Computational Investigation of the Chemical Modification of Polypropylene Through Fluorocarbon Ion Beam Deposition

Material Information

Computational Investigation of the Chemical Modification of Polypropylene Through Fluorocarbon Ion Beam Deposition
Fell, Christopher
Sinnott, Susan ( Mentor )
Place of Publication:
Gainesville, Fla.
University of Florida
Publication Date:


serial ( sobekcm )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.


This item has the following downloads:

Full Text

JOurn.31 or in.lnerra.3dua.- Re--search

.Oluini ', issue' 2 - r.o, ein er - ..' :,cen-ioer -.

Computational Investigation of the Chemical Modification of
Polypropylene Through Fluorocarbon Ion Beam Deposition

Christopher Fell, Wen-Dung Hsu, and Susan B. Sinnott


Classical molecular dynamics (MD) are used to study the effect of fluorocarbon (FC) ion beam deposition on an

a-isotactic polypropylene surface. The process of depositing C3F5+ ions of specific kinetic energy is used to

better understand the complex mechanisms associated with the processes used to grow polymer thin films

in plasmas. The goal is to investigate the difference in the ways different polyatomic ions chemically modify

a polypropylene substrate. In particular, the simulations allow for the comparison of the chemical products

produced, the penetration depths into the polypropylene surface, their reaction with the backbone chains, and

the amount of overall cross-linking in the polypropylene. The simulations indicate that the deposited C3F5+ ions

are unlikely to dissociate. However, the majority of the fluorocarbons are likely to embed into the surface.

These embedded fluorocarbons create a thin film near the surface of the a-isotactic polypropylene substrate.


Plastics have a wide range of uses in modern society and in industrial applications. One such polymer is a-

isotactic polypropylene (a-iPP). This polymer is utilized in packaging, containers, laboratory equipment, and

much more. Polypropylene is a rugged material with a higher stiffness, comparable density, and good

fatigue resistance when compared to other plastics.1-2

The growth of thin films on the surface of polymers through plasma processing is widely used in

industry. Fluorocarbon (FC) thin films have unique physical and chemical characteristics that can play essential

roles in microelectronics, antifouling, and medical applications. The growth of fluorinated thin films, in

particular, helps engineer the chemical and thermal resistances, the dielectric constant, and the coefficient

of friction.3-6

It is known that mass selected ion beam deposition isolates the effect of polyatomic ions and their contribution to

the chemical modification of the surface. Moreover, it is known that polyatomic ions help alter the top-most layers

of the surface. However, there is still much research needed to better understand the interactions between

the incident ions and the material surface. This interaction can lead to the production of new molecules,

penetration of ions into the material, and fragmentation. The incident ions can generate new products in the

material and alter the surface of the structure. Computer simulations of mass selected ion beam deposition

are utilized to provide important information concerning the chemical reactions and other processes that

occur between polyatomic ions in low-energy plasmas and the surface. These results are complementary

to experimental data and difficult to obtain directly through experimentation.

Molecular dynamics (MD) computer simulations7,8 are utilized to study the continuous deposition of the FC ion,

C3F5+, onto an a-isotactic polypropylene surface at experimental fluences. The deposition process is expected

to modify the chemical and electrochemical properties of the surface. The specific altercations that are examined

are the depth profiles of the incident atoms, the chemical products that are produced, the molecular

weight distribution following deposition, the deposition yield and uptake, and the etching effect and products

removed from the surface by etching.


The MD simulations used in this study numerically integrate Newton's equations of motion with a fourth

order Nordsieck predictor-corrector algorithm.9 They provide information concerning the position, velocity,

and acceleration of every atom in the system as a function of time. The second generation reactive bond order

(REBO)8,10 is used to calculate the short-ranged atomic interactions for fluorocarbon and hydrocarbon

interactions. The long-ranged interatomic interactions are calculated using the standard Lennard-Jones

(LJ) potential.11 Both of these potentials are connected smoothly by spline interpolation.12 The REBO

potentials utilized are capable of predicting new bond formation and bond breaking. These processes are crucial

to accurately model the polyatomic ion beam deposition. However, REBO is unable to effectively include

electronic excitations or true charging of the atoms due to the empirical and classical nature of the

potential; therefore, ions with positive charges are treated as reactive radicals. An ion that is truly charged may

be expected to react more aggressively than simulated radicals; however, it is also true that many incident ions

are neutralized as they approach the surface. REBO and Lennard-Jones potentials are expected to

provide qualitatively correct results and important insights into the modification of iPP by mass-selected

polyatomic FC ions.

The initial surface of a-isotactic polypropylene used in the simulations can be seen in Figure la and lb. It

is comprised of twelve layers containing six chains per layer. Each chain contains nine monomers of polypropylene

(-CH2-CHCH3-) and make up the long side of the surface at 41.58 ?. The short side of the surface slab is

comprised of six chains and is 35.7 ? long. The surface normal to the surface is 55.08 ?, and the iPP system as

a whole contains approximately 13,600 atoms. Periodic boundary conditions9 are applied within the surface plane

to mimic an infinite surface. Each polypropylene chain ends at the boundary and then effectively wraps around

on itself such that there are no surface slab edge effects.


* C atom in active region

H atom in actve region

* C atom in thermos-t region

48.47 ,0 6 - atom in therrm)o region

4- 35.70 -
Figure la. The initial setup of a-isotactic polypropylene substrate

4 35.70

S(C atom in active Tegon

41.58 H atom in aclve WTegion
31.60_ C atomn in lhermoSat regn

j H aloni in lherTmoslat Tregi3on

27.13 --

Figure lb. The surface slab of the a-isotactic polypropylene substrate

A thermostat is applied to approximately half of the atoms in the substrate to maintain the system temperature

at 300 K during deposition. The thermostat region consists of the three bottom layers of the substrate and its

edges such that the "active" area of the surface slab is about 32 x 27 ?2. The active area of the surface slab can

be seen in Figure lb. The atoms that are "active" can evolve freely in response to forces from the neighboring

atoms according to Newton's equation of motion without any additional constraints. In contrast, the

thermostat atoms have Langevin friction and stochastic forces applied to them9 and imitate the heat

dissipation process of much larger, real substrates that dissipate excess energy through atomic vibrations.

The thermostat atoms prevent the surface from translating in response to polyatomic ion beam deposition

and heating up. Prior to deposition, the iPP substrate is relaxed at 300 K for 20 ps at which point the system

potential energy fluctuates by 0.0033 eV/atom around a constant value as a function of time.

The fluorocarbon beam of C3F5+ ions (which are treated as radicals in these simulations) is deposited onto the

active area of the iPP substrate. The beam consists of 320 ions such that the total F fluence is 1.8 x 1016 F

atom/cm2. This value is comparable to experimental values.13 The total kinetic energy for each ion is 50 eV, and

the incident angle is normal to the PP surface. The C3F5+ ion is continuously deposited onto the a-

isotactic polypropylene surface at randomly selected locations within the active area and is randomly

orientated relative to the surface. The time interval between ion collisions with the surface is around 1.5 ps.

After every five ions are deposited, the entire system is equilibrated for 12.5 ps to maintain the surface

temperature around 300 K. After the ion beam deposition process is complete, the system is further equilibrated

for 220 ps at which point the system potential energy again fluctuates by 0.0033 eV/atom around a constant

value with time. The time step used in the simulation is 0.20 fs.


Figure 2 illustrates snapshots of the a-iPP following C3F5+ ion beam deposition. The figure indicates that the

chemical products generated as a result of the collision remain near the surface of the substrate. The C3F5+

ion beam chemically modified half of the structure, which includes the first six layers of the a-iPP substrate,

where each layer is approximately 5.0-6.0 A deep. The complete depth profile of incident carbon and fluorine

atoms is shown in Figure 3. The shape of the depth profile indicates a broad shaped curve with a maximum at

a depth of 0-6 A. The majority of the C3F5+ remains near the surface of the polypropylene because of the larger

size and mass and therefore lower incident velocity.

C atoms rmm CfFs'ions

SF atoDms ~mm CFs'iois

* C atom frm suhbitra

K H aom rom substrate

Figure 2. A snapshot from the simulations of the of a- isotactic polypropylene substrate after C3F5+

ion deposition

-12 L F



36 C3F5

0 5 10 15 20 25 30 35 40
Density of atomr o/cm)

Figure 3. Depth profile for C and F atoms from the incident C3Fs+ ions

The total numbers of carbon and fluorine atoms from the incident ions that embed themselves in, or bond with

the polypropylene surface are 49.06 and 81.24 (xl1020 cm3) for the C3F5+ ion beam deposition, respectively.

This result is accumulated over the entire deposition process, and indicates that F is 1.65 times more likely than C

to embed itself into the surface. The ratio of F to C atoms within the incident FC ions is 1.67. Thus, the ratio of F to

C atoms that embed themselves in or bond to the surface is very similar to the ratio of F to C atoms in the

incident ions. The fact that there is very little difference in the two ratios indicates that most of the ions remain

intact after deposition, and among those that do dissociate, very few F atoms leave the surface.

During deposition, varying products are formed after striking the polypropylene surface, which can be seen in

Figure 4. The products formed as a result of the C3F5+ ion beam deposition include C3F5, CnFm (where n>3

and m>5), CF2 C2Fn, C3Fn (where n>0 and *5), F and CF. The first two species account for 78% of the

chemical products formed, while C3F5 accounts for 59% and CnFm accounts for 19% of the chemical

products formed. This indicates that the majority of the products remain in species as large as or larger than

the original ions. Another significant product formed is CF2, which accounts for 5% of the chemical products that

are formed. This is significant because the CF2 radical is a major building block of fluorocarbon films.3,6 It is

also worth noting that 100% of the CF2 formed bond to the substrate on the time scales of these MD

simulations, whereas the total percentage of deposition products that bond to the surface is 22%.

50 -



Figure 4. Density of products that form from deposited C3Fs+ ions

Following completion of ion beam deposition and equilibration, the weight fraction molecular weight (WFMW)

is calculated for the polymer chains and fragments that make up the modified iPP substrate. This includes the

original polypropylene chains, polypropylene chain fragments, and atoms from the ion beam. Figure 5 illustrates

that there are mainly two peaks observed. One ranges from 120 to 140 (g/mol) and the other range from 600 to

700 (g/mol). The first peak indicates the fragmentation generated by the deposition process, while the second

peak means that some of the polypropylene chains remain intact following deposition. Most of the intact chains

are located farthest away from the polymer surface. It is also noted that there are two peaks with molecular

weights of around 760 and 820 g/mol, which indicate that cross-links between fragments and polypropylene

chain occurs.




u- 0.25

-9' 0.20




0.00 . l L , . . S , A I
0 100 200 300 400 500 600 700 800
Molecular Weight (gimol)
Figure 5. The distribution of molecular weights of the chemical products in the a-isotactic

polypropylene after C3F5+ ion beam deposition

The uptake of carbon and fluorine atoms throughout the deposition process is represented in Figure 6. As

the deposition progresses the uptake of these atoms increases monotonically until the end of the simulation, at

which point the uptake slightly decreases. The inset image in Figure 6 illustrates the ratio between the fluorine
and carbon uptake into the substrate, which is 1.666, a value that is very close to the initial ratio of fluorine to

carbon in the C3F5+ ion. The result here indicates that as the deposition progresses, the C3F5+ ions are unlikely
to dissociate and is consistent with the findings reported in Figure 4.

6.0- * F _uptaKe
5.5 C uptake

40 0 5 61 15 2C m
XK _ F lu e ncx ' 101F at F a&'Co si
V 3.5 '1

2.5 .
2.0' 81."
1.0 . a

0 2 4 6 8 10 12 14 16 18 20
Fluence(x 1015 F atoms/cnr)

Figure 6. Density of F and C atoms uptake during the C3F5+ ion beam deposition

The yield for ion beam deposition is charted in Figure 7. Initially the yield of the C3F5+ ion beam is at 100% from 0
to 20 ps for both carbon and fluorine atoms. The yield then converges to a value of about 27% towards the end
of the simulation, and then slightly decreases to 25% at the very end of the simulation. This slight decrease at
the end is consistent with Figure 6. The values of the yields for F and C are approximately equal throughout
the simulation.

-� 70

0 2 4 6 8 10 12 14 16 18 20
Fluence(x 1015 F atoms/cr)
Figure 7. Deposition yield of F and C atoms by C3F5+ ion beam deposition

Etching of the polypropylene substrate occurred during ion beam deposition and the etch curves can be seen

. C yield

SI . I I
- "? ^^^^^


in Figure 8. The amount of etching maintains a value close to zero throughout the deposition until late in the

process when it increases sharply and then flattens out. When examining both Figures 6 and 8, the depositing

ions and their fragments and particles remain on the surface of the polypropylene. This is confirmed by

the monotonically increasing uptake of C and F, and the almost zero degree of etching. This indicates that the

system is approaching a steady state, continuous film growth state.

6 0 - * C etch
S5 5 .... . .* H etch
50 0 "-
lc)"-- 5 -

3 5 - 2 5 10 i15 20
UJ 3 0 Fluencex 1015 F atamstAc
1.5 a.

0 2 4 6 8 12 14 W 18 20
Fluence(x 10 F atoms/cm)

Figure 8. Amount of etching on the a-isotactic polypropylene substrate by C3Fs+on beam deposition

The densities of the chemical products that are etched out of the system during deposition are illustrated in Figure

9. These chemical products are formed from the C3F5+ ions and are etched away from the polypropylene surface.

The majority of the etched species are C3F5, and the second most etched products are C2Fn and CF2. Combining

this information with the products formed that remain in the surface from Figure 2, the simulations predict that

C3F5 is the most common species formed. It is thus clear that the C3F5+ ions do not easily dissociate; however,

when it does dissociate CF2 and C2F3 are the most common fragments that are formed.

300 Ii Bond
C 1Nobond




50 -

Figure 9. Density of products formed from the incident C3F5+ ions and exited the substrate during

ion beam deposition

Chemical products were also etched from the surface of the polypropylene that did not involve bonding with C3F5

+ incident ions during the depositions. The most common products formed were H, CH3, and CH2. All of

these products indicate that the a-isotactic polypropylene chains are broken into smaller fragments by the ion

beam and gain enough kinetic energy from the ions to scatter away from the surface. Figure 10 also indicates

that there are a few large particles (that consist of more than three carbon atoms) that scatter away from

the surface. However, most of the etched substrate particles are small. This indicates that the depositing ions do

not transfer sufficient kinetic energy to etch large products. Most of the C3F5+ ions are incorporated into

the substrate and a majority of the chemical products are not etched. This is because the C3F5+ ions have a

large size, low velocity, and more degrees of freedom due to vibration, rotation, or torsional deformation.


30 -

0 20


0 L

Figure 10. Density of products formed from a-isotactic polypropylene chain fragments that exited

the substrate

The C3F5+ion beam deposition process is thus expected to be an efficient way of growing FC thin films due to

the large uptake of F and C atoms and the production of FC precursors. This production is expected to increase

the wear resistance of the a-isotactic polypropylene substrate.


The deposition of a fluorocarbon ion beam onto the a-isotactic polypropylene surface is considered here in

classical molecular dynamics simulations. The results illustrate that the C3F5+ ion beam is the most

prevalent species; however when the ions do dissociate it forms several important precursors needed for the

growth of fluorocarbon polymer films. These findings may be used to facilitate the engineering of the

surface properties of the a-isotactic polypropylene surface.


The authors gratefully acknowledge the support of the National Science Foundation through grant number



1. Paukkeri, R.; Vannen, T.; Lehtinen, A. Polymer 1993, 34, 2488.

2. Randall, J. Macromolecules 1997, 30, 803.

3. Takahashi, K.; Itoh, A.; Nakamura, T.; Tachibana, K. Thin Solid Films 2000, 374, 303-310.

4. Tanaka, K.; Inomata, T.; Kogoma, M. Thin Solid Films 2001, 386, 217-221.

5. Mehm da Costa, M.; Freire, F. L.; Jacobsohn, L. G.; Franceschini, D.; Mariotto, G.; Baumvol, I. R. J. Diam.

Relat. Mater. 2001, 10, 910-914.

6. Wang, J. H.; Chen, J. J.; Timmons, R. B. Chem. Mater. 1996, 8, 2212-2214.

7. Humbird, D and Graves, D.B., J. Appl. Phys. 2004, 96, 65.

8. Jang, I.K. and Sinnott, S.B., J. Phys. Chem. 2004, 108, 18993.

9. M.P. Allen and D.J. Tildesly, Computer Simulation of Liquids. Oxford Science Publications, 1986.

10. Brenner, D.W.; Shenderova, O.A.; Harrison, J.A.; Stuart, S.J.; Ni B.; and Sinnott, S.B., J. Phys. Condens.

Matt. 2002, 14, 783.

11. S.J.V. Frankland and D. W. Brenner, Chem. Phys. Letter. 2001, 334, 18.

12. S. B. Sinnott, 0. A. Shenderova, C. T. White, D. W. Brenner, Carbon 1998, 36, 1-9.

13. Wijesundara, M. B. J.; Ji, Y.; Ni, B.; Sinnott, S. B.; Hanley, L. J. Appl. Phys. 2000, 88, 5004-5016.


Back to the Journal of Undergraduate Research

College of Liberal Arts and Sciences I University Scholars Program I University of Florida | 1 l'l UNIVERSITY of

� University of Florida, Gainesville, FL 32611; (352) 846-2032.