structures with

different shapes and spherical holes were observed. Apparently, the material inside the templates

had been ejected from the core, but somehow remained connected through a ring in the

periphery. The core-shell particles on the other hand, together with the small particles observed

in the templates, showed core-shell particles with holes in their interior (Figure 4-11 C and D).

The size of these holes was similar to the diameter of the small particles observed. This may

indicate that when TMOS is added, it not only reacted on the surface of the particles, but in the

pores inside the templates, forming small spherical particles loosely attached to the core-shell

particles. These would explain the observations reported in the DLS analysis section that the

diameter of the particles did not changed after the first additions of TMOS. Moreover, this would









indicate that the reaction with APS would occur not only on the surface of the particles but in

their interior as well, resulting in the less positive (-potential values than expected, as observed.

These are interesting results that indicate that the particles might be used for the growth of metal

particles in their pores, but this would need to be studied in more detail.

4.4 Conclusions

The templates studied in Chapter 3 were used for the formation of core-shell particles. The

silanol groups on the surface of the templates were used as anchoring points to grow a shell. This

was done through the sol-gel chemistry of tetramethoxysilane, which placed a silicate network

on the surface of the templates, producing core-shell particles. Dynamic light scattering showed

that the coating process produced an increment in the average size of the particles. This was

confirmed by observing that it was possible to centrifuge the core-shell particles, contrary to the

templates, due to an increase in their density. TEM images were used to confirm that spherical

particles were obtained and that more dense particles, observed as darker species, were formed

with thicker coatings. The topographic features of the particles were studied by AFM, by

comparing the diameter and height of the templates and core-shell particles. Similarly to the DLS

results, an increment in the diameter and height of the particles was observed after the coating

process. We compared the DHR values of the templates and particles, finding that their rigidity

increased after reacting with TMOS. We believe this is produced by the less flexible and more

rigid silicate network surrounding the particles, which prevents them from spreading on the

substrate as much as the templates. Accordingly, lower DHR values were observed for particles

with thicker shells. Calculation of the volumes of the core-shell particles in solution and after

drying on a substrate showed that the templates contained a large amount of water, which is lost

after deposition on mica. Similarly, particles with a thick shell shrunk considerably, contrary to

the ones with a thin coating, suggesting that the shell is extensively hydrated. As a first step









towards the attachment of different species on the surface of the particles, they were reacted with

APS to place amino groups on their periphery. The derivatization of the particles was confirmed

through (-potential measurements, which reported changes in the charges on the surface of the

nanoparticles after the reaction with the amino moieties. This approach was used for the

derivatization of the particles with a fluorescent dye. A derivative of dansyl chloride with APS

was prepared to attach the dye to the exterior of the nanoparticles. Finally, preliminary studies to

obtain larger particles were performed. It was observed that when a solid dopant was used,

particles with diameters above 300 nm were obtained. Coating of these particles was difficult as

was found that only a thin shell could be grown and further addition of TMOS produced small

species not attached to the particles. These large templates and core-shell particles were

calcinated and the products imaged. It was found that a large number of core-shell particles

maintained their size and shape but they showed some small "holes" in their interior. Small

particles of similar dimensions to these holes were found in the images, which may indicate that

they were originally attached to the particles. It is possible that small silica particles have grown

in the pores of the templates while the shell was growing, but they were not chemically attached

to them. These are interesting findings that need to be studied further.



Table 4-1 Formulations used for the synthesis of templates and corresponding core-shell
particles.
SO t80 EB 1-d OTMS Py C153 TMOS
NEB 125 125 29 10 19 42
NCU 125 125 12 0.70 15
NPY 125 125 13 0.45 14
All quantities in mg.









Table 4-2 DLS characterization results for templates and the corresponding core-shell particles.
di d2 ST
NEB 175 203 14
NCU 80 120 20
NPY 74 200 63
di: templates diameter, d2: core-shell nanocapsules diameter, ST: shell thickness, all in nm.


Table 4-3 AFM characterization results for templates (1) and core-shell particles (2) prepared as
indicated in Table 4-1.


di hi DHR1 d2 h2 D
NEB 183 12 15.3 195 21 9.
NCU 63 5 12.6 87 21 4.
NPY 63 5 12.6 141 42 3.
Diameter (d) and height (h) values given in nm.


IHR2
3
1
4


Table 4-4 Volume of templates and coated particles based on DLS and AFM data
Vsuspension (DLS) V dried (AFM) temp coated
templates coated templates coated Vsol/Vdried Vsol/Vdried
NCU 2.68*105 9.05*105 1.56*104 1.25*105 17 7
NPY 2.12*105 4.19*106 1.56*104 3.28*105 14 13
Solution (DLS) obtained with the formula 4/3x(r 3), V dried (AFM) obtained with the formula 7cr2xh
both in nm


Table 4-5 Formulation and size characterization results for APS-coated nanoparticles (NPAPS)
SO T80 OTMS TMOS APS d1 DLS d2 DLS di (hi)AFM d2 (h2)AFM
126 125 17.7 21.6 11.7 76 120 72(6) 134(42)
All quantities in mg, d: diameter, h: height, (1): templates, (2) core-shell nanoparticles, all sizes
in nm.


Table 4-6 Formulations used for coating of the particles with APSDC1 and DLS characterization
results
SO T80 EB OTMS dl DLS TMOS d2 DLS
NPDCl-1A 127 125 -17.0 76 12.1 136
NPDC1-1B -76 45.8 175
NPDC1-2A 127 125 13 17.6 138 138
NPDC1-2B 138 20.3 236
All quantities in mg, d: diameter in nm.


I











OMe

i.-OMe
/Si\
MeO OMe

H20
A -MeOH
OH
I-OMe
n MeO-SiO

MeO




0
Si-O





,O
/O


OH

---OH
B /\
MeO OMe

H20
OH -MeOH

I .--OMe

MeO OMe


/


0
-O

SOSi

0

0O 0
o \
H










/ H

/---Si -O3S- iO

0
si- \o
0
St-\ /
0 Si

Si'o
,o /
OSi\ Si H
'o 1 0
/0


n H20

0-
Si-0--~ iOH
Si
/O \
Si-OP
O Si

P /O
nH20 Si
O O_ OH


Figure 4-1.Reaction scheme for the hydrolysis of TMOS, shown for the first A) and second B)
hydrolysis products. The reaction of these species with the silanol groups on the
surface of the templates is shown, as well as the final product after complete
hydrolysis at neutral pH.






























Figure 4-2.Crosslinking resulted when templates were not diluted before the addition of TMOS.
Scale bar 500 nm.


S00.














"0 0.25 O o 0.7;5 l.t0 1.2u


0 0o 0. 00 ?s 1._O im


Figure 4-3.TEM and AFM height images of NEB A) templates and B) core-shell particles. Scale
bars in all images 500 nm.












































0


0 100 200 3 DO r


>0


v.25 o~ac 0);5 .v


Figure 4-4.TEM and AFM height images of NCU A) templates and B) core-shell particles. Scale
bars in all images 300 nm.





































I 0 O.; i. l S
pt 0 0.5 0 .$0 cIA J.L) 1, SI.r'


Figure 4-5.TEM and AFM height images of NPY A) templates and B) core-shell particles. Scale
bars in all images 500 nm.



HO
o OH"
SSii O OH O\Si2 p9 7 OH
/ \> Si A (CH30)3Si-CH2CH2CH2NH2 pH 9, 70 oC /S\ Si \ /


Figure 4-6.Attachment of APS to core-shell nanoparticles.


*-V^A/V



























I ,


m..5 ]. 0 II;


Figure 4-7.AFM images for the templates and APS-coated nanoparticles reported on Table 4-4.


I









1


15-
10-
5-
0-


10-



-25-
-30-
-35-


~1



'-4;


S-4U 4 6 10 1
2 4 6 8 10 12 2 4 6 8 10 12


pH












. 1




Si'
2 4 6 8 10


15
10-
5-
0
-5
-10 -
-15-
-20-
-25-
-30-
-35-
i -40-
12


2 4 6 8 10 12


Figure 4-8. Comparison of (-potential results before (circles) and after (squares) derivatization

with APS: A) NPAPS; and APS-DCl: B) NPDCl-1A, C) NPDCl-1B and D) NPDC1-

2A (squares) and NPDC12B (triangles).




117


15-
10-
5 -
0 -
-5 -
-10-
S-15-
-20-
-25-
-30-
-35-
-40



15-
10-
5-
0-
-5 -
-10-
S-15-
-20 -
-25 -
-30 -
-35 -
An


!a













OMe
MeON I~ I 'Nl

MeCO


N(CTLCH' )
CHC12


Figure 4-9.Coupling between dansyl chloride and APS.


,- r'- .. .\


, .. I.
* < L.^r J,- .MJ


Figure 4-10.


r \.
o /
, I. II --
m 1


TEM and AFM images of N-benzyl maleimide-doped templates (A and C) and
core-shell particles (B and D). Scale bars 500 nm.


OMe

Mcd/
M~cO 6


+ NEtiHC[























tC t,,^i ^ :'- '4 4D





S ai -'. .*'. B .."
*.
8f


Figure 4-11.


TEM images ofN-benzyl maleimide-doped templates (A and B) and core-shell
particles (C and D) after calcination at 600 C. Scale bars 500 nm.









CHAPTER 5
SEQUESTRATION AND ENCAPSULATION OF FLUORESCENT DYES IN TEMPLATES
AND CORE-SHELL PARTICLES

5.1 Sequestration of Dyes by the Templates

5.1.1 Strategy

Based on the characterization studies discussed in the previous sections, the templates

obtained consist of a flexible core mainly formed by chemically connected OTMS molecules.

After addition of TMOS, a hydrophilic shell is built on the periphery of the templates, which was

observed to increase the rigidity of the particles. The flexibility and expansion of the templates

and core-shell particles after deposition on a substrate suggested that these particles contain

pores in their interior that could be used to accommodate guest molecules. We believe that the

polarity of the interior of the particles is lower than that of the solvent in which the particles are

suspended. This difference is intended to be used as a driving force for the sequestration of

species with low polarity, like drugs or pesticides. Similarly, particles for the encapsulation of

hydrophobic molecules, with low solubility in water, like fluorescent dyes or reactive species,

could be obtained by including the active molecules in the formulation used to synthesize the

particles. This would allow the use of these species in aqueous suspensions without precipitation.

This is especially important when applications in biotechnology are considered, since certain

dyes and other useful species are known to be toxic. In these cases, encapsulation of these

molecules assures that no unwanted reactions take place, since the molecules would have no

tendency to escape from the templates or nanoparticles due to the differences in polarities with

the media. As a consequence, we initiated studies related to the ability of these core-shell

structures to be used as nanocontainers and sequestration agents for hydrophobic small

molecules in aqueous systems.









As a first step, it was necessary to decide which technique would allow us to detect the

presence of different species in the interior of the templates and core-shell particles. To study the

sequestration of species from solution into particles, UV active species are commonly used.103-105

Typically, a volume of a known concentration of the absorbent is mixed with a suspension of the

particles. After a stabilization period, the particles are removed and the UV absorption of the

species left in the supernatant is analyzed, from which, the amount of molecules that were

sequestered is determined. Although being an effective and simple way to quantify the uptake

capacities of the particulate systems used, this technique gives no information on the

sequestration process, being unable to differentiate if the UV active species are adsorbed on the

surface or transported into the interior of the particles. Another alternative involves the

identification of the absorption maximum of the active species and to relate its changes to

variations in the environment where the probes are placed. Since the nanoparticles scatter light in

a broad area of the UV-vis region, it was not practical to use this technique for the

characterization of the loaded species. On the other hand, as shown in Figure 5-1, the

nanocapsules did not show any fluorescence when excited in the region that most dyes adsorb

and emit light, but scattering, due to the size of the particles, was observed. This was confirmed

using different wavelengths for excitation that produced changes in the position of the peak

observed.

Based on this, we started studies on the physical characterization of the templates using

fluorescent dyes. Pyrene was selected as the first probe due to its low polarity and well known

dependence of its emission on solvent polarity.67 Our first goal was to determine if the dye could

be incorporated in the templates. Then, we studied the influence of the amount of dopant, as well

as its polarity, in the environment sensed by the dye. Later, this strategy was extended to the









core-shell particles, to investigate whether the coating of the templates produced changes in the

internal structure of the particles. Finally, C153, another hydrophobic probe, was used to

investigate if the chemical structure of the dye used affects the results obtained.

5.1.2 Use of Pyrene as a Polarity and Rigidity Probe

Pyrene has very useful characteristics for the study presented here. Especially important

for us is the lack of functional groups in the molecule which prevented any kind of specific

interactions that are not due to the polarities of the probe and the different parts of the particles.

As explained in Chapter 2, the polarity index, defined as the ratio of the third and first peak

intensities (I3/11) in the emission of pyrene, has been qualitatively related to changes in polarity

in the environment where the dye is placed. This has been used by different groups to study the

polarity of different systems, which we used as a reference to compare our results to. Tedeschi et

al. used pyrene to determine the polarity of films obtained by the layer by layer deposition of

polyelectrolytes, observing that the polarity of the films obtained depended on the chemical

structure of the polymers used.106 Hugerth studied the aggregation of the hydrophobic drug

amitryptiline, using pyrene as a polarity sensor, based on the changes in the I3/11 value after

amitryptiline aggregates with low polarity were formed.107 Pandey et al. studied the solvation

environment of siloxane polysoaps dissolved in water. They observed that derivatization of the

polymers with hydrophobic side chains was correlated with a lower polarity sensed by the

probe.108 Danko et al. studied the polarity of interpenetrating polymer networks prepared with

different monomers to which pyrene had been chemically attached.109 A similar approach was

taken for the use of pyrene as a sensor for rigidity. As discussed in Chapter 2, the excimer

emission appears as a broad peak centered around 480 nm. Hence, the ratio of the intensities of

the emission of the monomer and excimer (I1/Iexc) is related to the ability of the dye molecules to

form dimers, and can be used as an indicator of changes in the viscosity of the media.65









Vanderkooi and Callis used pyrene as a probe to study the lateral diffusion of species in the

hydrophobic region of membranes.66 Jang and Oh observed the formation of excimers when the

loading of pyrene was increased in polymer core-shell nanoparticles.110 Due to its very low

solubility in water, pyrene has been used as a probe in studies of micellar systems as well,

observing that it is preferentially solubilized in the most external hydrophobic regions of these

aggregates, in contact with water molecules to a certain degree.67 Hence this probe has been used

for determination of CMC values. Pyrene would from aggregates in water if the surfactant

concentration is low enough to prevent the formation of micelles. Once the surfactant molecules

start aggregating, the probe will be incorporated in the micelles due its hydrophobicity, which is

observed as a dramatic shift in the emission of the dye and on the intensity of its emission.

5.1.3 Results

5.1.3.1 Determination of the optimal conditions for the study of the fluorescence of
sequestered pyrene

Figure 5-2 shows the emission spectra of pyrene in the two compounds used as dopants.

Changes in the intensity of the first and third peaks in the emission spectra of the dye (371 and

382 nm, respectively) were observed as a consequence of the difference in polarities of the

solvents. The polarity index value for pyrene dissolved in ethyl butyrate was determined to be

0.797 and 1.77 when dissolved in hexadecane. These values were expected based on the

literature polarity index for compounds with similar polarities, hexane (1.65) and methanol

(0.75). Based on this, we anticipated the templates to offer the encapsulated molecules

microenvironments with different polarities, had the dopant molecules stayed in the interior of

the templates after the purification steps. To find out the conditions to perform an accurate

determination of the I3/11 value, we carried out a titration of the templates with a methanolic

solution of pyrene. This was done with a 0.3 mg/mL template suspension in all the cases. This









concentration was selected to minimize scattering of light due to the size of the particles, whose

peaks could interfere with the fluorescence of the dye. As shown in Figure 5-4 A, the

concentration of the dye was increased from 0.1 aM, which was found to be too low, resulting in

a noisy signal for the emission of the dye, to 2.2 gM, which promoted the formation of excimers,

as observed by the appearance of a broad peak centered at approximately 475 nm. These results

showed that a concentration of the probe between 0.4 and 1 iM produced a clear signal without

excimers, thus, the I3/11 value could be obtained with reliability. Within this range, it was

observed in the titration curve (Figure 5-4 B) that a concentration of the probe of 0.5 iM is close

to the inflection point, which was taken as the I3/I1 value used to report the polarity of the media.

5.1.3.2 Sequestration of pyrene by ethyl butyrate-doped templates

Table 5-1 shows the polarity index obtained from sequestered pyrene for the two set of

templates investigated in the previous section. The EB-doped templates showed a I3/11 value

between 1.30 and 1.36, considerably higher than the value when the dye was dissolved in this

solvent, but not as high as when dissolved in the less polar hexadecane (I3/I1: 1.77). This is

interpreted as an apparent polarity in between that of the two pure dopants. Interestingly, when

the values for the three EB-doped systems were compared no significant differences were

observed, which indicated that the amount of dopant or OTMS used did not affect, to a

significant extent, the composition of the volume where the dye was placed.

There are two plausible explanations for the differences in the environment sensed by the

probe after internalization by the templates and when dissolved in the pure dopants. One

explanation is that the ethyl butyrate molecules where lost during the purification steps, leaving

pores in the interior of the templates whose polarity did not resemble that of EB. In this case the

dopant would be acting as a "porogen." The other explanation is that the EB-rich areas inside the

particles were not accessible to the dye molecules, probably due to their size or polarity. If we









consider that the polarity sensed by the dye molecules is lower than the dye suspended in

different micellar systems (typical I3/11 values between 0.74 and 0.95), where the dye is known to

reside in the outer volume of the aggregates, in proximity to the solvent front, we can conclude

that this was not our case. Based on this, we can assure that the dye molecules are in the interior

of the templates, somehow protected from the aqueous media where they were suspended. With

this, we believe the hypothesis based on the idea that the dye could not reach the interior of the

particles, where the dopant molecules would reside, is not valid. Figure 5-5 shows what is

believe to happen during the purification of the templates. Apparently, the dopant molecules

were removed during the washing used to separate the surfactants and unreacted materials from

the templates. After this, the external pores were filled with water molecules, which is less likely

to happen with the ones buried in the center of the particles. When pyrene was added to the

system, the dye molecules preferred to migrate to the more internal pores to avoid contact with

the solvent, which was observed as a low average polarity reported by the dye. Interestingly,

when the concentration of the dye was increased, an increase in the apparent polarity was not

observed but the formation of excimers did occur. This confirmed that the dye molecules

avoided the periphery of the templates, preferring to pack more tightly in the more protected

pores.

5.1.3.3 Sequestration of pyrene by hexadecane-doped templates

The polarity reported by pyrene sequestered by the hexadecane-doped templates was

slightly lower than the EB system. Surprisingly, these differences were not as dramatic as

expected based on the emission of the dye in each solvent, as shown in Table 5-1 (I3/I1: 1.4-1.6).

Confirming what was observed with the EB-system, the dye seemed to be placed in a nonpolar

environment, protected from the solvent. In this case, the 13/11 value was very close to the polarity

index when dissolved in the dopant. Apparently, the pores formed by hexadecane offer slightly









better protection than the ones formed by EB. This could be a consequence of the presence of a

small amount of dopant molecules that could not be removed during the purification due to its

low polarity.

5.1.3.4 Study on the retention of the dye after solvent exchange

To test the idea that the probes internalized by the templates are protected from the solvent,

we loaded two set of templates, one dopant-free and the second prepared with EB, with an excess

of pyrene, added as a solid. As shown in Figure 5-6, in both cases, the emission of the dye was

dominated by the excimer signal, as expected due to the high loading of the dye. After this,

dialysis was performed until no more dye could be detected by UV in the dialyzed media.

Subsequently, the fluorescence was recorded again. The emission of the dye was clearly

observed, confirming that the probe found pores in the interior of the particles that were not

reached by the solvent molecules, thus, avoiding being washed-out during the dialysis.

Moreover, the polarity index for both systems was very similar, 1.27 for the dopant-free

templates, and 1.19 for the EB-doped system, which indicated that they were localized in similar

environments.

5.2 Sequestration of Dyes by the Dopant-Free Core-Shell Particles

The fluorescence results showed above demonstrated the successful use of steady state

fluorescence spectroscopy to detect hydrophobic model compounds internalized by the

templates. Since our interest is to develop a method to synthesize core-shell nanoparticles for the

uptake of harmful compounds, we proceeded to study the sequestration of fluorescence probes

by the core-shell particles in a similar way. To study particles that could be use in biomedical

applications, we used dopant-free particles, which produced species with smaller sizes that the

ones obtained by using a dopant (Table 5-2). To expand this study and investigate if the results









obtained with pyrene were a general trend, C153, another fluorescent hydrophobic probe, was

used.

5.2.1 Sequestration of Pyrene by Dopant-Free Core-Shell Particles

The dye-free dopant-free core-shell nanoparticles were prepared similarly to the samples

D2 and D3 studied in the previous chapter. These particles were mixed with a methanolic

solution of pyrene and the fluorescence was recorded. As observed in Table 5-3, the

concentration of pyrene necessary to observe a clear emission but no aggregates was lower than

that for the doped-templates of a similar concentration (0.12 [M). It should be noted that the

concentration of particles used here are mass per volume, which means that despite the similar

concentrations used, the amount of the core-shell particles may be lower compare to the

templates, since the former are heavier. The I3/11 value observed was 1.38, similar to the ones

obtained with the dopant-filled templates, meaning that the polarity encountered by the dye was

similar to the previous case, indicating an effective protection from the solvent. This indicates

that the probe could diffuse to the interior of the particles through the shell, probably reaching

the core, which is mainly formed by what originally was the templates, or that it resides

somewhere in the shell where water molecules could not penetrate. These are promising results

for the use of these nanoparticles as sequestration agents for hydrophobic compounds. When the

concentration of the dye was increased to approximately 2 aM, the formation of excimers was

detected, as expected.

5.2.2 Sequestration of C153 by Dopant-Free Core-Shell Particles

5.2.2.1 Strategy

In order to investigate if these findings were a general trend we used C153, another

hydrophobic fluorescence probe, which has been used as a probe for polarity and rigidity due to

its large change in the dipole moment when excited,6870 as discussed in Chapter 2. There are a









number of reports of using C153 as a probe for the determination of polarity in different

environments. Kumbhakar et al. used C153 in poly(ethylene oxide)-based surfactants to study

the solvation dynamics in these systems. They observed a shift to the red in the emission of the

dye suspended in micelles of the surfactant with more ethylene oxide units, which they

interpreted as the probe sensing a more polar environment. Ferrer and Lianos111 determined the

maximum in the emission spectrum of the dye in different micellar systems, finding that a more

red shifted value corresponded to more polar environments in the micelles. Hof and Lianos112

incorporated C153 in the formulation for the synthesis of a Si02 matrix. They did not observe

any changes in the emission of the dye until the matrix was left drying for a period of two weeks.

They attributed this change to the increase in rigidity of the matrix where the dye resided, which

prevented the stabilization of the excited state by the polar surroundings. Seth et al.113 used this

probe for the characterization of ternary-micellar systems, including an ionic liquid, finding that

C153 resided in the surface of the microemulsions. For our studies, the important difference with

pyrene is that C153 contains functional groups that can potentially interact with the different

parts of the particles.

5.2.2.2 Results

Table 5-3 shows the emission maximum observed after loading of the dye from a

methanolic solution, similarly to what was done with pyrene. The polarity sensed by the probe

was higher than when the dye is dissolved in methanol (536 nm) and similar to the values

obtained by Stathatos in silica/poly(propylene oxide) composites.70 This value shifted slightly to

the blue from 543 to 541 nm with a ten-fold increment in the dye concentration. The emission

maximum observed indicated that the dye molecules did not penetrate the shell of the particles,

but stayed close to the surface. Such a high emission maximum has been identified as a sign of a

specific interaction, like hydrogen bonding. In the case studied here, this is probably due to the









interaction of the carbonyl and amino groups of C 153 with the silanol groups on the shell,114

similarly to what was observed by our group with an electroactive probe.115 When this is

compared to the results obtained from pyrene, as shown in Figure 5-7, it is clear that the

chemical structure of the probe to be sequestered influences its final position inside the particles.

This should be taken into consideration when designing an uptake system for a particular probe.

In the next chapter this idea will be introduced for the design of antenna systems.

5.3 Encapsulation of Dyes by the Templates and Core-Shell Particles

5.3.1 Strategy

We finally proceeded to study the changes in the environment that the probes experienced

when they were encapsulated in the templates and core-shell particles. We prepared these

systems by including the dyes in the formulation used to synthesize the particles, as explained in

the previous chapter (systems NCU and NPY). The aim of this study was to investigate the

changes occurring within the interior of the particles before and after the synthesis of the shell,

through changes in the fluorescence of the dyes used. This was compared to the AFM

information gathered in the characterization studies in Chapter 3 to determine whether they give

similar results.

5.3.2 Results

5.3.2.1 Encapsulation of pyrene in the templates and core-shell particles

A comparison of the fluorescence of pyrene in the templates and corresponding core-shell

particles for the system NPY is shown in Figure 5-8. It is expected, based on the results from the

previous sections, that some of the dye would be lost during the purification steps, for example,

dye molecules trapped in the shell during the synthesis could have been removed together with

the solvent, which would produce a decrease in the dye concentration. As a consequence, the

emission reported here corresponds to the molecules of the dye that remained in the system after









purification. As shown in Figure 5-8, the emission of the encapsulated dye in the templates was

dominated by the excimer fluorescence, observed as a broad peak centered at 475 nm, which

prevented us from using the I3/11 value to investigate the polarity of the interior of the particles.

As a consequence, we used the excimer to monomer emission ratio (I1/Iexc) to investigate the

rigidity of the media where the dye was placed. The I1/Iexc ratio changed from 4.4 in the

templates to 12.1 in the particles, which indicated that it is more difficult for the dye molecules

to form excimers with ground state monomers after excitation in the core-shell particles.110 This

is surprising since it would be easier for the templates to lose dye molecules than the core-shell

particles, since the latter would protect the dye better from contact with the solvent. A reduction

in the concentration of the dye would make it more difficult to form dimers. We believe that the

lower I1/Iexc in the core-shell particles is a consequence of the increased rigidity in the system

after the formation of the shell, which decreases the mobility of the probe. These results

correlated with the AFM observations, showing a decrease in the flexibility of the core-shell

particles compared to the templates as a consequence of the coating with the siloxane network.

5.3.2.2 Encapsulation of C153 in the templates and core-shell particles

Similarly to the work with the dye-free core-shell particles, we used C153 to investigate if

these findings were a general trend. First, to confirm that the emission detected was actually

from the encapsulated dye, we looked at the particles under the fluorescence microscope.

Unfortunately, this could not be done with the pyrene-doped particles due to the lack of the

proper filters in the microscopes we have access to. A fluorescence microscopy image of the

C153-encapsulated core-shell particles is presented in Figure 5-9 A. The size of the particles was

below the resolution limits of the microscope, so it was necessary to perform the imaging after

the solvent had evaporated to promote particle aggregation. The emission is observed to be

confined to spots with diameters in the few micrometers range, which corresponds to the









aggregates observed in the AFM images. We analyzed the templates and core-shell particles in

solution similarly to the previous cases. The difference in the emission after the coating is

observed qualitatively in Figure 5-9 B, as a shift to the blue after the shell growth. Table 5-4

shows fluorescence data for these templates and particles. The maximum in the emission shifted

from 533 to 525 nm after the shell growth, indicating a decrease in the polarity of the

environment where the dye was placed, similar to a change when the dye is dissolved in ethanol

(530 nm) and butanol (525 nm).68-70 We interpreted this decrease in the apparent polarity of the

environment surrounding the dye as a consequence of the shell acting as a barrier to isolate the

interior of the particles, where hydrophobic molecules tend to reside, from the polar solvent.

Moreover, these values are considerably lower than the ones obtained when the dye was

sequestered by dye-free particles. From this, we can conclude that the dye ended up placed in

different parts of the particles, sensing different environments.

5.4 Conclusions

Our results for the characterization of the structure of the templates and particles

synthesized in Chapters 3 and 4 are presented here. We used hydrophobic fluorescent dyes to

characterize the interior of the templates and core-shell particles. The dependence of the

fluorescence of pyrene on the polarity and rigidity of the microenvironments where it was placed

after its sequestration was used to study the environment in the interior of the particles. This was

done for the ethyl butyrate and hexadecane doped-templates. It was observed that the despite

slight differences, the polarity of the interior of the templates did not depend on the polarity of

the dopant. We interpreted this as an indication that the dopants were removed from the

templates during the purification steps. In average, the polarity of the interior of the templates

reported by pyrene was low, which means that dye molecules tended to migrate to the "dry"

pores, away from the polar solvent. Subsequently, we used dopant-free core-shell particles to









sequester pyrene from aqueous suspensions, finding that the dye ended up in an environment of

similar polarity than that of the doped-templates. A second probe was used, C153 which showed

different results. In this case the polarity sensed by the probe was higher and similar to the dye

dissolved in a polar solvent. Due to its low aqueous solubility we interpreted this information as

the dye being trapped in the shell due to hydrogen bond interaction. This suggested that the

structure of the dyes mandates the final position of the probe inside the particles. Finally, we

encapsulated these dyes in the templates and core-shell particles by adding the dye during their

synthesis. In this case, we intended to identify the changes in the polarity and rigidity in the

interior of the particles through changes in the emission of the dyes before and after the growth

of the shell. It was observed with pyrene that the rigidity of the volume where the dye was placed

increased after the coating of the particles. Coumarin 153, on the other hand, reported a decrease

in the polarity of the interior of the particles after the shell growth, showing that the siloxane

network isolates the interior of the particles from the polar solvent. We believe these findings are

promising for the design of scavengers for drugs and pesticides.



Table 5-1 13/11 values for pyrene in ethyl butyrate (EB) and hexadecane (Hx) and EB and Hx-
doped templates.
13/I1 13/11
EB 0.797 Hx 1.77
EBB 1.31 HXB 1.41
EBC 1.32 HXC 1.43
EBF 1.36 HXE 1.58



Table 5-2 Formulation used in the synthesis of core-shell particles for sequestration studies and
size characterization results obtained by DLS.
T80 SO OTMS TMOS di d2
NSL 125 125 14 12 98 200
All quantities in mg. Surfactants were suspended in 13.9 g of saline. di: templates, d2: core-shell
particles, both in nm.










Table 5-3 Fluorescence data for pyrene and C153 sequestered by dopant-free core-shell
particles.
pyrene [ 0.12 [M]b [2.1 tM]b C153 [0.13 [M]b [1.25 MM]b
13/11 1.38 maxa 543 541


I3/Iexcimer 1.9
a: excitation wavelength 420 nm, b: final dye concentration in the suspension


Table 5-4 Steady State Fluorescence data for templates and core-shell particles used to
encapsulate pyrene (NPY) and C153 (NCU).
C153 Pyrene
templates np templates np
maxa 533 525 I1/ex 4.4 12.1


80x106

60x106

40x106

20x106


wavelength (nm)


Figure 5-1.Fluorescence response detected after excitation of a dye-free nanocapsule suspension.
The peak observed is due to scattering of light by the particles, excitation wavelength
270 nm


pyrene


CF3



coumarin 153

coumarin 153


Figure 5-2.Chemical structures of the dyes used in this study.










7DE+05


S6DE+05
i5DJE+05
4JDE+05
3DE+05
2DJE+05
1DJE+05


360 370 380 390 400 410
wavelength nm)


360 370 380 390 400 410
wavelength nm)


Figure 5-3.Emission of a 1 kM pyrene solution in A) hexadecane and B) ethyl butyrate.


--aluM
--3 uM
a5 uM
a8uM
-1 DuM
- 2-uM


- 2.5
2
15.


360 410 460 510 560
wavelength (nm)


intensity


excimer

I\


0 0.5 1 15
[ pyrene ]


Figure 5-4.Determination of conditions to study the polarity of templates with pyrene. A) Pyrene
titration for hexadecane-doped templates, numbers in the box indicate pyrene
concentration, B) plot of the ratio of the intensities of the third and first emission
peaks of pyrene (13/11) at different dye concentrations.


55E+05
S45E+05
35E+05
S25E+05
15E+05
5DE+04


A 12E-07
1DE07
, 8flE.06


. 4flE.O 6


2 25


I'"--


r~'"'~-?










S pufication addition of
dopantt removal) dye




green: dopant-filled pores blue: water-filled pores red: dye-filled pores grey: empty pores


a'*
S '- *


Figure 5-5.Representation of the removal of dopant molecules during purification of the
templates. The positions preferred by pyrene molecules when sequestered by a
suspension of templates, is shown as well.


B 1

S0.6
0.4
0-2
0
n /\a4
J / \ i0
& / \oIC2
rk / \ ^


360 410 460 510
wavelength (nm)


360 410 460 510
wavelength (am)


Figure 5-6.Emission of pyrene loaded in A) dopant-free templates and B) EB-doped templates,
before (black line) and after (grey line) dialysis.


pyrene addition


C153 addition


red circles: sequestered pyrene


white circles: empty pores


orange circles: sequestered C153


Figure 5-7.Representation of the positions preferred by pyrene and C153 molecules when
sequestered by core-shell particles.


a

0

S.


A
1
j 0.8

04
0.2
0


K .





























380 400 420 440 460 480 500
wavelength (nm)






Figure 5-8.Normalized emission of pyrene encapsulated in templates and in the corresponding
core-shell particles (blue) for the NPY system.


.E -


0.2 -


.I.
I.aD


wavelength (nm)


Figure 5-9.Fluorescence microscopy images for C153 encapsulated in core-shell particles, scale
bar 10 am. Image was obtained in black and white, color was added for clarity B)
Normalized emission of C153 encapsulated in templates (red line) and in the
corresponding core-shell particles (blue line).









CHAPTER 6
STUDY OF THE COMPARTMENTALIZATION OF THE TEMPLATES AND CORE-SHELL
PARTICLES THROUGH ENERGY TRANSFER BETWEEN DYES

6.1 Energy Transfer in the Templates

6.1.1 Strategy

The information gathered in the previous chapters indicated that the templates and core-

shell particles obtained could be used as nanocontainers for different species of low polarity.

Two approaches were used for loading hydrophobic fluorescent dyes in the templates and core-

shell particles. The dye was included at the beginning of the synthesis of the particles in one case

encapsulationn), and after the coating of the templates in the second case (sequestration). It was

observed that the probes sensed a nonpolar environment in the former, and that the final position

in the particles differed, depending on the structure of the probe; in the latter. This gave us the

possibility of placing different dyes in different parts of the nanoparticles, based on their

polarities and interactions with the shell, and the method selected to load them in the particles.

In this Chapter, we use this knowledge to explore whether probes that are expected to be

placed in different parts of the particles can communicate through a resonance energy transfer

processes (RET). To do this, we took advantage of the fluorescence response of the two probes

used previously. As shown in Figure 6-1 A, the overlap between the excitation of C153 and the

emission of pyrene, makes this couple a very efficient acceptor-donor pair, especially due to the

high quantum yield of C153.116 This pair of probes has been used in the past in a few matrices,

including micellar systems117 and silica surfactant composites.11l In the study by Ferrer and

Lianos,111 a high efficiency for the RET process was observed. They proposed that both probes

were placed near the head of the surfactans used to prepare the composites, near the most

external methylene groups in the tail of the amphiphiles. In our case, we used the templates

reported on Chapter 3, prepared with two different dopants, to study the energy transfer in the









precursor of the nanoparticles by sequestering both probes from aqueous suspensions. After this,

a set of core-shell particles prepared with pyrene during the synthesis (system NPY) was used to

sequester C153. Based on the results in Chapter 5, we expected to see a low efficiency in this

system since it was observed that the probes ended in the different positions inside the particles.

We then we extended this approach to use another pair of probes. Based on the fact that

C153 is adsorbed on the shell of the particles when sequestered from aqueous suspension, we

selected another hydrophobic probe for RET studies, 4-(dicyanomethylene)-2-methyl-6-(p-

dimethylaminostyryl)-4H-pyran (DCM), known to place itself in the external part of micelles, in

proximity to the solvent front, which produces an emission with maximum above 600 nm.118-120

The emission of C153 and excitation of DCM overlap as shown in Figure 6-4 A. As a

consequence, this pair can undergo a resonance energy transfer (RET) process if they are located

in close proximity. This was used to study the ability of these probes to communicate through

RET based on the degree of internalization after their uptake from aqueous suspensions,

similarly to the previous system analyzed. To study this interaction between the probes in more

detail, we measured how the energy transfer process affected the fluorescence lifetime of the

donor. This was not possible with the previous RET pair due to limitations on the instrument

used, which prevented us from measuring the lifetime fluorescence of pyrene. We used the time-

correlated single photon counting technique (TCSPC) for these studies. As explained in Chapter

2, the fluorescence lifetime of the donor would be affected only if the interaction between the

probes happened after excitation and not by association of the probes before the donor was

excited. Based on this, if we were able to detect energy transfer between the probes by steady

state fluorescence and, correspondingly, to observe variations on the fluorescence lifetime of the

donor, we could conclude that there was a collisional process in between the probes. If no









variation in the lifetime could be observed, then the energy transfer was produced by a static

interaction.65

6.1.2 Results

The fluorescence of pyrene before the addition of C 153 was the typical response observed

for the dye sequestered in the templates, as reported in Chapter 5. Addition of the acceptor

produced a decrease in the intensity of the first peak (I1) of the donor emission, as reported in

Table 6-1 and observed qualitatively in Figure 6-1 B. This was confirmed by analyzing the third

peak (13) in the emission spectra of pyrene. In this case, similar trends were observed, although

the decrease in the intensity of the emission of the donor was slightly lower. It was not clear at

this point the reason for this discrepancy. We focused our discussion in the first emission peak,

since this signal is more separated from the emission peak of C153, which could overlap

(although minimally) with the emission of pyrene. As observed in Table 6-1, the emission of the

donor decreased to approximately 45% of its original value when the concentration of C153 was

increased to 2.45 aM. The values obtained at the highest C153 concentration studied were

probably not accurate, since it was observed that the suspensions became slightly turbid. This

could be a consequence of the saturation of the templates, which would prevent them from

sequestering more acceptor molecules, or the adsorption of the probe on the external part of the

particles, hiding the charges on their surface (as reported for the surfactants in Chapter 3), thus

promoting their aggregation. At the same time, the appearance of a peak above 500 nm was

observed clearly when the concentration of the acceptor was increased to approximately 0.6 lM,

confirming that C153 was excited through the emission of pyrene (Figure 6-1 B). This was

further confirmed by obtaining the excitation spectrum of C153 at 0.6 atM, as shown in Figure 6-

1 C, in which the excitation through pyrene is shown as a maximum of lower intensity at 338









nm. It is obvious the difference with Figure 6-1 A, where the peaks below 350 nm were not

present in a typical excitation spectrum of C 153 sequestered by the templates.

As discussed above, these experiments were designed to study whether the probes can

communicate when sequestered by the templates. The energy transfer observed confirmed that a

certain fraction of the probes were placed in close environments, since the Forster radius for this

pair has been calculated to be approximately 3.4 nm.111 Figure 6-1 D is a plot of the efficiency of

the process calculated by comparison of the emission of the donor before and after each addition

of the acceptor. It was observed that, within experimental error, the different templates used

showed a similar behavior. This supports the finding reported in the previous chapters that

suggested that the identity and quantity of the dopant used to prepare the particles does not affect

the polarity of the interior of the templates. Based on this, placement of the probes in different

positions inside the particles by using different dopants was not expected, which was confirmed

by the similarity of variation in the energy transfer when the concentration of the acceptor was

increased. The small variations in the efficiency reported at low concentrations of C153 could be

a consequence of the difference in the number of templates in the suspensions, since, as

explained earlier, despite having the same concentration in mass, the concentration of the

particles may differ due to their different mass per particle.

To understand this process better we tried fitting the energy transfer data to a linear Stern-

Volmer and to the modified Stern-Volmer plots, without success (see Appendix B). Obviously, a

linear plot was not expected to be satisfactory since we expect to have environments of different

polarities inside the templates. Surprisingly, the modified plot, which has been used to represent

energy transfer processes with probes placed in two or more different environments, did not offer

consistent results. It was observed that, in general, there was an improvement in the fitting









parameters with the modified plot, but the results were not good enough to accept them as

accurate. The fact that the data could not be fitted to these plots may be an indication that these

particles represent a complex system, in which there is a distribution of environments with

slightly different polarities, as a consequence of the degree of penetration of the solvent inside

the templates.

To have an idea what percentage of the molecules of acceptor were excited through the

energy transfer process, we excited the acceptor directly and compared the intensity of its

emission to that of the acceptor excited through the donor. This is shown in Table 6-2, together

with the emission maximum of C153 in each case. The intensity of the emission was determined

to be very similar in both cases, which indicated that most of the acceptor molecules were in

proximity to pyrene. This is assuming that the quantum yield of C153 is the same in the different

environments it is placed in the particles. One important control performed was exciting a

sample of templates used to sequester only C153 at the pyrene excitation maximum. It was

observed that the intensity was 2% of the intensity when excited at the C153 excitation

maximum, confirming that the acceptor is not directly excited in the RET experiments. The

maximum in the emission of the acceptor at higher concentration shifted to the red when excited

at its excitation maximum, which indicated that the excess acceptor molecules were

accommodated in the external part of the particles, close to the solvent or even in direct contact

with it, which explained the increase in the turbidity of the suspension at high acceptor

concentrations.

6.2 Energy Transfer in the Core-Shell Particles

6.2.1 Strategy

In this case, based on the results obtained with the templates, we performed a simple

experiment with the core-shell particles. We used a set of nanoparticles prepared with pyrene









during their synthesis (system NPY, Chapter 4), to sequester C153 at two different

concentrations. Interestingly, the low intensity-excimer peak observed during the

characterization reported on Chapter 4 was not observed when performing these studies. This

could be a consequence that the experiments reported in this section were performed a few

months after the particles were prepared. It is possible that during this time a small amount of

dye had leached from the particles or that the interior of the particles became more rigid. In any

case, the fluorescence spectra of the donor reported here were taken the same day to be able to

compare them. The concentration of the nanoparticles used was lower than the one used for the

templates (approximately 0.2 mg/mL), since the core-shell particles scattered more light due to

their larger size. As a consequence, the concentration of C153 loaded in the suspensions was

lowered as well.

In the second system investigated, we used C153 as the donor and DCM as the acceptor.

DCM has been used in micelles, observing that it tends to position itself in their periphery, near

the hydrophilic solvent.118' 119 Preliminary results with our templates and core-shell particles

showed that, similarly to C153, this probe was preferentially placed in contact with the solvent.

Based on this and the overlap between the emission of C153 and the excitation spectrum of

DCM (Figure 6-4 A), we believe both dyes could be sequestered by the particles to be placed in

the hydrated shell, close enough to observe an energy transfer between the dyes. This differs

from the previous system in which the most internalized molecules of C153 were expected to

take part of the energy transfer.

6.2.2 Results

6.2.2.1 Pyrene-C153 pair

Based on the fact that the Stern-Volmer plots did not offer any insights on the energy

transfer between the probes, we simplified the experiments performed with the core-shell









particles. Two different concentrations of the acceptor were used. At low concentrations (0.14

[M) no acceptor emission was detected, although a low pyrene emission quenching of

approximately 5% was observed (Figure 6-2 A, Table 6-3). When the acceptor concentration was

increased to 0.7 aM, the intensity of the donor emission diminished to approximately 68% (32%

efficiency). At the same time, the emission of the acceptor was clearly observed as a broad peak

centered at 527 nm, similar to the templates. When the acceptor was excited through pyrene (339

nm) the intensity of the emission of C 153 was 66% of the one when excited directly (420 nm), as

shown in Figure 6-2 B. This indicates that under these conditions, approximately 66% of the

acceptor molecules interact with the donor, despite the presence of the shell where an important

fraction of the acceptor molecules were supposed to be adsorbed. It is important to note that the

concentration of pyrene is unknown since it was included during the synthesis of the particles

and some of the probe was lost during their purification. The excitation of the acceptor through

interaction with the donor was confirmed by analysis of the C153 excitation spectrum, which

showed that a fraction of the dye molecules were excited by the donor. This confirms the RET

process in the dye-doped particles suggested by the pyrene emission decrease. This is obviously

not a very efficient system, but our goal was to investigate the possibility of observing an energy

transfer process based on the placement of the dyes inside the particles as a means to study the

compartmentalization of the interior of the core-shell particles.

One important detail in the data obtained is the difference in the maximum in the emission

spectrum of the acceptor after excitation through energy transfer from pyrene, (527 nm) and

when directly excited (534 nm), similarly to what was observed with the templates. Since the

emission of the dye shifts to the blue when the polarity of its surrounding decreases, the values

observed indicated that the acceptor molecules emitting through RET, which are the ones in









contact with the donor, are the ones better protected from the solvent. Base on these maxima, the

shift observed is similar to a change in the solvent where the dye is dissolved from methanol to

n-propanol. We interpreted this as an indication of the protection of the dye from water, meaning

that the acceptor molecules in contact with the donor are the ones more internalized, as expected

based on the polarity reported by pyrene.

It is clear from these results that only a fraction of the two different probes are in proximity

to each other, due to their differences in polarity and possibilities of noncovalent interactions

with the shell, as represented in Figure 6-3. We believe that in between the external part of the

shell filled with solvent molecules and the "dry core" region, where solvent molecules are not

present, there is an intermediate polarity region with a low concentration of solvent molecules.

This is the volume that the probes performing the energy transfer occupied which, based on the

C153 emission, is similar in polarity to n-propanol.

6.2.2.2 C153-DCM pair

Figure 6-4 B shows the emission of the nanoparticles loaded with C153 at a concentration

of 0.7 tM before and after sequestering the acceptor. The typical emission profile for the

sequestered donor when excited at 420 nm was observed in the absence of DCM, with emission

maximum at 539 nm, indicating, as reported on Chapter 5, that the dye was in proximity to the

solvent. After addition of a small concentration of the acceptor, the emission of the donor is

distorted. A broader emission, with a small peak at the red end of the spectrum was observed, as

shown in Figure 6-4 B, which suggested that the probes could communicate. Contrary to what

we expected, the emission of the donor did not decrease, but slightly increased. We believe this

is due to the emission from DCM excited by the C153 molecules, which involves a broad peak

starting at 500 nm. This may produce an increase in the apparent donor emission, if RET was

actually happening. When the concentration of DCM was increased to a similar value than the









donor, it was observed that the C153 emission was quenched to 83% of the value for the DCM-

free particles, confirming the energy transfer between the dyes. This indicated a transfer

efficiency of only 17%, but due to the superposition of the emission of the dyes, which may have

increased the apparent emission of the donor, this would be the lowest efficiency possible under

our experimental conditions. We then analyzed the excitation spectrum of the particles loaded

with equimolar amounts of the probes with the emission fixed at the acceptor maximum. The

graph obtained looked like a combination of the donor and acceptor excitation profiles, as shown

in Figure 6-4 C, which confirmed the RET process. To investigate how the emission of the

acceptor affects the intensity of the emission of the donor, we excited the C153-loaded particles

at the donor and acceptor excitation maximum, 420 and 475 nm, respectively. Figure 6-4 D

shows that the emission of DCM involves a broad peak, which overlaps with the emission of

C153. This may explain the increase in the intensity of the donor after DCM addition. We

studied the transfer process more deeply by measuring the fluorescence lifetime of the donor, as

shown in the next section.

6.3 Energy Transfer in the Core-Shell Particles Studied by Fluorescence Lifetime
Measurements

6.3.1 Strategy

Due to the overlap of the emission of the donor and acceptor, we decided to study the

changes in the lifetime of C153 after addition of DCM. In order to observe changes in the

fluorescence lifetime of the donor as a consequence of energy transfer to an acceptor, the process

has to involve a collisional encounter of the probes. The acceptor has to find the donor during the

time this is excited for the energy transfer to occur and quench the emission of the donor. Had

the probes made a complex before excitation, no changes in the fluorescence lifetime would be

observed. The studies presented in this section would give very important information regarding









the kind of process leading to the resonance energy transfer observed by steady state

fluorescence spectroscopy.

6.3.2 Results

The fluorescence lifetime of C 153 has been reported by different groups in different

conditions. In general, in pure solvents a single exponential is found to describe the fluorophore,

since it is well known that C153 has a single excited state responsible for its fluorescence.

Typical lifetime values are 4.3 ns in methanol,121 5.2 ns in toluene and 6.5 ns in acetonitrile.114

The value in micelles and vesicles usually involves 2 components, one in the 0.7-3.0 ns range

and the other similar to the one observed in pure solvents, approximately 4-5 ns.122 In our results,

a single exponential fit to the lifetime decay data was discarded due to the extremely high x2

value, which indicated that it did not represent the system properly (see Appendix C). A double

exponential was found to give a reasonable fit, together with the typical value about 4.5 ns, we

found a longer decay (Table 6-4). The relatively high x2 value obtained may be due to the

possibility of having the probes distributed over the different regions inside the particles, in

environments with slightly different polarities. Trying to fit this data to a set of just two different

families of fluorophores is, obviously, only an approximation to the real case. When a small

amount of acceptor was added to the C153-doped particles both lifetime values decreased,

although the longer one is affected the most. The contribution of both species remains the same

within experimental error. This produced a decrease in the average lifetime, which is the most

accurate comparison in systems with a nonsingle exponential decay, confirming the energy

transfer between the donor and acceptor. Interestingly, these results contradicted those obtained

with steady state fluorescence, in which a slight increase in the intensity of the emission of the

donor was observed. This may be a consequence of the overlap of the emission of the donor and

acceptor, as explained above. When an equimolar amount of DCM was added, the average









lifetime decreased to approximately 54% of its original value, giving an efficiency value of 46%,

higher than the one reported in the steady state studies, as shown in Table 6-4. These results

showed that the energy transfer between the probes is a collisional quenching of the fluorescence

lifetime of the donor. This indicates that the pair of probes was not associated within the core-

shell particles and more importantly, that the probes can diffuse inside the particles.

6.4 Use of the Fluorescence Lifetime of a Probe Chemically Attached to Study
Microenvironments Inside the Core-Shell Particles

6.4.1 Study of the Fluorescence of Dansyl Chloride Attached to the Periphery of the
Particles

6.4.1.1 Strategy

The system NPDC1-2A was prepared by reacting APSDC1 with the surface of the

templates without a coating, while the particles 2B were prepared by attaching the APSDC1

complex after the growth of a thick shell, as explained on Chapter 4. The idea behind this

approach was to study the solvation of a probe whose position has been fixed on the particles by

fluorescence lifetime measurements, as explained in Chapter 2.

Different groups have studied the distribution of dansyl in different environments when

incorporated in different matrices by the determination of its fluorescence lifetime.76' 77, 123 Two

decays have been identified matching the excited states of the dansyl moiety. A short decay time,

corresponding to the TICT state, where the dimethyl amino group and the naphthyl group are in

a nonplanar state in a hydrophilic environment, and a longer decay time, corresponding to the

probe in a coplanar state in a hydrophobic environment.77

6.4.1.2 Results

We start our discussion for the results obtained with the steady state fluorescence data.

First of all, the sample NPDC1 2A showed a weaker emission for a similar concentration of

particles than the system 2B, approximately 0.2 mg/mL. We believe this is due to the fact that









the sample 2B offered more groups for the APSDC1 complex to react with, due to the higher

surface area and the presence of internal silanol groups of the shell that can react with the dye.

Table 6-5 shows the emission maxima detected for each system when excited at 340 nm. As

explained earlier; this case is more complex that the ones previously studied due to the fact that

the emission of dansyl chloride is affected by both the polarity and rigidity of the media. The

system without a shell presented a red shifted signal (max= 523 nm) which may mean that the

dye is placed either in a polar or flexible environment. As expected, opposite results were

observed for the system 2B due to the presence of a shell which is a more rigid but hydrated

matrix (max= 510 nm). Based on the maximum detected in the emission of the dye, it seemed

like the polarity of the medium was lower compared to the system 2A or that the rigidity of the

environment surrounding the probe was higher due to the presence of the shell, although, at this

point, it was difficult to have a clear picture of which factor was more important. To clarify this

point we turned our attention to the study of the fluorescence lifetime of dansyl chloride attached

to the nanoparticles, as discussed below.

The fluorescence lifetime data showed interesting results on the factors affecting the

emission ofDC1. Both samples decays were fit to a double exponential as shown in the Table 6-

5. As explained for the antenna system formed with C153 and DCM, this is only an

approximation to the real case. In both cases, the major contribution detected corresponded to the

nonTICT-state, suggesting that the APS complex penetrated both, the templates and core-shell

particles to an important extent before reacting. Surprisingly, the system APSDC1 2A showed a

lower contribution of the fluorophore emitting through the TICT-state ('= 2.32 ns, 13%),

indicating that approximately 13% of the detected fluorophores were in contact with the solvent.

This suggests that most of the dye ended up connected to the interior of the templates where









access for the solvent molecules is more restricted, showing a major contribution from the

fluorophore emitting through the non-TICT state (z= 16.4 ns, 87%). System APSDC1 2B, on the

other hand, showed a shorter mean lifetime, indicating, that in average, the probe sensed a more

polar or less rigid environment. The higher contribution from the fluorophore in contact with the

solvent, emitting through the TICT-state ('= 2.93 ns, 25%) may be due to the presence of silanol

groups in the interior of the shell that the probe could have reacted with. This placed the dye not

only in the outer part of the shell, but in its interior, which, as determined previously, is

extensively hydrated, giving the probe a more polar environment. As a consequence, the

contribution of the protected dye ('= 15.13 ns, 75.4%), diminished.

6.4.2 Preliminary Studies on the Energy Transfer between Dansyl Chloride Attached to
the Periphery of the Particles and Sequestered DCM

6.4.2.1 Strategy

Based on the results from the lifetime fluorescence studies reported on the previous

section, we decided to study the energy transfer between dansyl choride attached to the particles,

partially expose to the solvent, and DCM, which was observed to sense a polar environment

when sequestered by the core-shell particles. The results showed in the following section are

preliminary since they were done in the search for the best conditions to obtain an energy

transfer between dansyl and DCM. As a consequence, each system was used under different

conditions and the results would be examined briefly. The most important reason to show them is

to state the feasibility of using the core-shell particles with a fixed probe in the design of antenna

systems.

6.4.2.2 Results

For the nanocapsule system APSDC1 2A, the acceptor was added as a solid and the

suspension sonicated for 30 minutes and stirred overnight. After this, the samples were filtered to









remove any excess dye. When the original nanoparticle suspension was excited at 370 nm the

typical emission from DC1 on the surface of the particles was obtained, as shown in Figure 6-5

A. When excited at 475 nm (the acceptor excitation maximum) no fluorescence was detected,

confirming that the donor did not interfere with the emission of the acceptor. Excitation of the

loaded particles at 475 nm showed the typical DCM fluorescence, with a maximum in its

emission of approximately 620 nm, an indication that the dye was in placed close to the solvent.

Excitation at 370 nm produced a broader peak, similar to a combination of the DCM and DCl

emissions, confirming that there is an energy transfer process between the dyes. We believe this

system could be improved to obtain a higher efficiency since both dyes would be placed in the

same volume in the particles, in contact with the solvated shell. Unfortunately, the emission of

the dyes was recorded at different conditions, so the intensities could not be compared to use

them to calculate the efficiency of the process.

The system APSDC1 2B was studied in a different way. DCM was loaded in the particles

from a methanolic solution to reach two different concentrations, and the emission of the donor

was analyzed. Figure 6-5 B shows a comparison of the emission before and after addition of the

acceptor. It was observed that the intensity of the emission of DC1 changed after incorporation of

DCM in the system, giving a transfer efficiency of 36%. It was not clear the reason for not

observing a peak for the emission DCM, but a signal dominated by scattering, as marked in

Figure 6-5 B. Interestingly, the decrease in the emission intensity of DC1 was very similar for

both concentrations of the acceptor added, with only a 3% increment in the transfer efficiency

observed after a 5-fold increased in the concentration of DCM. This is in agreement with the

results of the fluorescence lifetime studies for dansyl attached to the particles, where it was









determine that only 25% of the probe was exposed to the solvent. Based on this, the efficiency of

the transfer is not expected to increase after further addition of DCM.

6.5 Conclusions

This chapter showed results on studies concerning the possibility of designing antenna

systems to study the compartmentalization of the templates and core-shell particles synthesized

as reported in the previous chapters. The observation of energy transfer between pyrene and

C153 sequestered by the templates showed that the probes share a volume in their interior. Since

the Forster distance necessary for this pair of dyes to observe energy transfer is only 3.4 nm we

can conclude that the dyes were very close to each other. A more detailed study of this

phenomenon was tried by fitting the data to models representing systems that can accommodate

the probes in different environments, without success. We believe this is an indication of the

complex system these templates and particles represent. A similar study was performed for the

core-shell particles, using particles prepared by incorporation of pyrene during the synthesis. In

this case, a more simple approach was used, observing that energy transfer was possible with this

system as well. After establishing that energy transfer between different probes was possible in

the templates and core-shell particles we studied a system with the two probes expected to reside

in the shell, in a volume close to the periphery of the particles. The pair C153-DCM was used,

observing similar results to those obtained with the previous pair. In this case, in addition to the

steady state fluorescence studies, a study on the fluorescence lifetime of the donor was

performed to gain a better understanding of the energy transfer process observed. It was

determined that this was a collisional quenching of the emission of the donor by the acceptor,

which means that the probes were not associated but diffusing inside the particles. The efficiency

of the transfer was determined to be relatively high despite the use of low loadings, probably due

to the close proximity of the probes. These are promising findings that suggest that these









particles can be used as sensor for different quenchers. Finally, we studied the fluorescence

lifetime of dansyl chloride chemically connected to the templates after their preparation and to

the core-shell particles after the growth of a thick shell. The results showed that the templates

offered pores in their interior where the probe can react, which hides the dye from the solvent, as

observed with pyrene. The dye attached to the core-shell particles on the other hand, showed in

its fluorescence lifetime that a higher fraction of the fluorophore was exposed to water,

confirming the hydration of the shell, where most of the dansyl derivative was supposed to have

reacted. These dye-coated particles were used for preliminary studies to determine the feasibility

to use them for the design of antenna systems, observing that the probe could communicate with

DCM although more studies are necessary.


Table 6.1 Changes in the intensity of the first peak of sequestered pyrene emission (371 nm)
after addition of C 153
[C153] EBB EBC HXB HXC
I1 relative E I1 relative E I1 relative E I1 relative E
0 1 0 1 0 1 0 1 0
0.122 0.86 14 0.92 8 0.89 11 0.91 9
0.306 0.76 24 0.74 16 0.84 16 0.79 21
0.610 0.60 40 0.66 24 0.66 34 0.73 27
1.22 0.64 36 0.61 39 0.63 37 0.63 37
1.83 0.51 49 0.52 48 0.53 47 0.53 47
2.45 0.44 56 0.44 56 0.45 55 0.45 55
I1 relative calculated as the ratio between the intensity of the peak at 371 nm in the emission of
pyrene before and after addition of C153, after background correction. The efficiency is obtained
as (1- Il relative) X 100.


Table 6.2 Emission maximum of sequestered C153 in pyrene-doped particles and comparison
of the intensity of its emission obtained when excited at 339 and 420 nm
[C153]/ tM Xmax, exc 339 nm )max, exc 420 nm i339/1420 a
1.22 527 527 1.03
1.83 528 531 0.97
2.45 529 536 0.87
a values compared after background correction









Table 6-3 Steady state fluorescence data for the pyrene-C153 encapsulated pair (Xex= 339 nm).
[C153]/ pM 1371 relativee E" ,b C153 (527nm)
0 1.0 no
0.14 0.95 5 no
0.70 0.66 44 yes
a371 values compared after background correction,b E values calculated from (1- (1371 DA/I371 D)) X
100, D= donor, DA= donor in the presence of acceptor


Table 6-4 Steady state and fluorescence lifetime measurements results for the energy transfer
between C153 and DCM.
[DCM]/ pM Tr (ns) bl T2 (ns) b2 x2 Tmean(ns)a E(%)b
0 4.62 0.89 28.32 0.11 2.26 7.23
0.14 4.06 0.89 19.51 0.11 1.89 5.76 20
0.70 3.74 0.87 15.56 0.13 1.78 3.88 46
a average lifetime values obtained by Tmean= YbiTi, b efficiency obtained by fluorescence lifetime
measurements E= (1-(TDA/'D))O100, D= donor,DA= donor in the presence of acceptor

Table 6-5 Steady state and fluorescence lifetime measurements results for templates and
particles coated with dansyl chloride
I'qmax, exc 340 nm T 1 (%) T 2 (%) Tmean 2
APSDC1 2Aa 523 2.32 (13.1) 16.4 (86.9) 14.56 1.05
APSDC1 2Bb 510 2.93 (24.6) 15.1 (75.4) 12.11 1.76
T mean= Ibizi, Xexc 375 nm, aem 525 nm, b em 510 nm










B OE
1.OE+07


SwavelnEth (nm)


8.0E+06

S6.OE+06

S4.


2.0E+06

O.OE+00
360 410 460 510
D wvkethUnm)
D


300 350 400 450 500 0 0.5 1 1.5 2 2.5
wavelength (mn) [C13]



Figure 6-1.Energy transfer between pyrene and C153 in templates. A) Overlap of the emission of
sequestered pyrene and excitation of sequestered C153, B) changes in the
fluorescence spectrum of pyrene (1 pM) after addition of C153 C) typical excitation
spectra of sequestered C153 (0.6 pM, x,, 520 nm) in pyrene-loaded templates (1
pM), D) efficiency of the energy transfer observed for templates EBB (diamonds),
EBC (squares), HxB (triangles), HxC (circles).










6
3.0*10


6


12.0*10

1.0*10



0 -


400 500
wavelength (rnm)


400 500
wavelength (mm)


Figure 6-2.Energy transfer between pyrene and C153 in core-shell particles. A) Emission spectra
of encapsulated pyrene before (solid line), and after sequestration of to different
concentrations of C153 (0.14 aM-empty circles and 0.7 gM-empty triangles); B)
comparison between the emission of the doubled-doped nanoparticles excited at the
excitation maxima of pyrene (solid line) and C153 (empty circles).


/hydrated shell


C153
intermediate
polarity zone

pyrene
pyrene


"dry" core


Figure 6-3. Schematic representation of the energy transfer between probes based on their
placement inside the nanoparticles. The particles are divided in zones with different
polarities based on the penetration of the solvent.


6
2.0*10




6
1.0*10





0











1

0, o.8

S0.6

0.4

s .


wavelength (am)


400 500
wavelength (nm)


6
2.0*10





1.0*10





0

D


6
1.0*10 -


5
6.0*10



2.0*10


600
wavelength (am)


Figure 6-4.Energy transfer between C153 and DCM. A) Overlap between C153 emission (solid
line) and DCM excitation (empty circles); B) emission spectra of nanoparticles doped
with sequestered C153 (0.7 aM, empty circles), and after sequestering DCM (0.14
aM-empty triangles and 0.7 VM-solid line); C) excitation spectra for sequestered
C153 (0.7 WM, empty circles, emission 539 nm), sequestered DCM (0.7 aM, solid
line, emission 615 nm) and DCM in nanoparticles with both dyes sequestered (0.7
aM each, empty triangles, emission 610 nm); D) comparison between the emission of
the doubled-doped nanoparticles (0.7 aM each dye) excited at the excitation
maximum of C153 (420 nm, solid line) and DCM (475 nm, empty circles).


600
wavelength (am)


I 70
700










A B
1.0 a8.0'

0.8 partile scatterig
S- \ 6.0*10
1 0.6

S0-4- 4.0*10


2.0*10

500 600 400 500 600
wavelength (am) wavelength (urn)




Figure 6-5.Energy transfer between DC1 and DCM. A) Nanoparticles system APSDC1-2A, Xexc
370 nm (blue line), x,, 450 nm (black line); solid DCM added: X,, 370 nm (green
line), X,, 450 nm (red line). B) Nanoparticles system APSDC1 2B: X,, 340 nm (black
line); after sequestration of DCM: 0.5 vtM (red line), 2.5 vJM (green line) and 2.5 vIM
(X,, 450 nm).









CHAPTER 7
CONCLUSIONS AND SUGGESTIONS FOR FUTURE WORK

7.1 Conclusions

The work presented here took a different direction from the ideas originally discussed

when this part of the project was started. We believed it was necessary to take the time to

perform some basic studies to confirm that the ideas we based our understanding of the particles

we obtained were correct. The main goal of the project was to prepare nanoparticles with a

hydrophobic core and a hydrophilic shell that could sequester drugs and pesticides to be used as

a fast response in cases of intoxication. The particles were synthesized using a mixed surfactant

system. OTMS and t80 were mixed to obtain spherical particles, called templates. The use of t80

was necessary since it was observed that OTMS by itself tends to form large aggregates in

aqueous suspensions. The mixed-surfactant system was then fortified by hydrolyzing and

condensing the polar groups of OTMS. Subsequently, these particles were coated with a silicate

precursor. The product obtained was believed to be made of a hydrophobic core, mainly formed

by the tail of the surfactants; and a hydrophilic shell, formed by the silica coating. It was shown

that these particles could be prepared by the use of a microemulsion as well, obtained by the

addition of a small hydrophobic molecule (termed a dopant in this work) to the binary

amphiphile system. This was expected to tune the polarity of the core by trapping the dopant in

the interior of the particles by growing a shell on their periphery. The particles obtained by

previous work in the group were used successfully to sequester small organic molecules from

aqueous suspensions. This was interpreted as an indication of the core of the particles attracting

nonpolar molecules due to its low polarity, although no data was available to confirm this. At

this point, it was decided that it was necessary to elucidate if the driving force for this process









was the one proposed. As a consequence, a significant amount of the work presented here dealt

with the characterization of the templates and core-shell particles synthesized in our work.

The first step taken was to improve the yield of templates and particles obtained in our

synthetic approach, since with our previous methodology the low yield of particles obtained

made their purification and characterization tedious. This was done by adding an ionic surfactant

and reducing the amount of t80. The purification was changed from dialyzing the suspensions

obtained, to using small membranes with a diameter cut-off of 25 nm. This separated the

templates, with diameters of above 60 nm, from the micelles formed by the excess surfactant, of

sizes below 10 nm. After this, characterization of the templates and core-shell particles became

easier. DLS showed that the size of the templates could be tuned by varying the loading of the

surfactants. TEM images showed that the particles consisted of a very light material that was

difficult to image due to the low contrast in the microscope. It was observed on the atomic force

microscope that after deposition on a substrate they flattened considerably, forming disk-like

species. We believe this happens as the solvent evaporates, leaving the external volume of the

particles, which is hydrated in solution, empty. Similar studies were performed with the dopant-

filled templates. DLS results showed that the size of the doped-templates could be tuned by the

loading of the dopant. The AFM images showed that these templates flattened as well after

drying. Surprisingly, contrary to the templates prepared only with the amphiphiles, folded and

bent particles were observed, indicating that they were made of a material of higher flexibility

than the dopant-free templates. Preliminary studies on the strength of these templates were

performed by calcinating them at 600 OC. It was observed that the templates broke down, leaving

a fiber-like material.









The templates coated with the silicate shell were characterized similarly to the templates.

DLS data showed that particles with different shell thicknesses could be prepared. AFM images

showed that the coated particles were not as flexible as the templates and that the thickness of the

coating determines the extent the particles flattened on the substrate. This indicated that the shell

formed fortified the particles producing more rigid structures. Based on some simple

calculations, it was determined that the volume of the templates was dramatically reduced after

drying. This was suppressed when a thin coating was formed on the particles, but it increased

again with a thicker shell. We believe these were indications of the hydration of the core and the

shell. Calcination of these core-shell particles showed that, after coating, the particles did not

break down after heating to 600 OC, confirming that the siloxane shell fortified the particles.

We then turned our attention to the study of the sequestration of different probes by the

templates and core-shell particles. We designed an experimental approach that allow the study of

species trapped by the particles, contrary to typical studies that only quantify the uptake by

differences in the UV absorption of the non sequestered probes. Pyrene and C153, two

hydrophobic fluorescence dyes, were used in these studies as sensors of polarity and rigidity.

Pyrene fluorescence showed that sequestration by the templates placed it in a nonpolar

environment, protected from the solvent. Interestingly, the polarity reported by the probe did not

resemble the one from the dopant. This is believed to indicate that a significant fraction of the

dopant molecules were removed during purification, producing a more flexible material, which

correlates with the AFM data that showed that these particles could be folded without breaking.

A nonpolar environment was reported by pyrene for the core-shell particles as well, confirming

that the particles internalized this hydrophobic probe. C153 on the other hand, showed that the

final position of the probe inside the particles depended on its chemical structure, since due to









hydrogen bonding interaction, this molecule was trapped on the shell of the particles sensing a

polar environment. These results indicated that we can identify the position of the dyes based on

their emission profiles and that, by knowing the chemical structure of the probe to be

sequestered, we could predict its final position inside the core-shell particles.

As a next logical step, the information obtained with the fluorescent probes was used to

design antenna systems that could potentially be used as sensors for different purposes. We

accomplish this by using two pair of probes in which the emission of one of them overlapped

with the excitation spectrum of the other. The probes were chosen not only because of their

fluorescence response but for the position in the particles they were expected to place themselves

based on their structures. The pair pyrene-C 153 showed a transfer efficiency of approximately

50%, indicating that, despite sensing different polarities in average, an important number of the

probes can communicate and approximate each other, probably in an area of intermediate

polarity inside the particles. The third probe used, DCM, known to place itself on the periphery

of micelles was selected to act as an acceptor with C153. Contrary to what we expected, the

efficiency of the energy transfer obtained by steady state fluorescence spectroscopy was only

17%. We studied this process by analysis of the changes in the fluorescence lifetime of the

donor, obtaining an efficiency of 46%. This proved that the energy transfer was a collisional

process, showing that the probes were not associated in the particles, but that they approximate

each other after the donor has been excited. In the final part of this work, we studied templates

and particles reacted with dansyl chloride derivatized with silanol groups, which was expected to

place the fluorescence probe in their periphery. Study of the lifetime of the probe indicated that

most of the dye was somehow protected from the solvent, with a low percentage of fluorophores

in a polar, solvated environment. Apparently, a significant fraction of the probe molecules









diffused to the internal part of the shell before reacting. Finally, we showed preliminary results

on the used of these dye-coated-particles for energy transfer to a sequestered probe, which

broadens the choices for the design of particles for different applications.

We believed this work has confirmed that the templates and core-shell particles can be

obtained by the methodology develop by our group. The particles obtained are stable and they

can be characterized by different techniques. The ability to control their sizes allows tuning the

characteristics of the particles accordingly to the application in mind, makes them a versatile

system. The use of fluorescent probes was proved to be a successful methodology for the

characterization of the compartmentalization of particulate systems. The fluorescent probes used

in this study confirmed that molecules of low polarity can be sequestered by the nanoparticles

and that they were protected from the media. It was observed that the chemical structure of the

molecules sequestered determine their final position inside the particles. These findings were

used to place different probes inside the particles and study whether they were in contact.

Fluorescence resonance energy transfer was observed between three different pairs of probes.

Fluorescence lifetime measurements of the donor of one of these pairs demonstrated that the

energy transfer was produced by a collisional process. By attaching a dye to the templates and

core-shell particles we studied the degree of solvation of a probe reacted at the end of the

synthesis of the particles. It was observed that the reaction did not take place only on the most

external parts of the particles, but that the probe diffused to the interior, due to its low polarity,

before reacting.

7.2 Suggestions for Future Work

All the information gathered in the different steps of this work suggests a few important

ideas. First, the templates are strong, stable structures that can be used for different purposes,

especially for sequestering processes. They do not need to be coated. We believe the coating









should be performed in two cases. First, when a probe or dopant is included in the formulation

used for the synthesis of the particles. With this, the possibility of removing these species during

purification is lower, due to the isolation of the interior from the media by the shell. The second

case involves the attachment of different probes to the particles. Due to the well-known

derivatization strategies for silicate matrices, it would be useful to have a silica coating that could

be derivatized by different routes. Doping the particles with fluorescent dyes makes it possible to

use them as fluorescent labels for different applications, including biologically relevant systems.

We have demonstrated that the particles can be prepared with dyes that absorb and emit at

different wavelengths. This could be used to tune the response of the particles based on the needs

of particular applications. Finally, doping the particles with different dyes at the same time is a

suitable strategy for the design of sensors, in which a nonpolar dye can be used as a reference

signal, and a dye placed on the periphery of the particles, potentially able to interact with species

approaching the surface of the particles, could be used as a detection response.










APPENDIX A
APSDCL COMPLEX NMR


H
S/ H2 -OCH3
a i OCH3
aO=S-- ,' l S i j I cH
b d H2 H2 OCH3
g i
C
H3C NCH3
e


Figure A-1.


Chemical structure of the APSDC1 complex.


b d
a h





'^ .. ...... ulo W UJ



80 70 60 50 40 30 20 10 00
ppm (tl)


Figure A-2. NMR spectrum of the APSDC1 complex.












APPENDIX B
MODIFIED STERN-VOLMER PLOTS FOR EB AND HX TEMPLATES


EbB Modified Stern-Volmer


6

S4
Sy= 0.684x+1.732
2 R'= 0.977

0
0 2 4 6 8 10
1/[C153]


14 EbC Modified Stern-Vomer
12
10


S8 /
1.371x+ 0.922
4 / R2 = 0.962

O
-------------------
0
0 2 4 6 8 10
1/[C153]


10 HxB Modified Stern-Volmer


y= 0.727x+ 2.081
'6 R2=0.913
uL 4

2

0

0 5 10 15
1/[C153]


15 HxC Modified Stern-Volmer


L1
u 10

5 y= 0.898x+ 1.819
R2 = 0.993

0

0 5 1/[C153]10 15


Modified Ster-Volmer plots for hexadecane-doped templates.


Figure B-1.









APPENDIX C
FLUORESCENCE LIFETIME FITS FOR C153-DCM FRET

Table C-l Fluorescence lifetime fits for C153-doped templates used for FRET with DCM.


Ti (ns) s) 2s) T3 (ns)
C153 0.7 pM 5.25 6.01
4.62 (88.8%) 28.32 (11.1%) 2.26
0.588 (-12.6%) 4.26 (98.4%) 16.7 (14.16%) 1.09
C153 0.7 tM + 0.14 ,M DCM 4.58 5.26
4.06 (88.6%) 19.51 (11.4%) 1.89
0.393 (-6.6%) 3.79 (90.7%) 13.8 (16%) 1.18
C153 0.7 ,tM + 0.7 ,tM DCM 4.36 5.6
3.74 (87.2%) 15.56 (12.9%) 1.78
0.53 (-17.9%) 3.37 (98%) 11.1 (19.9%) 1.38









LIST OF REFERENCES

(1) Barrat, G. Cellular and Molecular Life Sciences 2003, 60, 21-37.

(2) Sukhorukov, G.; Fery, A.; Brumen, M.; Mohwald, H. PCCP 2004, 6, 4078-4089.

(3) Hof, F.; Craig, S.; Nuckolls, C.; Rebek Jr, J. Angew. Chem. Int. Ed2002, 41, 1488-1508.

(4) Djatali, R.; Samsom, J.; Matsui, H. J. Am. Chem. Soc. 2004, 126, 7935-7939.

(5) Kaanumalle, L.; Nithyanandhan, J.; Pattabiraman, M.; Jayaraman, N.; Ramamurthy, V. J
Am. Chem. Soc. 2004, 126, 8999-9006.

(6) Liu, S.; Weaver, J.; Save, M.; Armes, P. Langmuir 2002, 18, 8350-8357.

(7) Shchukin, D.; Sukhorukov, G. Adv. Mater. 2004, 16, 671-682.

(8) Lee, M.; Jan, C.; Ryu, J. J. Am. Chem. Soc. 2004, 126, 8082-8083.

(9) Wang, X. Q.; Dai, J. D.; Chen, Z.; Zhang, T.; Xia, G. M.; Nagai, T.; Zhang, Q. Journal of
Controlled Release 2004, 79, 421-429.

(10) Cai, T.; Hu, Z.; Ponder, B.; John, J.; Moro, D. Macromolecules 2003, 36, 6559-6564.

(11) Haag, R. Angew. Chem. Int. Ed 2004, 43, 278-282.

(12) Twyman, L.; King, A.; Martin, I. Chem. Soc. Rev. 2002, 31, 69-82.

(13) Underhill, R.; Jovanovic, A.; Carino, S.; Varshney, M.; Shah, D.; Dennis, D.; Morey, T.;
Duran, R. Chem. Mater. 2002, 14, 4919-4925.

(14) Zarur, A.; Ying, J. Nature 2000, 302, 65-67.

(15) Kim, M.; Seok, S.; Ahn, B.; Koo, S.; Paik, S. Journal ofSol-Gel Science and Technology
2003, 27, 355-361.

(16) Prichanont, S.; Leak, D. J.; Stuckey, D. C. Colloids and Surfaces A 2000, 166, 177-186.

(17) Allen, C.; Maysinger, D.; Eisenberg, A. Colloid and Surfaces B 1999, 16, 3-27.

(18) Miller, D.; Kabanov, A. Colloid and Surfaces B 1999, 16, 321-330.

(19) Kuang, M.; Duan, H.; Wang, J.; Chen, D.; Jiang, M. Chem. Commun. 2003, 4, 496-497.

(20) Sanji, T.; Kakatsuka, Y.; Ohnishi, S.; Sakurai, H. Macromolecules 2000, 33, 8524-8526.

(21) Hecht, S.; Frechet, J. Angew. Chem. Int. Ed 2001, 40, 74-91.










(22) Frechet, J. PNAS 2002, 99, 4782-4787.

(23) Cameron, C.; Gorman, C. Adv. Funct. Mater. 2002, 12, 17-20.

(24) Nakajima, R.; Tsuruta, M.; Higuchi, M.; Yamamoto, K. J. Am. Chem. Soc. 2004, 126,
1630-1631.

(25) Liu, L.; Breslow, R. J Am. Chem. Soc. 2004, 126, 12110-12111.

(26) Vogtle, F.; Gestermann, S.; Hesse, R.; Schwierz, H.; Windisch, B. Prog. Polym. Sci.
2000, 25, 987-1041.

(27) Aulenta, F.; Hayes, W.; Rannard, S. European Polymer Journal 2003, 39, 1741-1771.

(28) Couvreur, P.; Barrat, G.; Fattal, E.; Legrand, P.; Vauthier, C. Critical Reviews in
Therapeutic drug Carriers Systems 2002, 19, 99-134.

(29) Newkome, G.; Zhongqi, Y.; Baker, G.; Gupta, V. J Org. Chem. 1985, 50, 2003-2004.

(30) Morgan, M.; Carnahan, M.; Immons, C.; Ribeiro, A.; Finkelstein, S.; Lee, S.; Grinstaff,
M. J Am. Chem. Soc. 2003, 125, 15485-15489.

(31) Furuta, P.; Brooks, J.; Thompson, M.; Frechet, J. J. Am. Chem. Soc. 2003, 125, 13165-
13172.

(32) Furuta, P.; Frechet, J. J. Am. Chem. Soc. 2003, 125, 13173-13181.

(33) Sunder, A.; Kramer, M.; Hanselmann, R.; Mulhaupt, R.; Frey, H. Angew. Chem. Int. Ed
1999, 38, 3552-3555.

(34) Schappaper, M.; Deffieux, A. Polymer 2004, 45, 4633-4639.

(35) Chen, G.; Guan, Z. J. Am. Chem. Soc. 2004, 126, 2662-2663.

(36) Niu, Y.; Crooks, M. Chem. Mater. 2003, 15, 3462-3467.

(37) Soppimath, K.; Aminabhavi, T.; Kulkarni, A.; Rudzinski, W. J ControlledRelease 2001,
70, 1-20.

(38) Skirtach, A.; Antipov, A.; Shchukin, D.; Sukhorukov, G. Langmuir 2004, 20, 6988-6992.

(39) Peyratout, C.; Mohwald, H.; Dahne, L. Angew. Chem. Int. Ed2003, 115, 1722-1726.

(40) Khopade, A.; Caruso, F. Nano Letters 2002, 2, 415-418.









(41) Sun, B.; Chiu, D. Langmuir 2004, 20, 4614-4620.

(42) Wang, Y.; Caruso, F. Chem. Commun. 2004, 13, 1528-1529.

(43) Balabushevich, N. G.; Larionova, N. I. Biochemistry (Moscow) 2004, 69, 930-936.

(44) Turner, J.; Wooley, K. Nano Letters 2004, 4, 683-688.

(45) Hu, Y.; Jian, X.; Ding, Y.; Chen, Q.; Yang, C. Adv. Mater. 2004, 16, 933-937.

(46) Graff, A.; Winterhalter, M.; Meier, W. Langmuir 2001, 17, 919-923.

(47) Lawson, G.; Lee, Y.; Singh, A. Langmuir 2003, 19, 6401-6407.

(48) Bolinger, P.; Stamou, D.; Vogel, H. J. Am. Chem. Soc. 2004, 126, 8594-8595.

(49) Yang, X.; Zhang, Y. Langmuir 2004, 20, 6071-6073.

(50) Wang, H.; Chen, P.; Zheng, X. J. Mater. Chem. 2004, 14, 1648-1652.

(51) Jungmann, N.; Schmidt, M.; Ebenhoch, J.; Weis, J.; Maskos, M. Angew. Chem. Int. Ed
2003, 42, 1714-1717.

(52) Jungmann, N.; Schmidt, M.; Maskos, M.; Weis, J.; Ebenhoch, J. Macromolecules 2002,
35, 6851-6857.

(53) Kumar, R.; Chen, M.; Parmar, V.; Samuelson, M.; Kumar, J.; Nicolosi, R.; Yoganathan,
S.; Watterson, A. J. Am. Chem. Soc. 2004, 126, 10640-10644.

(54) Panyam, J.; Williams, D.; Dash, A.; Leslie-Pelecky, D.; Labhasetwar, V. Journal of
Pharmaceutical Sciences 2004, 37, 1804-1814.

(55) Bae, Y.; Fukushima, S.; Harada, A.; Kataoka, K. Angew. Chem. Int. Ed 2003, 42, 4640-
4643


(56) Chu, B. Laser Light Scattering 1991, Academic Press, Inc, San Diego.

(57) Precision Deconvolve32 User Manual 2003,

(58) Johnson, C., Gabriel, D. Laser Light Scattering 1981, Dover Publications, Inc., New
York.

(59) Nanoparticle Technology for Drug Delivery 2006, Taylor and Francis group, New York.

(60) Scanning Probe Microscopy Training Book 2000, Digital Instruments,










(61) Ebenstein, Y.; Nahum, E.; Banin, U. Nano Letters 2002, 2, 945-950.

(62) Zeta Potential Analyzer, Instruction Manual 2008, Brookhaven Instrument Corp, New
York.

(63) Jesionowski, T. Colloid and Surfaces A 2003, 222, 87-94.

(64) Wilhelm, M.; Zhao, C.-L.; Wang, Y.; Renliang Xu, a.; Winnik, M. A. Macromolecules
1991, 24, 1033-1040.

(65) Lakowicz, J. R. Principles ofFluorescence Spectroscopy 1999, Kluwer
Academic/Plenum Publishers, New York.

(66) Vanderkooi, J. M.; Callis, J. B. Biochemistry 1974, 13, 4000-4006.

(67) Kalyanasundaram, K.; Thomas, J. K. J. Am. Chem. Soc. 1977, 99, 2039-2044.

(68) Grant, C. D.; DeRitter, M. R.; Steege, K. E.; Fadeeva, T. A.; Castner Jr, E. W. Langmuir
2005, 21, 1745-1752.

(69) Kumbhakar, M.; Nath, S.; Mukherjee, T.; Pal, H. J. Chem. Phys. 2004, 121, 6026-6033.

(70) Stathatos, E.; Lianos, P.; Stangar, U. L.; Orel, B. Chem. Phys. Lett. 2001, 345, 381-385.

(71) Horng, M. L.; Gardecki, J. A.; Papazyan, A.; Maroncelli, M. J. Phys. Chem. 1995, 17311.

(72) Li, X.; McCarroll, M.; Kohli, P. Langmuir 2006, 22, 8615-8617.

(73) Wu, P.; Brand, L. Analytical Biochemistry 1994, 218, 1-13.

(74) Weber, G. Biochem. J. 1951, 51, 155-167.

(75) Li, Y.-H.; Chan, L.-M.; Tyer, M.; Moody, R. T.; Himel, C. M.; Hercules, D. J. Am. Chem. Soc.
1975, 97, 3118-3126.

(76) Ding, L.; Fang, Y.; Jiang, L.; Gao, L.; Yin, X. Thin SolidFilms 2005, 478, 318-325.

(77) Montalti, M.; Prodi, L.; Zaccheroni, N.; Falini, G. J. Am. Chem. Soc. 2002, 124, 13540-
13546.

(78) Leezenberg, P. B.; Frank, C. V. Chem. Mater. 1995, 7, 1784.

(79) Huang, Q. R.; Frank, C. V.; Mecerreyes, D.; Wolksen, W.; Miller, R. D. Chem. Mater.
2005, 17, 1521-1528.









(80) Fontaine, P.; Goldmann, M.; Rondelez, F. Langmuir 1999, 1348-1352.

(81) Linden, M.; Slotte, J. P.; Rosenholm, J. B. Langmuir 1996, 4449-4454.

(82) Vidon, S.; Leblanc, R. M. J. Phys. Chem. B 1998, 1279-1286.

(83) Jovanovic, A. V.; Underhill, R. S.; Bucholz, T. L.; Duran, R. S. Chem. Mater. 2005, 17,
3375-3383.

(84) Heydenreich, A. V.; Westmeier, R.; Pedersen, N.; Poulsen, H. S.; Kristensen, H. G.
International Journal ofPharmaceutics 2003, 83-87.

(85) Shinoda, K.; Arai, J. J. Phys. Chem. 1964, 68, 3485-3490.

(86) Gonzales-Perez, A.; Ruso, J. M.; Prieto, G.; Sarmiento, F. Langmuir 2004, 20, 2512-
2514.

(87) Morey, T. E.; Varshney, M.; Flint, J. A.; Seubert, C. N.; Smith, W. B.; Bjoraker, D. G.;
Shah, D. O.; Dennis, D. M. Journal ofNanoparticle Research 2004, 150-170.

(88) Katagiri, K.; Caruso, F. Macromolecules 2004, 37, 9947-9953.

(89) Bourgeat-Lami, E.; Tissot, I.; Lefebvre, F. Macromolecules 2002, 35, 6185-6191.

(90) ZetaPlus Potential Analyzer Instruccion Manual 2003,

(91) Acharya, K. R.; Bhattacharya, S. C.; Moulik, S. T. Journal of Photochemistry and
Photobiology A: Chemistry 1997, 109, 29-34.

(92) Huang, H.; Kowalewski, T.; Wooley, K. L. J Polym. Sci. Part A: Polym. Chem. 2003,
41, 1659-1668.

(93) Khopade, A. J.; Caruso, F. Nano Letters 2002, 2, 415-418.

(94) Sarid, D. in Scanning Force Microscopy, Lapp, M., Stark, H., Eds. 1991,

(95) Peyratout, C. S.; Mohwald, H.; Dahne, L. Adv. Mater. 2003, 15, 1722-1726.

(96) Brinker, C. J.; Schere, G. W. 1990, Academic Press, Inc., San Diego.

(97) Ni, K. F.; Shan, G. R.; Weng, Z. X.; Sheibath-Othman, N.; Fevotte, G.; Lefebvre, F.;
Bourgeat-Lami, E. Macromolecules 2005, 38, 7321-7329.

(98) Tissot, I.; Novat, C.; Lefebvre, F.; Bourgeat-Lami, E. Macromolecules 2001, 34, 5737-
5739.









(99) Jan, J.-S.; Lee, S.; Carr, C. S.; Shantz, D. F. Chem. Mater. 2005, 17, 4310-4317.

(100) Jovanovic, A. V.; Underhill, R. S.; Bucholz, T. L.; Duran, R. S. Chem. Mater. 2005, 17,
3375-3383.

(101) Huang, H.; Kowalewski, T.; Wooley, K. L. J Polym. Sci. Part A: Polym. Chem 2003,
41, 1659-1668.

(102) Koh, K.; Ohno, K.; Tsuji, Y.; Fukuda, T. Angew. Chem. Int. Ed. 2003, 42, 4194-4197.

(103) Meier, M. A. R.; Gohy J.-F.; Fustin, C.-A.; Schubert, U. S. J. Am. Chem. Soc. 2004,
11517-11521.

(104) Aathimanikandan, S. V.; Savariar, E. N.; Thayumanavan, S. J. Am. Chem. Soc. 2005,
127, 14922-14929.

(105) Tamano, K.; Imae, T.; Yusa, S.; Shimada, Y. J Phys. Chem. B. 2005, 1226-1230.

(106) Tedeschi, C.; Mohwald, H.; Kirstein, S. J. Am. Chem. Soc. 2001, 123, 954-960.

(107) Hugerth, A. M. Journal ofPharmaceutical Sciences 2001, 90, 1665-1677.

(108) Pandey S.; Redden R. A.; Hendricks A. E.; Fletcher K. A.; Palmer, C. P. J. Colloid
Interface Sci. 2003, 262,

(109) Danko, M.; Hrdlovic, P.; Borsig, E. J. Macrol. Sci.-Pure Appl. Chem., A 2001, 38, 476-
486.

(110) Jang, J.; Oh, J. H. Adv. Mater. 2003, 15, 977-980.

(111) Ferrer, M.; Lianos, P. Langmuir 1996, 12, 5620-5624.

(112) Hof, M.; Lianos, P. Langmuir 1997, 13, 290-294.

(113) Seth, D.; Chakraborty, A.; Setua, P.; Sarkar, N. Langmuir 2006, 22, 7768-7775.

(114) Kr6licki, R.; Jarzeba, W.; Mostafavi, M.; Lampre, I. J Phys. Chem. A 2002, 106, 1708-
1713.

(115) Joncheray, T. J.; Audebert, P.; Schwartz, E.; Jovanovic, A. V.; Ishaq, O.; Chavez, J. L.;
Pansu, R.; Duran, R. S. Langmuir 2006, 22, 8684-8689.

(116) Kumbhakar, M.; Nath, S.; Mukherjee, T.; Pal, H. J. Chem. Phys. 2004, 121, 6026-6033.

(117) Tummino, P. J.; Gafni, A. Biophysical Journal 1993, 64, 1580-1587.









(118) Pal, S. K.; Sukul, D.; Mandal, D.; Sen, S.; Bhattacharya, S. C. ChemicalPhysics Letters
2000, 327, 91-96.

(119) Pal, S. K.; Mandal, D.; Sukul, D.; Bhattacharya, S. C. ChemicalPhysics Letters 1999,
312, 178-184.

(120) Cheon, C. H.; Jo, S.-H.; Kim, K.; Jin, J.-I.; Shin, H.-W.; Kim, Y.-R. Macromolecules
2005, 38, 6336-6345.

(121) Boens, N. e. a. Anal. Chem. 2007, 79, 2137-2149.

(122) Sahu, K.; Ghosh, S.; Mondal, S. K.; Ghosh, B. C.; Sen, P.; Roy, D.; Bhattacharya, S. C.
J. Chem. Phys. 2006, 125, 44714-44721.

(123) Annaka, M.; Noda, H.; Motokawa, R.; Kakahira, T. Polymer 2001, 42, 9887-9894.









BIOGRAPHICAL SKETCH

Jorge L. Chavez, the second son of Manuel Chavez and Gladys Benavides, was born in

Lima, "the Pearl of the Pacific", on December, 1976. He spent his entire childhood in the same

city, trying to stay alive despite the craziness that ruled the place. In his early years he spent most

of his time playing soccer on the streets. During high school, he developed an interest for science

and rock music but, unfortunately, he discovered he lacked any kind of skills to play musical

instruments, so he decided to go to college. After graduating from high school, he joined the

Pontificia Universidad Cat6lica del Peru to study chemistry. There he obtained his B. Sc., taught

several classes and laboratories and performed research as an undergraduate in Dr. Javier

Nakamatsu's Laboratory. He moved to Gainesville in August 2002 to start working on his Ph. D.

in Dr. Randy Duran's research group. In 2006 he performed his most adventurous experiment,

marrying Sarah, with whom he had a daughter called Victoria Helena. What the future has for

him is uncertain now, but surely, it will be something exciting and, hopefully, he will be able to

make a living out of it.





PAGE 1

SYNTHESIS AND CHARACTERIZATION OF CORE-SHELL NANOPARTICLES, AND A STUDY OF THEIR USE AS ENCAPSULATION AGENTS By JORGE L. CHVEZ BENAVIDES A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008 1

PAGE 2

2008 Jorge L. Chvez Benavides 2

PAGE 3

To Gladys, Sarah and Victoria 3

PAGE 4

ACKNOWLEDGMENTS Many people helped me through the years this work was performed. The most important people supporting me, despite the distance, have always been my parents and brother. Their advice and faith have been crucial to fi nish this work without losing my mind. Working in the Department of Chemistry has been an experience that changed my life in different levels. I grew as a scientist and more importantly, as a person. In one of those long days, when the pleasure of working was fading, I had the opportunity to meet the perfect match for my unconventional soul. The lady that be came my wife and, later, the mother of my wonderful daughter, has been a sour ce of strength and inspiration. I had the chance to meet wonderful people in the Duran group, to all of them my respect for their work and gratitude for th eir help with my research and the time shared, especially those long hours in front of an instrument or a comput er that, without them, would have been tedious and difficult to stand. All these years in Gainesville have been a great opportuni ty to learn about how human beings from different parts of th e world are so similar in some as pects and so different in others. I have been fortunate to meet very interesting people, who have opened their hearts and minds to me, without asking for anything. I h ope my friendship has been for them as important as theirs to me. I thank Dr. Randy Duran for giving the chance to join his group and work in a project that I enjoyed very much. I thank the staff in the Depa rtment of Chemistry at UF, especially in the Butler Polymer Laboratories and the Graduate Progr am Office, for their help in so many things. A significant part of the work presented here was carried out in the Materials Chemistry Characterization Laboratory. I tha nk Dr. John R. Reynolds and Dr. Kirk Schanze for allow me to 4

PAGE 5

use their facilities. Finally, I thank my advisory committee for taking the time to review this contribution and for their comments and suggestions. 5

PAGE 6

TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ........10 LIST OF FIGURES.......................................................................................................................12 ABSTRACT...................................................................................................................................16 LIST OF ABBREVIATIONS........................................................................................................18 CHAPTER 1 REVIEW OF SOME OF THE MO ST IMPORTANT STRATEGIES OF ENCAPSULATION IN NANOPARTICLES........................................................................20 1.1 Introduction..................................................................................................................2 0 1.2 Different Strategies for Encapsulation.........................................................................21 1.2.1 Dendritic Architectures.....................................................................................21 1.2.2 Core-Shell and Hollow Particles......................................................................25 1.2.3 Microemulsions................................................................................................31 1.3 Conclusions..................................................................................................................36 2. METHODS AND EXPERIMENTAL PROCEDURES.........................................................39 2.1 Introduction..................................................................................................................3 9 2.2 Techniques and Methods.............................................................................................39 2.2.1 Dynamic Light Scattering.................................................................................39 2.2.1.1 Theory........................................................................................................39 2.2.1.2 Experimental settings................................................................................44 2.2.2 Transmission Electron Microscopy..................................................................45 2.2.2.1 Theory........................................................................................................45 2.2.2.2 Experimental settings................................................................................46 2.2.3 Atomic Force Microscopy................................................................................47 2.2.3.1 Theory........................................................................................................47 2.2.3.2 Experimental settings................................................................................48 2.2.4 -Potential.........................................................................................................50 2.2.4.1 Theory........................................................................................................50 2.2.4.2 Experimental settings................................................................................52 2.2.5 Study of the Polarity and Viscosity of a Matrix by Steady State Fluorescence Spectroscopy..............................................................................53 2.2.5.1 Use of pyrene as polarity and rigidity probe.............................................53 2.2.5.2 Use of C153 as a polarity probe................................................................54 2.2.6 Time-Domain Lifetim e Measurements............................................................55 6

PAGE 7

2.2.6.1 Fluorescence resonance energy transfer studied by time-resolved fluorescence...............................................................................................56 2.2.6.2 Time-correlated single photon counting (TCSPC)....................................59 2.2.6.3 Determination of the degree of so lvation of dansyl chloride by fluorescence lifetime measurements..........................................................60 2.2.6.4 Experimental settings................................................................................61 2.3 Experimental Procedures.............................................................................................61 2.3.1 Preparation of Templates..................................................................................61 2.3.2 Synthesis of Core-Shell Nanoparticles.............................................................62 2.3.3 Coating of the Particles with 3-aminopropylsilane..........................................63 2.3.4 Synthesis of the Complex between 3-aminopropylsilane and Dansyl Chloride...............................................................................................................63 2.3.5 Coating of the Templates and Core-Shell Particles with APSDCl Complex....................................................................................................................63 2.3.6 Loading the Dyes to Template s and Core-Shell Particles................................64 2.3.7 Fluorescence Images.........................................................................................64 2.3.8 Steady State Fluorescence Spectroscopy..........................................................64 3 SYNTHESIS AND CHARACTERIZATION OF SURFACTANT-BASED TEMPLATES FOR THE FORMATI ON OF CORE-SHELL PARTICLES.........................68 3.1 Preparation of Surfactant-Based Templates.................................................................68 3.1.1 Strategy.............................................................................................................68 3.1.2 Results..............................................................................................................68 3.1.2.1 Size analysis..............................................................................................70 3.1.2.2 TEM analysis.............................................................................................74 3.1.2.3 AFM analysis.............................................................................................76 3.2 Preparation of Doped Templates.................................................................................78 3.2.1 Results..............................................................................................................79 3.2.1.1 Size analysis of ethyl butyrate-doped templates........................................79 3.2.1.2 AFM analysis of ethyl butyrate-doped templates......................................79 3.2.1.3 Size analysis of hexade cane-doped templates...........................................80 3.2.1.4 AFM analysis of hexadecane-doped templates.........................................81 3.3 Other Systems Studied.................................................................................................81 3.3.1 Use of a Cationic Surfactant in the Synthesis of the templates........................81 3.4 Preliminary Studies of the St rength of the Templates.................................................82 3.4.1 Strategy.............................................................................................................82 3.4.2 Results..............................................................................................................82 3.4.2.1 Calcination of dopant-free templates.........................................................82 3.4.2.2 Calcination of ethyl but yrate-doped templates..........................................83 3.5 Conclusions..................................................................................................................83 4 TEMPLATE-BASED SYNTHESIS OF CORE-SHELL PARTICLES.................................97 4.1 Synthesis of Core-Shell Particles.................................................................................97 4.1.1 Strategy.............................................................................................................97 4.1.2 Results............................................................................................................100 7

PAGE 8

4.1.2.1 Size analysis of the core-shell particles...................................................101 4.1.2.2 TEM analysis of the core-shell particles.................................................101 4.1.2.3 AFM analysis of core-shell particles.......................................................102 4.1.2.4 Analysis of the dimensions of the templates and particles in suspension and after drying.....................................................................103 4.2 Derivatization of the Surf ace of the Nanoparticles....................................................105 4.2.1 Derivatization with 3-aminopropyltrimethoxysilane......................................105 4.2.2 Derivatization with dansyl chloride-APS complex........................................106 4.3 Preliminary Results on Other Systems Studied.........................................................108 4.3.1 Preparation of Large Templates and Co re-Shell Particles for Calcination Studies...............................................................................................................108 4.4 Conclusions................................................................................................................110 5 SEQUESTRATION AND ENCAPSULATION OF FLUORESCENT DYES IN TEMPLATES AND CORE-SHELL PARTICLES..............................................................120 5.1 Sequestration of Dyes by the Templates....................................................................120 5.1.1 Strategy...........................................................................................................120 5.1.2 Use of Pyrene as a Polarity and Rigidity Probe..............................................122 5.1.3 Results............................................................................................................123 5.1.3.1 Determination of the optimal conditions for the study of the fluorescence of sequestered pyrene.........................................................123 5.1.3.2 Sequestration of pyrene by ethyl butyrate-doped templates....................124 5.1.3.3 Sequestration of pyrene by he xadecane-doped templates.......................125 5.1.3.4 Study on the retention of the dye after solvent exchange........................126 5.2 Sequestration of Dyes by the D opant-Free Core-Shell Particles...............................126 5.2.1 Sequestration of Pyrene by Dopant-Free Core-Shell Particles.......................127 5.2.2 Sequestration of C153 by Dopant -Free Core-Shell Particles.........................127 5.2.2.1 Strategy....................................................................................................127 5.2.2.2 Results.....................................................................................................128 5.3 Encapsulation of Dyes by the Templa tes and Core-Shell Particles...........................129 5.3.1 Strategy...........................................................................................................129 5.3.2 Results............................................................................................................129 5.3.2.1 Encapsulation of pyrene in the temp lates and core-shell particles..........129 5.3.2.2 Encapsulation of C153 in the temp lates and core-shell particles............130 5.4 Conclusions................................................................................................................131 6 STUDY OF THE COMPARTMENTALI ZATION OF THE TEMPLATES AND CORE-SHELL PARTICLES THROUGH EN ERGY TRANSFER BETWEEN DYES.....137 6.1 Energy Transfer in the Templates..............................................................................137 6.1.1 Strategy...........................................................................................................137 6.1.2 Results............................................................................................................139 6.2 Energy Transfer in the Core-Shell Particles..............................................................141 6.2.1 Strategy...........................................................................................................141 6.2.2 Results............................................................................................................142 6.2.2.1 Pyrene-C153 pair.....................................................................................142 8

PAGE 9

6.2.2.2 C153-DCM pair.......................................................................................144 6.3 Energy Transfer in the Core-Shell Pa rticles Studied by Fluorescence Lifetime Measurements............................................................................................................145 6.3.1 Strategy...........................................................................................................145 6.3.2 Results............................................................................................................146 6.4 Use of the Fluorescence Lifetime of a Probe Chemically Attached to Study Microenvironments Inside the Core-Shell Particles..................................................147 6.4.1 Study of the Fluorescence of Dansyl Chloride Attached to the Periphery of the Particles................................................................................................147 6.4.1.1 Strategy....................................................................................................147 6.4.1.2 Results.....................................................................................................147 6.4.2 Preliminary Studies on the Energy Transfer between Dansyl Chloride Attached to the Periphery of th e Particles and Sequestered DCM...................149 6.4.2.1 Strategy....................................................................................................149 6.4.2.2 Results.....................................................................................................149 6.5 Conclusions................................................................................................................151 7 CONCLUSIONS AND SUGGESTI ONS FOR FUTURE WORK......................................158 7.1 Conclusions................................................................................................................158 7.2 Suggestions for Future Work.....................................................................................162 APPENDIX A APSDCl COMPLEX NMR..................................................................................................164 B MODIFIED STERN-VOLMER PLOT S FOR EB AND HX TEMPLATES......................165 C FLUORESCENCE LIFETIME FITS FOR C153-DCM FRET...........................................166 LIST OF REFERENCES.............................................................................................................167 BIOGRAPHICAL SKETCH.......................................................................................................174 9

PAGE 10

LIST OF TABLES Table page 3-1 Formulations used for the preparation of templates and characterization results obtained by DLS and TEM................................................................................................85 3-2 Size analysis results obtained by DLS and AFM for templates D2 and D3......................85 3-3 Formulations used for the formation of ethyl butyrate-doped templates, and size analysis results obtained by DLS.......................................................................................85 3-4 Topographic information for ethyl butyrate -doped templates obtained by AFM. DLS data is shown for comparison............................................................................................86 3-5 Formulations used for the formation of hexa decane-filled templates and size analysis results obtained by DLS and AFM....................................................................................86 4-1 Formulations used for the synthesis of templates and corresponding core-shell particles............................................................................................................................111 4-2 DLS characterization results for templates a nd the corresponding core-shell particles..112 4-3 AFM characterization results for templates (1 ) and core-shell particles (2) prepared as indicated in Table 4-1..................................................................................................112 4-4 Volume of templates and coated pa rticles based on DLS and AFM data.......................112 4-5 Formulation and size characte rization results for APS-coat ed nanoparticles (NPAPS)..112 4-6 Formulations used for coating of the particles with APS-DCl and DLS characterization results.....................................................................................................112 5-1 I3/I1 values for pyrene in ethyl butyrate (E B) and hexadecane (Hx) and EB and Hxdoped templates...............................................................................................................1 32 5-2 Formulation used in the synthesis of core-s hell particles for seque stration studies and size characterization re sults obtained by DLS.................................................................132 5-3 Fluorescence data for pyrene and C1 53 sequestered by dopant-free core-shell particles............................................................................................................................133 5-4 Steady State Fluorescence data for templa tes and core-shell particles used to encapsulate pyrene (NPY) and C153 (NCU)...................................................................133 6.1 Changes in the intensity of the first peak of sequestered pyrene emission (371 nm) after addition of C153......................................................................................................152 10

PAGE 11

6.2 Emission maximum of sequestered C153 in pyrene-doped particles and comparison of the intensity of its emission obt ained when excited at 339 and 420 nm.....................152 6-3 Steady state fluorescence data for the pyrene-C153 encapsulated pair ( ex= 339 nm)....153 6-4 Steady state and fluorescence lifetime measur ements results for the energy transfer between C153 and DCM..................................................................................................153 6-5 Steady state and fluorescence lifetime measurements results for templates and particles coated with dansyl chloride...............................................................................153 C-1 Fluorescence lifetime fits for C153-dope d templates used for FRET with DCM...........166 11

PAGE 12

LIST OF FIGURES Figure page 1-1 Synthesis of dendrimers for encapsulation. Divergent (A, D) and convergent (B, C) approaches to the synthesis of dendrimers for the encapsulation of active cores or active molecules............................................................................................................... ..37 1-2 General approach of the synthesis of hollo w particles using a removable core for the encapsulation of active species..........................................................................................37 1-3 Synthesis of hollow particles based on the use of lipids. These particles can encapsulate active molecules in th e water pool in their interior........................................38 1-4 Particle synthesis by micr oemulsion polymerization........................................................38 2-1 Possibilities for the interaction of a laser beam with a liquid sample. In a homogeneous medium, waves of the scattere d light interact dest ructively, producing no scattering (A); while in a heterogeneous medium scattering is produced (B)..............65 2-2 Comparison of the correlation function of two set of templates with different diameters as determined by DLS. The diam eters obtained were 47 nm (solid line) and 72 nm (empty circles)..................................................................................................66 2-3 A simple representation of the basic concept of a transmission electron microscope operating in the bright field mode......................................................................................66 2-4 A simple representation of the basic se tup of the atomic force microscope......................67 2-5 Data analysis of a Z-axis calibrati on grid performed in contact mode..............................67 3-1 Schematic representation of the s ynthesis of core-shell particles......................................86 3-2 Chemical structures for surfactants, am phiphiles and dopants used for this study...........87 3-3 Ideal schematic for the hydrolysis and c ondensation of the methoxy groups in OTMS at basic conditions............................................................................................................ ..88 3-4 Size distribution obtained by DLS for a typical set of templates A) before and B) after purification.................................................................................................................88 3-5 -potential variation as a f unction of pH for non pure (cir cles) and purified (squares) templates...................................................................................................................... ......89 3-6 Variation of the normalized correlation function obtained by diluting a suspension of particles to reach different light scattering intensities. A) templates, intensity: 3.5 x 105 kcps-purple line, 2.5 x105 kcps-green line, 1.1 x 105 kcps-red line, 0.5 x 105 12

PAGE 13

kcps-blue line, and B) SO/t80 micelles, intensity: 3.0 x 105 kcps-purple line, 2.0 x 105 kcps-green line, 1.0 x 105 kcps-red line, 0.5 x 105 kcps-blue line...............................89 3-7 TEM images for templates: a) B2, b) C2, c) D2. Scale bars 200 nm................................90 3-8 TEM images showing what are believed to be templates deposited with their longest axis perpendicular to the su rface of the substrate (indicat ed by the red boxes). Scale bars 500 nm........................................................................................................................90 3-9 AFM tapping mode images for samples A) D2 and B) D3. On the bottom part of each image the section analysis is shown..........................................................................91 3-10 AFM images and section analysis for temp lates A) EBB, B) EBC, C) EBE and D) EBF. The arrows indicate what are believed to be folded templates.................................92 3-11 AFM images and section analysis for temp lates A) HxB, B) HxC, C) HxE and D) HxF. The arrows indicate what are believed to be folded templates.................................93 3-12 AFM image of two set of templates pr epared with DTAB replacing SO..........................94 3-13 TEM images of dopant-free templates before (A) and after (B, C, D) calcination. Scale bars 200 nm.............................................................................................................. 94 3-14 AFM images of dopant-free templates before (A) and after (B, C, D) calcination...........95 3-15 TEM images of ethyl butyrate-doped temp lates before (A) and after (B, C, D) calcination. Scale bars 200 nm...........................................................................................95 3-16 AFM images of dopant-free templates be fore (A) and after (B) calcination.....................96 4-1 Reaction scheme for the hydrolysis of TMOS shown for the first A) and second B) hydrolysis products. The reaction of these species with the silanol groups on the surface of the templates is shown, as we ll as the final product after complete hydrolysis at neutral pH...................................................................................................113 4-2 Crosslinking resulted when templates were not diluted before the addition of TMOS. Scale bar 500 nm..............................................................................................................1 14 4-3 TEM and AFM height images of NEB A) te mplates and B) core-shell particles. Scale bars in all images 500 nm.......................................................................................114 4-4 TEM and AFM height images of NCU A) te mplates and B) core-shell particles. Scale bars in all images 300 nm.......................................................................................115 4-5 TEM and AFM height images of NPY A) te mplates and B) core-shell particles. Scale bars in all images 500 nm.......................................................................................116 4-6 Attachment of APS to core-shell nanoparticles...............................................................116 13

PAGE 14

4-7 AFM images for the templates and APS-co ated nanoparticles reported on Table 4-4....117 4-8 Comparison of -potential results before (circles) and after (squares) derivatization with APS: A) NPAPS; and APS-DCl: B) NPDCl-1A, C) NPDCl-1B and D) NPDCl2A (squares) and NPDCl2B (triangles)...........................................................................117 4-9 Coupling between dansyl chloride and APS....................................................................118 4-10 TEM and AFM images of N-benzyl maleimide-doped templates (A and C) and coreshell particles (B and D). Scale bars 500 nm...................................................................118 4-11 TEM images of N-benzyl maleimide-doped templates (A and B) and core-shell particles (C and D) after calcination at 600 C. Scale bars 500 nm.................................119 5-1 Fluorescence response detected after excitation of a dye-free nanocapsule suspension. The peak observed is due to scattering of light by the particles, excitation wavelength 270 nm..........................................................................................................133 5-2 Chemical structures of the dyes used in this study..........................................................133 5-3 Emission of a 1 M pyrene solution in A) hexade cane and B) ethyl butyrate................134 5-4 Determination of conditions to study the polarity of templates with pyrene. A) Pyrene titration for hexadecane-doped temp lates, numbers in the box indicate pyrene concentration, B) plot of th e ratio of the intensities of the third and first emission peaks of pyrene (I3/I1) at different dye concentrations....................................................134 5-5 Representation of the removal of dopant molecules during purification of the templates. The positions preferred by pyrene molecules when sequestered by a suspension of templates, is shown as well.......................................................................135 5-6 Emission of pyrene loaded in A) dopant-fr ee templates and B) EB-doped templates, before (black line) and after (grey line) dialysis..............................................................135 5-7 Representation of the positions preferred by pyrene and C153 molecules when sequestered by core-shell particles...................................................................................135 5-8 Normalized emission of pyrene encapsulate d in templates and in the corresponding core-shell particles (blue) for the NPY system................................................................136 5-9 Fluorescence microscopy images for C153 encapsulated in core-shell particles, scale bar 10 m. Image was obtained in black and white, color was added for clarity B) Normalized emission of C153 encapsulated in templates (red line) and in the corresponding core-shell particles (blue line)..................................................................136 6-1 Energy transfer between pyrene and C153 in templates. A) Overlap of the emission of sequestered pyrene and excitation of sequestered C153, B) changes in the fluorescence spectrum of pyrene (1 M) after addition of C 153 C) typical excitation 14

PAGE 15

spectra of sequestered C153 (0.6 M, exc 520 nm) in pyrene-loaded templates (1 M), D) effciency of the energy transfer observed for templates EBB (diamonds), EBC (squares), HxB (triangles), HxC (circles)...............................................................154 6-2 Energy transfer between pyrene and C153 in core-shell particles. A) Emission spectra of encapsulated pyrene before (so lid line), and after sequestration of to different concentrations of C153 (0.14 M-empty circles and 0.7 M-empty triangles); B) comparison between the em ission of the doubled-doped nanoparticles excited at the excitation maxima of pyrene (solid line) and C153 (empty circles).........155 6-3 Schematic representation of the energy transfer between probes based on their placement inside the nanoparticle s. The particles are divided in zones with different polarities based on the penetration of the solvent............................................................155 6-4 Energy transfer between C153 and DCM. A) Overlap between C153 emission (solid line) and DCM excitation (empty circles); B) emission spectra of nanoparticles doped with sequestered C153 (0.7 M, empty circles), and after sequestering DCM (0.14 M-empty triangles and 0.7 M-solid line); C) excitation spectra for sequestered C153 (0.7 M, empty circles, emission 539 nm), sequestered DCM (0.7 M, solid line, emission 615 nm) and DC M in nanoparticles with both dyes sequestered (0.7 M each, empty triangles, emission 610 nm); D) comparison between the emission of the doubled-doped nanoparticles (0.7 M each dye) excited at the excitation maximum of C153 (420 nm, solid line) and DCM (475 nm, empty circles). 156 6-5 Energy transfer between DCl and DCM. A) Nanoparticles system APSDCl-2A, exc 370 nm (blue line), exc 450 nm (black line); solid DCM added: exc 370 nm (green line), exc 450 nm (red line). B) Nanoparticles system APSDCl 2B: exc 340 nm (black line); after se questration of DCM: 0.5 M (red line), 2.5 M (green line) and 2.5 M ( exc 450 nm)........................................................................................................157 A-1 Chemical structure of the APS-DCl complex..................................................................164 A-2 NMR spectrum of th e APS-DCl complex.......................................................................164 B-1 Modified Stern-Volmer plots for hexadecane-doped templates......................................165 15

PAGE 16

Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy SYNTHESIS AND CHARACTERIZATION OF CORE-SHELL NANOPARTICLES, AND A STUDY OF THEIR USE AS ENCAPSULATION AGENTS By Jorge L. Chvez Benavides May 2008 Chair: Randolph S. Duran Major: Chemistry A mixed-surfactant system was used for the pr eparation of templates for the synthesis of core-shell particles. One of the amphiphiles used contained silanol groups that were condensed to fortify the templates. Subsequent ly, a silica coating was chemically attached, obtaining core-shell particles. It was observed by diffe rent techniques that the templates and core-shell particles were spherical and their diameter could be tuned by varying component composition. Atomic force microscopy studies after deposition on a hydrophilic substrate showed that these particles flattened dramatically, due to the formation of a flexible matrix in their interior. Formation of the shell increased their rigidity, preventing flattening to a certain degree. An experimental approach based on the use of hydrophobic fluorescent dyes wa s designed to study the polarity and rigidity of the templates and particles. By using a set of two different probes we found that the polarity of the interior of the particles d ecreased while its rigidity increas ed after coating. We used the fluorescence response of these dyes to st udy the uptake of hydrophobic compounds by the particles. It was determined that, based on their chemical structures, the sequestered probes were placed in different parts of the particles, se nsing different environments. We used this information to study the compartmentalization of the interior of the te mplates and particles by using probes that could undergo resonance energy transfer if they were placed in close 16

PAGE 17

environments. These studies suggested that the se questered probes could interact with each other producing a dynamic quenching of the donor. Fi nally, we covalently attached another hydrophobic fluorescent probe to the pa rticles to study the level of in teraction with the solvent, observing that the probe preferre d to react in the interior pores of the particles to reduce interaction with the polar media. 17

PAGE 18

LIST OF ABBREVIATIONS 1-d 1-dodecene AFM Atomic force microscopy APS 3-aminopropylsilane C153 Coumarin 153 d Diameter DCl Dansyl chloride DCM 4(dicyanomethylene)-2-methyl-6-( p-dimethylaminostyryl)-4H-pyran DHR Diameter to height ratio DLS Dynamic light scattering DTAB Dodecyltrimethylammonium bromide EB Ethyl butyrate FRET Fluorescence resonance energy transfer h Height kcps Kilocounts per second NCU Coumarin 153-doped particles NEB Ethyl butyrate-doped particles np Nanoparticles NPY Pyrene-doped particles OTMS Octadecyltrimethoxysilane RT Room temperature rpm Revolutions per minute SO Sodium octanoate ST Shell thickness t80 Poly(oxyethylene)[20]-sorbitan monooleate 18

PAGE 19

TCSPC Time-correlated single photon counting technique TEM Transmission electron microscopy TMOS Tetramethoxysilane UA Uranyl acetate 19

PAGE 20

CHAPTER 1 REVIEW OF SOME OF THE MOST IMPORTANT STRATEGIES OF ENCAPSULATION IN NANOPARTICLES 1.1 Introduction Encapsulation of a variety of species has been the focus of attent ion of many research groups in the last several years.1-3 Reaction vessels,4-8 drug delivery9-11 and catalysis12 are three examples of many potential app lications that encapsulati on systems offer. Different methodologies for the synthesis of such systems ha ve been presented in the literature, including microemulsions,13-16 polymers,17-20 and dendrimers.21-27 Especially important in this field is the study of core-shell particles. These systems of fer the possibility of combining two different environments in the same particle, which enhances the solubility of the encapsulated molecule if the polarity of the core is the opposite to that of the shell. Moreover, protection of the encapsulated agent from environmental changes that it could experience, li ke pH, ionic strength, and the like is another important property that can be accomplished by these architectures. When such encapsulating agents fit into the nanometer scale, new a nd exciting options for their use appear. This is especially important in applications related to medicine, since these devices can be injected in th e blood system without any risk of embolization. Moreover, since the encapsulated molecule is not in direct contact with the environment, irritation at the site of administration will be reduced.28 In this chapter, our goal is to review the most relevant contributions of different groups in this field in the last years. We have focused our attention on contributions based on several key synthetic approaches to sub-micron encapsula tion, a brief description of the synthetic methodology will be presented, and results that s how their potential toward novel applications will be discussed. 20

PAGE 21

1.2 Different Strategies for Encapsulation 1.2.1 Dendritic Architectures Since their introduction in 198529 dendrimers have become a powerful tool for the development of new technologies. Figure1-1 shows two typical approaches to the encapsulation in dendrimeric structures. Based on their synt hesis, the dendrimers can be prepared by a convergent or divergent approac h. The encapsulated molecule can be used as the core of the dendrimers or it can be incorporated after the macromolecule has been synthesized. In recent years, both strategies have been reported in the preparation of mo lecular capsules. In this section we will review the most important contributions in this area. In a convergent approach, Morgan et al.30 have synthesized different generations of poly(glycerol succinic acid) a nd tested them for encapsulation of a dye and drug delivery. The synthesis of these polymers was accomplished by first reacting succinic acid with cis 1,3-Obenzylidenglycerol (Bdg). Deprotection gave as a product a tetrahydroxy co re. The synthesis of the dendrons was performed reactin g Bdg with succinic anhydride in pyridine. Coupling of these two species gave the first generation dendrimer Higher generation dendrimers were synthesized in a similar way. Encapsulation of Reichardts dye (2,6-diphenyl-4-(2,4,6tryphenylpyridinio)phenolate) was achieved by addi ng the dye to a methanolic solution of the dendrimer. It was found that the dye could be encapsulated only w ith third and fourth generation dendrimers. The authors used UV-vis spectroscopy to show that the polarity of the core was similar to that of glycerol. NMR was used to elucidate the interacti on of the dye with the dendrimer, showing that the dye molecules e xperience restrictions in their motion. These macrostructures were tested as delivery ag ents for 10-hydroxycamptothecin (10HCPT), a topoisomerase I inhibitor. One important feature is that the physical prope rties of this molecule are different than Reichardts dye. This shows the versatility of the systems used. 1H-NMR NOE 21

PAGE 22

was used to prove the internaliz ation of 10HCPT within the dendrimer. The anticancer activity of the encapsulated inhibitor in human breast cancer cells was evaluated after being released from the capsules. In a more elegant, divergent approach, Furuta et al.31 have achieved the site isolation of two dyes, Coumarin 343 and pentatio phene, in the core of dendrimers. The goal of this work was to prevent quenching of the emissi on of the dyes when they were used in close proximity in the same device. This is needed in order to avoid energy transfer between dyes that leads to emission of solely the dye with the lowest HOMO-LUMO band gap. This group performed the synthesis by placing the dye molecule s at the center of a Fr chet-type dendrimer.32 Solubilization of the higher ge neration dendrimers was enhanced by alkyl ether functionalization of the periphery of the dendrimer. The surface of these molecules was further functionalized with hole-transporting triarylamine gr oups whose emission overlaps with the absorption bands of the dyes investigated. Films of the dendrimer were prepared with different thicknesses and showed to be uniform and aggregate-free. Thin film pho toluminescence (PL) experiments demonstrated that the isolation of the dye was more effectiv e as the size of the dendrons surrounding the cores increased. PL studies of the encapsulated dyes in corporated in a glassy polystyrene matrix showed that the emission of both dyes could be detected without significant quenching. This confirmed that the isolation of the dye by encapsulation in the de ndrimer reduced the extent of the energy transfer process. In a divergent approach, Sunder et al.33 synthesized hyperbranched polyglycerols based on the anionic ring-opening multibranched polymerization (ROMBP) of glycidol, obtaining polymers with narrow polydispersities. Further derivatization of the fr ee hydroxyl groups with different fatty acid chlorides resulted in dendrimers composed of hydrophilic cores and hydrophobic chains on the surface. Congo red, an anionic-water soluble dye, was chosen for 22

PAGE 23

encapsulation studies. After mixing an aqueous solution of the dye with a chloroform polymer solution, the probe was extracte d quantitatively below the satu ration concentration in the capsules. This behavior was shown to be general, as observed for other dye-solvent combinations. The response of the encapsulate d dye to environmental changes was studied by changing the pH of the water phase in contact w ith the organic solution. It was shown that the dye, despite being protected by a hydrophobic shell, was still able to respond to external stimulus. In a more elaborated process, Schappacher and Deffieux34 synthesized dendrimers by polymerizing -diethylacetal polystyrillithium (D EAPS) onto poly(chloroethylvinyl ether) (pCEVE) to form a comb-polymer core. After pur ification, the polymer chains were extended by cationic polymerization of CEVE. A second chain extension process was performed polymerizing DEAPS. The hydrophilic extension was performed by polymerizing, in sequence, 2-((tert-butyldimethylsilyl )-oxy)ethyl vinyl ether and di-O-isopropylidene-6-O-(2vinyloxyethyl)-D-galactopyranose. The final step involve s the deprotection of the poly(vinyl ether) blocks. As a result, a hyperbranched core-s hell architecture is achie ved when dissolved in an aqueous solution. Dynamic Light Scattering (D LS) measurement in water and a mixture of solvents showed that the dendrimer had a diam eter of approximately 100 nm. Zinc, cobalt and iron metallo-porphyrins were encapsulated by pouri ng a solution of the por phyrins in DMF into a large volume of a water/DMF solution containing the dendrigraft. Size exclusion chromatography was used to prove the encapsu lation of the metallo-porphyrins into the polymers. UV-vis absorption of the metallo dimer complex was used as the detection technique. The dendrimers synthesized by this group showed high encapsulation capacities. The authors observed that encapsulation produced significant changes in the si ze of the capsules, and, as a consequence, in their volume. The dendrigraft capacity for porphyrin complexation was shown 23

PAGE 24

to be dependent on the metal and the structure of the porphyrin used. The shift in the soret band of the porphyrins showed that the encapsulated sp ecies did not present si gnificant conformational restrictions. This is in agreement with the DLS results, which indicated that the porphyrin molecules occupy a high volume in the dendrim ers. To demonstrate the polarity of the environment in which the metallo-porphyrin co mplexes were located, UV absorption was used, determining that it was a poly(styrene)-like environment. Chen and Guan35 have accomplished the synthe sis of dendrimeric micelles by copolymerizing ethylene and its poly(ethylene glycol) derivative. Using the Brookhart palladium-diimine chain walking catalyst, they placed the poly(ethylene glycol) moiety at the end of the PE chain. In solution, these dendrim ers produce water-soluble unimolecular micelles. These macromolecules were tested as encapsula ting agents by dissolving nile red, a hydrophobic dye, in the core of these dendrimers. UV absorp tion of the encapsulated dye was compared to that of the probe in an aqueous suspension of sodium dodecyl sulfate (SDS), observing that the core of their capsules was more hydrophobic than th e SDS micelles. They also showed that the capacity of solubilizing nonpolar dyes was related to the mol ecular weight of the hydrophobic core. Niu and Crooks36 derivatized dendrimers with hydr ophobic groups on the exterior of the molecules and encapsulated metals nanoparticles in the core. The authors functionalized poly(propylene imine) 37 dendrimers by reacting the amine terminal groups with palmitoyl and hexanoyl functionalities. The derivatization was performed by dissolving the polymer in dry CH2Cl2 and mixing it with triethylamine under Ar. The final step involved the addition of palmitoyl chloride. A similar procedure was us ed for the hexanoyl derivative. The dendrimer encapsulated metal nanoparticles (DEMNs) were synthesized by dissolving the polymer in a 24

PAGE 25

mixture of CHCl3/MeOH. CuCl2 was dissolved in the same solvent mixture and this was added to the previously prepared dendrimer solution. Subsequently, the metal ion was reduced with NaBH4. The encapsulation of the metal ions was followed by UV; observing that the loaded capsules showed a different spectrum than the de ndrimer or ion alone. After the reduction of the metal ion, precipitation was not observed, wh ich showed the stabilization due to the encapsulation in the dendrimers. Moreover, the absorption properties of the complex changed. Since the encapsulation agents st udied by this group did not cont ain any peripheral amine groups and the interior amine groups did not seem to coordinate strongly with metal ions, the encapsulation process is not driven by c oordination to the metal ion. Based on these observations, the authors concluded that the encapsulation process is more likely to occur based on solubility differences. 1.2.2 Core-Shell and Hollow Particles Figure 1-2 shows a general approach to the sy nthesis of core-shell a nd hollow particles for the encapsulation of different species, based on a removable core-templating strategy. This can be done in the micro and nano scale, which offe rs the possibility of their use in different applications, as discussed below. Skirtach et al.38 used two different methods to synthe size capsules containing a core in the micrometer range (4-15 m) coated electrostatically. In the first case, multiple layers of alternating polyelectrolytes, poly(sodium 4styrenesulfonate) (PSS) and poly(allyl-amine hydrochloride) (PAH) were deposited and doped with Ag nanoparticles. In the second case, a cationic polyelectrolyte and an ionic dye (IR-806 ) were alternatively de posited. After deposition of the layers, the cores were dissolved to resu lt in hollow microcapsules with light absorption enhancers. When these systems were irradiated with a laser source, light in the IR region (830 25

PAGE 26

nm) was absorbed by the dye or the Ag nanoparticle s. The authors showed that irradiation of the doped capsules produced distortion of the shell, and depending on th e size of the particle and intensity of the laser, this eff ect could be perforation, deformation, or an ablative process. One important aspect of this work is that the wa velength of the laser used corresponded to the biologically friendly window (whe re the absorption of tissue, bi omedical substances, and water is minimal). Using the same system, Peyratout et al.39 promote the crystalli zation of the cationic dye, 1,1-diethyl-2,2-cyanine (PIC ) in the interior of the hollow particles. Monomeric PIC does not show any significant fluorescence but at high concentration it forms the fluorescent so-called J-aggregates. Since the formation of these fluor escence aggregates is promoted by the presence of negatively charged polyelectrolytes, PSS was s ynthesized in the interior of the capsules and subsequently PIC was added. In order to precipitate the dye forming the PIC-PSS complex, tetraphenyl borate (TPB) was added to the interior of the capsules. Changes in the absorption and emission spectra showed that a PIC-TPB complex was formed, sequestering the dye from the PSS matrix. Confocal laser scanning microscopy (CLSM) and atomic force microscopy (AFM) were used to characterize the species formed and their location in the hollow capsules. In a combination of the use of the layer by layer de position technique and the use of dendrimers as potential encapsulating agents, Khopade et al.40 have deposited PSS and a positively charge fourth generation poly(amidoamine) de ndrimer on spherical latex colloids. -potential was used to confirm the alternate deposition of the ch arged polymers on the surface of the growing particles. Carboxyfluorescein was loaded in the capsules by ex posure of the colloids to an aqueous solution of the fluorescent dye. The inte rnalization of the dye was studied by CLSM, showing that the dye placed itself in the shell of the dendrimer. The release of the dye in a NaCl aqueous solution was evaluated as a function of time. The authors observed a Fickian type 26

PAGE 27

release. Finally, hollow capsules were obtained by a similar appr oach, by using MF as the core and fortifying the final product with a PAH/PSS outer coating. Sun and Chiu41 prepared Ag nanopartic les with sizes between 25 and 200 nm by reducing silver nitrate in an ethylene glycol/poly(vinyilpyrrolidone) soluti on. Subsequently, these particles were coated with a silic a shell and the silver core was dissolved in an ammonia solution. Two techniques were used for encapsulation. The fi rst method consisted of mixing a nanocapsule aqueous suspension with a fluorescein solution in 2-propanol/chloroform. The second method consisted of drying the nanocapsule suspen sion, mixing with a 2-propanol/chloroform fluorescein solution and resuspension in borate bu ffer. They proved that these systems could be used for delivering small molecules using larger capsules, commercially available polystyreneacrylic-based hollow microcapsules coated with a silica shell. This process was followed by widefield fluorescence imaging and confocal point detection. The authors also investigated the delivery of carbachol from the nanocapsules. They trapped cells loaded with muscarinic acethylcholine receptor type I with a laser. Subs equently, a small volume of a suspension of the particles was added near the trapped cells. Since carbachol is an agonist of this receptor, the calcium signalling in the cells transfected with these receptors was observed as an indication of the delivery of the carbachol molecules. Taking advantage of the core-shell architectur e of the precursor used in this technique, encapsulation in the pores of the core material has been proposed as a new vehicle for encapsulation. Wang and Caruso42 have encapsulated enzymes in mesoporous silica particles and deposited polyelectrolyte layers on their surf ace to protect the encapsulated agent and avoid leaking. The authors studied silic a particles with two different pore size distributions. Bimodal mesoporous silica particles (BMS) with 23 nm and 10-40 nm pores sizes and 2-4 m diameter 27

PAGE 28

were used for these studies. BMS showed a hi gh enzyme-loading capacity due to their large pore size, although the amount of en zyme loaded depended on the mo lecular weight (MW) of the enzyme. CLSM was used to study the loading ability of the particles after derivatization with fluorescein isothiocyanate-label led peroxydase. It was shown that the enzyme adsorbed in BMS could be distributed on the surface and inside the particle. The coated capsules were obtained by alternate assembly of three la yer pairs of poly(diallylammoni um chloride) and 21 nm silica nanoparticles. The authors proved th e protective ability of the coati ng layers towards catalase (C100) by comparing its behavior ag ainst non-coated particles. BMS spheres showed that 25% of the enzyme had leaked after 6 h of being e xposed to a PBS solution, while no leaching was detected for the coated particles. Similarly, the encapsulated enzyme presented an enhanced stability against changes in pH and H2O2 degradation. When the stabil ity against proteolysis was studied, it was observed that the as-prepared capsules re tained 75% of the enzyme activity and total protection was obtained when eight layers of PAH-PSS were depos ited. Balabushevich et al.43 studied the release of the enzyme encapsulated by different methods in these particles. The one consisted of the formation of a complex between insulin and dextran sulfate (DS), while the second was based on the salting out of insulin mi croaggregates coated with DS and/or protein. Both systems showed that the protein could be release at pH values higher than 5, with a faster release when the pH was further increased. At th e same time, when more polyelectrolyte layers were deposited, the release of the enzyme s at the same pH was lowered. A different technique for the synthesis of hollow capsules is based on a core-templatedfree approach, where the use of a soluble core is avoided. This is preferentially used when the synthesis of the capsules involves a poly merization process. Turner and Wooley44 have synthesized a diblock copolymer formed with poly(tert-butyl acrylate) and polyisoprene and 28

PAGE 29

used them as templates for the formation of na nocage structures. Acidolis ys of the tert-butyl acrylate groups with a methanes ulfonic acid catalyst transfor med this macromolecule into amphiphilic structures that self-assembled in mi xed solvent environments. Crosslinking of the shell formed and subsequent ozonolysis of the co re resulted in hollow structures. Reduction of the intermediate ozonides was perf ormed with sodium sulphate. This gave as a result aldehyde and ketone groups in the interi or of the hollow capsules. Am ong other studies, the authors compared the capabilities of the core-shell structures and the ho llow particles for the encapsulation of BODIPY, a fluorescent dye. BODIPY was dispersed from methanol into an aqueous solution of the polymer. After ozonolysi s and reduction of the inner functional groups, the interior of the particles wa s less hydrophobic than the interior of the core-shell precursor. This was confirmed by the lack of encapsulating ability in the hollow particles. In another polymerizable system, Hu et al.45 synthesized hollow polymeric nanospheres with biocompatible and biodegradable polymers for the controlled release of an antitumor drug. The biopolymer chitosan (CS) with 90% deacetylation was dissolv ed in an acrylic acid (AA) aqueous solution. Due to their different charges, the AA monomers are attracted to the cationic biopolymers, being trapped in the CS micelles. AA was polymerized within the micelle s by initiation with K2S2O8, and CS was crosslinked with glutaraldehyde. DLS and -potential measurements were used to study the formation of the hollow capsules. After polymerization of AA an d crosslinking of CS, the micelles showed a dramatic decrease in their radio and -potential values. Transmission electron microscopy (TEM) showed the different species present at the di fferent stages of the synthetic process, confirming that hollow struct ures were formed at the end. In order to encapsulate doxorubicine, the drug was added to the nanospheres prior to the CS crosslinking. 29

PAGE 30

After purification, the release of doxorubicine was followed by steady state fluorescence, observing that the drug could be delivery for up to 15 days. In a simple and biocompatible design, lipids are used in solu tion to form vesicles to be used as encapsulating agents, as shown in Figure 13. In this case, the enca psulation is not used to enhance solubility, but to se parate reactive species from the solution. The walls of the hollow capsules are used as a se lective membrane for the intake of reactive compounds dissolved in the surroundings of the capsules. Graff et al.46 used a liposome system to form vesicles that were fortified by polymerizing a hydrophobic monomer in side the walls of the capsules. Membrane channels were added to the lipid bilayer in or der to enhance the permeability of the capsules. Preparation of the hollow particles involved dissolving th e lipid, 1-palmitoyl-2-oleoylsn glycero-3-phosphocoline (POPC) in chloroform. After evaporating the solvent, the membrane channel, OmpF, dissolved in an emulsion wa s added and vortexed. A film was obtained by drying the mixture and the enzyme -lactamase was added at pH 7.4. Purification and subsequent homogenization of the sample pr oduced hollow capsules with diameters of approximately 300 nm. A suspension of the capsules was mixed with an n-butyl methacrylate solution and ethylene glycol di methacrylate (the initiator). Polymerization was performed by irradiation of the sample with UV light (254 nm). The activity of the enzyme in the hydrolysis of ampicillin was analyzed for the capsules with and without the OmpF channels. The authors observed that there was no enzymatic activity if the channels were not present to enhance the permeability of the walls of the capsule toward s ampicillin. Moreover, the activity of the encapsulated enzyme was comparable to that of the free enzyme in high substrate concentration. The effect of the presence of the monomer on th e enzyme activity was studied as well. It was shown that, after the polymerizat ion, the activity of the func tionalized vesicles decreased 30

PAGE 31

considerably. The authors interpre ted this as a consequence of th e closure or expulsion of some of the channels in the liposomes. Similarly, Lawson et al.47 have used derivatized phospholipids for the preparation of vesicles to be used as encapsulation agents. The authors place 3,5-divinyl benzoyl, a polymerizable group, in the pol ar heads of two lipids 1,2-dipalmitoylsn -glycero-3phosphoethanolamine and 1,2-dilaurylsn -glycero-3-phosphoethanol amine, in order to polymerize them. The authors observed by TEM that the vesicles have the same diameter before and after the polymerization, approximately 70 nm. The same technique was used to prove that the presence of the polymerized units in the su rface stabilized the capsules compared to the nonpolymerized ones. A mixture of derivatized phos pholipids was used to prepare hollow capsules loaded with wild-type phosphotri esterase. The ability of the en capsulated enzyme to hydrolyze methyl paration (MPT) was studied, showing th at the encapsulated enzyme could hydrolyze MPT. In a more sophisticated approach, Bolinger et al.48 have encapsulated small vesicles (SV) in large vesicles (LV), in order to control the release of molecules from SVs into LVs. Addition of an anionic lipid in both types of vesicles (LV and SV) helped maintaining SVs suspended inside LVs. After encapsulation of a fluorescent dye at a self-quenching concentration in the interior of the SVs, the thermo tropic phase transition-induced rele ase of the dye was analyzed by fluorescence confocal microscopy. The authors obse rved an increase in the intensity of the fluorescence inside the LVs after changing the temperature of the system from 25 to 45 C. 1.2.3 Microemulsions Microemulsion systems have been used in the pa st due to their ability to enhance solubility of active agents in aqueous solutions, especially in biomedical applications. In the synthesis of encapsulating systems, organic, non-polar compounds are suspended in the presence of surfactants in order to react or crystallize, as shown in Figure1-4 for a suspension of a monomer. 31

PAGE 32

If polar reactive groups are placed on the surface, they can be used to grow a coating to fortify the particles. Yang and Zhang49 synthesized CdSe quantum dots (QD) in the microand nanometer range and encapsulated them with a polymer sh ell. Encapsulation was achieved by incorporating a toluene solution of QD into a CTAB em ulsion. A mixture of monomers, styrene, divinylbenzene and methacrylic acid, including the initiator (AIBN) was added. After 20 h of polymerization at 70 C, the polys tyrene particles in the emulsion were precipitated and washed with water and ethanol to remove the surfactan t. After the addition of NaOH, the negatively charged particles were coated with a silica she ll by mixing an acidic solu tion of tetraethoxysilane (labelled with a small amount of rhodamine) w ith an aqueous solution of the polymer-coated particles. Control on the size of the particles was achieved by changing the oil/water ratio and the amount of surfactant. The particles were char acterized by QD and rhodamine fluorescence. It was shown that the emission of th e QD was not quenched by coating. Wang et al.50 used a surfactant solution as a templa te for the synthesi s of core-shell and hollow nanoparticles. The core was synthesized in an aqueous basic solution of benzethonium chloride by polymerizing dimethoxydimethyls ilane (D). This process was followed by endcapping of the free hydroxyl groups with methoxytrimethylsilane (T). In order to synthesize core-shell particles, a mixture of D and T was added to the previously formed core. Free hydroxyl groups on the shell were encapped by mixi ng with T and hexamethyldisilazane. As a result, core-shell microparticles were obtaine d. Hollow nanocapsules were formed by dissolving the capsules in THF and subsequent extraction of the core. TEM showed that the architectures obtained were strong enough to main tain their shape after the extraction. The average size of the core-shell particles was approximately 300 nm. Afte r the removal of the core the average sizes of 32

PAGE 33

the hollow particles shrunk to approximately 200 nm. TEM images showed that the shell thickness was approximately 70 nm. AFM was used to confirm the hollow features of the capsules. Indomethacin (Idm) was encapsulated by dispersing the capsules in an aqueous bezethonium chloride solution swollen with an indomethacin/CH2Cl2 emulsion. TEM micrographs showed that Idm could diffuse into the capsules. Jungmann et al.51 used similar core-shell and hollow nanocapsules for the encaps ulation of dyes with different polarities and charges. These capsules contained a cationic additive in the core of the core-shell architectures or in the inner part of the hollow capsules. The synthesis is based on th e polymerization of T, (chloromethylphenyl)trymethoxysilane (ClBzT) and D using benzehtonium chloride as a surfactant.52 The core-shell particles become hydrophili c by quaternarization of ClBzT. In order to prepare hollow nanocapsules, the core was not covalently attached to the growing shell, so it could be removed after the shell was synthesize d. Solid-liquid and liquid-liquid phase transfer techniques were used to load dye molecules in the capsules. In th e former, a solid dye is added to the dispersion of the capsules in toluene, in the latter, an aqueous solution of the dye is covered with a layer of the nanocapsu les dispersion. Both techniqu es were showed to produce encapsulation. This process seemed to be depe ndent on the presence of the quaternary amino groups. This is especially important since it was observed that cationic dy es were able to be encapsulated (to a lower extent), despite elec trostatic repulsion c ould be thought as an unfavorable interaction. The mesh width of the nanocapsules wa s investigated. The lack of uptake of a sterically demanded dye showed that this is an important parameter that allows control over the uptake process. Kumar et al.53 have achieved the enzymatic synthesis of copolymers in order to form nanomicelles for the encapsulation and delivery of drugs. The synthesis of the copolymers was 33

PAGE 34

performed by mixing PEG (Mn 600) and dimethyl 5-hydroxyis ophtalate with the enzyme Novozyme-435 ( Candida Antarctica Lipase B), at 90 C for 48 h. The macromolecule obtained was reacted with different alkyl bromides in acetone, to functionalize the pol yester with different pendant groups. Due to the amphiphilic character of the copolymers, they formed micelles in hydrophilic and hydrophobic solvents. In order to encapsulate as pirin, the copolymer and the drug were dissolved in chloroform. After evaporati on of the solvent, the mi xture of both species was dissolved in water, resulting in drug loaded nanomicelles. 1H NMR longitudinal relaxation time studies were performed to study the micellization process. DLS and St atic Light Scattering (SLS) were used to determine the size and shape of the micelles formed. The particles obtained showed sizes between 15 nm for the different polyesters derivatives a nd their shapes varied from hollow to star-like spheres. SLS and UV absorption were used to confirm the encapsulation of aspirin and napronex. A sli ght increase in the capsules Rg compared to the hollow capsules was observed and the UV absorption of the load ed capsules showed the peaks of both, the nanospheres and the drug. The release of both activ e molecules was tested in animals, showing a significant efficiency in the reduction of in flammation compare to a similar dose of the commercially available drugs. Panyam et al.54 used poly(D,L-lactideco -glycolide) (PLGA) and pol ylactide (PLA), in an emulsion-solvent evaporation tec hnique in order to prepare micr o-and nanocapsules loaded with drugs. PLGA and PLA of different molecular wei ghts were mixed in different ratios with the drug in order to study its influence in the capsules synthesis. Afte r this, the loadi ng capacity of the particles and the in vitro release of the drugs was studied. Tritiated-labeled dexamethasone was dissolved in methanol and mixed with a polym er solution in chloroform. Polyvinyl alcohol was added as an emulsifier and the mixture wa s sonicated first and th en stirred for 18 h to 34

PAGE 35

evaporate the organic solvent. In the micropa rticles formation sonication was replaced by vortexing. Purification was performe d to wash the emulsifier out of solution and separate the non-encapsulated drug. DLS results showed that th e size of the nanoparticles varied between 240 and 270 nm. When low MW polymers were used, th e size of the particle s increased up to 740 nm. The authors found that polymer combinations with lower amount of polar groups produced capsules with higher drug loading. X-ray diffraction and differential scanning calorimetry were used to examine the physical state of the dr ug, observing that there was no crystalline drug present in most of the formulations, except when low MW polymers were used (and larger sizes were obtained). When studying the in vitro release of the drug, it was observed that capsules with a higher loading capacity had lower release rates. This was interpreted as a consequence of the lower partitioning of the drug fr om the polymer to the external aqueous phase. Moreover, the release was found to be faster when the differe nce between the solubili ty parameters of the matrix and the drug was higher. In a similar approach, Bae et al.55 have synthesized polymeric micelles based on poly(ethylene glycol)-poly(as partate-hydrazone-adria mycin) (PEG-PAHA). The synthesis was based on the ring opening polymerization of -benzyl-L-aspartate N carboxyanhydride with -methoxy-amino poly(ethylene glycol) in dimethylfomamide. Extension of the polymer chain ends was accomplished by adding hydrazide groups that were further derivatized with adriamycin ( ADR, a fluorescent, anticancer agen t) through an acid labile hydrazone bond. The amphiphilic polymer formed mi celles in a polar solv ent, placing the drug molecules in the core. DLS analysis showed that the diameter of the micelles was approximately 65 nm. The influence of the pH on the release of ADR from the polymeric matrix was studied by reversed-phase liquid chromatography. It was observed that, at pH below 5.5, the active molecule-polymer bond broke, delivering ADR. Th e loaded drug micelles and the drug by itself 35

PAGE 36

were incubated in a cell cultu re medium. CLSM was used to identify the intracellular localization of the micelles. Since the fluorescence of the drug is only possible when is outside the micelles (and not inside due to quenching fo r the high concentration when encapsulated), the authors showed that after 24 h of incubation the micelles were located in the cytoplasm and that ADR was released. 1.3 Conclusions A variety of synthetic techniques have emerged in recent years that greatly enhance the ability to achieve controlled encapsu lation of molecular species. In this Chapter, recent advances in several sub-micron encapsulati on strategies have been discu ssed. In particular, dendritic architectures, hollow particles, microemulsion-based systems, and core-shell nanocapsules have been highlighted. Together, thes e techniques show promise for th e encapsulation of a variety of species in a diverse range of conditions, which makes them a valuable tool for the design of nanocontainers, fluorescent tags and delivery systems. In the work presented in the following chap ters, we offer a different approach to the synthesis of nanoparticles. Our goal was to de sign a simple, fast and relatively cheap methodology for the synthesis of core-shell partic les for different applications. We decided to use a combination of the most practical aspects of the approaches reviewed here. We used a templating system based on the use of surfactants This approach takes advantage of the selfassembly of amphiphiles in aqueous suspensions producing species with a hydrophobic core and a hydrophilic periphery. We then used sol-gel chem istry to chemically co nnect the polar groups of one of the surfactants and used the free groups on the surface of the particles to attached different species. In order to form the core-s hell particles, a silica shell was grown on the periphery of the particles by a sol-gel process. 36

PAGE 37

Figure 1-1. Synthesis of dendrimers for encapsu lation. Divergent (A, D) and convergent (B, C) approaches to the synthesis of dendrimers for the encapsulation of active cores or active molecules Figure 1-2. General approach of the synthesis of hollow particles using a removable core for the encapsulation of active species. 37

PAGE 38

Figure 1-3. Synthesis of hollow particles base d on the use of lipids. These particles can encapsulate active molecules in th e water pool in their interior. Figure 1-4. Particle synthesis by microemulsion polymerization. 38

PAGE 39

CHAPTER 2 METHODS AND EXPERIMENTAL PROCEDURES 2.1 Introduction The templates and core-shell particles were characterized by dynamic light scattering, transmission electron microscopy, atomic force microscopy and -potential measurements. It was crucial for us to characterize the particles in detail, in order to learn how to control their size. This is especially important if we keep in mind that different applic ations may require the synthesis of particles with different sizes. In addi tion, to elucidate the polar ity of the interior of the particles, we used steady state fluo rescence spectroscopy and fluorescence lifetime measurements to study the response of different probes internalized by the templates and coreshell particles. Below we describe briefly the basic principles of the techniques used for the characterization studies, with emphasis on their a pplication to study partic ulate systems. At the end of each section we include the details of the experiments performed in this study. 2.2 Techniques and Methods 2.2.1 Dynamic Light Scattering 2.2.1.1 Theory The principle of the dynamic light scattering technique is that the scattering of light can be viewed as a result of microscopic heterogeneiti es within the illuminated volume. When a volume of a homogeneous sample is illuminated with a b eam of light, the scattere d waves will have the same amplitude and interfere destru ctively in all directions, except in the direction of the incident beam. If a heterogeneous sample is being analyz ed, the index of refraction would differ from the average value at some location, as a consequence, the wave that is scat tered at this location would not be compensated for and some light w ould be observed in directions other than the incident light and light scatteri ng occurs, as shown in Figure 2-1.56 39

PAGE 40

There are two ways to approach the phenome non of light scattered by particulate matter in solution. The one is to consider the suspensi on as a homogeneous medium and ascribe light scattering to the spatial fluctuations of the solute. This is appropriate for solutions of small molecules in which the average distance between th e center of the scatterers is small compared to the wavelength of light. The second is to consider each individual solute particle as a heterogeneity and therefore as a source of light scattering. The latter is more appropriate for solutions of large macromolecules and colloids, where the average dist ance between particles centers is comparable to the wavelength of light.57 In cases in which the size of the scatterers is not small compared to the wavelength of light, the interference of the elect romagnetic waves scattered by the solute is not all constructive and the phases of these waves must be taken into account. If the phase of a wave scattered at the origin is used as a reference, the phase of a wave scattered at a point with radius vector r is q.r The vector q is called the scattering vector, which is a fundamental ch aracteristic of any scattering process. The length of the vector is indicated in Equation 2-1.56 2 sin 4 qq (2-1) where is the refractive index of the medium is the wavelength of light is the scattering angle The essence of the DLS technique is to measure the temporal correlations in the fluctuations in the intensity of the scattered light and to reconstr uct from these data the physical characteristics of the scattere rs. In a suspension of particles, the scatterers are randomly distributed within the scattering volume. Since th e size of the scattering volume is much larger than q -1 (with the exception of nearly forward scattering, where q~ = 0), the phases of the waves scattered by different particles vary dramati cally. As a result, the average amplitude of the scattered light is proportional to N1/2, where N is the number of scatterers, and the average 40

PAGE 41

intensity is simply N times the intensity scattere d by an individual particle The local intensity, however, fluctuates from one poi nt to another around its average value. As the scattering particles move, the interference pattern changes in time resulting in temporal fluctuations in the intensity of light detected at the observation point.57 In a DLS measurement, the instrument de tects a random signal. The information is contained in the correlati on function of this signal i(t) which in the case of DLS is the photocurrent, defined as in Equation 2-2 )()(2 titiG (2-2) The notation G2( ) is introduced to distinguish the co rrelation function of the photocurrent from the correlation function of the electromagnetic field G1( ) (which is the Fourier transform of the light spectrum) )()()(* 1 tEtEG (2-3) The angular brackets denote an average over time t This time averaging, an inherent feature of the DLS method, is necessary to extr act information from the random fluctuations in the intensity of the scattered light. In the majority of practical cases, the scattered light is a sum of waves scattered by many independent particles and therefor e displays Gaussian statistics. This being the case, there is a relation between the intensity correlation function G2( ) and the electromagnetic field correlation function G1( ) : 2 2 21)( gIGo (2-4) Here g1( )= G1( )/ G1(0) is the normalized field correlation function, Io is the average intensity of the scattered light and is the intensity factor.57 41

PAGE 42

Temporal fluctuations of the intensity of the scattered light are caused by the Brownian motion of the scatterers. The speed of the particle s is related to their size, small particles move faster than large particles. Though each particle moves randomly; in a unit time more particles leave regions of high concentration than region of low concentration. This results in a net flux of particles along the concentration gradient. Browni an motion is then responsible for the diffusion of the solute, and is quantitatively ch aracterize by the diffusion coefficient, D Rigorous mathematical analysis of the process of light scattering by Br ownian particles leads to the following expression for the correlati on function of the scattered light: )exp(2 1 Dq g (2-5) The diffusion coefficient in an infinitely dilute solution is determined by particle geometry. For spherical particles, the relation between the radius R and its diffusion coefficient D is given by the Stokes-Einstein Equation: R Tk Db6 (2-6) where: kb is the Boltzman constant is the viscosity of the solution T is the absolute temperature Hence, from Equation 2-5 the diffusion coeffici ent can be obtained from the correlation function g1( ) Assuming that the scattere rs are spherical, Equation 26 can be used to obtain the hydrodynamic radius of the particles. We used th e imaging techniques to investigate whether the shape of the particles ma tched this requirement. In our studies, the suspensions obtained wh ere not monodisperse but rather, made of particles a range of sizes. This made the analysis of the data obtained more complicated. In the case of polydispersed samples, a different analysis of the correlation function is required. For a continuous distribution of scatteri ng particle size, the correlation function is obtained from the following equation 42

PAGE 43

dTDqDI I go 2 1exp 1 (2-7) where I(D)dD = N(D)Io(D)dD is the intensity of light sc attered by particles having their diffusion coefficient in the interval [D,D+dD] [N(D)dD] is the number of these particles in the scattering volume, Io(D) is the intensity of light scattered by each of them.58 The main goal of the software is to reconstruct as precisely as possible the distribution function I(D) from the experimentally measured function G2 exp( ). The main problem encountered is that dramatically different distributions I(D) lead to nearly identical correlation functions of the scattered light and therefore are equally acceptable fits of the experimental data. There are three typical approaches to solve this problem: Direct fit method: In which a functional form of I(D) is assumed a priori and the parameters that lead to the best fit of G2 theor( ) to G2 exp( ) are determined. Method of cumulants: This approach focus on the s table characteristics of the distribution, which are the mome nts of the distribution, or cl osely related quantities called cumulants. The first cumulant of the distribu tion that gives the average diffusion coefficient D*, can be determined from the initial slope of the field correlation function, as shown in Equation 2-8. The second cumulant, the width of the distribution, is obtained from the curvature of the initial part of the correlation function. 2* 2 )1()( 1 ln qDdDDqDI I g d do o (2-8) Regularization: Is a combination of the first previ ous methods mentioned, it assumes that the distribution I(D) is a smooth function and seeks a non negative distribution producing the best fit to the experimental data. This met hod is used in different approaches used to reconstruct the scattering partic le distribution from DLS data. The key point is the selection of the smoothness of the distribut ion. If the smoothing is too st rong, the distribution will be stable but will lack details. If the smoothing is too weak, false spikes will appear in the distribution. The moments of th e distribution reconstructed by the regularization procedure gives unbiased (apart from smoothing) info rmation on the shape of the distribution.58 The primary technique used to follow the effectiveness of the synthesis of the templates and particles was DLS, due to its ease of opera tion and the small amount of sample needed. As shown in Figure 2-2, the difference in size between different species can be visualized qualitative by a comparison of the normalized correlation function obtained. Similarly, a quantitative 43

PAGE 44

analysis can be performed in the case of coating particles with a laye r of other material by determining the diameter of the particles before and after the coating process. In our case, from the difference in sizes between the templates and core-shell particles, a shell thickness can be calculated using Equation 2-7. The size determination was performed afte r purification of the species obtained, to minimize the influence of the presence of small micelles in the determination of the correlation function. In addition, the suspen sions had to be filtered to eliminate dust or aggregated particles that could not be separated by sonication. This is especially important due to the de pendence of the intensity of the light scattered by a particle to the sixth power of its radius, which would result on the signal of the particles being obscured by the aggregates, producing an overe stimation of their diameter. 2/12ddST (2-9) where: d1 is the diameter of the templates d2 is the diameter of the core-shell particles 2.2.1.2 Experimental settings Throughout this work two instruments were us ed, a PDDLS/CoolBatch 90T detector with a PD2000DLSplus correlator, manu factured by Precision Detector s Inc, and a 90 Plus/BI-MAS detector with a BI 9000AT digital correlator manufactured by Brookhaven Inc. In both cases the detector was placed at 90 from the incident beam Each instrument works with its own software for size determination, the Precision Deconvol ve 32 in the first case, and the BI-ISDAW advanced size distribution software, in the second case. The templates and core shell particles were sonicated for approximately 30 minutes and filtered before DLS analysis. Sonication was intended to break any aggregates formed, which was observed after storing the samples for months, especially after purification, while filtering was used to remove aggregates th at could not be broken up. The samples needed to be diluted to avoid multiple scattering. The extent of dilution differed from templates to core-shell particles 44

PAGE 45

due to the differences in sizes and, as a conseque nce, on the intensity of light they scattered. On average, the concentration of templates and core -shell particles used was below 0.1 mg/mL. The appropriate concentration was determined on a sa mple by sample basis by examination of the intensity of the light scattered through the in strument. As indicated in the manuals, a total intensity of the scattered light between 100 to 300 kilocounts pe r second (kcps) was used; being the concentration of scatterers adjusted until these values were reached. Subsequently, the sample time was adjusted after a quick examina tion of the correlation function. The sample time is directly related to the size of the particles, so it can be adjusted to obtain the proper fit of the correlation function. After these settings were adjusted, the meas urements were performed in 10 minutes runs. This time was found to be optimal to obtain a good signal to noise ratio, resulting in stable and reproducible results. The final resu lts reported were obtained by the average of five measurements. 2.2.2 Transmission Electron Microscopy 2.2.2.1 Theory The need to use the electron microscope is due to the size of the particles to be imaged. Since their diameter is similar to the wavelength of the light used in common light microscopes, they cannot be characterized by this technique. In such cases it is necessary to use the transmission electron microscope, in which the use of light has been repl aced with a beam of electrons. This results in better resolution, due to the shorter wavele ngth of the electrons, compared to photons. In the electron microscope, electromagnetic lenses are used to focus the beam of electrons into a tight coherent beam, which is then focused on the sample. Data is collected after the beam has passed through the sa mple. Our imaging has been performed in the bright field mode. This means that after the b eam of electrons has passed the sample deposited on the substrate, an objective aperture has been in serted. This aperture allows the electrons in the 45

PAGE 46

transmitted field to pass and contribute to the resu lting bright field image, rejecting the electrons that had been scattered by the sample,59 as shown in Figure 2-3. One of the problems encountered with our templates and core-shell partic les is their lack of electron density. This could be due to the forma tion of a porous matrix in the interior of the particles or the formation of disk-like species, as will be discussed later. To overcome this problem different staining techniques were tested, including the use of OsO4 to stain unsaturations and uranyl acetate to stain the hyd rophilic part of the particles (mostly their periphery). This improved the qua lity of the images obtained, especially by the use of uranyl acetate. 2.2.2.2 Experimental settings The instrument used in this study was a Hit achi H-7000, with a maximum resolution of 0.2 nm and a maximum acceleration voltage of 125 kV Imaging the particles synthesized in this work was not an easy task due to their low elec tron density, as explained above. Moreover, due to their size, especially the templates, getting a decent focus that could be used to measure the size of the particles was difficu lt. Even trying higher accelerat ion of the electron beam through higher voltages failed, as a consequence, our im ages were taken with a voltage of 75 kV, the usual setting of the instrument. Considering that the instrument was not available for use at our will and the fee for its use, we decided to use th is technique to image our particles and confirm the geometries observed by AFM, technique then selected for the size ch aracterization of the particles, as explained below. Similarly to what was observed with DLS, the pa rticles needed to be diluted before loading them on the grids. If the concentration was too high, the particles were deposited as multilayer, making impossible their visualization. It was dete rmine that a concentration of about 0.1 mg/mL was optimal in most of the cases to load the grids. At the beginning of th ese studies, particles to 46

PAGE 47

be imaged were made with 1-dodecene to stain its unsaturation with OsO4 and visualize the core and shell of the particles. Afte r deposition of the suspensions on th e grid and evaporation of the solvent, the grids were placed in a small gl ass chamber where they were exposed to OsO4 vapors, for approximately 30 minutes. Using longer times made the images too dark and oxidization of the grids was observed in some cases. Before inse rtion of the samples containing the grids in the microscope, the samples were stained with a me thanolic solution of uranyl acetate (UA) for 1 minute. This is a common staining agent for biological samples, used primary to stain hydrophilic regions. Images were take n at different spots of the same grid to confirm that the species observed were present in most of the suspension and th at they were not an unusual finding. Size analysis was performed by comparison of the diameter of the particles and the scale bar in the image. This was done in the Adobe photoshop software and later with the Image J software. 2.2.3 Atomic Force Microscopy 2.2.3.1 Theory Atomic force microscopy, since its introduction in 1986, has been used in many different applications, especially because of the impr oved resolution compared to other microscopy techniques. In our case, the main reason to use this technique is the ability of AFM to obtain information on three dimensions without the need of any coatings or stains, an advantage over TEM. Especially important in our case is the determination of the particles height after deposition on a substrate, which we used as an in dication of the rigidity of the particles, and how the coating of the templates affects this. The technique is based on the interaction of the tip of a cantilever with the sample deposited on a substrate. The instrument measur es the forces between the tip of the flexible cantilever and the sample. The basic idea is that th e local or attractive for ces between the tip and 47

PAGE 48

the sample are converted into a bending, or deflection, of the cantilever. The key feature is that the force between the probe and the sample is ma intained constant while the probe is raster scanned across the surface. In order to detect th e probe bending, a laser beam is reflected from the back of the cantilever onto a de tector, in such a way that a small change in the bending angle of the cantilever is converted to a measurable large deflection in the position of the reflected spots. The deflection observed then is converted into an electrical signal, to produce an image of the surface. A simple representation of the instru ment setup is shown in Figure 2-4. In order to avoid any non desired interaction between the probe and the particles, the instrument was used in tapping mode. In this mode, the cantilever is osci llated close to its res onance frequency and the tip taps the surface only periodically, reducing signi ficantly the lateral force. This means that the probe is free to oscillate up and down at its resonance frequency as a consequence of the interaction with the substrate when it comes extr emely close to it. In tapping mode, the image is obtained by imaging the force of the oscillati ng contacts of the tip with the sample surface.60, 61 2.2.3.2 Experimental settings The instrument used in this study was a Nanoscope III, manufactured by Digital Instruments, Inc. Imaging was performed in tapping mode, using silicon probes (Nanosensors, with dimensions: T=3.8-4.5 m, W= 26-27 m, L= 128 m). The z-calibration was performed with a silicon grating (TGZ01, Mikromash), with a step height of 20 nm (accuracy 1 nm). Images were analyzed with the Nanoscope III software provided by the manufacturer. To determine how accurate the z-measurements were in the instrument, we analyzed a silicon grating calibration grid with a step height of 19.66 +/1 nm, as reported by the manufacturer. The instrument was used in contact mode. The area was scanned at 2.44 Hz and at 256 samples per line. The image was flattened wi th a first order fit, as suggested by the instrument manual. Figure 2-5 shows the bearing an alysis, in which the software determines the 48

PAGE 49

distribution of data points in the z-axis. The two spikes in the histogram correspond to two elevations on the surface: the bottom and top surface. The red box (Hist rel depth) indicates the distance between the two cursors on the histogr am (19.93 nm), which represent the height determined by the instrument. The result showed that the height was measured with an accuracy of 1.4 %, which confirmed that the measurements performed were accurate. The samples were deposited on mica by adding a couple of drops of the suspension and the solvent left to evaporate at room temperature. No other treatment was pe rformed before analysis. The images were obtained with the scanner E, which offers a maximum imaging area of 15 x 15 m. The exact conditions for the different parameters used varied from sample to sample, although average values or ranges ca n be reported. To image the samples, first, a fast scan was performed to identify areas were the sample was deposited, after which the tip was taken to one of this spots to zoom in and acquire an imag e of proper quality. Typically, we used 512 samples per line, the best quality possible with our in strument. Next, the amplitude setpoint was decreased slightly to optimize th e force use to scan the surface. This was stopped when the trace and retrace profiles in the section analysis looked similar. The scan rate was adjusted based on the dimension of the image, for areas between 2 x 2 2.5 x 2.5 m, a typical scan rate between 11.5 Hz was used, and this value had to be increas ed for smaller areas scanned, typically to 2-2.5 Hz. This was done to maintain good track of the tip on the z-axis, which allowed the visualization of the changes in the height profile. Finally, th e integral and proportional gains were adjusted to improve the quality of the imag es, with typical values of 0.35-0.45 and 0.4-0.6, respectively. In general, the templates were easier to image, since they tended to be deposited as layers of relatively homogeneous height, which made it easy for the tip to identify the topographic 49

PAGE 50

features of the species deposited. The opposite wa s observed for the core-shell particles, which were deposited as agglomerates of different heights and sizes, as a consequence, images of good quality were more difficult to obtain, but usuall y it was a matter of finding a spot with a low concentration of particles. In the first studies, a flattening of the images was performed through the software, but since we became interested in the topographic feat ures, this was not used in the study presented here. This was done to avoid any kind of artifacts that could affect the determination of the height of the particles, as a consequence of the le veling of the surface done through the flattening option. The section analysis was performed then with the images without any post-treatment. The particles used to measur e the diameter and height were selected from each image taken after zooming in the area of in terest and confirmation that both dimensions were possible to determine without ambiguities. Unless otherwise stated, the height images were used for the characterization of th e particles, although in many cas es it was easy to visualize the shape of the particles in the phase image, but, due to the lack of understanding on the meaning of these images in the literature, they were not used in our studies. 2.2.4 -Potential 2.2.4.1 Theory We used this technique to characterize changes in the charge of the su rface of the particles, after different treatments given to the template s and core-shell particle s. The electrostatic potential at the surface of shear of a kinetic unit is the -potential. The kinetic unit consists of the particle, ions adsorbed onto the surface, counter ions contained within th e surface of shear, plus solvent molecules strongly attached to the surface ions and counter ions in the double layer. This kinetic unit is also called th e hydrodynamic radius, obtained expe rimentally by DLS as explained in section 2.2.1. The magnitude of the electrostatic repulsive forces between particles is determined from -potential.62 50

PAGE 51

The technique works based on the detection of the mobility of the particles under the influence of an electric field. A laser beam passes through the samp le, the light scattered detected is Doppler shifted because of the motion of partic les under the influence of the electric field. A portion of the beam directed to the samples is sp lit off and then recombined with the scattered beam and modulated at 250Hz. With this, the Doppler sh ift of the signal, which is proportional to the velocity of the particles, is easy to detect as a shift in the frequency of the recombined beam. Two factors affect the motion of the particles, the sign of their charge, which determines the direction they move, and the valu e of this charge, which determines the speed they move at. Experimentally, it is the velocity of the char ged particles that is measured, from which the mobility and -potential are calculated. The calculation of mobility is st raight forward, contrary to the -potential.62, 63 The relationship between -potential and mobility depends on the theoretical model chosen. There are two classical models that results in two classical limits: the Smoluchowski and the Hckel equations. They each applied in two opposite limits. Simply stated, these limits have a common root: the magnitude of the dimensionless product where is the inverse of the double layer thickness and a the radius of the kinetic unit. In general, mobility is related to the potential as shown in Equation 2-10 ,3 2 afh ee (2-10) where f ( ka, ) is a model dependent function. In the simplest model, for 1:1 electrolytes 3 2330/5.37/5.42/3 aa a afaf (2-11) The Hckel limit is satisfied when << 1, typical for ions. In this limit -potential and mobility are related as shown in Equation 2-11 51

PAGE 52

3 2 ee (2-12) where: is the viscosity of the suspending liquid On the other hand, the Smoluchowski limit is satisfied when >>1, obtaining Equation 213 ee (2-13) For most colloidal particles it is impossible to satisfy the Hckel limit. As a consequence, in our studies, the Smoluchowski model was used To assure the limits of this approximation were satisfied to the best of our possibilities, the suspensions were analyzed suspended in a 10 mM KCl solution, to reduce the thickness of the double layer a nd increase the product .62, 63 The detection system is the same as the one explained in the dynamic light scattering section. The instrument used in this study wa s the Brookhaven instrument described in the DLS studies. The difference in this case was that the suspensions were placed in a cuvette to which an electrode was attached, and the detector placed at 15 from the original laser beam. 2.2.4.2 Experimental settings The suspensions of the particles were treated previous to the measurements similarly to the case for DLS studies. The samples were sonicated and filtered. Before the measurements, the particles were analyzed by DLS to confirm the size of the particles, assure the absence of aggregates and adjust the concentration to m odulate the intensity of the signal reaching the detector. Before each measurement, the electrode s had to be preconditioned. This was done as indicated by the manufacturer by running a NaCl solution in 3 repe titions of 10 cycles consisting of 10 runs each. With this, the electrodes became black, producing a dark brown solid that was discarded. The electrodes were rinsed with Mi llipore water before the measurements. This was 52

PAGE 53

found to be a very critical step in the setting of the conditions for the measurement. If this preconditioning was not done properly, the valu es obtained were diffi cult to reproduce. We usually started our measurem ents at the pH the samples we re stored, approximately, 7. We then increased the pH by addition of the proper volume of NaOH 0.5 M. After reaching a pH of around 10, the measurement at the first pH value tried was repeated. If the values were off, the preconditioning was considered improperly done, th e suspension discarded and the measurement redone. Lower pH values were obtained by adding HC l 0.5 M. With the temp lates, the pH of the suspension could be taken to a lower value comp ared to the core-shell particles. The ones derivatized with amino groups, tend ed to precipitate when the pH was below 3. In all the cases, below pH 3, addition of larger volumes of the acid was necessary to achieve changes in the pH of the suspensions, since protonation of the gr oups on the surface takes placed preferentially under these conditions. The measurem ent at each pH was repeated five times and the mean value was used in the plots presented through the text. 2.2.5 Study of the Polarity and Viscosity of a Matrix by Steady State Fluorescence Spectroscopy 2.2.5.1 Use of pyrene as polarity and rigidity probe Pyrene has been used extensively as a probe for the study of polarity of different matrices. It has been established that the intensities of its various vibron ic bands show a strong dependence on the solvent environment. In th e presence of polar solvents there is an enhancement of the 0-0 band at the expense of others. The polarity index, defined as the ra tio of the third and first peak intensities (I3/I1) has been qualitatively related to change s in polarity in the environment where the dye is placed.64, 65 This is a consequence of the perturbations of the intensities of the vibrational fine structures in its emission spectrum, due to the extent of interaction between the solvent dipoles and the excited si nglet states of pyrene. When there is minimal coupling, the 53

PAGE 54

polarity index is 2.0 (as in perfluoromethylcycl ohexane). However, efficient dipolar coupling, with solvents such as acetonitrile or dimethyl sulfoxide, causes a drop in the I3/I1 value to as low as 0.5. Pyrene has been used as a probe for rigidity as well. The photoexcited molecule, during its lifetime, may approach an unexcited dye molecule to form a collisional complex called excimer. Since this process is dominated by translational diffusion, it has been related to the viscosity of the media where the dye resides.66 The excimer emission appears as a broad peak centered around 480 nm. Hence, the ratio of the intensitie s of the emission of the monomer and excimer (I1/Iexc) is related to the ability of the dye molecu les to form dimers, and can be used as an indicator of changes in th e viscosity of the media.65 Due to its very low solubility in water, pyrene has been used as a probe in studies of micellar systems as well, observing that it is preferentially solubilized in the most extern al hydrophobic regions of these aggregates, in contact with water molecules to a certain degree.67 Hence this probe has been used for determination of CMC values. Pyrene would be di ssolved in water if the surfactant concentration is low enough to prevent the formation of micelles. Once the surfactant molecules start aggregating, the probe will be incorporated in the micelles due its hydrophobicity, which is observed as a dramatic shift in the emission of the dye and on the intensity of its emission. 2.2.5.2 Use of C153 as a polarity probe C153 has been used extensively in different sy stems as a probe for polarity due to its large change in the dipole moment when excited.68-70 After excitation, the dipole moment of the molecule increases from 6.6 in the ground state to a value between 14.2 and 16, depending on the solvent.71 When the probe is excited, reorganization of the solvent molecules leads to a relaxed state of lower energy. The relaxation is larger wh en the solvent polarity is larger. Only one single excited state ( S1 state) of intramolecular ch arge transfer is responsib le for the fluorescence of 54

PAGE 55

C153.69 Thus, complications arising due to the participation of multiple fluorescence states are avoided. As a result, the dye emits with a ma ximum between 450 nm to 550 nm. This excitation and emission wavelengths are very similar to dyes commonly used in protein labeling which makes it useful for fluorescence particles intende d to label biologically relevant species. 2.2.6 Time-Domain Lifetime Measurements Time-resolved fluorescence data usually contai n more information than is available from the steady state studies. For inst ance, it can give information on the presence of one fluorophore in different environments, by reporting different decay times. Another important use of this technique deals with the differentiation between static and dynamic quenching. Two methods of measuring time-resolved fluorescence are common, the time-domain and the frequency-domain. In our case we worked with the former, as a c onsequence, we will focus our discussion on this technique. In time-domain, the sample is excited w ith a pulse of light, made as short as possible, preferentially much shorter than the decay ti me of the sample to be examined. The time dependent intensity is measured following the excitation pulse, and the decay time is calculated from the slope of a plot of logI( t ) versus t or from the time at which the intensity decreases to 1/ e of the value at t =0.65 In general, the inverse of the lifetime is th e sum of the rates whic h depopulate the excited state. The lifetime of a fluorophore is also inte rpreted as the average amount of time a probe remains in the excited stated following excitation. This lifetime is a statistical average, and fluorophores emit randomly throughout the decay. For a large number of fluorophores, some emit quickly following the excitation, and some w ill emit at times longer th an the lifetime. This time distribution of emitted photons is the intensity decay.65 55

PAGE 56

2.2.6.1 Fluorescence resonance energy transfer st udied by time-resolved fluorescence Fluorescence resonance energy transfer (FRET) is transfer of the exc ited-state energy from the initially excited donor (D) to an acceptor (A). The rate of energy transfer depends upon the extent of spectral overlap of the emission spec trum of the donor with th e absorption spectrum of the acceptor, the quantum yield of the donor, the relative orientation of the donor and acceptor transition dipoles, and the distance between the donor and acceptor molecu les. The extent of energy transfer is also influenced by the pres ence of the donor-to-acceptor diffusion during the donor lifetime, which is usually studied by time-resolved measurements. Resonance energy transfer (RET) is a th rough space interaction which is mostly independent of the intervening solvent. The term RET is preferred over FRET since this process does not involve emission and reabsorption of photons. The theory of RET is based on the concept of a fluorophore as an oscillating dipole, which can exchange energy with another dipole with a similar resonance frequency. In principle, the orientation of the donor and acceptor can prevent energy transfer between a closely space D-A pair, but such a result is rare. Hence one can assume that RET will occur if the spectral properties are suitable and the D-A distance is short enough.65 In the presence of an acceptor, the donor decay becomes significantly nonexponential, especially at higher concentrati ons of the acceptor. The origin of the nonexponential decay is the time-dependent distribution of acceptors around the excited donors. At short times, following excitation, there exist more donors with nearby acceptors. These decay more rapidly because of the distance dependence of energy transfer. At la ter times, the donors with nearby acceptors have decayed, and emission results preferentially fr om donor without nearby acceptors. The decay of these donors is longer, owing to a slower rate of RET to more distant acceptors. The distance at which RET is 50% efficient, called the Frste r distance, is typically in the range 20-60 The 56

PAGE 57

rate of energy transfer from donor to acceptor ( kT) is given by Equation 2-14, where D is the decay time of the donor in the absence of acceptor, r is the donor-to-acceptor distance, and R0 is the Frster distance 6 01 r R kD T (2-14) The Frster distance is define d as in Equation 2-15, where 2 is the orientation factor, is the refractive index of the medium and QD is the quantum yield of the donor in the absence of the acceptor 6/1 42211.0JQ RD o (2-15) The normalized fluorescence intensity ( FD) of the donor in the absence of the acceptor and the extinction coefficient of the acceptor ( A) are related to J( ) as shown in Equation 2-15 0 0 4)( )()( )( dF d F JD AD (2-16) The transfer efficiency is typically measured using the relative fluor escence intensity of the donor, in the absence ( FD) and presence ( FDA) of acceptor. The transfer efficiency can also be calculated from their lifetimes ( D and DA), as shown in Equations 2-17 and 2-18 D DAF F E 1 (2-17) D DAE 1 (2-18) In Equations 2-17 and 2-18, it was assumed th at the lifetime of the donor was not altered by binding of the acceptor, other than by the rate of energy transfer. Allo steric interactions between the donor and acceptor sites could alter the donor lifetime by enhancing other decay 57

PAGE 58

processes. Under these circumstances, more complex analysis of the apparent transfer efficiency is required, typically by a co mparison of the apparent e fficiency by donor quenching and enhanced acceptor emission.65, 72, 73 Quenching of the fluorescence lifetime of the donor can be used to study RET, considering a few assumptions, including the following: Formation of static ground state complexe s does not decrease the decay times of the uncomplexed fluorophores, because only the unquenched probes are observed in a fluorescence lifetime experiment. Dynamic quenching is a rate process acting on th e entire excited-state population and thus decreases the mean decay time of the excited-state population. A fluorophore usually displays the same radiative decay in different environments. There are two cases for quenching of the lifetime of a fluorophore, static and dynamic quenching. Static quenching is based on the form ation of a nonfluorescent complex between the donor and the quencher. When this complex abso rbs light, it immediatel y returns to the ground state without emission of a photon. For static quenching, there are no changes in the fluorescence lifetime of the donor ( D/ DA =1), since the uncomplexed fraction is unperturbed. In a collisional quenching process, a decrease in the lifetime of the donor occurs because quenching is an additional rate process th at depopulates the excited state. At th e same time, there is a decrease in the yield of the fluorescence because quenching depopulates the excited state without emission. As a consequence, the decrease in the intensity is pr oportional to the change in the fluorescence lifetime ( D/ DA= FD/ FDA). Collisional quenching of fluorescence is described by the SternVolmer (S-V) Equation ][1][10 0QKQk F FD q (2-19) In Equation 2-19, F0 and F are the fluorescence intensities in the absence and presence of the quencher, respectively, kq is the bimolecular quenching constant, 0 is the lifetime of the 58

PAGE 59

fluorophore in the absence of quenc her, and [Q] is the concentration of the quencher. If the quenching is known to be dynamic, KD is used. A linear S-V plot is indicative of a single class of fluorophores, all equally accessible to the quencher. If two fluorophore populations are present, and one is not accessible to the quencher, then th e S-V plots deviate from linearity towards the xaxis. Observation of a linear S-V plot does not prove a collisional quenchi ng, since, under some circumstances, static quenching al so results in linear S-V. The be st way to distinguish them is by lifetime measurements.65 2.2.6.2 Time-correlated single photon counting (TCSPC) In this technique, the sample is repeatedly excited using a pulse light source. The time measurement clock starts when the sample is ex cited. The start signal is used to trigger the voltage ramp of the time-to-amplitude convert er (TAC) and it is stopped when the first fluorescence photon of the sample is detected. Th e TAC provides an output pulse whose voltage is proportional to the time between the start and stop signals. Subsequently, a multichannel analyzer converts this voltage to a time cha nnel using an analog to digital converter. The experiment is continued until more than 10 000 c ounts are collected in the peak channel, to obtain an optimal signal to noise ratio. The instrument used in this study is a home-made singlephoton-counting system. For data acquisition, the hardware (control electronics and the multichannel buffer (MCB)) is from PRA and the software used was Maestro-32 (version 6.05) from Ortec. For data analysis (curve-fitting), the software used was DAS6 from IBH, which is based on the Nonlinear Least-Squares analysis (NLLS). Its goal is to obtain estimates of parameter values which have the highest probability of being correc t. These values would provide the best match between the data N(t) and the calculated decay Nc(t). This is accomplished by minimizing the 59

PAGE 60

goodness-of-fit-parameter 2, where n is number of channels, and is the standard deviation of each data point n c n ctN tNtN tNtN1 2 2 1 2 2)( )()( )()( 1 (2-20) A multiexponential model was used to fit the da ta, assuming that the decay is the sum of individual single-exponential decays, as shown in Equation 2-21, where i is the amplitude of the components at t = 0, i is the decay times and n is the number of decay times n i i it tI1)/exp( )( (2-21) This is the most common used model, but the meaning of the parameters (i and i) depends on the system studied. The most obviou s case is to a mixture of fluorophores, each displaying one of the decay times, i. If the probe can exist in two environments, such as exposed and shielded from the solvent, th en a decay time can be assigned to each of these states. In this case, which is the one we worked with, it is gene rally safe to assume that the fluorophore has the same radiative decay rate in each environment and the i values represent the fraction of the molecule in each conformation at t =0, which corresponds to the ground state equilibrium.65 2.2.6.3 Determination of the degree of solvation of dansyl chloride by fluorescence lifetime measurements Dansyl chloride has been extensively used as a fluorescence probe after its introduction by Weber.74 The family of dansyl probes shows characteristic changes in its fluorescence maxima and intensity, which have permitted their use to determine polar and nonpolar sites in macromolecules.75 The fluorescence band of dansyl chlori de is associated with a transition involving an excited state having a considerable charge-transfer ch aracter, due to the promotion of a lone-pair electron of the amino group into an antibonding orbital of the naphthalene ring, 60

PAGE 61

called twisted intramolecular ch arge transfer (TICT). TICT phenomenon occurred in some especial fluorophores, which may have two parts in their structure, a donor and an acceptor part. When excited, the donor part may rotate around a single bond and transfer an electron to the acceptor part. Therefore there exists two excited states, one is the TICT and the other is the non charge transfer state.76 Because of its nature, th e energy of the TICT state is very sensitive to the polarity of the solvent.77 It has been identified that changes in polarity of the media where the dyes resides can be correlated to variations in the emission maximum of the dye, which has been used extensively in the literature.65 Similarly, dansyl chloride can be used to monitor the local viscosity of different media, as demonstrated by Leezenberg and Frank.78 If a dansyl probe is locked in a rigid environment, solvent dipoles may not relax around the excited molecules and lower its emission. Moreover, a rigid environment hinders the r earrangement of the naphthalene group, increasing the energy of th e excited state, observed as a blue shift in its emission.79 2.2.6.4 Experimental settings The samples were characterized by steadystate fluorescence spect roscopy prior to the lifetime measurements, to obtain the maximum in the excitation and em ission spectra of the encapsulated probe. In some cases, the intensity of the doped particles wa s slightly low, which required performing the experiments for longer than 30 minutes to obtain a good signal to noise response. The data obtained was analyzed with different exponentials, taking the 2 value as an indication of the good fit between th e experimental and obtained data being the valu es closer to one considered the best. 2.3 Experimental Procedures 2.3.1 Preparation of Templates Poly(ethylene[20]-sorbitan monool eate) (t80) and sodium octani ate (SO) were mixed with 13.4 mL of saline (9 g NaCl in 1 L deionized wa ter) at room temperature (RT) in an orbital 61

PAGE 62

shaker (J-Kem Scientific, model BTS 1500) at 300 rpm. After 4 h, octadecyltrimethoxysilane (OTMS) was added. The mixtures were placed in an orbital shaker at 300 rpm, and heated to 75 0.5 C overnight to promote the inclusion of OTMS in the system. After cooling down to RT, the pH of the solution was increased to approx imately 10 using a 0.5 M NaOH aqueous solution. After 30 min, the pH was decreased to 7.4 usi ng a 0.5 M HCl aqueous solution. Prior to any characterization or fluorescence studies, the temp lates were purified through 25 nm pore size membranes (Millipore). Finally, the concentrated suspensions were diluted to their original volume, sonicated for 15 min and filtered with 450 nm pore size filters (Fisherbrand). For dopedtemplates, the surfactants were mixed with the dopant and heated until a homogenous suspension was obtained. Subsequently, OTMS was added and the synthesis continued as explained for the dopant-free templates. 2.3.2 Synthesis of Core-Shell Nanoparticles The core-shell particle synthesis started with the templates formation as explained in the previous section. For dye-load ed particles, the dye, pyrene (system NPY) or coumarin 153 (system NCU), was mixed with the surfactants at room temperature in the first step of the synthesis. For the NEB system, ethyl butyrat e (EB) and 1-dodecene were mixed with the surfactants at 75 0.5 C for 4 h. This was followed by OTMS addition as explained for the synthesis of the templates. Once the templates were obtained, a 5-fold dilution of the original templates suspension was necessary before adding tetramethoxysilane (TMOS), in order to avoid crosslinking. TMOS was added in small portio ns (approximately 20 mg every 24 h) and the solution was allowed to stir until the size reach ed the desired range, as observed by DLS. Unlike the templates, the core-shell particles could be centrifuged. Based on this, the nanocapsules were centrifugated in a Sorvall RC 5B centrifuge (G MI, Inc) at 2990xg for 80 minutes to remove excess reagents. The nanocapsules were recovered as a pellet and resuspended by sonication for 62

PAGE 63

approximately 1 h. This process was repeated twice. After resuspension, the samples were filtered with 450 nm pore size filters. 2.3.3 Coating of the Particles with 3-aminopropylsilane The particles were not purified before r eacting with 3-aminopr opylsilane (APS). The suspensions were diluted 5 times with Millipore water and the pH was adjusted to approximately 9.5 with a NaOH 0.5 M solution. The suspension was placed in an oil bath and the temperature was increased to 70 C. APS was added in sma ll portion of approximately 15 mg every 4 h until the desired amount to add was reached. After this the suspensions were purified as explained earlier by centrifugation. 2.3.4 Synthesis of the Complex between 3aminopropylsilane and Dansyl Chloride Dansyl chloride (DCl, 1.00 g, 3.71x10-3 mol) was dissolved in 80 mL of CH2Cl2. Separately; APS (0.8 mL, 4.6x10-3 mol) was dissolved in 20 ml CH2Cl2. The two solutions were combined in a 250 mL round bottom flask under Ar. Finally, triethylamine (0.6 mL, 2.5x10-3 mol) was added to neutralize the acid formed. The mixture was stirred at RT for 3 h, after which the solvent and excess triethylamine were evapor ated under vacuum. The so lid left was dissolved in THF and the insoluble hydrochloride salt wa s removed by filtration. The product (APSDCl complex) was obtained as a yellow viscous liqui d. The excess DCl that did not react was not separated, since due to the silanol groups attach ed to the probe, column chromatography was not possible to use. After reaction with the particle s, we expected the nonderivatized dye to be removed from the suspensions during the dialysis, as observed by UV. 2.3.5 Coating of the Templates and CoreShell Particles with APSDCl Complex. A suspension of the particles or templates before purification was mixed with excess of the APSDCl complex at pH 9.5. The suspensions were left stirring in the dark for three days, after which the excess solid dye was removed by filtration. The nonreacted probe and nonderivatized 63

PAGE 64

dye were removed by dialysis with membrane s with a 6-8000 Da MWCO (Spectrum Labs). After this, the core-shell particles were centrif ugated and the templates we re purified with the 25 nm pore size membranes. Subsequently, the sa mples were resuspended and diluted to their original volume. 2.3.6 Loading the Dyes to Templates and Core-Shell Particles A volume of the nanocapsule or templates suspensions (purified and resuspended to its original volume) was mixed with the proper volume of a stock so lution of the dye dissolved in methanol, to obtain the desired final concentrat ion of the dye. The samples were vortexed and then stirred for 1 h. In the cas e of energy transfer studies, the donor was added to the suspension and the system vortexed and stirred for 30 min. Subsequently, an aliquot of the acceptor was added to reach the desired concentration, stir red for 15 minutes and the fluorescence recorded. This was repeated until the maximum concentration was reached. 2.3.7 Fluorescence Images A drop of the nanocapsule suspension was pl aced on a clean glass s lide (Corning Inc.) and dried at room temperature. Imaging was pe rformed on a fluorescence microscope (Olympus, model IX70) with excitation maximum at 360 nm (+/20 nm) and emission maximum at 525 nm (+/25 nm). 2.3.8 Steady State Fluorescence Spectroscopy Steady state fluorescence spectroscopy m easurements were carried out in a FluorologMax-3 (Horiba Jobin Yvon) with the fo llowing set up: 2 mm slits and 1 s integration time. The dye-doped particles were dialyzed until no more dye could be removed. After this, they were stored at room temperature. Before th e measurements, they were diluted five times to reduce scattering of light due to the size of the pa rticles. It was not possible to avoid the presence of the scattering peaks, as obser ved in the emission spectra of C153, but these appeared far from 64

PAGE 65

the maximum. As a consequence, no complications in the determination of the maximum in the emission of the dye were encountered. The susp ensions were stirred for one hour before the measurement to homogenize them. The dye-loade d particles were diluted and homogenized similarly to the doped particles. After addition of the dye, the sa mples were vortexed and stirred for one hour before measuring their emission. Figure 2-1. Possibilities for th e interaction of a laser beam with a liquid sample. In a homogeneous medium, waves of the scattere d light interact dest ructively, producing no scattering (A); while in a heterogene ous medium scattering is produced (B). 65

PAGE 66

0.51.01.52.02.53.0 0.0 0.2 0.4 0.6 0.8 1.0 C(t)log t Figure 2-2. Comparison of the correlation functi on of two set of templates with different diameters as determined by DLS. The diameters obtained were 47 nm (solid line) and 72 nm (empty circles). Figure 2-3. A simple representation of the basic concept of a transmission electron microscope operating in the bright field mode. 66

PAGE 67

Figure 2-4. A simple representation of the basic setup of the atom ic force microscope. Figure 2-5. Data analysis of a Z-axis ca libration grid performed in contact mode. 67

PAGE 68

CHAPTER 3 SYNTHESIS AND CHARACTERIZATION OF SURFACTANT-BASED TEMPLATES FOR THE FORMATION OF CO RE-SHELL PARTICLES 3.1 Preparation of Surfactant-Based Templates 3.1.1 Strategy The synthesis of core-shell particles proposed in this work is based on the formation of templates with reactive groups on th eir surface that can be used to grow a shell, as shown in Figure 3-1. The first step of the synthesis involves the formation of spherical micelles, to which a reactive amphiphile is added. Si nce micelles can be broken down by dilution, the reactive groups in the micelles will be chemically connected in subsequent steps to fortify them. The unreacted groups left after this process can act as anchor ing points to grow a she ll that is chemically connected to the core, fixing the shape and size of the species obtaine d. This would produce more robust particles which would not be affected by changes in concentr ation, ionic strength, pH or the like. We decided to use OTMS as the reactive amphiphile due to its well-known chemistry (see Figure 3-2 for its chemical structure). The chemical properties of OTMS have been extensively studied, especially at the airwater interface, with empha sis in the influence of the pH in the hydrolysis and conde nsation of its methoxy head group.80-82 It has been established at the air-water interface that lo w pH (3.9-4.8) and high pH (10.3-11.7) promoted the hydrolysis and condensation of its methoxy groups, althoug h totally hydrolyzed or totally condensed species are not obtained unless the pH is lower than 3.9 or higher than 11.7. 3.1.2 Results Based on the information about the reactivity of OTMS discussed above, we decided to use this molecule to fix the structure of the te mplates obtained through its hydrolysis-condensation chemistry. In previous work in our group,83 t80 was used to produce spherical micelles with diameters of approximately 7 nm, as observed by DLS.84 Subsequently, OTMS was added to the 68

PAGE 69

suspension to produce the templates. The first pr oblem encountered was the low miscibility of t80 and OTMS. To overcome this, the system was heated to 75 C afte r the addition of the chemically active amphiphile. At this temperatur e, the ethylene glycol chains in t80 become more hydrophobic which promoted the incorpora tion of OTMS into the micelles already formed.85 This approach formed the desired te mplates but the yield was low and excess surfactant was left in the suspension, which ma de the purification process tedious. Preliminary studies showed that using increasing amounts of OTMS in the suspension did not produce more templates, but instead resulted in phase separation. To overcome this problem, a combination of t80 with sodium octanoate was selected. This sm all surfactant is well know n for its extreme self assembly behavior, forming small aggregates an d having a high critical micelle concentration (CMC).86 Based on this, it has been used in different biomedical applications, especially to promote mixing of the different components of microemulsions.87 In the study presented here, the concentration of t80 was kept above its CMC, to assure that spherical micelles were present in the suspension, while the concentration of SO was maintained below its CMC to promote its addition to the already formed t80 micelles. Afte r addition of OTMS and heating, the suspension changed from transparent to white-blue, which is a clear indication of the increase in the size of the species obtained. Subsequently, the me thoxy groups in OTMS were hydrolyzed and condensed by increasing the pH of the suspen sion to approximately 10.5 for 30 minutes, as shown in Figure 3-3. We chose to work under basic conditions due to the presence of sodium octanoate in the formulation, whic h would precipitate at low pH. As explain earlier, at this pH the condensation process would not produce a totally condensed netw ork, but we decided not to use a higher pH to avoid harsh c onditions that could affect the other components of the system. After this reaction, a lose network had been formed, producing the templates. The particles 69

PAGE 70

obtained are strong enough that they could be diluted wit hout disruption of the system, as observed by DLS. In any case, the condensati on process would continue at a lower pace, producing a tighter network at pH 7.4, conditions at which the templates are stored. 3.1.2.1 Size analysis Different SO/t80/OTMS mixtures were prepared to study the influence of each component on the size and shape of the templa tes obtained. Table I shows the formulations used and the size characterization results. Samples A, B and C were prepared with different SO/t80 mol ratios, and each one was mixed with three different amount s of OTMS. DLS was used for size analysis before the addition of OTMS to the three original samples, showing low intensity of the scattered light and diameter values between 6-8 nm in all cases, corres ponding to t80 micelles.84 This indicates that the inclus ion of SO in the system did not perturb the size of the micelles, probably due to the small size of the anionic amphiphile which allows its packing in between the poly(ethylene glycol) (PEG) chai ns of the t80 polar head. Af ter addition of the reactive surfactant, the transparent suspensions becam e white to various degrees, depending on the amount of OTMS added. The rou tine size analysis of the templa tes and core-shell particles was performed by DLS. This is a more convenient wa y to characterize the particles compared to the microscopy techniques, since DLS characterizati on is performed in solution, with conditions similar to what we expect the particles to be us ed in different applications. Before analysis, the suspensions were purified by passing them through a cellulose membrane w ith a size cut-off of 25 nm. Subsequently, the suspensi ons were diluted to their orig inal volume and sonicated to separate any aggregates formed. This proce ss removed the excess surf actants used for the synthesis of the templates and the ions present in the original mixture, as observed by DLS in Figure 3-4 and -potential, in Figure 3-5. It was obse rved that if the particles were not completely dried during the purification, they were easily resuspended after 15 minutes of 70

PAGE 71

sonication. If all the solvent was removed, the templates formed a paste which required extensive sonication to be resuspended, and a significant mass of particles was lost when the suspensions were filtered. The DLS graphs show that after purification of the templates the signal for the micelles was lost. This indicated that the surfactant mo lecules were removed of the system, which was immediately noticeable when the suspensions were shaken, as no bubbles were formed. This was also confirmed by analyzing the electrophoretic mobility of the samples, as shown in Figures 35. In a typical -potential measurement, the particles are suspended in a soluti on of an electrolyte and the velocity at which they moved when an el ectric field is applied is measured. From this, the electrophoretic mobility of the particles is directly calculated and this value is used to calculate the -potential of the particles. The sign of its value co rresponds to the charge of the particles and its absolute value is an indication of the stability of the particles in suspension. Large -potential values indicate a low tendency for aggregation. In this study, it was observed that the -potential of the templates varied differently when the pH of the suspensions was changed before and after purification. The pure templates showed more negative (less positive) values than the nonpure ones. The -potential of the pure templates was similar to the ones obtained by Katagiri et al. for colloids coated with a lipid bearing a triethoxysilil moiety as the polar head,88 and more negative that the ones obtained by Bourgeat-Lami et al. for SiOH functionalized polymer latexes.89 The difference after purifying the templates is believed to be a consequence of the presence of the non ionic surfactant (t80) adsorbed on the surface of the templates before purification, which hides th e charges of the silano l groups, due to its poly(ethylene glycol) polar head.90 This resulted in the diffusion of the templates being less 71

PAGE 72

affected by changes in the external fiel d applied to the su spensions, producing -potential values closer to zero. Size analysis by DLS showed that the samples labeled as 1 produced a small amount of templates, easily recognizable due to the pale white color of the suspensions and confirmed by the low intensity of the scattered light detect ed. Moreover, the size distribution was largely dominated by the micelles. Considering that the in tensity of the light scat tered by a particle is related to its radius to the sixth power,57 and the large difference in size between the micelles and templates, the high intensity of the signal observed for the micelles indicated that they outnumber the templates by a large factor. Samples labeled as 2 and 3, showed a white-blue color as well as high intensity of scattered light, indicating that th e number of aggregates, as well as their sizes, were higher than th e ones obtained in the sample set labeled as 1. In this case, the size distribution is largely domina ted by the templates, with just a small signal for the micelles, as shown in Figure 3-4 A. These results are an indication of the incorporation of OTMS into the aggregates, which suggests that SO is promoting this process, in creasing the yield of templates obtained. Samples A2 and A3 still included a high percentage of t80 in the formulation, but since one of our goals was to reduce the mass fraction of this surfactant, they were not further analyzed. As explained in the previous chapter, in the DLS technique, the diffusion coefficient of the scatterers is obtained from the va riation of the intensity of the li ght scattered by the particles due to Brownian motion. From this information a correlation function is ob tained, and the diffusion coefficient of the particles determined. This value is used in the Stokes-Einstein Equation to calculate the diameter of the particles, assumi ng that the particles are spherical. The values obtained by this technique are considered hydrodynamic, meaning that the size estimated 72

PAGE 73

includes any species adsorbed on the surface of th e particles, like ions, surfactant molecules and the like, which influence their motion in solution.57 As can be observed in the DLS results reported in Table I, the incorpora tion of OTMS in the system produ ced an increase in the size of the aggregates obtained of approxi mately ten times, compared to the SO/t80 micelles. This is clearly an indication of the in corporation of the reactive amph iphile in the micelles, although such dramatic effect on their diameter is not completely understood. It is known that the aggregation number in t80 micelles increases with the concentration of the surfactant by association of four primary micelles.91 The incorporation of OTMS in the system may be promoting this association, producing larger aggregates. The results for the samples B and C indicated that, for a fixed SO/t80 combination, larg er aggregates were obta ined by the addition of higher amounts of OTMS. In both cases, by incr easing the OTMS mol fraction four times an increment of approximately 40% in the diameter of the templates was observed. This suggested a mean to control the size of the particles. The stability of the templates was compared to that of the SO/t 80 micelles by analyzing their correlation function in the DLS instrument The suspension of the micelles without OTMS was used as is for the first measurement, detecting an intensity of scattered light of approximately 3.0 x 105 kcps. The suspensions of the templa tes was diluted to obtain a similar intensity of scattered light and the correlation f unction of each of these was obtained, as shown in Figure 3-6. Both samples were diluted until the signal was weak but still strong enough to obtain a good fit for the co rrelation function, which was observed to be 0.5 x 105 kcps. The data for the templates (Figure 3-6 A) shows that, de spite changes in the ba ckground signal for the correlation function, the results for the templates did not change, an indica tion that the particles did not break after dilution. Diffe rent results were observed for th e mixed micelles, as shown in 73

PAGE 74

Figure 3-6 B. In this case, the typical si gnal for the micelles was obtained at 3.0 x 105 kcps, a distortion of the normalized correlation func tion was observed upon dilution. The correlation function showed that the mice lles broke down and bigger species were formed. These results showed the effect of the fortif ication of the particle s after the condensati on of OTMS which was one of the goals of this work. To validate the results obtained by DLS we imaged the particles in the electron microscope, the results of which will be discussed in more detail in the next section. In all cases, spherical particles with low polydispersity were observed, independently of the mol ratio of surfactants used, as shown for samples B2, C2 and D2 in Figure 3-7. This confirmed that the particles were spherical; and that their diam eter can be obtained with the Stokes-Einstein Equation with the DLS software. As can be observed in Table I, a 1:1 mass ratio of SO/t80 produced smaller templates after the addition of OTMS; as a consequence, this combination was selected to study the formation of core-shell nanoparticles. The sa mple set D was prepared based on this combination but with a lower total mass of surfactant in order to avoid complications when purifying the nanocapsules. 3.1.2.2 TEM analysis To investigate the geometry of the templates we used the electron microscope. Samples for imaging were prepared by placi ng a drop of a dilute suspension of the templates on a formvar coated carbon grid and the solvent evaporated Figure 3-7 presents typical images of the templates. The size of the particles was determined from the images using Adobe Photoshop software and Image-J software, by comparison of th e diameter of the par ticles and the scale bar (Table 3-1). The values obtained were similar to the ones observed in the DLS analysis, although a slight overestimation of the diameter was observed by TEM, especially for samples D. This is interesting since DLS data is biased to the larg er particles, while TEM gives a number average. 74

PAGE 75

Based on this, larger values would be expected in the DLS results, th e opposite trend observed may indicate that the particles expanded after being deposited on the s ubstrate. One important detail for these images is the low electron densit y of the templates, which gave a low contrast when imaging. This means that the electron beam did not encounter en ough electron density when passing through the particles, which indicate s that the templates are porous. In the images showed in Figure 3-7 and 3-8, ur anyl acetate was used to stain the templates prior to imaging. This staining agent is used to interact with hydr ophilic regions in biolog ical systems, increasing the electron density of these areas.92 It was observed that the st aining improved the ease of imaging by making the periphery of the particles more defined; hen ce, size analysis was easier to perform. Figure 3-8 shows evidence that the part icles flattened after the solvent is evaporated. The red boxes showed what are believed to be pa rticles that were deposite d with their longer axis perpendicular to the surface of the grid, showing a side view of the templates. It is clear in these images that the particles show a disk-like geomet ry when dried. Moreover, this does not seem to depend on the use of a dopant in the synthesis of the templates, since the sample on Figure 3-8 B was prepared with ethyl butyrate as a dopant, as explained in the ne xt sections. Interestingly, the longer axis of these particles is similar on average to the spherical ones, confirming that they are the same kind of species. It is important to menti on that this side view of the templates was not commonly found when imaging, and they seemed to occur when the concentration of the templates in the suspension used to load the TEM grid was high. Anot her significant detail comes from the fact that these images are different from the ones typically obtained for hollow particles, in which a thin skin, th at usually collapses, is visualized.93 This may indicate that the OTMS condensation is not restricted to the surf ace of the templates but it occurred in their interior as well. 75

PAGE 76

We performed a simple calculation to have an idea of the extent the particles expanded on a substrate. We calculated the area of the templa tes D2, considering them as spheres, using the formula 4 r2, where r is the hydrodynamic radius of the templates (in this case, 32 nm). We then obtained the radius of the maxi mum area possible to obtain afte r flattening, considering the formation of two circular sheets, with total area of 2 r2. The radius of the totally flattened particles was determined to be 45 nm. Comp arison to the value obtained by TEM (41 nm) showed that, after drying on the substrate; the diameter of the templates reached 91% of the maximum possible, indicating a significant expansion. 3.1.2.3 AFM analysis Atomic force microscopy was found to provi de more information on the shape of the particles due to its ability to construct a 3-D image of the sample. Analysis of the images obtained by TEM suggested that the particles flattened after deposit ion on the substrate; hence, we decided to study the morphology of the templates into more detail. One of the advantages of AFM over TEM is that no staining is necessary to image the particles, so the probability of the images containing artifacts is reduced. Moreover, TEM imaging is performed under vacuum, while AFM imaging was performed under standard conditions. One significant difference in this analysis was the use of mica, a hydrophilic material as a substrate for AFM imaging, contrary to TEM analysis, in which hydrophobic gr ids were used. As a conseque nce, the interaction between the particles and substrate is expe cted to differ in each technique.61 Imaging was performed in tapping mode to avoid direct inte ractions between the tip and the particles that can cause damage to the sample (or the tip). In this mode, the tip of the cantilever is set to interact with the sample through a constant force, without touching the surface but interacting with it probably through the layer of water adsorbed on the surface.94 76

PAGE 77

Typical images showed that the particles form ed aggregates of a few micrometers in size after drying, similarly to what Jungmann et al. observed in the characterization of nanospheres made of poly(organosiloxanes).52 We believe these are formed during the solvent evaporation, since these same suspensions were analyzed by DLS and no aggregates were detected. The only difficulty encountered during imaging was related to the sample concentra tion, if this was too high the particles were deposited as multilayers. Th is produced large differences in the height of different parts of the sample, whic h was difficult for the tip of the cantilever to track, resulting in images with poor resolution. This was avoide d by using very dilute suspensions. Similar problems arose when templates and core-shell part icles with large diameters were imaged, since the differences in the features on the z-axis be came difficult for the inst rument to track. Both problems were more frequent with the core-shell particles than with the templates. Templates D2 and D3 were selected to study their topographic featur es by AFM, as shown in Figure 3-9, to complement the results obt ained with TEM and DLS. The images were analyzed with the Nanoscope 5.3r software, obs erving spherical particle s when a top view was used. The section analysis option in the software allows an accurate esti mation of the diameter and height of the particles. We used this feat ure to study the flattening of the particles after deposition on the substrate suggested by the TEM images. As expected, it was observed that the height of the templates was considerably lower than their diameter and the latter was larger than the ones obtained in the DLS st udies. Confirming what the TEM images suggested, the templates did not look like hollow part icles made of a thin skin, which usua lly appear as a thin film of the material that collapses.95 Interestingly, when the diamet er observed for D2 and D3 were compared, a larger diameter for D3 was matched by a larger height value, as shown in Table 3-2. We believe this is a consequence of the highe r amount of OTMS includ ed in the formulation 77

PAGE 78

used to prepare the templates labeled as D3. Hi gher loadings of OTMS offered more head groups to condense which would produce a more rigid system, preventing the templates from spreading as much on the substrate. We used this flatte ning of the templates as an indication of their flexibility, expressed as the diameter to height ra tio (DHR). This value will be compared later to the corresponding core-shell particles to study the e ffect of the coating of the particles on their flexibility. The DHR values are 10.4 for D2 and 8. 9 for D3, showing an in crease in the rigidity (lower DHR) as a consequence of the increas e in the OTMS loading in the templates. Comparison of these values to the ones obtained by Wang et al. who obtained DHR values above 20 for hollow particles, indicated that, despite being flexible, these templates are not hollow.50 3.2 Preparation of Doped Templates The templates studied above can have different potential applicati ons, including their use as nanoreactors, drug delivery systems, agents for sequestration and encapsulation, and the like. It is desirable to have control over the size of the particles to meet the requirements of the application in mind. It was obser ved on the results showed in the previous sections that changing the mol fraction of OTMS offered some degree of control over the diameter of the templates, but this was limited by phase separation if the fractio n of OTMS was high. As an alternative, the templates could be prepared incl uding a dopant. Molecules of diffe rent polarities and sizes could be used to fill the interior of the droplets formed before the addition of OTMS, which would allow control over the diameter of the precursor s of the templates. After purification, these molecules, depending on their polarity, could be removed by dialysis or other means, leaving pores in the interior of the templates that could be used to trap other species, especially if they are nonpolar. To explore these ideas we prepared templates with two different dopants, one 78

PAGE 79

moderately polar and small, ethyl butyrate, a nd one nonpolar and larger, hexadecane (see Figure 3-2 for structures), as shown in Table 3-3. 3.2.1 Results 3.2.1.1 Size analysis of ethyl butyrate-doped templates The formulations used to prepare the ethyl butyrate-doped template s (EB) are shown in Table 3-3, as well as the size characterization results obtained by DLS. The EB templates were prepared similarly to the templates labeled as D2, but the dopant was mixed with the surfactants in the first step of the synthesis. It was observe d that the suspension needed to be heated up to obtain a homogeneous suspension and that the tim e needed depended on the dopant used, in this case 2.5 h. After this, OTMS was added and the mixture heated overnight as explained above. Samples G, H and I were not considered suc cessful systems since a significant amount of material was lost when the templates were pa ssed through filters with 450 nm pore size. As a consequence, they were analyzed by DLS only. Th e size analysis reported in Table 3-3 showed that the size of the templates co uld be varied by changing the c oncentration of the dopant. An increase in size of approximately 30% was obs erved when the concentration of the filler molecule was doubled, and 50% when it was tripled. Differently to the te mplates studied above, the concentration of OTMS did not seem to have a significant effect on the diameter of the templates, but on the amount that were formed. Th is was clearly observed by the scattering of the samples after the formation of the templates was finished. We then used the atomic force microscope to study the topographi c features of these templates. 3.2.1.2 AFM analysis of ethyl butyrate-doped templates The ethyl butyrate-doped templates were imag ed with the atomic force microscope, and analyzed similarly to the set D, as shown in Table 3-4 and Figure 3-10. The size increase detected by DLS was observed in a similar way in the AFM images. The templates showed, in 79

PAGE 80

all cases, spherical particles and a higher polyd ispersity compared to D2 and D3. The data on Table 3-4 showed that the increase in the diameter of the particles matched an increase in their height after deposition on the s ubstrate. Contrary to the templates studied before, a significant number of hollow particles were identified in the images, as indicated with black arrows in Figure 3-10. We believe these are templates that are squeezed in between other ones. This is an indication of the influence of the dopant in th e final properties of the templates, producing particles that retained their geometry after purif ication, but are flexible enough so that they can be folded without breaking. This opposes the te ndency observed with the DHR values, which are lower than the D set, indicating a more rigid system. Nonetheless, in each pair when more OTMS is used in the synthesis of the templates, a more rigid system is obt ained, as explained for the D templates. This disagreement may be a consequence of the larger diameter of the templates obtained when dopants are load ed in the templates. 3.2.1.3 Size analysis of hexadecane-doped templates The hexadecane system showed a similar trend that the EB templates (Table 3-5). In this case, the amount of the dopant us ed was lower, since an incremen t of its concentration did not produced a homogeneous suspension, even ove r longer periods of heating. Hence, the concentration of hexadecane was kept at approxima tely one order of magnitude lower than ethyl butyrate, despite using the same concentration of surfactants. In this case, the samples were heated for 3.5 h. DLS analysis showed that th e size range obtained was similar to the ones prepared with ethyl butyrate. Moreover, when the hexadecane concentration was doubled, a 30% increment in the diameter was observed. This shows that the diameter of the templates can be tuned by varying the concentration of the dopant. 80

PAGE 81

3.2.1.4 AFM analysis of hexadecane-doped templates Images obtain with AFM and the corresponding section analysis for the hexadecane-doped templates are shown in Figure 3-11. These template s looked similar to the ethyl butyrate system, with a significant number of hollo w particles observed in all the cases. This indicates that this is a trend for the templates obtained when a dopant is used in their synthesis. Comparison of the results for the HxE and HxF templates suggested that, at a higher hexade cane concentration, the amount of OTMS plays more control over the flexibility of the particles. Despite having the same diameters in solution, as observed by DLS, their dimensions are considerably different when imaged by AFM. Templates HxE showed a larger diameter and height, which seemed to be due to the lower OTMS content. 3.3 Other Systems Studied 3.3.1 Use of a Cationic Surfactant in the Synthesis of the Templates In an attempt to develop a robust approach to the synthesis of the templates we tried to extend the choices of surfactants that could be used in this methodology. We believe that the encapsulation of anionic species would be promot ed by the use of a cationic surfactant. As a consequence, we explored the use of dodecy ltrimethylammonium bromide (DTAB) for the formation of the precursors of the templates instead of sodium octa noate. Figure 3-12 shows AFM images of two sets of templates obtained w ith different surfactant loadings, and diameters of 53 nm (A) and 34 nm (B), as obtained by DLS. These templates looked very similar to the ones obtained with SO without dopan ts. Work is needed to determine if these templates could be used for the synthesis of core-shell particles. 81

PAGE 82

3.4 Preliminary Studies of the Strength of the Templates 3.4.1 Strategy The templates synthesized in this work were formed by the alkyl chain of the OTMS molecules and their chemically attached h ead groups. We believed, based on the DLS and potential data (Chapter 2), that most of th e t80 and SO molecules are separated from the suspensions during the purificati on steps. The particles flattened but did not collapse when deposited on a substrate, which i ndicates that they are made of a thick skin of inorganic material, probably with pores in their in terior where the tails of OTMS could be accommodated. These pores were probably formed by aggregation of the surfactant and dopant molecules during the condensation of OTMS. To investig ate the strength of th e structures formed, they were heated up slowly to temperatures at which the organic tails of the amph iphile would be removed. We investigate what kind of species were obtai ned after this process by TEM and AFM. 3.4.2 Results 3.4.2.1 Calcination of dopant-free templates Figure 3-13 shows TEM images of dopantfr ee templates with diameter of 60 nm, as obtained by DLS. Figure A shows the typical image of templates as reported in the previous sections. Figures B, C and D showed what was found in the crucible after the calcination was done. The solid left was scratched although not all of it could be removed, which means than some other species could be present in the final mixture. In any case, most of what was observed looked like fibers (Figure 3-13 B and C), although a few dark, s pherical species with different sizes were observed as well (Figure 3-13 D). In general, it looked like the templates were destroyed and what was left was a fiber-like mate rial, probably made of what originally was the siloxane network formed by the polar heads of OTMS, since the hydrocarbon part of the particles was removed. Similarly to the templates, thes e fibers showed a very low electron density. 82

PAGE 83

Interestingly, in the AFM images (Figure 314), more species similar to the original templates were observed. Although they are not comp letely spherical, there seems to be a variety of particles with different sizes, some of which have the same dimension of the templates. This may indicate that the templates broke into smalle r pieces during calcination. It is not clear why these species were observed more frequently on AFM compared to TEM. It is possible that the fibers observed in the electron microscopes were very thin (as indicated by their low electron density) and thus, difficult for th e tip of the cantilever to dete ct them, although Figure 3-14 D look similar to the images obtained by TEM. 3.4.2.2 Calcination of ethyl butyrate-doped templates Similar results to the ones obtained with dopant-free template s were observed when using ethyl butyrate as a dopant. Figure 3-15 A shows th e original spherical templates, which were not found after the calcination, as shown in Figures B, C and D. Small particles with different sizes as well as fibers were observed. Images obtai ned by AFM (Figure 3-16) showed particles of different sizes and shapes, similar to the previous case, confirming that the particles tend to break apart at high temperatures. In the Chapter 4 results for similar studies for the core-shell particles will be examined. 3.5 Conclusions This chapter focused on the synthesis and char acterization of templates for the preparation of core-shell particles. Our goa l was to obtain spherical partic les with chemical groups on the surface that could be reacted in subsequent step s with a coating agent to build a shell on their periphery. A ternary amphiphile system was used, including octadecyltrimethoxysilane as a chemically active surfactant. The heads groups of OTMS were covalently connected to fix the shape of the structures obtained. Dynamic light scattering showed that templates of sizes in between 60 to 80 nm were obtained. Transmissi on electron microscopy images were used to 83

PAGE 84

confirm that the particles were spherical. Thes e images suggested that after deposition on a substrate and solvent evaporation, the particles flattened, giving disk-lik e particles. This was confirmed using the atomic force microscope, which let us obtained more details on the geometry of the particles through the section analysis option in its software. It was observed that the size of the particles could be varied by the amount of OTMS loaded in the formulation used to synthesize the particles, although this was li mited by phase separation. To have more control on the size of the particles obtained, dopant molecu les of different polarities were used in the synthesis of the templates. Analysis by DLS showed that this produced an increase in the diameter of the templates, as well as in their pol ydispersity. It was observe d that the polarity of the dopant used affected its maximum loading in the formulations, obtaining, in average, particles with hydrodynamic diamet ers in between 150 to 300 nm. Im aging with the atomic force microscope showed that the doped-templates were more flexible than the dopant-free ones. We were able to observe folded pa rticles which suggested that the dopant molecules were lost after purification leaving a porous and flexible material. In this case, the amount of doped-templates obtained seemed to be influenced by the amount of OTMS loaded in th e system. We started studies on the used of different su rfactants in the synthe sis of the core-shell particles, showing that DTAB could be used instead of SO. This may facilitate the use of anionic dopants through ionic interactions although this needs more work to be confirmed. In general, the templates can be diluted without affecting thei r size or shape, as observed by DLS and AFM, suggesting they are robust species. To test this we calcinated templates prepar ed with and without dopants. Preliminary studies based on AFM and TEM images showed that the partic les break apart after heating them up to 600 C. We believe the result s summarized here indicate that the templates are composed of a thick skin of a siloxane netw ork formed by the hydrolysis and condensation of 84

PAGE 85

the OTMS polar groups. This skin retains its spherical shape in solution but deflates after removal of the dopant, just like a soccer ball after removing air from its interior, producing flexible structures. We believe these particles are st rong enough to be used in solution in different applications. In the next chapter, we explore the coating of the templates with a silicate precursor and study how this affects their flexibility. Table 3-1 Formulations used for the preparat ion of templates and characterization results obtained by DLS and TEM. A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 tween80 (g) 1.0 1.0 1.0 0.7 0.7 0.7 0.4 0.4 0.4 0.5 0.5 0.5 SO (g) 0.4 0.4 0.4 0.7 0.7 0.7 1.0 1.0 1.0 0.5 0.5 0.5 tween80/SO (mol) 1/3 1/8 1/19 1/8 OTMS (mg) 11 20 42 10 22 43 7 22 45 11 31 45 tween80/OTMS (mol) 26 14 7 20 9 5 16 5 3 13 5 3 dDLS (nm) 81 65 47 54 64 63 75 88 50 64 72 dTEM (nm) 62 69 68 84 82 96 Table 3-2 Size analysis results obtained by DLS and AFM for templates D2 and D3. dDLS dAFM hAFM DHR D2 64 80 8 10.4 D3 76 86 10 8.9 All values in nm, d: diameter, h: he ight, DHR: diameter to height ratio. Table 3-3 Formulations used for the formation of ethyl butyrate-dope d templates, and size analysis results obtained by DLS. EBA EBB EBC EBD EBE EBF EBG EBH EBI SO 125 126 126 126 126 128 124 128 124 t80 125 123 128 123 128 129 126 124 126 EB 13 13 13 26 26 26 40 40 40 OTMS 13 26 44 13 26 44 13 26 44 dDLS 180 158 176 243 205 232 251 272 261 All quantities in mg, except diameter (d) in nm. 85

PAGE 86

Table 3-4 Topographic information for ethyl bu tyrate-doped templates obtained by AFM. DLS data is shown for comparison. EBB EBC EBE EBF dDLS 158 176 205 232 dAFM 184 232 245 260 hAFM 23 43 35 40 DHR 8.0 5.4 7.0 6.5 Table 3-5 Formulations used for the formation of hexadecane-filled temp lates and size analysis results obtained by DLS and AFM. SO t80 Hx OTMS dDLS dAFM hAFM DHR HXB 126 123 6 26 141 140 15 9.3 HXC 124 128 6 44 160 142 17 8.4 HXE 126 128 12 26 204 178 44 4.0 HXF 128 129 12 44 203 146 18 8.1 All quantities in mg, except diameter (d) and height (h ) values, in nm. Figure 3-1. Schematic representation of the synthesis of core-shell particles. 86

PAGE 87

Figure 3-2. Chemical structures for surfactan ts, amphiphiles and dopants used for this study. 87

PAGE 88

Figure 3-3. Ideal schematic for the hydrolysis and condensation of the methoxy groups in OTMS at basic conditions. Figure 3-4. Size distributi on obtained by DLS for a typical set of templates A) before and B) after purification. 88

PAGE 89

234567891011 -50 -40 -30 -20 -10 0 mVpH Figure 3-5. -potential variation as a f unction of pH for non pure (cir cles) and purified (squares) templates. Figure 3-6. Variation of the normalized correla tion function obtained by diluting a suspension of particles to reach different light scattering intensities. A) templates, intensity: 3.5 x 105 kcps-purple line, 2.5 x105 kcps-green line, 1.1 x 105 kcps-red line, 0.5 x 105 kcpsblue line, and B) SO/t80 micelles, intensity: 3.0 x 105 kcps-purple line, 2.0 x 105 kcps-green line, 1.0 x 105 kcps-red line, 0.5 x 105 kcps-blue line. 89

PAGE 90

Figure 3-7. TEM images for templates: a) B2, b) C2, c) D2. Scale bars 200 nm. Figure 3-8. TEM images showing what are believed to be templates deposited with their longest axis perpendicular to the su rface of the substrate (indicat ed by the red boxes). Scale bars 500 nm. 90

PAGE 91

Figure 3-9. AFM tapping mode images for samples A) D2 and B) D3. On the bottom part of each image the section analysis is shown. 91

PAGE 92

Figure 3-10. AFM images and section analysis for templates A) EBB, B) EBC, C) EBE and D) EBF. The arrows indicate what are believed to be folded templates. 92

PAGE 93

Figure 3-11. AFM images and section analysis fo r templates A) HxB, B) HxC, C) HxE and D) HxF. The arrows indicate what are believed to be folded templates. 93

PAGE 94

Figure 3-12. AFM image of two set of temp lates prepared with DTAB replacing SO. Figure 3-13. TEM images of dopant-free templates before (A) and after (B, C, D) calcination. Scale bars 200 nm 94

PAGE 95

Figure 3-14. AFM images of dopant-free templates before (A) and after (B, C, D) calcination. Figure 3-15. TEM images of ethyl butyrate-dope d templates before (A) and after (B, C, D) calcination. Scale bars 200 nm. 95

PAGE 96

Figure 3-16. AFM images of dopant-free templa tes before (A) and after (B) calcination. 96

PAGE 97

CHAPTER 4 TEMPLATE-BASED SYNTHESIS OF CORE-SHELL PARTICLES 4.1 Synthesis of Core-Shell Particles 4.1.1 Strategy The templates studied on Chapter 3 were form ed with the idea of having reactive groups on their surface that could be used as anchoring points for the growth of a sh ell. In order to build a hydrophilic coating that would produce more r obust particles, we turn ed our attention to alkoxysilanes. This would allow us to work with the same kind of chemistry used for the synthesis of the templates. Our cas e would be similar to the synthesi s of a silicate, but instead of letting the monomers hydrolyze a nd condense to form particles, we needed to promote the reaction of the hydrolyzed monomers with the sila nol groups already present in the templates. As a consequence, we based the design of our synthesis on the knowledge of the preparation of silicate gels. In general, silicate gels are s ynthesized by hydrolyzing monomeric, tetrafunctional alkoxide precursors, employing a mineral acid or base as a catalyst. Under most conditions, condensation commences before hydrolysis is comp lete, as shown in Figure 4-1. Because water and alkoxysilanes are immiscible, alcohols are used to homogenize the mixtures. However, gels can be prepared without a hom ogenizer, since the alcohol produ ced as the by-product of the hydrolysis reaction is sufficient to homogenize the system. The most common tetraalkoxysilanes used in the sol-gel process ar e tetraethoxysilane and tetramethoxys ilane. The steric bulk and the electron donor characteristic of the alkoxide or the organic substituent attached to silicon will largely determine the kinetics of the hydrolysis a nd condensation reactions. Therefore, the use of specific precursors is often dictated by kinetics considerations or compatibility with precursor or other network-forming elements in multicomponent silicate gel synthesis. Numerous investigations have shown that variations in the condi tions of the synthesi s cause modifications 97

PAGE 98

in the structure and properties of the polys ilicate products, although a consistent trend is apparent. Acid-catalyzed hydrolysis with low H2O:Si ratios produced weakly branched polymeric sols, whereas base ca talyzed hydrolysis with large H2O:Si ratios produced highly condensed particulate sols. Inte rmediate conditions produce stru ctures intermediate to these extremes.96, 97 In this study, we selected tetramethoxysilane as the silicate precursor to coat the templates. This was selected over tetraethoxysilane due to th e dependence of the rate of the hydrolysis of the alkoxysilanes on the ster ics of the substituents attached to the silicon center. We believe a faster hydrolysis would produce a rapid homogeni zation of the system, promoting the reaction between the silanol groups on the surface of the templates and the silicate precursor.97 The hydrolysis product, salicylic acid is known to gr ow primarily by addition of monomers to highly condensed particles, rather than by particle aggregation, at pH above 7. Moreover, as explained earlier, the condensation of the alkoxysilanes is expected to st art before the hydrolysis is completed, as shown in Figure 4-1. TMOS, being smaller, would encounter less steric hindrance to approach the surface of the templates and react with its silanol groups. This is important, since a competition with the reaction between the alkoxysilane molecules is expected, which would produce a gel and, possibly, smaller particles. We carried out the reac tion between TMOS and the hydrolyzed and condensed polar groups in OTMS on the surface of the templates at pH 7.4, since at this and higher pH th e reaction between larger, more highly condensed species, which contain acidic silanols, and smaller, le ss weakly branched species is favored.96 In order to investigate the effectiveness of the coating of the templates to produce coreshell particles, different nanoparticle systems were prepared to be charac terized before and after the shell growth by DLS, AFM and TEM. Based on the study of the formation of the templates 98

PAGE 99

discussed in the previous chapter, a combination of equal mass es of surfactants was selected for the synthesis of core-shell part icles (Table 4-1). The shell growth was determined by DLS, comparing the diameter of the te mplates and the coated particles.98, 99 An increase in the average size of the particles was consider ed as an indication of a successf ul coating. Once this process was finished, the nanocapsule suspensions were centrifuged at 1600xg twice, followed by sonication for approximately 1 h in order to resuspend the particles. Similarly to the templates, it was observed that after sonication and filtration, most of the original mass of particles was recovered. The diameter of the templates and the particles obtained by DLS after purification were compared to obtain the shell thickness (ST). Equation 4-1 was used assuming that one template produced one core-shell particle, where d1 is the diameter of template and d2 is the diameter of the core-shell particle. ST = (d2 d1) / 2 (4-1) We selected three representative systems for the physical characterization of the core-shell particles obtained. The systems NCU and NPY were prepared with formulations similar to the template set D. They included small amount s of hydrophobic dyes for fluorescence studies, whose results will be presented in the next chapters. Due to the low electron density observed for the templates studied in the previous chapter, we decided to try a differe nt staining technique to improve the quality of the images obtained. Th e system NEB was prepared with a combination of ethyl butyrate and 1-dodecene. With this co mbination of dopants we intended to obtain large templates for ease of imaging. 1-dodecene was selected due to the possibility of reacting its unsaturation with OsO4 through an oxidative addition.100 A highly hydrophobic molecule as 1dodecene would be difficult to remove from the core -shell particles during the purification steps, especially due to the expected protection of the interior of the particles from the solvent after 99

PAGE 100

their coating. The reaction of the unsaturation of 1-dodecene woul d also fix the metal center in the particles, producing an increase in the contrast of the images obtained. As a second step, the three systems were stained with UA to visualize the periphery of the partic les, as discussed for the templates in the previous chapter.101 All the systems were analyzed by AFM in a similar approach to the one used for the characterization of the templates presented in Chapter 3. In this case, we focused our attention on the differences in the values of the diameter and height obtained in the section analysis before and after th e synthesis of the shell, as shown in Table 4-3. 4.1.2 Results The templates were used for the formation of core-shell particle s without any further purification, since it was observed that the presence of the surfactants used in their synthesis was necessary for the shell to grow. Interestingly, when the coating of the particles was performed with the templates suspensions as is, precipit ation occurred, due to chemical crosslinking and formation of a gel, as shown in Figure 4-2. To prevent this, a 5-fold dilution of the original suspension was used and TMOS was added in sm all portions of approximately 20 mg every 24 h, under vigorous stirring. The dilution assured that the templates were separated enough in solution to prevent their crossli nking during the growth of the sh ell. The addition of TMOS in small volumes prevented the silicate precursor molecules from reacting in between them to produce a gel. As mention above, the core-shell particles could be centrifuged, contrary to the templates. This indicates that the templates are ve ry light and that there is an increment in their mass after the coating, as expected due to the gr owth of the silicate network. This gave us an effective way to separate the core-shell particle s from unreacted materials and templates that did not form a shell. After centrifugation, the particles were resusp ended by sonication for approximately 1 h. It was observed by DLS that the size of the particles was not affected by sonication, confirming that th e core and shell of the part icles were firmly connected. 100

PAGE 101

4.1.2.1 Size analysis of the core-shell particles The size analysis report here was performed similarly to the study of the templates discussed in Chapter 3. The system NEB produced particles with diameter values twice the ones of the dye-doped templates, as expected due to the large difference in the concentration of the dopant. For the three systems shown in Table 4-1 th ere was an increase in the average size of the particles after addition of TMOS, as shown in the characterization results on Table 4-2. The shell thickness, calculated with Equation 4-1, was surp risingly higher for NCU than for NPY. Both systems were prepared similarly, including equimolar amounts of dyes, which produced templates with similar dimensions, although th e NCU templates were slightly larger. The difference in the shell thickness may indicate that pyrene, be ing a more hydrophobic molecule, promoted the formation of a lower amount of templates compared to C153. Since a similar amount of TMOS was added to both systems, the one with a lower number of templates will be coated with a thicker shell. A thin shell was pr oduced in the case of NEB, despite the use of approximately three times the amount of TMOS used for NCU and NPY. This is expected due to the larger radius of these particles which result s in a larger area per particle to cover. An interesting general observation is that the increm ent in size is not immediately observed after TMOS addition. Despite observing that the suspensions became whiter, only after a certain amount of TMOS was added the diameter of the particles changed. It was observed that the initial mass of the silicate precu rsor necessary to produce this change was larger for bigger templates. 4.1.2.2 TEM analysis of th e core-shell particles A number of authors have used TEM for the characterization of core-shell particles, especially in cases where the shell was prepared with silicate precursors.35, 44, 102 Figure 4-3, 4-4 and 4-5 (top) show TEM images of the template s and the corresponding core-shell particles. In 101

PAGE 102

all cases spherical particles were observed, al though no significant differe nces in the contrast between the core and shell were found, as has been observed by other groups.97 The images of the NEB templates looked similar to the ones ob tained for the sample set D, although imaging was easier due to their larger size. The eff ect of doping the particles with 1-dodecene was observed in the higher quality of the images obt ained. The particles l ooked homogeneous, which indicates an even distribution of the dopant inside the templates. Moreover, a darker ring on the outside can be observed on their periphery as a consequence of the use of UA. The core-shell particles seemed to show some differences in th e electron density of the different parts of the particles, but no trend could be identified, since some of them showed a darker interior while others showed a higher electron dens ity in their external part. We believe this is not related to the presence of the dopant but that it is an arti fact obtained while manipulating the instrument. Systems NCU and NPY did not showed any speci al features as expe cted, although the pyreneloaded core-shell particles looked darker on the microscope compared to the other particles studied, probably as a consequence of the thicke r shell obtained (Figure 4-5). At the same time, small particles were observed on the images, whic h seem to be produced when a thicker shell was formed. Due to the larger difference in size between the templates and core-shell particles it was possible to observed a clear difference in their sizes in the TEM images, which was not the case with the other two systems with thinner shells. 4.1.2.3 AFM analysis of core-shell particles AFM was used to study the influence of the coating process in the rigidity of the particles, by comparing the diameter to hei ght ratio (DHR) values for each system before and after the coating of the templates, as s hown in Table 4-3. As observed in the previous chapter for the doped-templates, the NEB templates showed hollow-like structures, with a DHR of 15.3, indicating a very flexible system. After the shell growth, the images showed remarkable 102

PAGE 103

differences in the section analysis, where no hollow-like structures were observed. The average height of the particles increased, producing a lower DHR value (9.3), indicating a significant increase in the rigidity of the system. Based on these results, we believe the shell built on the periphery prevented the particles from deforming as much as the templates after deposition on a substrate, producing a more robust system. Apparently, the shell being a chemically connect network, is less flexible than the material the templates are made of. This is confirmed by the results obtained from Figures 4-4 and 4-5 for the NCU and NPY systems, respectively. In both cases, the same diameter and height values for the templates were obtained, 63 and 5 nm, with a DHR value of 12.6. This va lue is lower than the one for the NEB system, meaning that these templates are more rigid, pr obably due to the small amount of dye in their interior. After the shell growth, both, the diamet er and height, increased considerably for both dye-doped systems. As mentioned above, despite using similar amounts of TMOS, NPY grew a thicker shell than NCU, observed as differe nt diameter results by DLS (200 and 120 nm, respectively). These values correlate to the diameters obtained in the AFM analysis (141 and 87 nm, respectively), and more importantly, to the height values, which increased to 41 and 21 nm, resulting in a decrease in the DHR values to 3.4 for NPY and 4.1 for NCU. This is a more dramatic change than the NEB te mplates to particles since the dye -doped particles grew a thicker shell and thus, become more rigid. 4.1.2.4 Analysis of the dimensions of the templa tes and particles in suspension and after drying In order to understand the cha nges in the dimensions of the templates and particles in solution and when dried, we proceeded to analyz e the changes in the volume of the particles based on the data obtained by DLS and AFM, resp ectively. We selected the systems NPY and NCU due to the similar dimensions of their temp lates but different shel l thicknesses. Since the 103

PAGE 104

radius of the particles calculated by DLS is obtained by fitting the data to the presence of spherical scatterers, we calculated their volume with the formula 4/3 x ( r3), with r being the hydrodynamic radius. When the particles were deposited on mica and the solvent evaporated, disc-like structures were observed, as a cons equence, the volume was obtained by approximating the particles to a cylinder, using the formula ( r2) x h, where r is half the diameter and h the height obtained through the section analysis in AFM. Table 4-4 shows the results of these calculations. It should be menti oned that the values obtained ar e approximate and used only for qualitative comparisons. For both set of templates similar results were obtained, as expected based on the similarities of thei r dimensions. There is a dramatic decrease for the volume of both, the NCU and NPY templates, after drying (17 and 14-fold decrease, respectively). This indicates that the templates contained a large amount of water molecu les that were removed during drying. Interestingly, af ter coating, the volume of th e particles changed differently depending on the thickness of the shell. The system NCU, made of particles with a thin shell, shrunk less than the templates after evaporation of the solvent (7-fold de crease in their volume). This may indicate that the coating is preventing so lvent molecules to penetrate to the interior of the particles and that only the shell is hydrated. This would resu lt in less water molecules being evaporated, producing a smaller change in volume. The system NPY with a thicker shell, shrunk 13 times, as expected due to the larger volume ava ilable in the shell to store solvent molecules. These are interesting results that can be used for the design of carrier systems for different species. For instance, species placed on the shell of the particles are expected to be solvated, even if they are not in the pe riphery. On the other hand, hydro phobic species that are included in the mixture used to synthesize the particles could be expected to be preferentially located in their 104

PAGE 105

interior, avoiding contact with solvent molecules and, as a consequence, they would be difficult to remove during the purification. 4.2 Derivatization of the Surface of the Nanoparticles 4.2.1 Derivatization with 3-aminopropyltrimethoxysilane One of the advantages of coating the templates with a siloxane matrix is the different options for further derivatization of this ma terial. 3-aminopropylsilane (APS) is the common choice for attaching amino groups in silica particles and surfaces co ated with silicate precursors. The advantage of using this molecule is relate d to the ability of th e amino group to act in subsequent reactions as a nucle ophile, which provides a means to attach different groups on the surface of the particles by relativ ely simple chemistry, as shown in Figure 4-6. Based on this, we prepared core-shell particles and let them react with APS (system NPAP S). For the reaction to occur, it was necessary to raise the pH to 9.5 a nd the temperature to 70 C. Similarly to what was observed for the reaction with TMOS, the templates or particles could not be purified for the reaction to work and the suspensi ons had to be diluted five time s. To avoid crosslinking, the coating agent had to be added in small vol umes under strong agitation. The products were dialyzed after the reaction took pl aced to separate any unreacted material. Size analysis was used as explained before to confirm the growth of the shell, as shown in Table 4-5. It was observed that after coating of the particles with TMOS the addition of APS di d not produce a significant change in the diameter of the particles, suggesting that only a thin layer of APS was deposited on the surface of the particles. Typical AFM images for the templates and corresponding APS-coated core-shell particles are shown in Figure 4-7. As exp ected based on the results discussed in the previous sections, an increment in the diameter and height of the part icles was observed, correlated with the changes reported by DLS, as shown in the se ction analysis results reported in Table 4-5. Interestingly, the 105

PAGE 106

APS-coated particles showed a lower tendency to aggregate after deposition on mica, compared to the templates and part icles coated with TMOS. To confirm the presence of amino groups on th e surface of the particles, we studied the changes in the overall charge of the particles at different pH conditions. Figure 4-8 A shows typical -potential plots for the system NPAPS before and after the reaction with APS. As explained in the previous chapter, the sign of the -potential corresponds to the charge of the particles and its absolute value is an indication of the stability of the particles in suspension. In general, it was observed that the total charge of the particles wa s dictated by the silanol groups on the surface, which are deprotonated, producing negative -potential values. The particles seemed to be stable when the pH of the suspen sions is above 4, as observed by the high absolute value of their -potential. The response of the APS-coated particles is slightly different than the particles without the amino groups at the same pH, although these differences are not dramatic. The less negative -potential values after coating are produ ced by the reduction of the number of silanol groups, which reduced the number of ne gative charges on the surf ace of the particles and the incorporation of more basic amino groups. Apparently, the partic les were coating with a thin layer of APS, resulting in a low density of am ino groups. This is in agreement with the DLS data, which showed no differences in size after the coating. 4.2.2 Derivatization with Dansyl Chloride-APS Complex Dansyl chloride, a hydrophobic fluorescent dye, wa s attached to the surface of the particles to study its fluorescence, which will be discussed in the next chapters. In this case, a reaction between APS and dansyl chloride was performed as shown in Figure 4-9 to create a silane derivative of the dye (APSDCl complex). The pro duct obtained was a yellow thick paste, which was kept under argon until used. Two set of templa tes were prepared for derivatization with the APSDCl complex. The system NPDCl-1 was obt ained from templates prepared without a 106

PAGE 107

dopant, while the system NPDCl-2 was doped with EB. The suspensions of the templates were divided in two, to grow coatings of different thicknesses with TMOS, before the attachment of the APSDCl complex, as shown in Table 4-6. Once the desired size ra nge was obtained, an excess of the APSDCl complex was added as a so lid to the suspensions, and left stirring for a week under strong agitation. After this time, the excess complex was separated through filtration and any hydrolyzed APSDCl that did not react wi th the surface was sepa rated through dialysis, until no more DCl was detected by UV. Figure 4-8 B and C show the -potential values at different pH for particles prepared with the same templates containing no dopants (systems NPDCl-1A and -1B). The system 1A showed clear differences at low pH valu es. This is due to the presence of amino groups on the surface of the particles, which were pr otonated at acidic conditions, resulting in more positive -potential values. These differences became more noticeab le for the sample NPDCl-1B, which grew a thicker shell. We believe this is a consequence of the larger surface area available for coating for the bigger particles. This would result in more amino groups avai lable for protonation at low pH. Moreover, the differences are more significant th an the ones for the syst em NPAPS since each APS-DCl complex contains two amino groups, compared to only one in APS. This doubles the number of amino groups added to the surface of the particles per molecu le of the dye complex attached. In both cases, aggregation of the amino-coated particle s was observed when the pH of the suspensions was close to their isoelectric point. This made the accurate determination of the charge of the APS-coated particles cha llenging, at pH valu es close to 2. Some interesting results were obtained for the particle set NPDCl2, as shown in Figure 4-6 D. Similarly to the previous case, both dye-coate d particles aggregated when the pH was reduced to 3 or lower values, a clear in dication of the presence of amino groups on the surface. In this 107

PAGE 108

case, the first comparison is made between the templates and particles coated with APSDCl but no TMOS. Surprisingly, the templates showed a more positive charge than the amino-coated particles, which contradicts the fact that the dy e-coated particles precipitated at low pH and the templates did not. These results are not comple tely understood, although it could be explain by the presence of excess surfactant in the templates that was not completely removed despite the extensive purification pe rformed. The particles coated with TMOS and the dye, on the other hand, showed the expected results based on th eir larger size, which can accommodate more APSDCl molecules, resulting in more positive -potential values. 4.3 Preliminary Results on Other Systems Studied 4.3.1 Preparation of Large Templates and Co re-Shell Particles for Calcination Studies We started to investigate dopants that can produce larger templates and core-shell particles than the ones obtained with hexadecane and ethy l butyrate. N-benzyl maleimide was used during the synthesis of the particles as a dopant with surp rising results. It was difficult to disperse this material in the original surfactan t mixture, even after long periods of heating. Despite this, large templates and core-shell particles were obt ained with diameters of 355 nm and 410 nm, respectively, as observed by DLS. Similarly to th e previous systems, the polydispersity of both, the templates and core-shell particles, was high. Due to the large size of the templates, the coating of the particles was difficu lt. As observed previously for ot her large templates, the initial addition of TMOS did not produce an increase in the average size of the particles. After this, a first small increased in their diameter was obtai ned, after which no more changes were observed with DLS, even after adding la rge amounts of TMOS. Apparently when trying to coat large particles, only a thin shell grows, after which small particles are formed Surprisingly, despite using the same staining techniques used for the other templates, the TEM images showed very dark particles, as observed in Figure 4-10 A. Th is may indicate that the dopant was not removed 108

PAGE 109

from the system during the purification steps, due to its very low solubility in water, increasing the electron density of th e particles. Moreover, in this case due to the ease of imaging, the coating is observed as a skin on the surface of th e particles (Figure 4-10 B). On the other hand, it was difficult to image the particles in the atomic microscope due to their large size, as observed in Figure 4-10 C and D. The tip of the cant ilever had problems track ing the topology of the surface due to the large differences of the featur es on the Z-axis, which made the size analysis difficult. Approximate values were obtained for the diameter and height of the templates (450 and 95 nm, respectively), and for core-shell pa rticles (550 and 300 nm, respectively). Despite these difficulties, spherical particles were obser ved with only a few hollow particles in both cases, as expected due to the presence of the dopant in the interior of the core-shell particles. Calcination of these particles was performed similarly to the templates studied in Chapter 3. The images obtained for the templates (Figur e 4-11 A and B) show the presence of small particles, as observed with the smaller template s studied before. Interestingly, ring structures with sizes similar to the templates were observe d. At the same time, darker structures with different shapes and spherical hol es were observed. Apparently, the material inside the templates had been ejected from the core, but somehow remained connected through a ring in the periphery. The core-shell particle s on the other hand, together w ith the small particles observed in the templates, showed core-shell particles w ith holes in their interi or (Figure 4-11 C and D). The size of these holes was similar to the diamet er of the small particles observed. This may indicate that when TMOS is added, it not only reacted on the surface of th e particles, but in the pores inside the templates, forming small spheri cal particles loosely attached to the core-shell particles. These would explain th e observations reported in the DLS analysis section that the diameter of the particles did not changed after the first additions of TMOS. Moreover, this would 109

PAGE 110

indicate that the react ion with APS would occur not only on the surface of the particles but in their interior as well, resulting in the less positive -potential values than expected, as observed. These are interesting results that indicate that the particles might be used for the growth of metal particles in their pores, but this would need to be studied in more detail. 4.4 Conclusions The templates studied in Chapter 3 were used for the formation of core-shell particles. The silanol groups on the surface of the templates were us ed as anchoring points to grow a shell. This was done through the sol-gel chemistry of tetram ethoxysilane, which placed a silicate network on the surface of the templates, producing core-shell particles. Dynamic light scattering showed that the coating process produced an increment in the average size of the particles. This was confirmed by observing that it was po ssible to centrifuge the core-she ll particles, contrary to the templates, due to an increase in their density. TE M images were used to confirm that spherical particles were obtained and that more dense part icles, observed as darker species, were formed with thicker coatings. The topographic features of the particles were studied by AFM, by comparing the diameter and height of the templates and core-shell particles. Similarly to the DLS results, an increment in the diameter and height of the particles was observed after the coating process. We compared the DHR values of the templates and particles, finding that their rigidity increased after reacting with TMOS. We believe this is produced by the less flexible and more rigid silicate network surrounding the particles, which prevents them from spreading on the substrate as much as the templates. Accordingly, lower DHR values were observed for particles with thicker shells. Calculation of the volumes of the core-shell particles in solution and after drying on a substrate showed that the templates contained a large am ount of water, which is lost after deposition on mica. Similarly, particles with a thick shell sh runk considerably, contrary to the ones with a thin coating, suggesting that the shell is extensively hydra ted. As a first step 110

PAGE 111

towards the attachment of different species on the surface of the particles, they were reacted with APS to place amino groups on their periphery. The derivatization of the pa rticles was confirmed through -potential measurements, which reported ch anges in the charges on the surface of the nanoparticles after the reaction with the amino moieties. This approach was used for the derivatization of the particles with a fluorescent dye. A derivative of dans yl chloride with APS was prepared to attach the dye to the exterior of the nanoparticles. Finally, preliminary studies to obtain larger particles were performed. It wa s observed that when a solid dopant was used, particles with diameters above 300 nm were obtained. Coating of these particles was difficult as was found that only a thin shell could be grow n and further addition of TMOS produced small species not attached to the particles. These large templates and core-shell particles were calcinated and the products imaged. It was found that a large number of core-shell particles maintained their size and shape but they showed some small holes in their interior. Small particles of similar dimensions to these holes were found in the images, which may indicate that they were originally attached to the particles. It is possible that small silica particles have grown in the pores of the templates while the shell was growing, but they were not chemically attached to them. These are interesting findings that need to be studied further. Table 4-1 Formulations used for the synthesis of templates and corresponding core-shell particles. SO t80 EB 1-d OTMS Py C153 TMOS NEB 125 125 29 10 19 42 NCU 125 125 12 0.70 15 NPY 125 125 13 0.45 14 All quantities in mg. 111

PAGE 112

Table 4-2 DLS characterization results for templa tes and the corresponding core-shell particles. d1 d2 ST NEB 175 203 14 NCU 80 120 20 NPY 74 200 63 d1: templates diameter, d2: core-shell nanocapsules diameter ST: shell thickness, all in nm. Table 4-3 AFM characterization results for template s (1) and core-shell part icles (2) prepared as indicated in Table 4-1. d1 h1 DHR1 d2 h2 DHR2 NEB 183 12 15.3 195 21 9.3 NCU 63 5 12.6 87 21 4.1 NPY 63 5 12.6 141 42 3.4 Diameter (d) and height (h) values given in nm. Table 4-4 Volume of templates and coat ed particles based on DLS and AFM data Vsuspension (DLS) V dried (AFM) temp coated templates coated templates coated Vsol/Vdried Vsol/Vdried NCU 2.68*105 9.05*105 1.56*104 1.25*105 17 7 NPY 2.12*105 4.19*106 1.56*104 3.28*105 14 13 Vsolution (DLS) obtained with the formula 4/3x( r3), V dried (AFM) obtained with the formula r2xh both in nm3 Table 4-5 Formulation and size characterization resu lts for APS-coated nanoparticles (NPAPS) SO T80 OTMS TMOS APS d1 DLS d2 DLS d1 (h1)AFM d2 (h2)AFM 126 125 17.7 21.6 11.7 76 120 72 (6) 134 (42) All quantities in mg, d: diameter, h: height, (1): templates, (2) co re-shell nanoparticles, all sizes in nm. Table 4-6 Formulations used for coating of the particles with APSDCl and DLS characterization results SO T80 EB OTMS d1 DLS TMOS d2 DLS NPDCl-1A 127 125 17.0 76 12.1 136 NPDCl-1B 76 45.8 175 NPDCl-2A 127 125 13 17.6 138 138 NPDCl-2B 138 20.3 236 All quantities in mg, d: diameter in nm. 112

PAGE 113

H2O Si OH MeO OMe OMe -MeOH Si OH MeO OMe OH Si OH MeO MeO OMe Si O Si O Si OH O-OH O Si O O-MeOH Si O Si O Si O O O Si O Si O O Si O Si O O O H O Si OMe MeO OMe OMe n Si O Si O Si O O O Si O Si O O Si O Si O O O H O H H O O n n Si O Si O Si O O O Si O Si O O Si O Si OH OH OOH2O H2O H2OA B Figure 4-1. Reaction scheme for the hydrolysis of TMOS, shown fo r the first A) and second B) hydrolysis products. The reaction of these species with the silanol groups on the surface of the templates is shown, as well as the final product after complete hydrolysis at neutral pH. 113

PAGE 114

Figure 4-2. Crosslinking resulted when template s were not diluted before the addition of TMOS. Scale bar 500 nm. Figure 4-3. TEM and AFM he ight images of NEB A) templates a nd B) core-shell particles. Scale bars in all images 500 nm. 114

PAGE 115

Figure 4-4. TEM and AFM he ight images of NCU A) templates and B) core-shell particles. Scale bars in all images 300 nm. 115

PAGE 116

Figure 4-5. TEM and AFM he ight images of NPY A) templates a nd B) core-shell particles. Scale bars in all images 500 nm. Si O O Si O O Si OH OH (CH3O)3SiCH2CH2CH2NH2 Si O O Si O O Si Si O O Si Si O O OH NH2 OH HO pH9,70C Figure 4-6. Attachment of A PS to core-shell nanoparticles. 116

PAGE 117

Figure 4-7. AFM images for the templates and A PS-coated nanoparticles re ported on Table 4-4. Figure 4-8. Comparison of -potential results before (circles) and after (squares) derivatization with APS: A) NPAPS; and APS-DCl: B) NPDCl-1A, C) NPDCl-1B and D) NPDCl2A (squares) and NPDCl2B (triangles). 117

PAGE 118

Figure 4-9. Coupling between dansyl chloride and APS. Figure 4-10. TEM and AFM images of N-benzyl maleimide-doped templates (A and C) and core-shell particles (B and D). Scale bars 500 nm. 118

PAGE 119

Figure 4-11. TEM images of N-benzyl maleimid e-doped templates (A and B) and core-shell particles (C and D) after calcina tion at 600 C. Scale bars 500 nm. 119

PAGE 120

CHAPTER 5 SEQUESTRATION AND ENCAPSULATION OF FLUORESCENT DYES IN TEMPLATES AND CORE-SHELL PARTICLES 5.1 Sequestration of Dyes by the Templates 5.1.1 Strategy Based on the characterization studies discusse d in the previous sections, the templates obtained consist of a flexible core mainly formed by chemically connected OTMS molecules. After addition of TMOS, a hydrophilic shell is built on the periphery of the templates, which was observed to increase the rigidity of the particles. The flexibilit y and expansion of the templates and core-shell particles after deposition on a substrate suggest ed that these particles contain pores in their interior that could be used to accommodate guest molecules. We believe that the polarity of the interior of the particles is lower th an that of the solvent in which the particles are suspended. This difference is intended to be used as a driving force for the sequestration of species with low polarity, like drugs or pesticides Similarly, particles for the encapsulation of hydrophobic molecules, with low solubility in wa ter, like fluorescent dye s or reactive species, could be obtained by including the active molecule s in the formulation used to synthesize the particles. This would allow the use of these spec ies in aqueous suspension s without precipitation. This is especially important when applications in biotechnology are considered, since certain dyes and other useful species are known to be to xic. In these cases, encapsulation of these molecules assures that no unwanted reactions take place, since the molecules would have no tendency to escape from the templates or nanopartic les due to the differences in polarities with the media. As a consequence, we initiated studi es related to the ability of these core-shell structures to be used as nanocontainers and sequestra tion agents for hydrophobic small molecules in aqueous systems. 120

PAGE 121

As a first step, it was necessary to decide wh ich technique would allo w us to detect the presence of different species in the interior of the te mplates and core-shell particles. To study the sequestration of species from solution into particles, UV active species are commonly used.103-105 Typically, a volume of a known con centration of the abso rbent is mixed with a suspension of the particles. After a stabilization period, the particles are removed and the UV absorption of the species left in the supernatant is analyzed, fr om which, the amount of molecules that were sequestered is determined. Although being an e ffective and simple way to quantify the uptake capacities of the particulate systems used, this technique gives no information on the sequestration process, being unabl e to differentiate if the UV active species are adsorbed on the surface or transported into the interior of the particles. A nother alternative involves the identification of the absorption maximum of the active species and to relate its changes to variations in the environment wh ere the probes are placed. Since th e nanoparticles scatter light in a broad area of the UV-vis re gion, it was not practical to use this technique for the characterization of the loaded species. On the other hand, as shown in Figure 5-1, the nanocapsules did not show any fluorescence when excited in the region that most dyes adsorb and emit light, but scattering, due to the size of the particles, wa s observed. This was confirmed using different wavelengths for excitation that produced changes in the position of the peak observed. Based on this, we started studies on the physical characterization of the templates using fluorescent dyes. Pyrene was selected as the fi rst probe due to its low polarity and well known dependence of its emission on solvent polarity.67 Our first goal was to determine if the dye could be incorporated in the templates. Then, we studi ed the influence of the amount of dopant, as well as its polarity, in the environment sensed by th e dye. Later, this strategy was extended to the 121

PAGE 122

core-shell particles, to investigate whether the coating of the templates produced changes in the internal structure of the pa rticles. Finally, C153, another hydrophobic probe, was used to investigate if the chemical structure of the dye used affects the results obtained. 5.1.2 Use of Pyrene as a Polarity and Rigidity Probe Pyrene has very useful characteristics for th e study presented here. Especially important for us is the lack of functional groups in th e molecule which prevented any kind of specific interactions that are not due to the polarities of the probe and the different parts of the particles. As explained in Chapter 2, the polarity index, de fined as the ratio of the third and first peak intensities (I3/I1) in the emission of pyrene, has been qual itatively related to changes in polarity in the environment where the dye is placed. This has been used by different groups to study the polarity of different systems, which we used as a reference to compare our results to. Tedeschi et al. used pyrene to determine the polarity of films obtained by the layer by layer deposition of polyelectrolytes, observing that the polarity of the films obt ained depended on the chemical structure of the polymers used.106 Hugerth studied the aggr egation of the hydrophobic drug amitryptiline, using pyrene as a polarity sensor, based on the changes in the I3/I1 value after amitryptiline aggregates with low polarity were formed.107 Pandey et al. studied the solvation environment of siloxane polysoaps dissolved in wa ter. They observed that derivatization of the polymers with hydrophobic side chains was corre lated with a lower polarity sensed by the probe.108 Danko et al. studied the polarity of interpenetrating polymer networks prepared with different monomers to which pyrene had been chemically attached.109 A similar approach was taken for the use of pyrene as a sensor for rigi dity. As discussed in Chapter 2, the excimer emission appears as a broad peak centered around 480 nm. Hence, th e ratio of the intensities of the emission of the monomer and excimer (I1/Iexc) is related to the ability of the dye molecules to form dimers, and can be used as an indicator of changes in the vi scosity of the media.65 122

PAGE 123

Vanderkooi and Callis used pyrene as a probe to study the latera l diffusion of species in the hydrophobic region of membranes.66 Jang and Oh observed the formation of excimers when the loading of pyrene was increased in polymer core-shell nanoparticles.110 Due to its very low solubility in water, pyrene has been used as a probe in studies of micellar systems as well, observing that it is pref erentially solubili zed in the most external hydrophobic regions of these aggregates, in contact with wate r molecules to a certain degree.67 Hence this probe has been used for determination of CMC values. Pyrene would from aggregates in wa ter if the surfactant concentration is low enough to prevent the form ation of micelles. Once the surfactant molecules start aggregating, the probe will be incorporated in the micelles due its hydrophobicity, which is observed as a dramatic shift in the emission of the dye and on the intensity of its emission. 5.1.3 Results 5.1.3.1 Determination of the optimal condition s for the study of the fluorescence of sequestered pyrene Figure 5-2 shows the emission spectra of pyrene in the tw o compounds used as dopants. Changes in the intensity of the first and third peaks in the emission spect ra of the dye (371 and 382 nm, respectively) were observed as a conseq uence of the difference in polarities of the solvents. The polarity index value for pyrene dissolved in ethyl butyrate was determined to be 0.797 and 1.77 when dissolved in hexadecane. These values were expected based on the literature polarity index for compounds with si milar polarities, hexane (1.65) and methanol (0.75). Based on this, we anticipated the te mplates to offer the encapsulated molecules microenvironments with different polarities, had the dopant molecules stayed in the interior of the templates after the purification steps. To find out the conditions to perform an accurate determination of the I3/I1 value, we carried out a titration of the templates with a methanolic solution of pyrene. This was done with a 0.3 mg/m L template suspension in all the cases. This 123

PAGE 124

concentration was selected to minimize scattering of light due to the size of the particles, whose peaks could interfere with the fluorescence of the dye. As shown in Figure 5-4 A, the concentration of the dye was increased from 0.1 M, which was found to be too low, resulting in a noisy signal for the emission of the dye, to 2.2 M, which promoted the formation of excimers, as observed by the appearance of a broad peak ce ntered at approximately 475 nm. These results showed that a concentration of the probe between 0.4 and 1 M produced a clear signal without excimers, thus, the I3/I1 value could be obtained with reliability. Within this range, it was observed in the titration curve (Figure 5-4 B) that a concentration of the probe of 0.5 M is close to the inflection point, which was taken as the I3/I1 value used to report th e polarity of the media. 5.1.3.2 Sequestration of pyrene by ethyl butyrate-doped templates Table 5-1 shows the polarity index obtained from sequestered pyrene for the two set of templates investigated in the previous section. Th e EB-doped templates showed a I3/I1 value between 1.30 and 1.36, considerably higher than the value when the dye was dissolved in this solvent, but not as high as when di ssolved in the less polar hexadecane (I3/I1: 1.77). This is interpreted as an apparent polar ity in between that of the two pure dopants. Interestingly, when the values for the three EB-doped systems were compared no significant differences were observed, which indicated that the amount of dopant or OTMS used did not affect, to a significant extent, the com position of the volume where the dye was placed. There are two plausible explanations for the differences in the environment sensed by the probe after internalizat ion by the templates and when disso lved in the pure dopants. One explanation is that the ethyl but yrate molecules where lost during the purification steps, leaving pores in the interior of the templates whose polarity did not resemble that of EB. In this case the dopant would be acting as a porogen. The other e xplanation is that the EB -rich areas inside the particles were not accessible to the dye molecules, probably due to their si ze or polarity. If we 124

PAGE 125

consider that the polarity sensed by the dye molecules is lo wer than the dye suspended in different micellar systems (typical I3/I1 values between 0.74 and 0.95), where the dye is known to reside in the outer volume of th e aggregates, in proximity to the solvent front, we can conclude that this was not our case. Based on this, we can assure that the dye molecules are in the interior of the templates, somehow protected from the a queous media where they were suspended. With this, we believe the hypothesis based on the idea th at the dye could not reac h the interior of the particles, where the dopant mol ecules would reside, is not valid. Figure 5-5 shows what is believe to happen during the purif ication of the templates. Ap parently, the dopant molecules were removed during the washing used to separate the surfactants and unreacted materials from the templates. After this, the external pores were filled with water molecules, which is less likely to happen with the ones buried in the center of the particles. When pyrene was added to the system, the dye molecules preferred to migrate to the more internal pores to avoid contact with the solvent, which was observed as a low averag e polarity reported by the dye. Interestingly, when the concentration of the dye was increased an increase in the apparent polarity was not observed but the formation of excimers did oc cur. This confirmed that the dye molecules avoided the periphery of the temp lates, preferring to pack more tightly in the more protected pores. 5.1.3.3 Sequestration of pyrene by hexadecane-doped templates The polarity reported by pyrene sequestere d by the hexadecane-doped templates was slightly lower than the EB system. Surprisingly, these differences were not as dramatic as expected based on the emission of the dye in each solvent, as shown in Table 5-1 (I3/I1: 1.4-1.6). Confirming what was observed with the EB-system, the dye seemed to be placed in a nonpolar environment, protected from the solvent. In this case, the I3/I1 value was very close to the polarity index when dissolved in the dopant. Apparently, the pores formed by hexadecane offer slightly 125

PAGE 126

better protection than the ones formed by EB. This could be a consequence of the presence of a small amount of dopant molecule s that could not be removed dur ing the purification due to its low polarity. 5.1.3.4 Study on the retention of th e dye after solvent exchange To test the idea that the probes internalized by the templates are protected from the solvent, we loaded two set of templates, one dopant-free and the second prepared w ith EB, with an excess of pyrene, added as a solid. As shown in Figure 5-6, in both cases, the emission of the dye was dominated by the excimer signal, as expected due to the high loading of the dye. After this, dialysis was performed until no more dye could be detected by UV in the dialyzed media. Subsequently, the fluorescence was recorded again. The emission of the dye was clearly observed, confirming that the probe found pores in the interior of the particles that were not reached by the solvent molecules, thus, avoi ding being washed-out during the dialysis. Moreover, the polarity index for both systems was very similar, 1.27 for the dopant-free templates, and 1.19 for the EB-doped system, which indicated that they were localized in similar environments. 5.2 Sequestration of Dyes by the Dopant-Free Core-Shell Particles The fluorescence results showed above demons trated the successful use of steady state fluorescence spectroscopy to detect hydr ophobic model compounds in ternalized by the templates. Since our interest is to develop a method to synthesize core-s hell nanoparticles for the uptake of harmful compounds, we proceeded to study the sequestration of fluorescence probes by the core-shell particles in a similar way. To study particles that could be use in biomedical applications, we used dopant-free particles, whic h produced species with smaller sizes that the ones obtained by using a dopant (Tab le 5-2). To expand this study a nd investigate if the results 126

PAGE 127

obtained with pyrene were a general trend, C153, another fluorescent hydrophobic probe, was used. 5.2.1 Sequestration of Pyrene by Dopant-Free Core-Shell Particles The dye-free dopant-free core-she ll nanoparticles were prepared similarly to the samples D2 and D3 studied in the previous chapter. These particles were mixed with a methanolic solution of pyrene and the fluorescence was recorded. As observed in Table 5-3, the concentration of pyrene necessary to observe a clear emission but no aggregates was lower than that for the doped-templates of a similar concentration (0.12 M). It should be noted that the concentration of particles used here are mass per volume, which means that despite the similar concentrations used, the amount of the core-shell particles may be lower compare to the templates, since the former are heavier. The I3/I1 value observed was 1.38, similar to the ones obtained with the dopant-filled templates, meaning that the polarity encountered by the dye was similar to the previous case, indicating an effect ive protection from the solvent. This indicates that the probe could diffuse to the interior of the particles through the shell, probably reaching the core, which is mainly formed by what orig inally was the templates, or that it resides somewhere in the shell where water molecules c ould not penetrate. These are promising results for the use of these nanoparticles as sequest ration agents for hydrophobic compounds. When the concentration of the dye was increased to approximately 2 M, the formation of excimers was detected, as expected. 5.2.2 Sequestration of C153 by Dopant-Free Core-Shell Particles 5.2.2.1 Strategy In order to investigate if these findings were a general trend we used C153, another hydrophobic fluorescence probe, which ha s been used as a probe for polarity and rigidity due to its large change in the dipole moment when excited,68-70 as discussed in Chapter 2. There are a 127

PAGE 128

number of reports of using C153 as a probe fo r the determination of polarity in different environments. Kumbhakar et al. used C153 in poly(ethylene oxid e)-based surfactants to study the solvation dynamics in these systems. They observed a shift to the red in the emission of the dye suspended in micelles of the surfactant with more ethylene oxide units, which they interpreted as the probe sensing a more polar environment. Ferrer and Lianos111 determined the maximum in the emission spectrum of the dye in different micellar system s, finding that a more red shifted value corresponded to more polar e nvironments in the micelles. Hof and Lianos112 incorporated C153 in the formulation for the synthesis of a SiO2 matrix. They did not observe any changes in the emission of the dye until the matrix was left drying for a period of two weeks. They attributed this change to the increase in ri gidity of the matrix where the dye resided, which prevented the stabilizat ion of the excited state by the polar surroundings. Seth et al.113 used this probe for the characterization of ternary-micellar systems, includi ng an ionic liqui d, finding that C153 resided in the surface of the microemulsions For our studies, the important difference with pyrene is that C153 contains func tional groups that can potentiall y interact with the different parts of the particles. 5.2.2.2 Results Table 5-3 shows the emission maximum observed after loading of the dye from a methanolic solution, similarly to what was done with pyrene. The polarity sensed by the probe was higher than when the dye is dissolved in methanol (536 nm) and similar to the values obtained by Stathatos in silica/po ly(propylene oxide) composites.70 This value shifted slightly to the blue from 543 to 541 nm with a ten-fold in crement in the dye concentration. The emission maximum observed indicated that the dye molecule s did not penetrat e the shell of the particles, but stayed close to the surface. Such a high emi ssion maximum has been identified as a sign of a specific interaction, like hydrogen bonding. In the case studied here this is probably due to the 128

PAGE 129

interaction of the carbonyl and amino groups of C153 with the silanol groups on the shell,114 similarly to what was observed by ou r group with an electroactive probe.115 When this is compared to the results obtained from pyrene, as shown in Figure 5-7, it is clear that the chemical structure of the probe to be sequestered influences its fi nal position inside the particles. This should be taken into consideration when de signing an uptake system for a particular probe. In the next chapter this idea will be intr oduced for the design of antenna systems. 5.3 Encapsulation of Dyes by the Templates and Core-Shell Particles 5.3.1 Strategy We finally proceeded to study the changes in the environment that the probes experienced when they were encapsulated in the templates and core-shell particles. We prepared these systems by including the dyes in the formulation used to synthesize th e particles, as explained in the previous chapter (systems NCU and NPY). The aim of this study was to investigate the changes occurring within the interior of the partic les before and after the synthesis of the shell, through changes in the fluorescence of the dyes used. This was compared to the AFM information gathered in the characterization studies in Chapter 3 to determine whether they give similar results. 5.3.2 Results 5.3.2.1 Encapsulation of pyrene in the templates and core-shell particles A comparison of the fluorescence of pyrene in the templates and co rresponding core-shell particles for the system NPY is shown in Figure 58. It is expected, base d on the results from the previous sections, that some of the dye would be lost during the purification steps, for example, dye molecules trapped in the shell during the synt hesis could have been removed together with the solvent, which would produce a decrease in the dye concentr ation. As a consequence, the emission reported here corresponds to the molecules of the dye that remained in the system after 129

PAGE 130

purification. As shown in Figure 5-8, the emission of the encapsulated dye in the templates was dominated by the excimer fluorescence, observe d as a broad peak centered at 475 nm, which prevented us from using the I3/I1 value to investigate the polarity of the interior of the particles. As a consequence, we used the excimer to monomer emission ratio (I1/Iexc) to investigate the rigidity of the media where the dye was placed. The I1/Iexc ratio changed from 4.4 in the templates to 12.1 in the particles, which indicate d that it is more difficult for the dye molecules to form excimers with ground state monomers after excitation in the core-shell particles.110 This is surprising since it would be easier for the temp lates to lose dye molecules than the core-shell particles, since the latter would protect the dye better from cont act with the solvent. A reduction in the concentration of the dye would make it more difficult to form dimers. We believe that the lower I1/Iexc in the core-shell particles is a consequence of the increased rigidity in the system after the formation of the shell, which decreases the mobility of the probe. These results correlated with the AFM observation s, showing a decrease in the flexibility of the core-shell particles compared to the templates as a conseque nce of the coating with the siloxane network. 5.3.2.2 Encapsulation of C153 in the templates and core-shell particles Similarly to the work with the dye-free core-shell pa rticles, we used C153 to investigate if these findings were a general tre nd. First, to confirm that the emission detected was actually from the encapsulated dye, we looked at th e particles under the fluorescence microscope. Unfortunately, this could not be done with the pyrenedoped particles due to the lack of the proper filters in the microscopes we have access to. A fluorescence microscopy image of the C153-encapsulated core-shell partic les is presented in Figure 5-9 A. The size of the particles was below the resolution limits of the microscope, so it was necessary to perform the imaging after the solvent had evaporated to promote particle aggregation. The emi ssion is observed to be confined to spots with diameters in the fe w micrometers range, which corresponds to the 130

PAGE 131

aggregates observed in the AFM images. We anal yzed the templates and core-shell particles in solution similarly to the previous cases. The difference in the emissi on after the coating is observed qualitatively in Figure 59 B, as a shift to the blue after the shell growth. Table 5-4 shows fluorescence data for these templates and particles. The maximum in the emission shifted from 533 to 525 nm after the shell growth, i ndicating a decrease in the polarity of the environment where the dye was placed, similar to a change when the dye is dissolved in ethanol (530 nm) and butanol (525 nm).68-70 We interpreted this decrease in the apparent polarity of the environment surrounding the dye as a consequence of the shell acting as a barrier to isolate the interior of the particles, wh ere hydrophobic molecules tend to re side, from the polar solvent. Moreover, these values are considerably lo wer than the ones obtained when the dye was sequestered by dye-free particles. From this, we can conclude that the dye ended up placed in different parts of the particles, sensing different environments. 5.4 Conclusions Our results for the characterization of the structure of the templates and particles synthesized in Chapters 3 and 4 are presen ted here. We used hydrophobic fluorescent dyes to characterize the interior of the templates and core-shell particles. The dependence of the fluorescence of pyrene on the polarity and rigidity of the microenvironments where it was placed after its sequestration was used to study the environment in the interior of the particles. This was done for the ethyl butyrate and hexadecane dopedtemplates. It was observed that the despite slight differences, the polarity of the interior of the templates did not depend on the polarity of the dopant. We interpreted this as an indica tion that the dopants we re removed from the templates during the purification steps. In average, the polarity of the interior of the templates reported by pyrene was low, which means that dye molecules tended to migrate to the dry pores, away from the polar solvent. Subsequent ly, we used dopant-free core-shell particles to 131

PAGE 132

sequester pyrene from aqueous suspensions, finding that the dye ended up in an environment of similar polarity than that of the doped-template s. A second probe was used, C153 which showed different results. In this case th e polarity sensed by the probe wa s higher and similar to the dye dissolved in a polar solvent. Due to its low aqueou s solubility we interpreted this information as the dye being trapped in the shell due to hydrogen bond interacti on. This suggested that the structure of the dyes mandates th e final position of the probe insi de the particles. Finally, we encapsulated these dyes in the templates and core -shell particles by adding the dye during their synthesis. In this case, we intended to identify the changes in the polarity and rigidity in the interior of the particles through changes in the emission of th e dyes before and after the growth of the shell. It was observed w ith pyrene that the rigidity of the volume where the dye was placed increased after the coating of the particles. Coumarin 153, on the other hand, reported a decrease in the polarity of the interior of the particles after the shell growth, showing that the siloxane network isolates the interior of the particles from the polar solvent. We believe these findings are promising for the design of scavengers for drugs and pesticides. Table 5-1 I3/I1 values for pyrene in ethyl butyrate (E B) and hexadecane (Hx) and EB and Hxdoped templates. I3/I1 I3/I1 EB 0.797 Hx 1.77 EBB 1.31 HXB 1.41 EBC 1.32 HXC 1.43 EBF 1.36 HXE 1.58 Table 5-2 Formulation used in the synthesis of co re-shell particles for se questration studies and size characterization re sults obtained by DLS. T80 SO OTMS TMOS d1 d2 NSL 125 125 14 12 98 200 All quantities in mg. Surfactants were suspended in 13.9 g of saline. d1: templates, d2: core-shell particles, both in nm. 132

PAGE 133

Table 5-3 Fluorescence data for pyrene a nd C153 sequestered by dopant-free core-shell particles. pyrene [ 0.12 M]b [2.1 M]b C153 [0.13 M]b [1.25 M]b I3/I1 1.38 max a 543 541 I3/Iexcimer 1.9 a: excitation wavelength 420 nm, b: fina l dye concentration in the suspension Table 5-4 Steady State Fluorescen ce data for templates and core-shell particles used to encapsulate pyrene (NPY) and C153 (NCU). C153 Pyrene templates np templates np max a 533 525 I1/Iex 4.4 12.1 350 400 450 500 550 0.0 2.0x1064.0x1066.0x1068.0x1061.0x1071.2x107 emission intensity (a. u.)wavelength (nm) Figure 5-1. Fluorescence response detected after excitation of a dye-free nanocapsule suspension. The peak observed is due to scattering of light by the particles, excitation wavelength 270 nm pyrene coumarin153 N O CF3 O Figure 5-2. Chemical structures of the dyes used in this study. 133

PAGE 134

Figure 5-3. Emission of a 1 M pyrene solution in A) hexade cane and B) ethyl butyrate. Figure 5-4. Determination of cond itions to study the polarity of temp lates with pyrene. A) Pyrene titration for hexadecane-doped templates, numbers in the box indicate pyrene concentration, B) plot of th e ratio of the intensities of the third and first emission peaks of pyrene (I3/I1) at different dye concentrations. 134

PAGE 135

Figure 5-5. Representation of the removal of dopant molecules during purification of the templates. The positions preferred by pyrene molecules when sequestered by a suspension of templates, is shown as well. Figure 5-6. Emission of pyrene lo aded in A) dopant-free template s and B) EB-doped templates, before (black line) and after (grey line) dialysis. Figure 5-7. Representation of the positions preferred by pyrene and C153 molecules when sequestered by core-shell particles. 135

PAGE 136

380400420440460480500 0.0 0.2 0.4 0.6 0.8 1.0 normalized intensitywavelength (nm) Figure 5-8. Normalized emissi on of pyrene encapsulated in templates and in the corresponding core-shell particles (blue) for the NPY system. Figure 5-9. Fluorescence microscopy images for C 153 encapsulated in core-shell particles, scale bar 10 m. Image was obtained in black and white, color was added for clarity B) Normalized emission of C153 encapsulated in templates (red line) and in the corresponding core-shell particles (blue line). 136

PAGE 137

CHAPTER 6 STUDY OF THE COMPARTMENTALIZATION OF THE TEMPLATES AND CORE-SHELL PARTICLES THROUGH ENERGY TRANSFER BETWEEN DYES 6.1 Energy Transfer in the Templates 6.1.1 Strategy The information gathered in the previous chap ters indicated that the templates and coreshell particles obtained could be used as nanoc ontainers for different species of low polarity. Two approaches were used for loading hydrophobic fluorescent dyes in th e templates and coreshell particles. The dye was included at the beginni ng of the synthesis of th e particles in one case (encapsulation), and after the coating of the templates in the second case (sequestration). It was observed that the probes sensed a nonpolar environment in the form er, and that the final position in the particles differed, depending on the structure of the probe; in the latter. This gave us the possibility of placing different dyes in differe nt parts of the nanoparticles, based on their polarities and interactions with the shell, and the method selected to load them in the particles. In this Chapter, we use this knowledge to e xplore whether probes that are expected to be placed in different parts of the particles can communicate through a resonance energy transfer processes (RET). To do this, we took advantage of the fluorescence response of the two probes used previously. As shown in Figure 6-1 A, th e overlap between the excitation of C153 and the emission of pyrene, makes this couple a very e fficient acceptor-donor pair, especially due to the high quantum yield of C153.116 This pair of probes has been us ed in the past in a few matrices, including micellar systems117 and silica surfactant composites.111 In the study by Ferrer and Lianos,111 a high efficiency for the RET process was observed. They proposed that both probes were placed near the head of the surfactans us ed to prepare the composites, near the most external methylene groups in the tail of the am phiphiles. In our case, we used the templates reported on Chapter 3, prepared with two differe nt dopants, to study the energy transfer in the 137

PAGE 138

precursor of the nanoparticles by sequestering both probes from aqueous suspensions. After this, a set of core-shell particles pr epared with pyrene during the synt hesis (system NPY) was used to sequester C153. Based on the result s in Chapter 5, we expected to see a low efficiency in this system since it was observed that th e probes ended in the different pos itions inside the particles. We then we extended this approach to use anot her pair of probes. Based on the fact that C153 is adsorbed on the shell of the particles when sequestered from aqueous suspension, we selected another hydrophobic probe for RET studies, 4-(dicyanomethylene)-2-methyl-6-( p dimethylaminostyryl)-4H-pyran (DCM), known to place itself in the ex ternal part of micelles, in proximity to the solvent front, which produces an emission with maximum above 600 nm.118-120 The emission of C153 and excitation of DCM overlap as shown in Figure 6-4 A. As a consequence, this pair can undergo a resonance en ergy transfer (RET) proce ss if they are located in close proximity. This was used to study th e ability of these probes to communicate through RET based on the degree of internalization af ter their uptake from aqueous suspensions, similarly to the previous system analyzed. To st udy this interaction between the probes in more detail, we measured how the energy transfer pr ocess affected the fluorescence lifetime of the donor. This was not possible with the previous RET pair due to limitations on the instrument used, which prevented us from measuring the lif etime fluorescence of pyrene. We used the timecorrelated single photon counting te chnique (TCSPC) for these studies. As explained in Chapter 2, the fluorescence lifetime of the donor would be affected only if the in teraction between the probes happened after excitation and not by asso ciation of the probes before the donor was excited. Based on this, if we were able to de tect energy transfer between the probes by steady state fluorescence and, correspondin gly, to observe variations on the fluorescence lifetime of the donor, we could conclude that there was a coll isional process in betw een the probes. If no 138

PAGE 139

variation in the lifetime could be observed, th en the energy transfer was produced by a static interaction.65 6.1.2 Results The fluorescence of pyrene before the addition of C153 was the typical response observed for the dye sequestered in the templates, as reported in Chapter 5. Addition of the acceptor produced a decrease in the in tensity of the first peak (I1) of the donor emission, as reported in Table 6-1 and observed qualitatively in Figure 6-1 B. This was confirmed by analyzing the third peak (I3) in the emission spectra of pyrene. In this case, similar trends were observed, although the decrease in the intensity of the emission of th e donor was slightly lower. It was not clear at this point the reason for this discrepancy. We focused our discussion in the first emission peak, since this signal is more separated from the emission peak of C153, which could overlap (although minimally) with the emission of pyrene. As observed in Table 6-1, the emission of the donor decreased to approximately 45% of its orig inal value when the co ncentration of C153 was increased to 2.45 M. The values obtained at the highe st C153 concentration studied were probably not accurate, since it was observed that the suspensions became slightly turbid. This could be a consequence of the saturation of the templates, which would prevent them from sequestering more acceptor molecules, or the adsorp tion of the probe on the external part of the particles, hiding the charges on their surface (as re ported for the surfactants in Chapter 3), thus promoting their aggregation. At the same ti me, the appearance of a peak above 500 nm was observed clearly when the concentration of th e acceptor was increased to approximately 0.6 M, confirming that C153 was excited through the em ission of pyrene (Figur e 6-1 B). This was further confirmed by obtaining the ex citation spectrum of C153 at 0.6 M, as shown in Figure 61 C, in which the excitation through pyrene is shown as a maximum of lower intensity at 338 139

PAGE 140

nm. It is obvious the difference with Figure 6-1 A, where the peaks below 350 nm were not present in a typical excita tion spectrum of C153 sequest ered by the templates. As discussed above, these experiments were designed to study whether the probes can communicate when sequestered by the templates. The energy transfer observed confirmed that a certain fraction of the probes were placed in close environments, since the Frster radius for this pair has been calculated to be approximately 3.4 nm.111 Figure 6-1 D is a plot of the efficiency of the process calculated by compar ison of the emission of the donor before and after each addition of the acceptor. It was observed that, within e xperimental error, the different templates used showed a similar behavior. This supports the finding reported in the previous chapters that suggested that the identity and qua ntity of the dopant used to prep are the particles does not affect the polarity of the interior of the templates. Based on this, placement of the probes in different positions inside the particles by using different dopants was not expected, which was confirmed by the similarity of variation in the energy transfer when the concentration of the acceptor was increased. The small variations in the efficiency reported at low concentr ations of C153 could be a consequence of the difference in the number of templates in the suspensions, since, as explained earlier, despite having the same con centration in mass, the concentration of the particles may differ due to their different mass per particle. To understand this process better we tried fittin g the energy transfer da ta to a linear SternVolmer and to the modified Stern-Volmer plot s, without success (see Appendix B). Obviously, a linear plot was not expected to be satisfactory since we expect to have environments of different polarities inside the templates. Surprisingly, the modified plot, which has been used to represent energy transfer processes with probes placed in tw o or more different environments, did not offer consistent results. It was observe d that, in general, there was an improvement in the fitting 140

PAGE 141

parameters with the modified plot, but the results were not good enough to accept them as accurate. The fact that the data co uld not be fitted to these plots may be an indication that these particles represent a complex system, in which there is a distribution of environments with slightly different polarities, as a consequence of the degree of pe netration of the solvent inside the templates. To have an idea what percentage of the molecules of acceptor were excited through the energy transfer process, we excited the acceptor directly and compared the intensity of its emission to that of the acceptor excited through th e donor. This is shown in Table 6-2, together with the emission maximum of C153 in each case. The intensity of the emission was determined to be very similar in both cases, which indicate d that most of the accep tor molecules were in proximity to pyrene. This is assuming that the qu antum yield of C153 is the same in the different environments it is placed in the particles. One important control performed was exciting a sample of templates used to sequester only C153 at the pyrene excitation maximum. It was observed that the intensity wa s 2% of the intensity when excited at the C153 excitation maximum, confirming that the acceptor is not directly excited in the RET experiments. The maximum in the emission of the acceptor at highe r concentration shifted to the red when excited at its excitation maximum, which indicated that the excess accep tor molecules were accommodated in the external part of the particles, close to the solvent or even in direct contact with it, which explained the increase in th e turbidity of the suspension at high acceptor concentrations. 6.2 Energy Transfer in the Core-Shell Particles 6.2.1 Strategy In this case, based on the results obtained w ith the templates, we performed a simple experiment with the core-shell particles. We used a set of nanoparticle s prepared with pyrene 141

PAGE 142

during their synthesis (system NPY, Chapte r 4), to sequester C 153 at two different concentrations. Interestingly, the low in tensity-excimer peak observed during the characterization reported on Chapter 4 was not observed when performing these studies. This could be a consequence that the experiments re ported in this secti on were performed a few months after the particles were prepared. It is possible that during this time a small amount of dye had leached from the particles or that the inte rior of the particles beca me more rigid. In any case, the fluorescence spectra of th e donor reported here were taken the same day to be able to compare them. The concentration of the nanoparticles used was lo wer than the one used for the templates (approximately 0.2 mg/mL), since the core -shell particles scattered more light due to their larger size. As a consequence, the concen tration of C153 loaded in the suspensions was lowered as well. In the second system investigated, we used C153 as the donor and DCM as the acceptor. DCM has been used in micelles, observing that it tends to position itself in their periphery, near the hydrophilic solvent.118, 119 Preliminary results with our templates and core-shell particles showed that, similarly to C153, this probe was pr eferentially placed in contact with the solvent. Based on this and the overlap between the em ission of C153 and the excitation spectrum of DCM (Figure 6-4 A), we believe both dyes could be sequestered by the part icles to be placed in the hydrated shell, close enough to observe an en ergy transfer between the dyes. This differs from the previous system in which the most in ternalized molecules of C153 were expected to take part of the energy transfer. 6.2.2 Results 6.2.2.1 Pyrene-C153 pair Based on the fact that the Stern-Volmer plot s did not offer any insights on the energy transfer between the probes, we simplified th e experiments performed with the core-shell 142

PAGE 143

particles. Two different concentrations of the acceptor were used. At low concentrations (0.14 M) no acceptor emission was detected, although a low pyrene emission quenching of approximately 5% was observed (Figure 6-2 A, Table 6-3). When the acceptor concentration was increased to 0.7 M, the intensity of the donor emission di minished to approximately 68% (32% efficiency). At the same time, the emission of the acceptor was clearly observed as a broad peak centered at 527 nm, similar to the templates. When the acceptor was excited through pyrene (339 nm) the intensity of the emission of C153 was 66% of the one when excited directly (420 nm), as shown in Figure 6-2 B. This indicates that unde r these conditions, appr oximately 66% of the acceptor molecules interact with the donor, despite the presence of the she ll where an important fraction of the acceptor molecules were supposed to be adsorbed. It is important to note that the concentration of pyrene is unknown since it was included during the synthesis of the particles and some of the probe was lost during their pur ification. The excitation of the acceptor through interaction with the don or was confirmed by analysis of the C153 excitation spectrum, which showed that a fraction of the dye molecules we re excited by the donor. This confirms the RET process in the dye-doped particle s suggested by the pyrene emissi on decrease. This is obviously not a very efficient system, but our goal was to investigate the po ssibility of observing an energy transfer process based on the placement of the dyes inside the pa rticles as a means to study the compartmentalization of the interior of the core-shell particles. One important detail in the data obtained is the difference in the maximum in the emission spectrum of the accepto r after excitation through energy transfer from pyrene, (527 nm) and when directly excited (534 nm), similarly to wh at was observed with the templates. Since the emission of the dye shifts to the blue when th e polarity of its surrounding decreases, the values observed indicated that the acceptor molecule s emitting through RET, which are the ones in 143

PAGE 144

contact with the donor, are the ones better protected from the solvent. Base on these maxima, the shift observed is similar to a change in the solv ent where the dye is diss olved from methanol to n -propanol. We interpreted this as an indication of the protection of the dye from water, meaning that the acceptor molecules in contact with the do nor are the ones more internalized, as expected based on the polarity reported by pyrene. It is clear from these results that only a fracti on of the two different pr obes are in proximity to each other, due to their differences in polar ity and possibilities of noncovalent interactions with the shell, as represented in Figure 6-3. We believe that in between the external part of the shell filled with solvent molecules and the dry core region, where solvent molecules are not present, there is an intermediate polarity region with a low concentration of solvent molecules. This is the volume that the probes performing th e energy transfer occupied which, based on the C153 emission, is similar in polarity to n-propanol. 6.2.2.2 C153-DCM pair Figure 6-4 B shows the emission of the nanopart icles loaded with C 153 at a concentration of 0.7 M before and after sequestering the acceptor. The typical emission profile for the sequestered donor when excited at 420 nm was ob served in the absence of DCM, with emission maximum at 539 nm, indicating, as reported on Chap ter 5, that the dye was in proximity to the solvent. After addition of a small concentrati on of the acceptor, the emission of the donor is distorted. A broader emission, with a small peak at the red end of the spectrum was observed, as shown in Figure 6-4 B, which suggested that th e probes could communicate. Contrary to what we expected, the emission of the donor did not decr ease, but slightly increased. We believe this is due to the emission from DCM excited by the C153 molecules, which involves a broad peak starting at 500 nm. This may produce an increas e in the apparent donor emission, if RET was actually happening. When the concentration of DCM was increased to a similar value than the 144

PAGE 145

donor, it was observed that the C153 emission wa s quenched to 83% of the value for the DCMfree particles, confirming the energy transfer between the dyes. This indicated a transfer efficiency of only 17%, but due to the superposition of the emissi on of the dyes, which may have increased the apparent emission of the donor, this would be the lowest efficiency possible under our experimental conditions We then analyzed the excitation spectrum of the particles loaded with equimolar amounts of the probes with the emission fixed at the acceptor maximum. The graph obtained looked like a combination of the donor and acceptor excitation profiles, as shown in Figure 6-4 C, which confirmed the RET pro cess. To investigate how the emission of the acceptor affects the intensity of the emission of the donor, we exci ted the C153-loaded particles at the donor and acceptor excitation maximum, 420 and 475 nm, respectively. Figure 6-4 D shows that the emission of DCM involves a br oad peak, which overlaps with the emission of C153. This may explain the increase in the in tensity of the donor after DCM addition. We studied the transfer process more deeply by m easuring the fluorescence lifetime of the donor, as shown in the next section. 6.3 Energy Transfer in the Core-Shell Partic les Studied by Fluorescence Lifetime Measurements 6.3.1 Strategy Due to the overlap of the emission of th e donor and acceptor, we decided to study the changes in the lifetime of C153 after addition of DCM. In order to observe changes in the fluorescence lifetime of the donor as a consequence of energy transfer to an acceptor, the process has to involve a collisional encounter of the pr obes. The acceptor has to find the donor during the time this is excited for the energy transfer to occur and quench the emission of the donor. Had the probes made a complex before excitation, no changes in th e fluorescence lifetime would be observed. The studies presented in this section would give very important information regarding 145

PAGE 146

the kind of process leading to the resonance energy transfer observed by steady state fluorescence spectroscopy. 6.3.2 Results The fluorescence lifetime of C153 has been reported by different groups in different conditions. In general, in pure so lvents a single exponential is found to describe the fluorophore, since it is well known that C153 has a single excited state res ponsible for its fluorescence. Typical lifetime values are 4.3 ns in methanol,121 5.2 ns in toluene and 6.5 ns in acetonitrile.114 The value in micelles and vesicl es usually involves 2 components one in the 0.7-3.0 ns range and the other similar to the one observed in pure solvents, approximately 4-5 ns.122 In our results, a single exponential fit to the lifetime decay data was discarded due to the extremely high 2 value, which indicated that it did not represent the system properly (see Appendix C). A double exponential was found to give a reas onable fit, together with the typical value about 4.5 ns, we found a longer decay (Table 6-4). The relatively high 2 value obtained may be due to the possibility of having the probes di stributed over the different regi ons inside the particles, in environments with slightly different polarities. Trying to fit this data to a set of just two different families of fluorophores is, obviously, only an ap proximation to the real case. When a small amount of acceptor was added to the C153-doped particles both lifetim e values decreased, although the longer one is affected the most. The contribution of both species remains the same within experimental error. This produced a decrease in the average lifetime, which is the most accurate comparison in systems with a nonsi ngle exponential decay, confirming the energy transfer between the donor and acceptor. Interest ingly, these results contradicted those obtained with steady state fluorescence, in which a slight increase in the intensity of the emission of the donor was observed. This may be a consequence of the overlap of the emission of the donor and acceptor, as explained above. When an equimo lar amount of DCM was added, the average 146

PAGE 147

lifetime decreased to approximately 54% of its orig inal value, giving an e fficiency value of 46%, higher than the one reported in the steady state studies, as shown in Table 6-4. These results showed that the energy transfer between the prob es is a collisional quenching of the fluorescence lifetime of the donor. This indicates that the pair of probes was not associated w ithin the coreshell particles and more importantly, that th e probes can diffuse inside the particles. 6.4 Use of the Fluorescence Lifetime of a Probe Chemically Attached to Study Microenvironments Inside the Core-Shell Particles 6.4.1 Study of the Fluorescence of Dansyl Chlo ride Attached to the Periphery of the Particles 6.4.1.1 Strategy The system NPDCl-2A was prepared by r eacting APSDCl with the surface of the templates without a coating, while the particle s 2B were prepared by attaching the APSDCl complex after the growth of a thick shell, as explained on Chapter 4. The idea behind this approach was to study the solva tion of a probe whose position ha s been fixed on the particles by fluorescence lifetime measurements, as explained in Chapter 2. Different groups have studied th e distribution of dansyl in different environments when incorporated in different matrices by the determination of its fluorescence lifetime.76, 77, 123 Two decays have been identified matchi ng the excited states of the dans yl moiety. A short decay time, corresponding to the TICT state, where the dimethyl amino gr oup and the naphthyl group are in a nonplanar state in a hydrophilic environment, and a longer decay time, corresponding to the probe in a coplanar state in a hydrophobic environment.77 6.4.1.2 Results We start our discussion for the results obtai ned with the steady state fluorescence data. First of all, the sample NPDCl 2A showed a weaker emission for a similar concentration of particles than the system 2B, approximately 0.2 mg /mL. We believe this is due to the fact that 147

PAGE 148

the sample 2B offered more groups for the APSD Cl complex to react with, due to the higher surface area and the presence of internal silanol groups of the shell that can r eact with the dye. Table 6-5 shows the emission maxima detected for each system when excited at 340 nm. As explained earlier; this case is more complex that the ones previously studie d due to the fact that the emission of dansyl chloride is affected by both the polarity and rigidity of the media. The system without a shell presented a red shifted signal ( max= 523 nm) which may mean that the dye is placed either in a polar or flexible environment. As expected, opposite results were observed for the system 2B due to the presence of a shell which is a more rigid but hydrated matrix ( max= 510 nm). Based on the maximum detected in the emission of the dye, it seemed like the polarity of the medium was lower compared to the system 2A or that the rigidity of the environment surrounding the probe was higher due to the presence of the she ll, although, at this point, it was difficult to have a clear picture of which factor was more important. To clarify this point we turned our attention to the study of the fluorescence lifetim e of dansyl chloride attached to the nanoparticles, as discussed below. The fluorescence lifetime data showed intere sting results on the f actors affecting the emission of DCl. Both samples decays were fit to a double exponential as shown in the Table 65. As explained for the antenna system fo rmed with C153 and DCM, this is only an approximation to the real case. In both cases, the major contribution detected corresponded to the nonTICT-state, suggesting that the APS complex penetrated both, the templates and core-shell particles to an important extent before reacting. Surprisingly, the system APSDCl 2A showed a lower contribution of the fluorophore emitting through the TICT-state ( = 2.32 ns, 13%), indicating that appr oximately 13% of the detect ed fluorophores were in co ntact with the solvent. This suggests that most of the dye ended up conn ected to the interior of the templates where 148

PAGE 149

access for the solvent molecules is more restri cted, showing a major contribution from the fluorophore emitting through the non-TICT state ( = 16.4 ns, 87%). System APSDCl 2B, on the other hand, showed a shorter mean lifetime, indica ting, that in average, the probe sensed a more polar or less rigid environment. The higher contribution from the fluorophore in contact with the solvent, emitting through the TICT-state ( = 2.93 ns, 25%) may be due to the presence of silanol groups in the interior of the she ll that the probe could have reacted with. This placed the dye not only in the outer part of the shell, but in its interior, which, as determined previously, is extensively hydrated, giving the probe a more polar environment. As a consequence, the contribution of the protected dye ( = 15.13 ns, 75.4%), diminished. 6.4.2 Preliminary Studies on the Energy Transf er between Dansyl Chloride Attached to the Periphery of the Particles and Sequestered DCM 6.4.2.1 Strategy Based on the results from the lifetime fluor escence studies reported on the previous section, we decided to study the energy transfer be tween dansyl choride att ached to the particles, partially expose to the solvent, and DCM, wh ich was observed to sense a polar environment when sequestered by the core-shell particles. The results showed in the following section are preliminary since they were done in the search for the best conditions to obtain an energy transfer between dansyl and DCM. As a cons equence, each system was used under different conditions and the results would be examined briefl y. The most important reason to show them is to state the feasibility of using the core-shell par ticles with a fixed probe in the design of antenna systems. 6.4.2.2 Results For the nanocapsule system APSDCl 2A, th e acceptor was added as a solid and the suspension sonicated for 30 minutes and stirred over night. After this, the samples were filtered to 149

PAGE 150

remove any excess dye. When the original nanop article suspension was excited at 370 nm the typical emission from DCl on the surface of the particles was obtai ned, as shown in Figure 6-5 A. When excited at 475 nm (the acceptor ex citation maximum) no fluorescence was detected, confirming that the donor did not interfere with the emission of the acceptor. Excitation of the loaded particles at 475 nm showed the typica l DCM fluorescence, with a maximum in its emission of approximately 620 nm, an indication that the dye was in placed close to the solvent. Excitation at 370 nm produced a broader peak, si milar to a combination of the DCM and DCl emissions, confirming that there is an energy tran sfer process between the dyes. We believe this system could be improved to obtain a higher efficiency since both dyes would be placed in the same volume in the particles, in contact with th e solvated shell. Unfort unately, the emission of the dyes was recorded at different conditions, so the intensities could not be compared to use them to calculate the efficiency of the process. The system APSDCl 2B was studied in a diffe rent way. DCM was loaded in the particles from a methanolic solution to reach two differe nt concentrations, and the emission of the donor was analyzed. Figure 6-5 B shows a comparison of the emission before and after addition of the acceptor. It was observed that the intensity of th e emission of DCl changed after incorporation of DCM in the system, giving a transfer efficiency of 36%. It was not clear the reason for not observing a peak for the emission DCM, but a si gnal dominated by scattering, as marked in Figure 6-5 B. Interestingly, the decrease in the emission intensity of DCl was very similar for both concentrations of the acceptor added, with on ly a 3% increment in the transfer efficiency observed after a 5-fold increased in the concentr ation of DCM. This is in agreement with the results of the fluorescence lifetime studies for dans yl attached to the particles, where it was 150

PAGE 151

determine that only 25% of the probe was exposed to the solvent. Based on this, the efficiency of the transfer is not expected to increase after further addition of DCM. 6.5 Conclusions This chapter showed results on studies concerning the possibility of designing antenna systems to study the compartmentalization of the templates and core-shell particles synthesized as reported in the previous chapters. The obser vation of energy transfer between pyrene and C153 sequestered by the templates showed that the probes share a volume in their interior. Since the Frster distance necessary for this pair of dye s to observe energy transfer is only 3.4 nm we can conclude that the dyes were very close to each other. A more detailed study of this phenomenon was tried by fitting the data to mode ls representing systems that can accommodate the probes in different environments, without succe ss. We believe this is an indication of the complex system these templates and particles represent. A sim ilar study was performed for the core-shell particles, using particles prepared by incorporation of pyrene dur ing the synthesis. In this case, a more simple approach was used, observing that energy transfer was possible with this system as well. After establishing that energy tr ansfer between different probes was possible in the templates and core-shell particles we studied a system with the two probes expected to reside in the shell, in a volume close to the periphery of the particles. The pair C153-DCM was used, observing similar results to those obtained with the previous pair. In this case, in addition to the steady state fluorescence studi es, a study on the fluorescence lifetime of the donor was performed to gain a better understanding of the energy transfer process observed. It was determined that this was a collisional quenchi ng of the emission of the donor by the acceptor, which means that the probes were not associated but diffusing inside the part icles. The efficiency of the transfer was determined to be relatively high despite the use of low loadings, probably due to the close proximity of the probes. These ar e promising findings that suggest that these 151

PAGE 152

particles can be used as sensor for different quenchers. Finally, we studied the fluorescence lifetime of dansyl chloride chemically connected to the templates after their preparation and to the core-shell particles after the growth of a thic k shell. The results showed that the templates offered pores in their interior where the probe can react, which hides the dye from the solvent, as observed with pyrene. The dye attached to the co re-shell particles on the other hand, showed in its fluorescence lifetime that a higher fracti on of the fluorophore was exposed to water, confirming the hydration of the shell, where most of the dansyl derivative was supposed to have reacted. These dye-coated particles were used for preliminary studies to determine the feasibility to use them for the design of antenna systems, observing that the probe could communicate with DCM although more studies are necessary. Table 6.1 Changes in the intensity of the firs t peak of sequestered pyrene emission (371 nm) after addition of C153 [C153] EBB EBC HXB HXC I1 relative E I1 relative E I1 relative E I1 relativeE 0 1 0 1 0 1 0 1 0 0.122 0.86 14 0.92 8 0.89 11 0.91 9 0.306 0.76 24 0.74 16 0.84 16 0.79 21 0.610 0.60 40 0.66 24 0.66 34 0.73 27 1.22 0.64 36 0.61 39 0.63 37 0.63 37 1.83 0.51 49 0.52 48 0.53 47 0.53 47 2.45 0.44 56 0.44 56 0.45 55 0.45 55 I1 relative calculated as the ratio between the intensity of the peak at 371 nm in the emission of pyrene before and after addition of C153, after background correction. The efficiency is obtained as (1I1 relative) x 100. Table 6.2 Emission maximum of sequestered C153 in pyrene-doped particles and comparison of the intensity of its emission obtai ned when excited at 339 and 420 nm [C153]/ M max, exc 339 nm max, exc 420 nm I339/I420 a 1.22 527 527 1.03 1.83 528 531 0.97 2.45 529 536 0.87 a values compared after background correction 152

PAGE 153

Table 6-3 Steady state fluor escence data for the pyrene-C153 encapsulated pair ( ex= 339 nm). [C153]/ M I371 (relative) a E(%)b I C153 (527 nm) 0 1.0 no 0.14 0.95 5 no 0.70 0.66 44 yes a I371 values compared afte r background correction,b E values calculated from (1(I371 DA/I371 D)) x 100, D= donor, DA= donor in the presence of acceptor Table 6-4 Steady state and fluores cence lifetime measurements results for the energy transfer between C153 and DCM. [DCM]/ M 1 (ns) b1 2 (ns) b2 2 mean(ns)a E(%)b 0 4.62 0.89 28.32 0.11 2.26 7.23 0.14 4.06 0.89 19.51 0.11 1.89 5.76 20 0.70 3.74 0.87 15.56 0.13 1.78 3.88 46 a average lifetime values obtained by mean= bii, b efficiency obtained by fluorescence lifetime measurements E= (1-( DA/ D))x100, D= donor ,DA= donor in the presence of acceptor Table 6-5 Steady state and fluor escence lifetime measurements results for templates and particles coated with dansyl chloride max, exc 340 nm 1 (%) 2 (%) mean 2 APSDCl 2Aa 523 2.32 (13.1) 16.4 (86.9) 14.56 1.05 APSDCl 2Bb 510 2.93 (24.6) 15.1 (75.4) 12.11 1.76 mean= bi i, exc 375 nm, a em 525 nm, b em 510 nm 153

PAGE 154

Figure 6-1. Energy transfer between pyrene and C153 in templates. A) Overlap of the emission of sequestered pyrene and excitation of sequestered C153, B) changes in the fluorescence spectrum of pyrene (1 M) after addition of C 153 C) typical excitation spectra of sequestered C153 (0.6 M, exc 520 nm) in pyrene-loaded templates (1 M), D) effciency of the energy transfer observed for templates EBB (diamonds), EBC (squares), HxB (triangles), HxC (circles). 154

PAGE 155

Figure 6-2. Energy transfer between pyrene and C153 in core-shell particles. A) Emission spectra of encapsulated pyrene before (solid line), and after sequestration of to different concentrations of C153 (0.14 M-empty circles and 0.7 M-empty triangles); B) comparison between the emission of the doubled-doped nanoparticles excited at the excitation maxima of pyrene (solid line) and C153 (empty circles). Figure 6-3. Schematic representation of the energy transfer between probes based on their placement inside the nanoparticle s. The particles are divided in zones with different polarities based on the penetr ation of the solvent. 155

PAGE 156

Figure 6-4. Energy transfer between C153 and DCM. A) Overla p between C153 emission (solid line) and DCM excitation (empty circles); B) emission spectra of nanoparticles doped with sequestered C153 (0.7 M, empty circles), and af ter sequestering DCM (0.14 M-empty triangles and 0.7 M-solid line); C) excitation spectra for sequestered C153 (0.7 M, empty circles, emission 539 nm), sequestered DCM (0.7 M, solid line, emission 615 nm) and DCM in nanopart icles with both dyes sequestered (0.7 M each, empty triangles, emission 610 nm); D) comparison between the emission of the doubled-doped nanoparticles (0.7 M each dye) excited at the excitation maximum of C153 (420 nm, solid line) and DCM (475 nm, empty circles). 156

PAGE 157

Figure 6-5. Energy transfer between DCl and DCM. A) Nanopa rticles system APSDCl-2A, exc 370 nm (blue line), exc 450 nm (black line); solid DCM added: exc 370 nm (green line), exc 450 nm (red line). B) Nanoparticles system APSDCl 2B: exc 340 nm (black line); after sequestration of DCM: 0.5 M (red line), 2.5 M (green line) and 2.5 M ( exc 450 nm). 157

PAGE 158

CHAPTER 7 CONCLUSIONS AND SUGGESTIONS FOR FUTURE WORK 7.1 Conclusions The work presented here took a different dire ction from the ideas originally discussed when this part of the project was started. We believed it was necessary to take the time to perform some basic studies to confirm that the ideas we based our understanding of the particles we obtained were correct. The main goal of th e project was to prepar e nanoparticles with a hydrophobic core and a hydrophilic shell that could sequester drugs a nd pesticides to be used as a fast response in cases of intoxication. The pa rticles were synthesized using a mixed surfactant system. OTMS and t80 were mixed to obtain spherical particles, called templates. The use of t80 was necessary since it was observed that OTMS by itself tends to form large aggregates in aqueous suspensions. The mixed-surfactant sy stem was then fortif ied by hydrolyzing and condensing the polar groups of OTMS. Subsequently these particles were co ated with a silicate precursor. The product obtained was believed to be made of a hydrophobic core, mainly formed by the tail of the surfactants; a nd a hydrophilic shell, formed by th e silica coating. It was shown that these particles could be prepared by the use of a microemulsion as well, obtained by the addition of a small hydrophobic molecule (terme d a dopant in this wo rk) to the binary amphiphile system. This was expected to tune the polarity of the core by trapping the dopant in the interior of the particles by growing a shell on their peri phery. The particles obtained by previous work in the group were used successful ly to sequester small organic molecules from aqueous suspensions. This was interpreted as an i ndication of the core of the particles attracting nonpolar molecules due to its low polarity, althou gh no data was available to confirm this. At this point, it was decided that it was necessary to elucidate if the driving force for this process 158

PAGE 159

was the one proposed. As a consequence, a signif icant amount of the work presented here dealt with the characterization of the templates and core-shell particles synthesized in our work. The first step taken was to improve the yield of templates and particles obtained in our synthetic approach, since with our previous methodology the low yield of particles obtained made their purification and characterization tedi ous. This was done by adding an ionic surfactant and reducing the amount of t80. The purification was changed fr om dialyzing the suspensions obtained, to using small membranes with a di ameter cut-off of 25 nm. This separated the templates, with diameters of above 60 nm, from the micelles formed by th e excess surfactant, of sizes below 10 nm. After this, ch aracterization of the templates and core-shell particles became easier. DLS showed that the size of the templates could be tuned by varying the loading of the surfactants. TEM images showed th at the particles consisted of a very light material that was difficult to image due to the low contrast in th e microscope. It was observed on the atomic force microscope that after deposition on a substrate they flattened considerably, forming disk-like species. We believe this happens as the solvent evaporates, leaving the external volume of the particles, which is hydrated in solution, empty. Similar studies were performed with the dopantfilled templates. DLS results showed that the size of the doped-templates could be tuned by the loading of the dopant. The AFM images showed that these templates flattened as well after drying. Surprisingly, contrary to the templates prepared only with the amphiphiles, folded and bent particles were observed, indi cating that they were made of a material of higher flexibility than the dopant-free templates. Preliminary studies on the strength of these templates were performed by calcinating them at 600 C. It was observed that the templates broke down, leaving a fiber-like material. 159

PAGE 160

The templates coated with the silicate shell we re characterized similarly to the templates. DLS data showed that particles with different shell thicknesses could be prepared. AFM images showed that the coated particles were not as flexible as the templates and that the thickness of the coating determines the extent the particles flattene d on the substrate. This indicated that the shell formed fortified the particles producing more rigid structures. Based on some simple calculations, it was determined that the volume of the templates was dramatically reduced after drying. This was suppressed when a thin coati ng was formed on the par ticles, but it increased again with a thicker shell. We be lieve these were indications of the hydration of the core and the shell. Calcination of these core-shell particles showed that, after coating, the particles did not break down after heating to 600 C, confirming that the siloxane shell fortified the particles. We then turned our attention to the study of the sequestration of different probes by the templates and core-shell particle s. We designed an experimental approach that allow the study of species trapped by the particles, contrary to typical studies that only quantify the uptake by differences in the UV absorption of the non sequestered probes. Pyrene and C153, two hydrophobic fluorescence dyes, were used in these st udies as sensors of polarity and rigidity. Pyrene fluorescence showed that sequestration by the templates placed it in a nonpolar environment, protected from the solvent. Interest ingly, the polarity reported by the probe did not resemble the one from the dopant. This is believ ed to indicate that a significant fraction of the dopant molecules were removed during purificati on, producing a more flexible material, which correlates with the AFM da ta that showed that th ese particles could be fo lded without breaking. A nonpolar environment was reporte d by pyrene for the core-shell pa rticles as well, confirming that the particles internalized this hydrophobic probe. C153 on the other hand, showed that the final position of the probe inside the particles depended on its chemical structure, since due to 160

PAGE 161

hydrogen bonding interaction, this molecule was tr apped on the shell of the particles sensing a polar environment. These results indicated that we can identify the position of the dyes based on their emission profiles and that by knowing the chemical stru cture of the probe to be sequestered, we could predict its final pos ition inside the core-shell particles. As a next logical step, the in formation obtained with the fluo rescent probes was used to design antenna systems that could potentially be used as sensors for different purposes. We accomplish this by using two pair of probes in which the emission of one of them overlapped with the excitation spectrum of the other. The probes were chosen not only because of their fluorescence response but for the position in the part icles they were expected to place themselves based on their structures. The pair pyrene-C153 s howed a transfer efficiency of approximately 50%, indicating that, despite sensi ng different polarities in averag e, an important number of the probes can communicate and approxi mate each other, probably in an area of intermediate polarity inside the particles. The third probe used, DCM, known to place itself on the periphery of micelles was selected to act as an acceptor wi th C153. Contrary to what we expected, the efficiency of the energy transfer obtained by steady state fluorescence spectroscopy was only 17%. We studied this process by analysis of the changes in the fluor escence lifetime of the donor, obtaining an efficiency of 46%. This proved that the energy transfer was a collisional process, showing that the probes we re not associated in the particles, but that they approximate each other after the donor has been excited. In the final part of this work, we studied templates and particles reacted with dansyl chloride derivatized with sila nol groups, which was expected to place the fluorescence probe in their periphery. St udy of the lifetime of the probe indicated that most of the dye was somehow protected from the solvent, with a low pe rcentage of fluorophores in a polar, solvated environment. Apparently a significant fraction of the probe molecules 161

PAGE 162

diffused to the internal part of the shell before reacting. Finally, we showed preliminary results on the used of these dye-coated-particles for energy transfer to a sequestered probe, which broadens the choices for the design of particles for different applications. We believed this work has confirmed that th e templates and core-shell particles can be obtained by the methodology develop by our group. The particles obtained are stable and they can be characterized by different techniques. The ability to contro l their sizes allows tuning the characteristics of the particles accordingly to the application in mind, makes them a versatile system. The use of fluorescent probes was proved to be a successful methodology for the characterization of the compartmen talization of particulate systems. The fluorescent probes used in this study confirmed that molecules of low polarity can be sequester ed by the nanoparticles and that they were protected from the media. It was observed that the chemical structure of the molecules sequestered determine their final position inside the particles. These findings were used to place different probes inside the par ticles and study whether they were in contact. Fluorescence resonance energy transfer was obser ved between three differe nt pairs of probes. Fluorescence lifetime measurements of the donor of one of these pairs demonstrated that the energy transfer was produced by a collisional process. By attaching a dye to the templates and core-shell particles we studied the degree of so lvation of a probe reacted at the end of the synthesis of the particles. It was observed that the reaction did not take place only on the most external parts of the particles, but that the probe diffused to the interior, due to its low polarity, before reacting. 7.2 Suggestions for Future Work All the information gathered in the different steps of this work suggests a few important ideas. First, the templates are strong, stable stru ctures that can be used for different purposes, especially for sequestering processes. They do not need to be coated. We believe the coating 162

PAGE 163

should be performed in two cases. First, when a probe or dopant is included in the formulation used for the synthesis of the particles. With th is, the possibility of removing these species during purification is lower, due to the isolation of the interior from the media by the shell. The second case involves the attachment of different probes to the particles. Due to the well-known derivatization strategies for silicate matrices, it w ould be useful to have a silica coating that could be derivatized by different routes. Doping the part icles with fluorescent dyes makes it possible to use them as fluorescent labels for different applic ations, including biologi cally relevant systems. We have demonstrated that the particles can be prepared with dyes that absorb and emit at different wavelengths. This could be used to tune the response of the particles based on the needs of particular applications. Fina lly, doping the particles with diffe rent dyes at the same time is a suitable strategy for the design of sensors, in which a nonpolar dye can be used as a reference signal, and a dye placed on the periphery of the part icles, potentially able to interact with species approaching the surface of the particles, c ould be used as a detection response. 163

PAGE 164

APPENDIX A APSDCL COMPLEX NMR N S H3C CH3 O N O C H2 H2C C H2 Si OCH3 OCH3 OCH3 abcdef ghijH Figure A-1. Chemical structure of the APSDCl complex. ppm (t1 ) 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 0. 0 1. 0 2. 0 3. 0 4. 0 5. 0 6. 0 1.00 1.95 2.24 0.98 5.98 0.73 0.45 9.18 0.81 8.03 1.26 0.75 1.76 c b d a CDCL 3 f j e g h i f Figure A-2. NMR spectrum of the APSDCl complex. 164

PAGE 165

APPENDIX B MODIFIED STERN-VOLMER PLOT S FOR EB AND HX TEMPLATES Figure B-1. Modified Stern-Volmer pl ots for hexadecane-doped templates. 165

PAGE 166

APPENDIX C FLUORESCENCE LIFETIME FITS FOR C153-DCM FRET Table C-1 Fluorescence lifetime fits for C 153-doped templates used for FRET with DCM. 1 (ns) 2 (ns) 3 (ns) 2 C153 0.7 M 5.25 6.01 4.62 (88.8%) 28.32 (11.1%) 2.26 0.588 (-12.6%) 4.26 (98.4%) 16.7 (14.16%) 1.09 C153 0.7 M + 0.14 M DCM 4.58 5.26 4.06 (88.6%) 19.51 (11.4%) 1.89 0.393 (-6.6%) 3.79 (90.7%) 13.8 (16%) 1.18 C153 0.7 M + 0.7 M DCM 4.36 5.6 3.74 (87.2%) 15.56 (12.9%) 1.78 0.53 (-17.9%) 3.37 (98%) 11.1 (19.9%) 1.38 166

PAGE 167

LIST OF REFERENCES (1) Barrat, G. Cellular and Molecular Life Sciences 2003, 60 21-37. (2) Sukhorukov, G.; Fery, A.; Brumen, M.; Mhwald, H. PCCP 2004, 6 4078-4089. (3) Hof, F.; Craig, S.; Nuckolls, C.; Rebek Jr, J. Angew. Chem. Int. Ed 2002, 41, 1488-1508. (4) Djatali, R.; Samsom, J.; Matsui, H. J. Am. Chem. Soc. 2004, 126, 7935-7939. (5) Kaanumalle, L.; Nithyanandhan, J.; Pattabiraman, M.; Jayaraman, N.; Ramamurthy, V. J. Am. Chem. Soc. 2004, 126, 8999-9006. (6) Liu, S.; Weaver, J.; Save, M.; Armes, P. Langmuir 2002, 18, 8350-8357. (7) Shchukin, D.; Sukhorukov, G. Adv. Mater. 2004, 16, 671-682. (8) Lee, M.; Jan, C.; Ryu, J. J. Am. Chem. Soc. 2004, 126, 8082-8083. (9) Wang, X. Q.; Dai, J. D.; Chen, Z.; Zhang, T.; Xia, G. M.; Nagai, T.; Zhang, Q. Journal of Controlled Release 2004, 79, 421-429. (10) Cai, T.; Hu, Z.; Ponder, B.; John, J.; Moro, D. Macromolecules 2003, 36, 6559-6564. (11) Haag, R. Angew. Chem. Int. Ed 2004, 43, 278-282. (12) Twyman, L.; King, A.; Martin, I. Chem. Soc. Rev. 2002, 31, 69-82. (13) Underhill, R.; Jovanovic, A.; Carino, S.; Varshney, M.; Shah, D.; Dennis, D.; Morey, T.; Duran, R. Chem. Mater. 2002, 14, 4919-4925. (14) Zarur, A.; Ying, J. Nature 2000, 302, 65-67. (15) Kim, M.; Seok, S.; Ahn, B.; Koo, S.; Paik, S. Journal of Sol-Gel Science and Technology 2003, 27, 355-361. (16) Prichanont, S.; Leak, D. J.; Stuckey, D. C. Colloids and Surfaces A 2000, 166, 177-186. (17) Allen, C.; Maysinger, D.; Eisenberg, A. Colloid and Surfaces B 1999, 16, 3-27. (18) Miller, D.; Kabanov, A. Colloid and Surfaces B 1999, 16, 321-330. (19) Kuang, M.; Duan, H.; Wang, J.; Chen, D.; Jiang, M. Chem. Commun. 2003, 4 496-497. (20) Sanji, T.; Kakatsuka, Y.; Ohnishi, S.; Sakurai, H. Macromolecules 2000, 33, 8524-8526. (21) Hecht, S.; Frechet, J. Angew. Chem. Int. Ed 2001, 40, 74-91. 167

PAGE 168

(22) Frechet, J. PNAS 2002, 99, 4782-4787. (23) Cameron, C.; Gorman, C. Adv. Funct. Mater. 2002, 12 17-20. (24) Nakajima, R.; Tsurut a, M.; Higuchi, M.; Yamamoto, K. J. Am. Chem. Soc. 2004, 126, 1630-1631. (25) Liu, L.; Breslow, R. J. Am. Chem. Soc. 2004, 126, 12110-12111. (26) Vogtle, F.; Gestermann, S.; Hesse, R.; Schwierz, H.; Windisch, B. Prog. Polym. Sci. 2000, 25, 987-1041. (27) Aulenta, F.; Hayes, W.; Rannard, S. European Polymer Journal 2003, 39, 1741-1771. (28) Couvreur, P.; Barrat, G.; Fattal, E.; Legrand, P.; Vauthier, C. Critical Reviews in Therapeutic drug Carriers Systems 2002, 19, 99-134. (29) Newkome, G.; Zhongqi, Y.; Baker, G.; Gupta, V. J. Org. Chem. 1985, 50, 2003-2004. (30) Morgan, M.; Carnahan, M.; Immons, C.; Ribeiro, A.; Finkelstein, S.; Lee, S.; Grinstaff, M. J. Am. Chem. Soc. 2003, 125, 15485-15489. (31) Furuta, P.; Brooks, J.; Thompson, M.; Frechet, J. J. Am. Chem. Soc. 2003, 125, 1316513172. (32) Furuta, P.; Frechet, J. J. Am. Chem. Soc. 2003, 125, 13173-13181. (33) Sunder, A.; Kramer, M.; Ha nselmann, R.; Mulhaupt, R.; Frey, H. Angew. Chem. Int. Ed 1999, 38, 3552-3555. (34) Schappaper, M.; Deffieux, A. Polymer 2004, 45, 4633-4639. (35) Chen, G.; Guan, Z. J. Am. Chem. Soc. 2004, 126, 2662-2663. (36) Niu, Y.; Crooks, M. Chem. Mater. 2003, 15 3462-3467. (37) Soppimath, K.; Aminabhavi, T.; Kulkarni, A.; Rudzinski, W. J. Controlled Release 2001, 70, 1-20. (38) Skirtach, A.; Antipov, A.; Shchukin, D.; Sukhorukov, G. Langmuir 2004, 20, 6988-6992. (39) Peyratout, C.; Mohwald, H.; Dahne, L. Angew. Chem. Int. Ed 2003, 115, 1722-1726. (40) Khopade, A.; Caruso, F. Nano Letters 2002, 2 415-418. 168

PAGE 169

(41) Sun, B.; Chiu, D. Langmuir 2004, 20, 4614-4620. (42) Wang, Y.; Caruso, F. Chem. Commun. 2004, 13, 1528-1529. (43) Balabushevich, N. G.; Larionova, N. I. Biochemistry (Moscow) 2004, 69, 930-936. (44) Turner, J.; Wooley, K. Nano Letters 2004, 4 683-688. (45) Hu, Y.; Jian, X.; Ding, Y.; Chen, Q.; Yang, C. Adv. Mater. 2004, 16 933-937. (46) Graff, A.; Winterhalter, M.; Meier, W. Langmuir 2001, 17, 919-923. (47) Lawson, G.; Lee, Y.; Singh, A. Langmuir 2003, 19, 6401-6407. (48) Bolinger, P.; Stamou, D.; Vogel, H. J. Am. Chem. Soc. 2004, 126, 8594-8595. (49) Yang, X.; Zhang, Y. Langmuir 2004, 20, 6071-6073. (50) Wang, H.; Chen, P.; Zheng, X. J. Mater. Chem. 2004, 14, 1648-1652. (51) Jungmann, N.; Schmidt, M.; Ebenhoch, J.; Weis, J.; Maskos, M. Angew. Chem. Int. Ed 2003, 42, 1714-1717. (52) Jungmann, N.; Schmidt, M.; Maskos, M.; Weis, J.; Ebenhoch, J. Macromolecules 2002, 35, 6851-6857. (53) Kumar, R.; Chen, M.; Parmar, V.; Samuelson, M.; Kumar, J.; Nicolosi, R.; Yoganathan, S.; Watterson, A. J. Am. Chem. Soc. 2004, 126, 10640-10644. (54) Panyam, J.; Williams, D.; Dash, A.; Leslie-Pelecky, D.; Labhasetwar, V. Journal of Pharmaceutical Sciences 2004, 37, 1804-1814. (55) Bae, Y.; Fukushima, S.; Harada, A.; Kataoka, K. Angew. Chem. Int. Ed 2003, 42, 46404643 (56) Chu, B. Laser Light Scattering 1991, Academic Press, Inc, San Diego. (57) Precision Deconvolve32 User Manual 2003, (58) Johnson, C., Gabriel, D. Laser Light Scattering 1981, Dover Publications, Inc., New York. (59) Nanoparticle Technology for Drug Delivery 2006, Taylor and Francis group, New York. (60) Scanning Probe Micros copy Training Book 2000, Digital Instruments, 169

PAGE 170

(61) Ebenstein, Y.; Nahum, E.; Banin, U. Nano Letters 2002, 2 945-950. (62) Zeta Potential Analyzer, Instruction Manual 2008, Brookhaven Instrument Corp, New York. (63) Jesionowski, T. Colloid and Surfaces A 2003, 222, 87-94. (64) Wilhelm, M.; Zhao, C.-L.; Wang, Y.; Renliang Xu, a.; Winnik, M. A. Macromolecules 1991, 24, 1033-1040. (65) Lakowicz, J. R. Principles of Fluorescence Spectroscopy 1999, Kluwer Academic/Plenum Publishers, New York. (66) Vanderkooi, J. M.; Callis, J. B. Biochemistry 1974, 13, 4000-4006. (67) Kalyanasundaram, K.; Thomas, J. K. J. Am. Chem. Soc. 1977, 99, 2039-2044. (68) Grant, C. D.; DeRitter, M. R.; Steege, K. E.; Fadeeva, T. A.; Castner Jr, E. W. Langmuir 2005, 21, 1745-1752. (69) Kumbhakar, M.; Nath, S.; Mukherjee, T.; Pal, H. J. Chem. Phys. 2004, 121, 6026-6033. (70) Stathatos, E.; Lianos, P.; Stangar, U. L.; Orel, B. Chem. Phys. Lett. 2001, 345, 381-385. (71) Horng, M. L.; Gardecki, J. A.; Papazyan, A.; Maroncelli, M. J. Phys. Chem. 1995, 17311. (72) Li, X.; McCarroll, M.; Kohli, P. Langmuir 2006, 22, 8615-8617. (73) Wu, P.; Brand, L. Analytical Biochemistry 1994, 218, 1-13. (74) Weber, G. Biochem. J. 1951, 51, 155-167. (75) Li, Y.-H.; Chan, L.-M.; Tyer, M.; Moody, R. T.; Himel, C. M.; Hercules, D. J. Am. Chem. Soc. 1975, 97, 3118-3126. (76) Ding, L.; Fang, Y.; Jiang, L.; Gao, L.; Yin, X. Thin Solid Films 2005, 478, 318-325. (77) Montalti, M.; Prodi, L.; Zaccheroni, N.; Falini, G. J. Am. Chem. Soc. 2002, 124, 1354013546. (78) Leezenberg, P. B.; Frank, C. V. Chem. Mater. 1995, 7 1784. (79) Huang, Q. R.; Frank, C. V.; Mecerreyes, D.; Wolksen, W.; Miller, R. D. Chem. Mater. 2005, 17, 1521-1528. 170

PAGE 171

(80) Fontaine, P.; Goldmann, M.; Rondelez, F. Langmuir 1999, 1348-1352. (81) Lindn, M.; Slotte, J. P.; Rosenholm, J. B. Langmuir 1996, 4449-4454. (82) Vidon, S.; Leblanc, R. M. J. Phys. Chem. B 1998, 1279-1286. (83) Jovanovic, A. V.; Underhill, R. S.; Bucholz, T. L.; Duran, R. S. Chem. Mater. 2005, 17, 3375-3383. (84) Heydenreich, A. V.; Westmeier, R.; Pedersen, N.; Poulsen, H. S.; Kristensen, H. G. International Journal of Pharmaceutics 2003, 83-87. (85) Shinoda, K.; Arai, J. J. Phys. Chem. 1964, 68, 3485-3490. (86) Gonzles-Prez, A.; Ruso, J. M.; Prieto, G.; Sarmiento, F. Langmuir 2004, 20 25122514. (87) Morey, T. E.; Varshney, M.; Flint, J. A.; Seubert, C. N.; Smith, W. B.; Bjoraker, D. G.; Shah, D. O.; Dennis, D. M. Journal of Nanoparticle Research 2004, 150-170. (88) Katagiri, K.; Caruso, F. Macromolecules 2004, 37, 9947-9953. (89) Bourgeat-Lami, E.; Tissot, I.; Lefebvre, F. Macromolecules 2002, 35, 6185-6191. (90) ZetaPlus Potential Anal yzer Instruccion Manual 2003, (91) Acharya, K. R.; Bhattacharya, S. C.; Moulik, S. T. Journal of Photochemistry and Photobiology A: Chemistry 1997, 109 29-34. (92) Huang, H.; Kowalewski, T.; Wooley, K. L. J. Polym. Sci. Part A: Polym. Chem. 2003, 41, 1659-1668. (93) Khopade, A. J.; Caruso, F. Nano Letters 2002, 2 415-418. (94) Sarid, D. in Scanning Force Microscopy, Lapp, M., Stark, H., Eds. 1991, (95) Peyratout, C. S.; Mhwald, H.; Dhne, L. Adv. Mater. 2003, 15, 1722-1726. (96) Brinker, C. J.; Schere, G. W. 1990, Academic Press, Inc., San Diego. (97) Ni, K. F.; Shan, G. R.; Weng, Z. X. ; Sheibath-Othman, N.; Fevotte, G.; Lefebvre, F.; Bourgeat-Lami, E. Macromolecules 2005, 38, 7321-7329. (98) Tissot, I.; Novat, C. ; Lefebvre, F.; Bourgeat-Lami, E. Macromolecules 2001, 34 57375739. 171

PAGE 172

(99) Jan, J.-S.; Lee, S.; Carr, C. S.; Shantz, D. F. Chem. Mater. 2005, 17 4310-4317. (100) Jovanovic, A. V.; Underhill, R. S.; Bucholz, T. L.; Duran, R. S. Chem. Mater. 2005, 17, 3375-3383. (101) Huang, H.; Kowalewski, T.; Wooley, K. L. J. Polym. Sci. Part A: Polym. Chem 2003, 41, 1659-1668. (102) Koh, K.; Ohno, K.; Tsuji, Y.; Fukuda, T. Angew. Chem. Int. Ed. 2003, 42, 4194-4197. (103) Meier, M. A. R.; Gohy J.-F.; Fustin, C.-A.; Schubert, U. S. J. Am. Chem. Soc. 2004, 11517-11521. (104) Aathimanikandan, S. V.; Savariar, E. N.; Thayumanavan, S. J. Am. Chem. Soc. 2005, 127, 14922-14929. (105) Tamano, K.; Imae, T.; Yusa, S.; Shimada, Y. J. Phys. Chem. B. 2005, 1226-1230. (106) Tedeschi, C.; Mhwald, H.; Kirstein, S. J. Am. Chem. Soc. 2001, 123, 954-960. (107) Hugerth, A. M. Journal of Pharmaceutical Sciences 2001, 90, 1665-1677. (108) Pandey S.; Redden R. A.; Hendric ks A. E.; Fletcher K. A.; Palmer, C. P. J. Colloid Interface Sci. 2003, 262, (109) Danko, M.; Hrdlovic, P.; Borsig, E. J. Macrol. Sci.-Pure Appl. Chem., A 2001, 38, 476486. (110) Jang, J.; Oh, J. H. Adv. Mater. 2003, 15, 977-980. (111) Ferrer, M.; Lianos, P. Langmuir 1996, 12, 5620-5624. (112) Hof, M.; Lianos, P. Langmuir 1997, 13, 290-294. (113) Seth, D.; Chakraborty, A.; Setua, P.; Sarkar, N. Langmuir 2006, 22, 7768-7775. (114) Krlicki, R.; Jarzba, W.; Mostafavi, M.; Lampre, I. J. Phys. Chem. A 2002, 106 17081713. (115) Joncheray, T. J.; Audebert, P.; Schwartz E.; Jovanovic, A. V.; Ishaq, O.; Chvez, J. L.; Pansu, R.; Duran, R. S. Langmuir 2006, 22, 8684-8689. (116) Kumbhakar, M.; Nath, S.; Mukherjee, T.; Pal, H. J. Chem. Phys. 2004, 121, 6026-6033. (117) Tummino, P. J.; Gafni, A. Biophysical Journal 1993, 64, 1580-1587. 172

PAGE 173

(118) Pal, S. K.; Sukul, D.; Mandal, D.; Sen, S.; Bhattacharya, S. C. Chemical Physics Letters 2000, 327, 91-96. (119) Pal, S. K.; Mandal, D.; Sukul, D.; Bhattacharya, S. C. Chemical Physics Letters 1999, 312, 178-184. (120) Cheon, C. H.; Jo, S.-H.; Kim, K.; Jin, J.-I.; Shin, H.-W.; Kim, Y.-R. Macromolecules 2005, 38, 6336-6345. (121) Boens, N. e. a. Anal. Chem. 2007, 79, 2137-2149. (122) Sahu, K.; Ghosh, S.; Mondal, S. K.; Ghosh, B. C.; Sen, P.; Roy, D.; Bhattacharya, S. C. J. Chem. Phys. 2006, 125, 44714-44721. (123) Annaka, M.; Noda, H.; Motokawa, R.; Kakahira, T. Polymer 2001, 42, 9887-9894. 173

PAGE 174

BIOGRAPHICAL SKETCH Jorge L. Chvez, the second son of Manuel Chavez and Gladys Benavides, was born in Lima, the Pearl of the Pacific, on December, 1976. He spent his entire childhood in the same city, trying to stay alive despite the craziness that ruled the place. In his early years he spent most of his time playing soccer on the st reets. During high school, he deve loped an interest for science and rock music but, unfortunately, he discovered he lacked any kind of skills to play musical instruments, so he decided to go to college. After graduating from high school, he joined the Pontificia Universidad Catlica del Peru to study chemistry. There he obtained his B. Sc., taught several classes and laboratories and performed research as an undergraduate in Dr. Javier Nakamatsus Laboratory. He moved to Gainesville in August 2002 to start working on his Ph. D. in Dr. Randy Durans research group. In 2006 he performed his most adventurous experiment, marrying Sarah, with whom he had a daughter cal led Victoria Helena. Wh at the future has for him is uncertain now, but surely, it will be some thing exciting and, hopefully, he will be able to make a living out of it. 174