Citation
CO Adsorption on Various Oxide Phases Grown on Palladium(111) and Isothermal Oxidation on PdO(101)

Material Information

Title:
CO Adsorption on Various Oxide Phases Grown on Palladium(111) and Isothermal Oxidation on PdO(101)
Creator:
Li, Tao
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (47 p.)

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Chemical Engineering
Committee Chair:
WEAVER,JASON F
Committee Co-Chair:
HAGELIN,HELENA AE
Graduation Date:
5/3/2014

Subjects

Subjects / Keywords:
Adsorption ( jstor )
Atoms ( jstor )
Desorption ( jstor )
Oxidation ( jstor )
Oxides ( jstor )
Oxygen ( jstor )
Palladium ( jstor )
Plasmas ( jstor )
Pumps ( jstor )
Surface temperature ( jstor )
Chemical Engineering -- Dissertations, Academic -- UF
co -- isothermal -- oxide -- palladium -- rairs
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Chemical Engineering thesis, M.S.

Notes

Abstract:
CO molecule, due to its simple vibrational and electronic structure, makes itself an ideal candidate for the investigation of vibrational spectroscopy on transition metals and their oxide surfaces. In this thesis, I will present two studies that were conducted during my two years master research at University of Florida. The first study focuses on the investigation of oxidation on Pd(111) surface. Temperature programmed desorption (TPD) of two-dimensional (2D) oxide and PdO(101) were obtained under the Ultra high vacuum (UHV) condition and oxygen uptake curve was collected at the temperature before and after precursor desorption peak. Because the initial formation of PdO(101) surface is mediated by the stability of precursor state, and at high temperature the uptake curve abruptly slows down at ~0.7 ML. Thus we used this phenomenon to calibrate the oxygen coverage scale in this experiment. The second study focuses on the Reflection-Absorption Infrared Spectroscopy (RAIRS) spectra of CO adsorption on Pd(111) and oxidized surface as well as isothermal oxidation of CO on PdO(101) surface. It turns out that at 100 K CO adsorbs on bridge and hollow sites on Pd(111) at the saturation coverage of 0.5 ML, whereas for Pd5O4 and PdO(101) surface, CO mainly adsorbs on mixture of atop and bridge sites at saturation coverage of 0.46 ML and 0.35 ML, respectively. Isothermal CO oxidation experiment shows the change of CO characteristic adsorption peak as the decrease of oxygen surface coverage. ( en )
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 (M.S.)--University of Florida, 2014.
Local:
Adviser: WEAVER,JASON F.
Local:
Co-adviser: HAGELIN,HELENA AE.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2016-05-31
Statement of Responsibility:
by Tao Li.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Embargo Date:
5/31/2016
Resource Identifier:
908645383 ( OCLC )
Classification:
LD1780 2014 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

CO ADSORPTION ON VARIOUS OXIDE PHASES GROWN ON PALLADIUM(111) AND ISOTHERMAL OXIDATION ON PdO(101) By TAO LI A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENT S FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2014

PAGE 2

2014 Tao Li

PAGE 3

To Quanning, Feng, and my family

PAGE 4

4 ACKNOWLEDGMENTS I sincerely thank to my advisor, Dr. Jason Weaver for his guidance and encouragement. I gratefully acknowledge my committee member, Dr. Helena Hagelin Weaver for her interest in my education. I particularly thank Feng Zhang for his helpful discussion and experimental support I also want to acknowledge John T. Diulus Rahul R ai, Dr. Can Hakanoglu for their valuable advices. On a personal point, I thank my family for their financial and mental support throughout these years. I also thank my boyfriend for his strong encouragement and letting me know what is truly important in my life. I thank all my friends I met during these two years master study, you guys are unbelievable.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF FIGURES ................................ ................................ ................................ ......................... 6 ABSTRACT ................................ ................................ ................................ ................................ ..... 7 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................... 9 Background and Motivation ................................ ................................ ................................ ..... 9 Plasma Source Characte rization ................................ ................................ ............................... 9 Experimental Techniques ................................ ................................ ................................ ....... 10 TPD ................................ ................................ ................................ ................................ .. 11 RAIRS ................................ ................................ ................................ ............................. 12 Crystal S tructure of PdO(101) S urface ................................ ................................ ................... 13 2 THE OXIDATION OF PALLADIUM (111) IN ULTRAHIGH VACUUM ........................ 18 Palladium (111) Oxidation Motivation ................................ ................................ ................... 18 Experimental Methods ................................ ................................ ................................ ............ 18 Result and Discussion ................................ ................................ ................................ ............. 20 Calibration of O xygen C overage ................................ ................................ ..................... 20 TPD of Various Oxide Phases ................................ ................................ ......................... 21 Atomic O xygen U ptake C ur ve ................................ ................................ ........................ 22 Summary ................................ ................................ ................................ ................................ 23 3 ISOTHERMAL CO OXIDATION ON PdO (101) AT 373K ................................ ................ 29 CO Surface Interaction Motivation ................................ ................................ ........................ 29 Experimental Methods ................................ ................................ ................................ ............ 30 Result and Discussion ................................ ................................ ................................ ............. 32 RAIRS of CO A dsorption on Pd(111), Pd 5 O 4 and PdO(101) S urface ............................ 32 Isothermal CO O xidation on PdO(101) at 373 K ................................ ............................ 34 Summary ................................ ................................ ................................ ................................ 36 4 CONCLUSION ................................ ................................ ................................ ....................... 43 LIST OF REFERENCES ................................ ................................ ................................ ............... 44 BIOGRAPHICAL SKE TCH ................................ ................................ ................................ ......... 47

PAGE 6

6 LIST OF FIGURES Figure page 1 1 O atom fraction in the beam as a function of O 2 partial pressu re for different RF input power ................................ ................................ ................................ ........................ 15 1 2 O atom flux estimate sample surface as a function of O 2 partial pressu re for different RF input power ................................ ................................ ................................ .................. 16 1 3 Crystal s tructure of PdO(101) thin film ................................ ................................ ............. 17 2 1 O 2 TPD spectra was obtained by exposing Pd(111) surface to atomic oxygen under various exposure time a t surface temperature of 583K ................................ ..................... 24 2 2 Uptake curve plotted as oxygen coverage vs. atomic oxygen exposure at the surface temperatur e of 583K ................................ ................................ ................................ .......... 25 2 3 O 2 TPD short time exposure spectra was obtained by exposi ng Pd(111) surf ace to atomic oxygen under 433K ................................ ................................ ................................ 26 2 4 O 2 TPD long time exposure spectra was obtained by exposing Pd(111) surf ace to atomic oxygen under 433K ................................ ................................ ................................ 27 2 5 Uptake curve plotted as oxygen coverage vs. atomic oxygen exposure at the surface temperature of 433K ................................ ................................ ................................ .......... 28 3 1 RAIRS of CO on Pd(111), Pd 5 O 4 PdO(101) surface. I R spectra collected at 100K with 1.0 10 8 torr CO background do s ing ................................ ................................ ........ 37 3 2 IR frequencies (cm 1 ) and intensities (in parenthesis) of CO on different sit es on pristine PdO(101) surface ................................ ................................ ................................ .. 38 3 3 IR frequencies (cm 1 ) and intensities (in parenthesis) of CO on different sites on PdO(101) w i th one oxygen vacancy ................................ ................................ .................. 39 3 4 Isotherma l CO oxidation on PdO(101) surface and TPD of residual oxygen and CO 2 ..... 40 3 5 Isothermal IR of CO reducti on on PdO(101) surface at 373 K ................................ ......... 42

PAGE 7

7 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science CO ADSORPTION ON VARIOUS OXIDE PHASES GROWN ON PALLADIUM(111) AND ISOTHERMAL OXIDATION ON PdO(101) By Tao Li M ay 2014 Chair: Jason F. Weaver Major: Chemical Engineering CO molecule, due to its simple vibrational and electronic structure, makes itself an ideal candidate for the investigation of vibrational spectroscopy on transition metals and their oxide surfaces. In this thesis, I will present two studies that were conducted during my two years master research at University of Florida. The first study focuses on the investigation of oxidation on Pd(111) surface. Temperature programmed desorption (T PD) of two dimensional (2D) oxide and PdO(101) were obtained under the Ultra high vacuum (UHV) condition and oxygen uptake curve was collected at the temperature before and after precursor desorption peak. Because the initial formation of PdO(101) surface is mediated by the stability of precursor state, and at high temperature the uptake curve abruptly slows down at ~0.7ML. Thus we used this phenomenon to calibrate the oxygen coverage scale in this experiment. The second study focuses on the Reflection Abso rption Infr ar ed Spectroscopy (RAIRS) spectra of CO adsorption on Pd(111) and oxidized surface as well as isothermal oxidation of CO on PdO(101) surface. It turns out that at 100 K CO adsorbs on bridge and hollow sites on Pd(111) at the saturation coverage of 0.5 ML, whereas for Pd 5 O 4 and PdO(101) surface, CO mainly adsorbs on mixture of atop and bridge sites at saturation coverage of 0.46 ML and 0.35 ML,

PAGE 8

8 respectively. Isothermal CO oxidation experiment shows the change of CO characteristic adsorption peak a s the decrease of oxygen surface coverage.

PAGE 9

9 CHAPTER 1 INTRODUCTION Background and Motivation Palladium oxide is one of the important heterogeneous oxidation catalysts for commercial oxidation processes such as catalytic combustion of methane in lean gas turbines, oxidation of CO in automotive exhausts and so forth. Improving the performance of those processes as well as the necessity of energy conservation motivate us to fully investigate the fundamental understanding of the mechanism of palladium and its oxide growth, as well as the chemical property of oxide phase on metal surface. CO oxidation is one of the simplest and most widely studied catalytic reactions in heterogeneous catalysis, its highly facile reactivity towards palladium and palladium oxide makes it an ideal model for the study of elementary surface processes. Various techniques and methods can be employed in the characterization of palladium and its oxide as well as the study of CO oxidation. In this study, TPD, RAIRS and density functional theory (DFT) calculations are used. Other techniques such as low energy electron diffraction (LEED), scanning tunneling microscopy (STM) will be employed in the future work. By using atomic oxygen beam, we have been now successfully extend the range of oxy gen concentrations on palladium surface under UHV condition which enable us to conduct detail investigation on bulk oxide phase, therefor obtain new information about the fundamental details of the properties of oxidized surface and the its essentially ro le in the oxidation catalysis. Plasma Source Characterization As one of the important facilities, plasma source is used for routinely sample cleaning and making oxide. Thus, the characterization of plasma source will give us a clear idea about the range o f operational condition for utilization. We fed pure O 2 to generate plasma and

PAGE 10

10 characterized the resulting beams using line of sight mass spectrometry to determine the beam composition as a function of oxygen pressure. We also investigated the oxidation of Pd(111) at specific plasma conditions (200W, 9 m illit orr O 2 ) to estimate the O atom flux at the sample. Figure 1 1 shows the O atom fraction in the beam as a function of O 2 partial pressure in the plasma for 150 W to 240 W radio frequency (RF) power inpu t. The O atom fraction represents the fraction of the total species (O and O 2 ) in the beam that are O atoms. As we can see from Figure 1 1, the O atom fraction decreases with O 2 partial pressure in the plasma chamber and increase with RF power, but latter will influence more comparing with former. Note that we limit the RF power to 240 W to be conservati ve and partial pressure to 10 m illit orr at 150 W in case that plasma would extinguish. Figure 1 2 shows the O atom flux estimated at sample surface. The sa mple was held at an angle of about 45 29.2 cm away from the plasma source. As we can see, the atomic flux increases monotonically with O 2 partial pressure above 200 W whereas appears to pass through a maximum at 200 W and lower RF power. At our normal op eration condition (9 m illit orr and 200 W), the estimated O atom flux is 0.0 043 ML/s, where 1ML=1.5 10 15 atoms/cm 2 Experimental Techniques UHV environment is the starting point for utilization of all other facilities. Several pumping techniques are used fo r achieving the base pressure of 2 10 10 torr, mainly due to the different operation range of different types of pumps as well as large pressure change that occurs in the chambers. There are four types of pumps utilized in the lab, along with an additional bake out procedure to obtain low base pressure. The first level of obtaining UHV condition is by setting roughing pumps to achieving 1 10 3 torr such that it is suitable for the implementation of other high vacuum pumps. Roughing

PAGE 11

11 pump operates by apply ing compression mechanism to displace gas from the vacuum system. The second level is achieved by using turbomolecular pump, which is connected to the roughing pump for reaching base pressure of 1 10 7 torr 1 10 8 torr. Turbomolecular pump rotates a set o f blades at a high speed and the gas molecules are hit from the inlet towards the outlet of the pump to create higher vacuum. Another two pumps, ion pumps and titanium sublimation pump (TSP), help to achieve the third level of UHV. Former one ionizes gas m olecules in a vessel which is applied by strong electrical potential, and this high potential accelerates the ionized molecules to be captured by the wall of the pump. This type of pump is capable of reaching base pressure of 10 11 torr. TSP consists of th ree titanium filament s and by sublimating them, a thin coat film will be coated surrounding the chamber wall. Since clean titanium is very reactive, it will react with gas molecules colliding on the wall and form a stable product, thus maintaining the low pressure. TPD TPD is a surface technique that quantitatively provide s the information about kinetics of desorption molecules from the surface when surface temperature is increased. Whereas another similar technique, temperature program reaction spectrosco py (TPRS), monitors desorption species with additional reactions occurring on the surface as increasing surface temperature. These two techniques are carried out by first dosing the species of interest on the surface, followed by heating the surface in a c onstant rate (1 K/s) while monitoring the desorbents that either directly desorbs from the surface or desorbs after reactions. A quadropole mass spectrometer (QMS) is used for detecting and monitoring in this case. Typically, the signal of desorbents is pl otted as a function of surface temperature in TPD spectrum, thus the activation energies can be deduced because desorption temperature reveals the amount of energy that is required for breaking surface bond.

PAGE 12

12 Due to the large pumping speed and low heating r ate, desorption rate can be assumed to proportional to the desorption pressure [1]. Redhead provided a method for providing the kinetics of surface processes in his 1963 paper [2 3], in which he showed that desorption rate could be expressed by the followi ng equation: = = (1 1) W here is desorption rate of species A, is surface sites concentration, is the coverage of species A, t is time, is pre exponential of ra te constant, n is desorption order, is activation energy of desorption reaction, K is Boltzmann s constant, T is surface temperature. After taking the first derivative of equation (1 1) and rearranging, the maximum desorption rate can be found in t he following equation: (1 2) Where is maximum desorption peak temperature, is heating rate, is the coverage at peak maxima. From e quation 1 2, if n=1, then the peak temperature would be independent with surface coverage, whereas if n=2, the initial coverage would influence the desorption process, thus result in a shift of the temperature of desorption peak. TPD spectra can provide us quantitative information about surface coverage. By integrating the entire curve with respect to surface temperature, and comparing with known saturation coverage that has been calibrated for the surface, one can obtain surface coverage of the adsorbent. RAIRS RAIRS is a surface technique that perform vibration spectrum of absorbed molecules on metal surface. Besides, due to the high sensitivity and resolution of this technique, it is especially

PAGE 13

13 suitable for the study of interaction between adsorbents and metal surface as well as intermolecular interaction within the adsorbed layers. Hoffmann [4] has already summarized IRAIS studies in a macroscopic perspective, and three considerations are important: 1. Incident light must have p polarized component (i.e. the component parallel to the plane of incidence), and this component only vibrates with a dipole moment that is perpendicular to the metal surface. 2. A grazing incidence is necessary regarding incident light. 3. For adsorption on transition metals, usage of s ingle reflection is preferred, thus studies should be carried out on oriented single crystals. As illustrated by Hoffmann [4], excitation of vibrational transition by infrared light follows the rule that only vibrational transitions associated to the chang e in dipole moment are IR active. The intensity of IR peak can be expressed in following equation: (1 3) W here is molecule dipole moment, is the normal coordinate associated to the k th vibrational mode. For ad sorbed molecule, the number of IR active vibration is determined by both nonzero dipole moment change and number of internal (3N 6) modes plus hindered translational and rotational modes due to the adsorption on metal surface, where N is the number of mole cule. Crystal S tructure of PdO(101) S urface As presented in Figure 1 3, bulk PdO crystal has tetragonal unit cell which consists of square planar units of Pd atoms coordinating with four fold oxygen atoms. The bulk terminated PdO(101) surface investigated in this study is defined by rectangular unit cell with dimensions of a=3.043 and b=6.142 and consists of closed packed rows of three fold Pd (Pd 3f ) and four fold Pd (Pd 4f ) arranged alternately coordinating with three fold or four fold oxygen atoms,

PAGE 14

14 res pectively. A and b lattice vectors of rectangular unit cell are labeled in the [010] and [ 01] directions in F igure 1 3 B Thus, half of the Pd atoms on the surface are c oordinately unsaturated Pd cus which likely to be more reactive than Pd 4f regarding b inding with adsorbed molecules. The density of palladium and oxygen surface atom is 0.7 ML, so 0.35 ML cor responds to the coverage of Pd cus and O cus

PAGE 15

15 Figure 1 1. O atom fraction in the beam as a function of O 2 partial pressure for different RF input p ower.

PAGE 16

16 Figure 1 2. O atom flux estimate sample surface as a function of O 2 partial pressure for different RF input power.

PAGE 17

17 Figure 1 3. Crystal s tructure of PdO(101) thin film. A ) Top view and B ) Side view The red and blue atoms represent O and Pd at oms, respectively. Directions a and b labeled in A ) correspond to the [010] and [ 01] direction of PdO. 3 fold coordinated O cus and Pd cus rows are indicated in B ).

PAGE 18

18 CHAPTER 2 THE OXIDATION OF PALLADIUM (111) IN ULTRAHIGH VACUUM Palladium (111) Oxidation Motivation The oxidation of Pd(111) surface to 2D oxide and PdO(101) bulk oxide plays a significant role in the determination of property of palladium catalysts in oxygen rich applications, which include the exhaust gas such as alkane and CO remediation i n automobiles [5 8], power generation plants and fuel cell catalysis. The development of oxidation of Pd(111) has been characterized well [9 17], with mainly working on the 2D surface oxide [18 26]. It was shown that a p ( ) structure is first arranged by chemisorbed oxygen atoms at the total oxygen coverage of 0.25ML. Further surface reconstruction and formation of various 2D oxide are found at higher coverage. Lundgren and Han [ 27 28 ] reported that 2D oxide has a preferred Pd 5 O 4 stoichiometry with ( ) structure, aligning with the [ 1 1 0] directions of Pd (111) substrate whereas other oxygen phase appear to be meta stable relative to Pd 5 O 4 surface. Furthermore, previous studies [29 32] have investigate distinct feature and property about precursor in TPD spectra of Pd(111) oxide. Briefly, a broad peak between 500K and 650K in the desorption feature arises from precursor state and initially appear at ~0.7 ML, which is the saturation coverage of 2D oxide. Atomic oxygen uptake curve has also been done by Kan [29] in different incident fluxes and surface temperature, which reveals a strongly dependency of kinetics for Pd(111) oxidation on the thermal stability of precursor state. In this study, we investigated the TPD spectra of 2D oxide and PdO(101) su rface, obtained atomic oxygen uptake curves under different temperature as well as quantif ied oxygen surface coverage Experimental Methods The experiments were conducted in an UHV system that reaches the base pressure less than Torr. The chamber is evacuated by a turbomolecular pump, an ion pump and a

PAGE 19

19 titanium sublimation pump that is inserted into the liquid nitrogen cooled cryoshield. A shielded QMS a single stage differentially pumped beam chamber that houses a RF power plasma source for generating atomic oxygen beam and three leak valves which are used for gas dosing are assembled in the middle of the main chamber. One side of the chamber houses a LEED / Auger electron spectroscopy (AES) A Bruker Tensor 27 FTIR is set ou tside the UHV chamber and sends the mid infrared (MIR) beam that is transmitted within a sealed box to the differentially pumped potassium bromide ( KBr ) window. Inside the sealed box which is continuously purged with H 2 O and CO 2 free compressed air a set of flat mirrors direct the MIR beam to a parabolic mirror that is used to focus the MIR beam onto the sample. Inside the UHV chamber, focused MIR beam impinges on the sample surface at an angle of relative to the normal direction. The reflected MIR beam exits the chamber through a second KBr window and is detected by liquid N 2 cooled HgCdTe (MCT) detector using another set of focusing mirrors, all of which are in a second sealed box that is continuously purged with H 2 O and CO 2 free compressed air T he Pd(111) crystal that was used in this study is a circular disk attached to a type K thermocouple wire for temperature measurement and a high voltage wire at the backside of the crystal for electron bombardment heating. Crystal is attached to two a coppe r sample holder in thermal contact with a liquid nitrogen (LN) cooled reservoir A tungsten wire is set under the back of sample without touching for radiative heating which is controlled by a proportional integral derivative ( PID ) controller that varies the output of a programmable direct current ( DC ) power supply to maintain or linearly ramp the sample temperature from 91 K to 1250 K. Sample was initially cleaned by sputtering with Ar + ions at 700 K, followed by annealing to 1100 K. After several cycles argon sputtering, sample was exposed to an atomic oxygen beam at 823K for routinely cleaning followed by flashing the sample to 900K. Previous study [30] has already

PAGE 20

20 discussed that the limitation of sample temperature at 900K will maintain oxygen saturati on in the subsurface and will remove carbon impurities. This minimizes the possibility of losing chemisorbed oxygen because of the bulk dissolution dur ing following TPD experiments [13 ], s o that the reproducibility of oxygen coverage obtained by TPD will b e ensured. The surface was considered to be clean when we obtained sharp LEED patterns consistent with Pd(111) surface. In this system, gaseous oxygen atoms were produced by partially dissociating pure oxygen ( Airgas 99.999% ) supplied to the quartz tube at end of the plasma. We used RF power of 200 W and 9 mtorr oxygen pressure in this study. The source to target distance was about 29.2 cm and the sample was held at an angle of about with respect to the tube axis for the measurements. We characterize d the Pd(111) surface after oxidation by atomic oxygen beam at various exposure time and surface temperatures, then immediately adjusted the sample position such that the sample would face the entrance of a QMS ionizer at the distance of 2 mm followed by increasing the temperature at a linear rate of 1 K/s. Result and Discussion Calibration of O xygen C overage We quantify the 2D oxide and PdO(101) oxygen surface coverage by comparing integrated TPD signal areas to the area obtained after averaging several surface oxide saturation coverage that were done at similar exposure time at 583 K. Previous research [29] has already shown that at high temperature, the uptake curve abruptly slows down at ~0.7 ML surface oxide saturation coverage, which implies that the PdO formation is initially limited by the competitive desorption from precursor state. The rate of oxidation then gradually increases with atomic oxygen exposure, followed by decreasing to the PdO layer saturation coverage. An autocatalytic mechanism [29] for PdO growth can explain the acceleratory kinetics after ~0.7 ML. More specifically, as the PdO precursors grow in size, they become more stable such that less of them

PAGE 21

21 undergo decomposition process and more PdO particles grow bigger, which revealed by t he accelerating rate before PdO layer saturation. In this study, we obtained the O 2 TPD spectra by exposing the Pd(111) surface held at 583 K to the atomic oxygen beam under various exposure time and integrated O 2 TPD signal areas. The areas between 190 s 240 s are relatively close and we assume that this is the surface oxide saturation coverage 0.7 ML. The coverage scale is 1.1244 10 7 ML/C, which is calculated by using 0.7 ML divided by the average area of 190 s 240 s. Figure 2 1 shows the TPD spectra aft er unit transformation of y axis from counts per second to monolayer per second. Figure 2 2 shows the oxygen coverage as a function of atomic oxygen exposure at 583 K. We only did the oxygen coverage up to 0.775 ML, because the uptake curve has already giv en us the TPD areas relative to ~0.7 ML to calculate the coverage scale. The estimation of oxygen atom beam flux was done by assuming the unit probability of gas phase oxygen atoms adsorption up to 0.7 ML and fitting the early part of oxygen uptake curve w ith linear function. Previous studies [33] have shown that this procedure gives a reasonable estimation of beam flux. TPD of Various Oxide P hases Figure 2 3 shows a series of short time exposure O 2 TPD spectra collected by exposing the Pd(111) surface to gas phase atomic oxygen at a surface temperature of 433 K. As mentioned above, the oxygen atoms will first chemisorb into a p (2 2) structure below 0.25 ML, which is revealed by the broad peak produced through recombinative desorption between 550 K and 775 K. As the oxygen coverage increase above 0.25 ML, we observe a sharp peak that shifts from 638 K to 644 K, which originates from the autocatalytic decomposition process of ordered 2D oxide [11 12]. A leading edge appears at 670 K of sharp 2D oxide peak at 0.5 ML and gradually develops into a separate peak as the increasing of desorption rate. T he broad peak between 433 K

PAGE 22

22 and 583 K is the feature of precursor state and previous studies [29] has already demonstrated that first appearance of this state is at ~0.7 ML saturation coverage of 2D oxide, which agrees well with this experiment. The peak near 620 K is attributed to the decomposition of bulk like PdO domains and autocatalytic decomposition of coexisting surface oxide results in the peak near 662 K [34] Figure 2 4 shows a series of long time exposure O 2 TPD spectra collected by exposing the Pd(111) surface to gas phase atomic oxygen at a surface temperature of 433 K. From (a) we can see that t he precursor state peak gradually diminishes as the bulk like PdO domain continues to grow. The peak at 665 K both intensifies and shifts towards higher temperature as the PdO grows. Further increasing of the oxygen coverage leads to the intensification of bulk like PdO domain and the peak at ~693 K without shifting toward higher temperature. Atomic O xygen U ptake C urve Figure 2 2 and 2 5 reveal that thermal stability of precursor has a strong influence on the kinetics of Pd(111) oxidation. The temperatures carried out in two uptake curves lie below and above the pre cursor desorption peak in TPD spectra (Fig ure 2 3 B Fig ure 2 4 A ). Comparing with temperature at 583 K, the oxygen uptake increases monotonically with exposure at 433K until the saturation coverage at ~4.0 ML. As illustrated above, the beam flux is calculat ed by fitting the early part of oxygen uptake curve with linear function and the beam flux is 0.00429 ML s 1 The uptake curve at 583 K, however, abruptly slows down at ~0.7 ML, which is the saturation coverage of 2D oxide, and then gradually increases wit h the atomic oxygen exposure. These phenomena can be explained that initially PdO formation is limited by the competitive desorption from precursor, as this state becomes more stable, the self accelerating kinetics occurs, which leads to the increase of ox idation rate.

PAGE 23

23 Summary We calculated the oxygen calibration coverage and investigated the TPD of 2D and PdO(101) as well as oxygen uptake curve under low and high temperature using atomic oxygen beam under UHV condition. The O 2 TPD spectra characterized the evolution of 2D oxide and PdO(101) and show the development of precursor state under different oxygen coverage. The oxygen uptake curves collected at the temperature before and after precursor desorption peak give us the evidence that initial formation of PdO on the surface is mediated by the stability of precursor state, more specifically, at low temperature, the oxygen coverage increases monotonically with exposure, whereas at high temperature, the oxygen coverage initially is a linear function of exposu re, but abruptly slows down at ~0.7 ML, which is the saturation coverage of 2D oxide, and then gradually increases with exposure until PdO layer saturation coverage.

PAGE 24

24 Figure 2 1. O 2 TPD spectra was obtained by exposing Pd(111) surface to atomic oxygen under various exposure time at surface temperature of 583K. The heating rate is 1K s 1 for all of the TPD measurements. The panel illustrates the exposure time and the relative oxygen coverage.

PAGE 25

25 Figure 2 2. Uptake curve plotted as oxygen coverage vs. a tomic oxygen exposure at the surface temperature of 583K. The unit of x axis is ML. Beam flux illustrated in the panel is 0.00129ML s 1

PAGE 26

26 Figure 2 3. O 2 TPD short time exposure spectra was obtained by exposing Pd(111) surface to atomic oxygen under 433K The heating rate is 1K s 1 A) Oxygen coverage from c hemisorbed oxygen to 2D oxide. B) 2D oxide coverage t o its saturation coverage

PAGE 27

27 Figure 2 4. O 2 TPD long time exposure spectra was obtained by exposing Pd(111) surface to atomic oxygen under 433K. The heating rate is 1K s 1

PAGE 28

28 Figure 2 5. Uptake curve plotted as oxygen coverage vs. atomic oxygen exp osure at the surface temperature of 433K. The unit of x axis is ML. Beam flux illustrated in the panel is 0.00429ML s 1

PAGE 29

29 CHAPTER 3 ISOTHERMAL CO OXIDATION ON PdO (101) AT 373K CO Surface Interaction Motivation CO oxidation is one of the most extensively investigated reactions in heterogeneous catalysis and surface science Understanding the mechanism of this catalytic reaction is essential for reducing the amount of pollutants release by automobile exhaust and increasing the fuel economy of automobiles. N umerous studies have already been done on substrate supported [35 36] and singe crystal metal surface [37 39] as well as metal oxide [40], in which some of the general conclusion have been drawn. For oxidation reaction particularly, CO follows concerted me chanism [41 42] and molecularly adsorbs to the late transition metals surface via carbon atom. The oxidation proceeds between chemisorbed CO and O atoms via Langmuir Hinshelwood mechanism and CO adsorption is the rate determining step. Surface reactivity c an be probed through chemical reduction, during which oxide surface configuration can be changed. A variety of previous studies [43 45] have been done concerning CO adsorption on Pd single crystal and RAIRS has proved itself to be a powerful technique for providing vibrational frequency of different C O bonding sites on the surface. Bradshaw and Hoffmann [45] assigned 2120 2050 cm 1 as the frequency range for CO adsorbing on the top of Pd (atop). 1950 1880 cm 1 is the frequency range of bridge CO interactin g with two Pd atoms. 1880 1800 cm 1 is referred to the hollow site frequency of which CO interacts with three Pd atoms. It has been well characterized that surface structure of CO overlayers on Pd(111) are coverage dependent [43, 45 46], where CO forms and c(4 2) 2CO at 0.33 ML and 0.5 ML respectively, saturates at 0.75 ML by forming (2 2) 3CO compressed structure. Previous studies [4, 43, 45] have indicated that at 0.33 ML, CO adsorbs in 3 fold hollow sites whereas at 0.75 ML, CO occupies bo th 3 fold hollow and atop sites. However, the conclusion

PAGE 30

30 about surface structure at 0.5ML was questioned in the later studies [47 48]. Regarding on this argument, recent study [49] shows that at 0.55 ML, CO adsorbs on both bridge and 3 fold hollow sites on Pd(111) at 100 K. Comparing with the investigation in metal, less work has been done on palladium oxide due to the difficulty of generating well ordered palladium oxide surfaces under UHV condition. As introduced above, the apparatus we have allow us to expose the Pd(111) crystal to highly reactive atomic oxygen and prepare the well ordered PdO(101) surface ~3.3 ML under the main chamber pressure of ~5 10 9 torr for 30 minutes exposure. In this study, we investigated the isothermal oxidation of CO on PdO( 101) surface under 373 K by using TPD and RAIRS, from which the reaction coverage and vibration frequencies were obtained. The comparison of them reveals the change in bonding site during the reaction as well as provides us more information about the kinet ics of isothermal CO oxidation on PdO(101) surface. Experimental Methods Details about the UHV chamber and preparation of clean Pd(111) surface have already been discussed above. In the RAIRS experiment of CO adsorption on Pd(111) and oxidized Pd(111) sur face, Pd(111) was cleaned by exposing to the plasma generated atomic oxygen beam at 823 K, followed by flashing to 900 K to remove carbon contamination. Pd 5 O 4 was prepared by exposing the cleaned Pd(111) surface to the atomic oxygen for 220 s at 583 K fol lowed by annealing to 600 K As discussed in the chapter, this condition made the surface coverage in the so called metastable regime and oxygen coverage is 0.7 ML. PdO(101) layers was prepared by exposing the cleaned Pd(111) surface to atomic oxygen for 3 0 min at 433 K and oxygen coverage is ~3.3 ML. RAIRS data were collected at 100 K after exposing the three types of surfaces to a saturate dose of CO (~10 L).

PAGE 31

31 In isothermal CO oxidation experiment the PdO(101) layer was prepared by exposing the Pd(111) crystal to the atomic oxygen for 30min at 433 K, followed by annealing to 533 K to make the surface more well ordered, then dropped down to 100 K. During decreasing the surface temperature, the sample was positioned at a distance about 3 mm from aperture o f mass spectrometer. CO was charged through a leak valve once the temperature dropped down to 100 K at main chamber pressure of 5 10 9 torr, and CO 2 signal was monitored to establish a baseline for 4 minutes. Then the temperature was increased at a constan t rate of 1 K/s to 373 K and held for 3 minutes, followed by decreasing to 100 K. We repeated this cycle for 7 times until most of the oxide was reacted by CO. The residual oxygen cover age was collected by another TP RS experiment which was taken immediatel y after isothermal experiment. Sample temperature was increased at the rate of 1 K/s to 900 K at the CO background of 5 10 9 torr, O 2 and CO 2 signal were monitored. In isothermal RAIRS experiment, PdO(101) layer was prepared as discussed above and sample w as positioned so that mid IR beam can be reflected through the sample surface to the detector. After the surface temperature was dropped down to 100 K, CO was charged through a leak valve once the temperature dropped down to 100 K at main chamber pressure of 5 10 9 torr, and CO 2 signal was monitored by the mass spectroscopy at the same time. After for running for 5 minutes at 100 K at 5 10 9 torr, scan the IR spectrum, followed by heating the sample to 373 K and holding for 3 minutes, then dropping back to 100 K and did another IR scan. Repeat the cycle for another 6 times, and every IR scan was obtained during 100 K sample temperature. After the 7 IR spectrum scan, sample was flashed to 900 K for 3 minutes and then dropped to 100 K to obtain the IR spectrum of CO on Pd(111) metal surface by charging CO to main chamber pressure for 5 10 9 torr.

PAGE 32

32 Result and Discussion RAIRS of CO A dsorption on Pd(111), Pd 5 O 4 and PdO(101) S urface RAIRS experiments of CO adsorption on metal and oxidized surface were performed to determine the site assignment at CO saturation coverage. The results are shown in Figure 3 1. For CO on Pd(111) surface, a strong peak is observed at 1960 cm 1 and three small peaks are apparent at 2118, 2109 and 1899 cm 1 TPD measurement under this case gave us 0.53 ML CO saturation coverage and the 1960 cm 1 feature in our spectra has been assigned previously [4, 43 44] to the CO adsorbed on bridging sites within c(4 2) phase at 0.5 ML CO coverage. Notices that CO saturation coverage can reach 0.75 ML b y forming c(2 2) 3CO phase at 100 K reported by previous studies [43 44], and the spectrum is dominated by atop peak at 2111 cm 1 and 1896 cm 1 hollow peak for CO binding within the (2 2) 3CO phase. In our case, 0.75 ML can only be achieved after annealing Pd(111) surface to 150 K while exposing to CO background and took IR scan when sample temperature dropped back to 100 K. This phenomenon was also observed by Kuhn et al [43] before. The small peaks at 2118, 2109 and 1899 cm 1 in this spectra indicate ther e are small c(2 2) 3CO compressed phase regimes within the c(4 2) phase. The DFT calculations reported recently [49] shows that the frequencies of c(4 2) bridge+hollow structure close to our experimental results 1960 cm 1 and 1867 cm 1 For Pd 5 O 4 surface, prior work shows that annealing the temperature to 600 K after the preparation of oxide transforms so called metastable 2D oxide to the root 6 on Pd(111) surface. Figure 3 4 shows that the saturation coverage of CO on Pd 5 O 4 is 0.46 ML at 100 K. As shown in Figure 3 1, at CO saturation coverage, IR spectra give us a strong peak at 2131 cm 1 and a shoulder peak at 2111 cm 1 some broad peaks at 2010, 1984 and 1956 cm 1 and two small peaks at 1867 and 1826 cm 1 DFT frequency calculations [49] of CO adsorptio n on Pd 5 O 4 at low and high coverage show that the dominant 2131 cm 1 is consistent to the atop CO, whereas 2010,

PAGE 33

33 1956 and 1867 cm 1 is close to the frequencies of bridge CO 1826 cm 1 low intensity peak can be assigned to the hollow site according to the D FT calculation. The shoulder peak appeared at 2111 cm 1 has been observed as atop CO on Pd(111) metal surface at compressed c(2 2) 3CO phase [43, 46] at saturation coverage 0.75 ML, which may indicate that Pd 5 O 4 is imperfect and some Pd(111) regimes may ex ist on the surface. For PdO(101) surface, previous TPRS study [50] has shown that CO saturation coverage is 0.35 ML, which is consistent with the concentration of coordinate ly unsaturated Pd (Pd 3f ) atoms on PdO(101) surface. This coincidence indicates that CO preferentially adsorbs on the Pd 3f sites, and all these sites are occupied under the saturation coverage. In Figure 3 1, a dominant 2127 cm 1 and a broad 1954 cm 1 peak are observed, which correspond to atop and bridging CO respectively according to th e previous [49] and recent DFT calculation (Figure 3 2). Atop to bridge peaks ratio is 1.65 in this case. Specifically, recent DFT calculation has been used to explore possible configurations on pristine PdO(101) surface and the preferred ones are CO adsor bing on the top of Pd 3f (all atop) and atop bridge atop CO (t b t) at high coverage. According to DFT calculation, the in tensity of atop vs. bridge peak is 1.09 vs. 5.73 for t b t configuration with which atop CO frequency is at 2143 cm 1 and bridge freque ncy is at 1953 cm 1 For all atop case, the intensity of atop CO peak is 1.05 with frequency at 2149 cm 1 Based on these data, we assume that t b t phase and all atop phase co exist at saturation coverage and 90% of the species are atop CO. 36% of the sur face coverage is t b t phase, the rest is all atop surface. This assumption gives us 0.32 ML CO saturation coverage which is reasonably close to previous TPRS result (0.35 ML). Notice that the frequency for atop CO in these two cases is ~2143 cm 1 which is larger than 2132 cm 1 in spectra. The red shift of the frequency may be either caused by

PAGE 34

34 the mixture of atop and bridge CO, which results in a stronger adsorption for atop CO, or the uncertainty of experimental data or DFT calculation. Isothermal CO O xi dation on PdO(101) at 373 K The isothermal CO oxidation spectra (Figure 3 4 ) were collected at the same condition under 5 10 9 torr CO background. In Figure 3 4 A the surface was heated to 373 K and reaction took place for 3 minutes, followed by cooling do wn to 100 K, wher ein IR was collected (Figure 3 5 ). The reaction was repeated for 7 cycles until almost all the oxygen coverage w as consumed (96.47%). Figure 3 4 B was collected immediately after isothermal experiment, where sample was heated to 900 K at ra te of 1 K/s with 5 10 9 torr CO background. The residual oxygen and reacted CO 2 was 0.08 ML and 0.03 ML respectively. In this case, reacted CO 2 coverage is equal to the initial oxygen coverage (3.17 ML) minus addition of residual O 2 and CO 2 coverage. The c alculation of reacted CO 2 coverage, residual oxygen coverage as well as oxygen coverage consumption percentage for each reaction cycle are listed on the top of each pe ak in Figure 3 4 A Figure 3 5 shows the IR spectra we collected after each cycle of isoth ermal oxidation The bottom one (black) was obtained before the 1 st cycle, which corresponds to CO adsorption on pristine PdO(101) surface, whereas the top one (dark blue) was obtained at 100 K under 5 10 9 torr CO background after annealing the surface to 900 K, which corresponds to CO adsorption on Pd(111) surface. Based on Figure 3 4 A first three cycles of isothermal CO reaction removed more than half of the oxygen coverage, which leads to the remarkably increas e of oxygen vacancy. Furthermore, due to t he continuous CO background dosing, we assume that surface was covered by high CO coverage throughout these 7 reaction cycles. As illustrated above, recent DFT calculation suggests that 36% of t b t phase and 64% of the all atop phase compose of CO adsorbi ng on pristine PdO(101) surface.

PAGE 35

35 Comparing IR of CO adsorbing on pristine PdO(101) surface with IR after 1 st cycle reaction, 2132 cm 1 atop peak blue shifts and red shifts to 2140 cm 1 and 2094 cm 1 respectively The 21 40 cm 1 peak is consistent with a top CO adsorbs on oxide phase broad peak at 2094 cm 1 is composed by several peaks overlapped together and is related to atop CO next to oxygen vacancy (N t) site (Figure 3 3) This assumption is based on the fact that after 1 st cycle, 0.463 ML (14.66%) o xygen was reacted, which is more than cus O (0.35 ML) from the first layer, and due to the diffusion of oxygen atom from deeper layer, we can assume that the surface still remain large domain of PdO(101) with relatively less oxygen vacancy. Note that a sma ll peak appeared at 2111 cm 1 which is consistent with atop CO on Pd(111) (2 2) 3CO compressed phase. This demonstrates a co existence of metallic and PdO(101) domains during reaction. After 2 nd cycle reaction, the intensity of 2140 cm 1 peak decreases wh ereas 2094 cm 1 peak blue shifts to 2097 cm 1 with almost the same intensity, which indicates the coverage decrease of former type of atop CO and influence of oxygen vacancy on latter type of atop CO. IR spectrum which was collected after 3 rd cycle shows t hat 2140 cm 1 peak further blue shifts to 2142 cm 1 and diminishes, accompanying with the increase of 2111 cm 1 peak. Combining wi th the information in Figure 3 4 A 52.69% oxygen coverage has been consumed at this time, the change in the spectrum shows tha t more oxygen vacancies have been generated and Pd(111) regimes increase. From cycle 4 to 7, atop CO peak on oxidized phase has completely disappear and 2111 cm 1 is the only dominant peak. Noticed that a small peak at 2095 cm 1 appears after fifth cycle, which consists with atop CO on Pd(111) that persists until the (2 2) 3CO phase saturate [39]. As already discussed above, bridging CO in t b t phase cause 1953 cm 1 peak. After 1 st and 2 nd reaction cycles, some single oxygen vacancies are formed. In order to reduce the

PAGE 36

36 repulsion, some distribution of atop and bridge CO may cause blue shift of bridging CO from 1953 cm 1 to 1969 cm 1 The spectrum collected after 3 rd cycle reaction appears a new peak located at 1896 cm 1 and Figure 3 4 A shows that 52.96% of the oxygen atom has been reacted. Previous DFT calculation [46] mentioned that 1896 cm 1 corresponds to the Pd(111) surface hollow site within (2 2) 3CO phase the appearance of this peak shows that due to the generation of large number of oxygen vacancy, Pd(111) domain has been formed. Besides, a redshift of the peak from 1969 cm 1 to 1967 cm 1 can be observed. After 4 th cycle reaction, oxygen coverage only remained 0.663 ML, which is close to the saturation coverage of 2D oxide. A new peak appears at 2006 cm 1 which matches to the frequen cy of bridge CO on Pd 5 O 4 according to the previous DFT calculation [46]. Furthermore, trailing edge of peak located at 1967 cm 1 becomes broad and this trend lasts until to the last reaction cycle. Summary In this chapte r, we have investigated the RAIRS spectra of CO adsorption on Pd(111), Pd 5 O 4 and PdO(101) surfaces as well as isothermal CO oxidation on PdO(101) surface. It turns out that in this case, CO adsorbs on bridge and hollow sites on Pd(111) at the coverage of 0.5 ML at surface temperature 100 K, whereas for Pd 5 O 4 and PdO(101) surface, CO mainly adsorbs on a mixture of atop and bridge site that are formed by Pd 2f and Pd 3f respectively. In the isothermal CO oxidation experiment, more than 50% of the oxygen cover age is reacted by CO at first 3 cycles, and since then CO adsorption on metal domain within the oxidized surface becomes evident. As the oxygen is almost removed by CO after 7 cycles, IR spectra shows similar peaks comparing with CO adsorption on Pd(111) s urface.

PAGE 37

37 Figure 3 1. RAIRS of CO on Pd(111), Pd 5 O 4 PdO(101) surface. IR spectra collected at 100K with 1.0 10 8 torr CO background do s ing.

PAGE 38

38 IR frequenc ies on pristine PdO(101) t b t all atop Frequency/cm 1 (intensity) 2143(1.09) 1953(5.73) 21 49(1.05) Figure 3 2. IR frequencies (cm 1 ) and intensities (in parenthesis) of CO on different sites on pristine PdO(101) surface.

PAGE 39

39 IR frequencies on PdO(101) with one oxygen vacancies b t Ov t b t Ov b t t Ov t t t t Ov t t t Ov t Frequency/ cm 1 (intensity) 2139 (0.68) 2106 (0.26) 1953 (0.79) 2117 (0.55) 1991 (0.96) 1967 (0.02) 2134 (1.11) 2119 (0.01) 2094 (0.12) 2135 (1.12) 2114 (0.02) 2096 (0.21) 2140 2114 2078 Figure 3 3. IR frequencies (cm 1 ) and intensities (in parenthesis) of CO on di fferent sites on PdO(101) w i th one oxygen vacancy. For each case, we only take frequencies with large intensities (red) into account.

PAGE 40

40 Figure 3 4 Isothermal CO o xidation on PdO(101) surface and TPD of residual oxygen and CO 2 A ) Isothermal CO o xidation on PdO(101) surface at 373K. CO background pressure and dwell time is illustrated in the panel. The CO 2 generation coverage is shown on top of each peak (blue), the remaining oxygen coverage and consumption percentage of total oxygen coverage after each re action cycle is listed below the CO 2 coverage (green). B ) TPD experiment of residual oxygen and CO 2 after stopping the CO background dosing. The coverage of residual O 2 and CO 2 is listed in the panel.

PAGE 41

41

PAGE 42

42 Figure 3 5 Isothermal IR of CO reduction on P dO(101) surface at 373 K. IR was taken at 100 K after each cycle corresponding to the Figure 3 2 (a). The last one (dark blue) was collected at 100 K after annealing the sample to 900 K for 3 minutes and CO background dosing is P co =5 10 9 torr.

PAGE 43

43 CHAPTER 4 CONCLUSION In conclusion, these studies illustrate utility of atomic oxygen beam in the study of oxidized Pd(111) surface under UHV condition, CO adsorption on metal and oxidized Pd(111) surface as well as CO reduction mechanism on PdO(101) surface. It i s evident from current and previous studies that CO adsorption and reaction is both complicated and rich thinking when it comes to the assignment of adsorption sites and geometries of adsorption and reaction structure. Future study will be carried on at: 1 ) Isothermal reduction of CO on bulk and 2D oxide at high, medium and low temperature. 2) STM studies on real surface for in situ investigating of CO reaction on metal surface as well as oxidized surface. 3) LEED AES study of metal and palladium oxide.

PAGE 44

44 L IST OF REFERENCES [1] D.A. King, Surf. Sci., 47 (1975) 384 402. [2] P.A. Redhead, Vacuum, 12 203 211. [3] R.I. Masel, Principles of Adosrption and Reaction on Solid Surfaces, Wiley, New York, 1996. [4] F. M. Hoffmann, Surf. Sci. 3 (1983) 107 [5] C.J. Zhang P. Hu, J. Am. Chem. Soc. 123 (2001) 1166. [6] K. Nakao, S. I. Ito, K. Tomishige, K. Kunimori, Catal. Today 111 (2006) 316. [7] H. Gabasch, A. Knop Gericke, R. Schloegl, M. Borasio, C. Weilach, G. Rupprechter, S. Penner, B. Jenewein, K. Hayek, B. Kloetzer Phys. Chem. Chem. Phys. 9 (2007) 533. [8] E.E. Tornau, V. Petrauskas, G. Zvejnieks, Catal. Today 116 (2006) 62. [9] D. Zemlyanov, B. Aszalos Kiss, E. Kleimenov, D. Teschner, S. Zafeiratos, M. Haevecker, A. Knop Gericke, R. Schloegl, H. Gabasch, W. Unterb erger, K. Hayek, B. Kloetzer, Surf. Sci. 600 (2006) 983. [10] H. Gabasch, W. Unterberger, K. Hayek, B. Kloetzer, E. Kleimenov, D. Teschner, S. Zafeiratos, M. Haevecker, A. Knop Gericke, R. Schloegl, J. Han, F.H. Ribeiro, B. Aszalos Kiss, T. Curtin, D. Zeml yanov, Surf. Sci. 600 (2006) 2980. [11] H. Gabasch, W. Unterberger, K. Hayek, B. Kloetzer, G. Kresse, C. Klein, M. Schmid, P. Varga, Surf. Sci. 600 (2005) 205. [12] B. Klotzer, K. Hayek, C. Konvicka, E. Lundgren, P. Varga, Surf. Sci. 482 485 (2001) 237. [1 3] F.P. Leisenberger, G. Koller, M. Sock, S. Surnev, M.G. Ramsey, F.P. Netzer, B. Klotzer, K. Hayek, Surf. Sci. 445 (2000) 380. [14] E.H. Voogt, A.J.M. Mens, O.L.J. Gijzeman, J.W. Geus, Surf. Sci. 373 (1997) 210. [15] K.W. Kolasinski, F. Cemic, E. Hasselbr ink, Chem. Phys. Lett. 219 (1994) 113. [16] X. Guo, A. Hoffman, J.T. Yates, Jr., J. Chem. Phys. 90 (1989) 5787. [17] T. Matsushima, Surf. Sci. 157 (1985) 297. [18] D. Zemlyanov et al., Surf. Sci. 600 (2006) 983. [19] H. Gabasch et al., Surf. Sci. 600 (2006 ) 2980. [20] B. Klotzer, K. Hayek, C. Konvicka, E. Lundgren, P. Varga, Surf. Sci. 482 485 (2001) 237.

PAGE 45

45 [21] H. Gabasch, W. Unterberger, K. Hayek, B. Kloetzer, G. Kresse, C. Klein, M.Schmid, P. Varga, Surf. Sci. 600 (2005) 205. [22] X. Guo, A. Hoffman, J.T. Yates Jr., J. Chem. Phys. 90 (1989) 5787. [23] E.H. Voogt, A.J.M. Mens, O.L.J. Gijzeman, J.W. Geus, Surf. Sci. 373 (1997) 210. [24] K.W. Kolasinski, F. Cemic, E. Hasselbrink, Chem. Phys. Lett. 219 (1994) 113. [25] T. Matsushima, Surf. Sci. 157 (1985) 297. [26] D.L. Weissman Wenocur, M.L. Shek, P.M. Stefan, I. Lindau, W.E. Spicer, Surf. Sci.127 (1983) 513. [27] E. Lundgren, G. Kresse, C. Klein, M. Borg, J.N. Andersen, M. De Santis, Y. Gauthier, C. Konvicka, M. Schmid, P. Varga, Phys. Rev. Lett. 88 (2002) 24 6103/1. [28] H.H. Kan, J.F. Weaver, Surf. Sci. 602 (2008) L53 [29] H.H. Kan, J.F. Weaver, Surf. Sci. 603 (2009) 2671. [30] H.H. Kan, J.F. Weaver, Surf. Sci. 602 (2008) 1337. [31] H. Gabasch, K. Hayek, B. Kloetzer, W. Unterberger, E. Kleimenov, D. Teschner S. Zafeiratos, M. Haevecker, A. Knop Gericke, R. Schloegl, B. Aszalos Kiss, D. Zemlyanov, J. Phys. Chem. C 111 (2007) 7957. [32] M.S. Chen, Y. Cai, Z. Yan, K.K. Gath, S. Axnanda, D.W. Goodman, Surf. Sci. 601 (2007) 5326. [33] R.B. Shumbera, H.H. Kan, J.F. Weaver, Surf. Sci. 601 (2007) 4809 [34] G. Zheng, E.I. Altman, Surf. Sci. 462 (2000) 151 [35] S. H. Oh, J. Catal. 124 (1990) 477 [36] N.W. Cant, P.C. Hicks, and B.S. Lennon, J. Catal. 54 (1978) 372 [37] C. H. F. Peden and J.E. Houston, J. Catal. 128 (1991) 405 [38] A.D. Logan and M.T. Paffett, J. Catal. 133 (1992) 179 [39] D.W. Goodman and C. H. F. Peden, J. Phys. Chem. 90 (1986) 4839 [40] J. Gustafson, R. Westerstr m, O. Balmes, A. Resta, R. van Rijn, X. Torrelles, C.T. H erbschleb, J. W. M. Frenken, E. Lundgren, J. Phys. Chem. C, 114 (2010) 4580 [41] G. Blyholder, J. Phys. Chem. 68 (1964) 2772

PAGE 46

46 [42] D. F. Shriver, P. W. Atkins, C. H. Langford, Inorganic Chemistry, 2 nd ed.; Oxford University Press: Oxford/New York, 1994 [43] W. K. Kuhn, J. Szanyi and D.W. Goodman, Surf. Sci. 274 (1992) L611 [44] X. C. Guo and J. T. Yates, J. Chem. Phys. 90 (1989) 6761 [45] A. M. Bradshaw and F. M. Hoffmann, Surf. Sci. 72 (1978) 513 [46] E. Ozensoy, D. C. Meier and D. W. Goodman, J. Phys. Chem B, 106 (2002) 9367 [47] T. Gieel: et al. Surf. Sci. 406 (1998) 90 [48] M. K. Rose, T. Mitsui, J. Dunphy, A. Borg, D. F. Ogletree, M. Salmeron, P. Sautet, Surf. Sci. 512 (2002) 48 [49] N. M. Martin, et al. J. Phys. Chem. C 118 (2014) 1118 [50] J. A., Jr. ,Hinojosa, H.H. Kan, J. F. Weaver, J. Phys. Chem. C 112 (2008) 8324

PAGE 47

47 BIOGRAPHICAL SKETCH Tao Li earned her BS degree in chemical engineering from Jilin U niversity, China, in 201 2, where she did undergraduate research under Dr. Zhan Shi. After gradua tion, she began her master s studies at the University of Florida in the Department of Chemical Engineering under Dr. Jason F. Weaver in spring 2013. She has coauthored three manuscripts in peer reviewed journals.


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID ED2KL16LQ_FHZYIE INGEST_TIME 2014-10-03T21:42:24Z PACKAGE UFE0046847_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES