Tuning the Structure of Pt Nanoparticles Through Support Interactions

6 hours ago - Interactions of nanoparticles (NPs) with their environment may have a pronounced effect on their structure and shape as well as on their...
0 downloads 0 Views 960KB Size
Subscriber access provided by UNIV OF LOUISIANA

C: Physical Processes in Nanomaterials and Nanostructures

Tuning the Structure of Pt Nanoparticles Through Support Interactions: An In Situ Polarized X-ray Absorption Study Coupled with Atomistic Simulations Mahdi Ahmadi, Janis Timoshenko, Farzad Behafarid, and Beatriz Roldan Cuenya J. Phys. Chem. C, Just Accepted Manuscript • Publication Date (Web): 27 Mar 2019 Downloaded from http://pubs.acs.org on March 27, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

1

Tuning the Structure of Pt Nanoparticles Through Support Interactions: An in situ Polarized

2

X-ray Absorption Study Coupled with Atomistic Simulations

3

M. Ahmadi1⸸, J. Timoshenko2⸸, F. Behafarid1, B. Roldan Cuenya1,2*

4

1Department

of Physics, University of Central Florida, Orlando, FL 32816, USA

5

2Department

of Interface Science, Fritz-Haber-Institute of the Max Planck Society, 14195 Berlin,

6

Germany

7

*e-mail: [email protected]

8

⸸equally

contributing authors

9 10

Abstract

11

Interactions of nanoparticles (NPs) with their environment may have a pronounced effect on their

12

structure and shape as well as on their functionality in applications such as catalysis. It is therefore

13

crucial to disentangle the particle-adsorbate and particle-support interaction effects on particle

14

shape, its local structure, atomic dynamics and their possible anisotropies. In order to gain insight

15

into the support effect, we carried out an X-ray absorption fine-structure spectroscopy (XAFS)

16

investigation of adsorbate- and ligand-free size-selected Pt NPs deposited on two different

17

supports in ultrahigh vacuum (UHV). Polarization-dependent XAFS measurements, neural

18

network-based analysis of X-ray absorption near-edge structure (XANES) data and reverse Monte

19

Carlo (RMC) simulations of extended X-ray absorption fine structure (EXAFS) were used to

20

resolve the 3D shape of the NPs and details of their local structure. A synergetic combination of

21

advanced in-situ XAFS analysis with atomic force microscopy (AFM) and scanning tunneling

22

microscopy (STM) imaging provides uniquely detailed information about the particle-support

23

interactions and the NP/support buried interface, not accessible to any experimental technique,

24

when considered alone. In particular, our combined approach reveals differences in the structure

25

of Pt NPs deposited on TiO2(110) and SiO2/Si(111). Pt NPs on SiO2 assume a spherical-like 3D

26

shape and weakly interact with the support. In contrast, the effective shape of analogously

27

synthesized Pt NPs on TiO2(110) after annealing at 600°C is found to be a truncated octahedron

28

with (100) top and interfacial facets that are encapsulated by the TiO2 support. Modeling disorder 1 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

29

effects in these NPs using an RMC approach reveals differences in bond-length distributions for

30

NPs on different supports, and allows us to analyze their anisotropy, which may be crucial for the

31

interpretation of support-dependent atomic dynamics and can have an impact on the understanding

32

of the catalytic properties of these NPs.

33 34

Introduction

35

Metal nanoparticles (NPs) supported on oxides have tremendous applications in the field of

36

heterogeneous catalysis. Numerous studies have been dedicated to explore the parameters

37

affecting the catalytic performance of nanocatalysts such as their particle size, shape, oxidation

38

state and interaction with the support.1-12 Nevertheless, in many cases it is hard to isolate the

39

respective contribution of these parameters. For example, it was shown that the NP shape can be

40

affected by the size, adsorbates, ligands and support, while the NP/support interaction itself is also

41

influenced by the NP size,1,3,13 all of which could subsequently affect the NPs activity and

42

selectivity.14-18 For understanding structure-properties relationship in heterogeneous catalysts and

43

for the rational design of novel catalyst materials, it is necessary to disentangle these many degrees

44

of freedom. For this purpose, studies of well-defined model systems are useful, since they allow

45

us to single out the importance of each of these catalyst characteristics. Here we focus on particle-

46

support interactions and their effect on particle shape and local structure.

47

The NP support can affect the NP structure by (i) changing the electronic structure of the NPs,19-

48

21

49

the NP and changing the surface energy.24-26 Interactions with the support may have an impact also

50

on atomic dynamics, especially for smaller NPs,27 resulting in non-bulk-like distributions of

51

interatomic distances, which, again, can result in unique functionality.5 The presence of particle-

52

support interactions introduces naturally also the anisotropy in the NP structure, which may

53

manifest itself in unusual vibrational properties and strain distributions.27

54

To rationalize the importance of these factors, tools are needed to extract detailed information

55

about the geometry and internal structure of NPs on different supports and under different

56

experimental conditions. For this purpose, different microscopy techniques can be used, including

57

scanning tunneling microscopy (STM), atomic force microscopy (AFM), and transmission

(ii) by inducing strain in the NP lattice due to interfacial mismatch,22,23 and (iii) by encapsulating

2 ACS Paragon Plus Environment

Page 2 of 31

Page 3 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

58

electron microscopy (TEM).3,28-31 These techniques can provide insight into the shape of individual

59

NPs and their local arrangement on relatively small substrate areas (hundreds of µm2), but, when

60

applied to small NPs, tip-convolution effects in some of these methods can obscure the real NP

61

structure and the amount of quantitative information that can be extracted is limited. On the other

62

hand, complementary spectroscopic techniques such as grazing incidence small-angle X-ray

63

scattering (GISAXS) and X-ray absorption fine-structure spectroscopy (XAFS) can provide

64

information about the average NP size, shape and structure.4,32-36 An important advantage of these

65

spectroscopic techniques is their ability to resolve transformations of nanocatalysts in situ and

66

under operando reaction conditions, including high temperatures and elevated gas pressures.

67

Nevertheless, due to ensemble-averaging, the latter techniques provide the most meaningful

68

answers when they are applied to model systems with narrow particle size, shape and

69

compositional distributions (as verified by microscopy techniques). When this condition is not

70

met, extreme care needs to be taken in the interpretation of the obtained results, since the analysis

71

will yield only an effective, ensemble-averaged particle 3D structure. In some cases, such effective

72

NPs model can, nevertheless, be successfully correlated with sample functionality.37

73

Among the different structure-sensitive techniques, extended X-ray absorption fine-structure

74

(EXAFS) analysis stands out as an element-specific method that is very sensitive to the local

75

structure around the absorbing atom.38 For samples containing monodispersed NPs, EXAFS

76

analysis allows the extraction of very detailed information about NP size, shape and structure.39-42

77

For example, we previously reported the possibility to determine the shape of size-selected Pt NPs

78

supported on nanocrystalline Al2O3 powders by combining the information on the 1th to 4th

79

coordination numbers (CNs) extracted from multiple-scattering (MS) EXAFS analysis, with the

80

knowledge of the NP diameter determined by TEM.4,16,34 Note, however, that our previous studies

81

were conducted under H2-atmosphere to prevent NP oxidation, and therefore, the NP shapes

82

observed might have been affected by interactions with adsorbed hydrogen.43 It is well known

83

that adsorbates can affect the stability of the otherwise more thermodynamically favorable NP

84

shapes.43-45 Furthermore, on the nanocrystalline supports that were previously employed, epitaxial

85

relationships between the NPs and the support could not be ensured. Since the NPs were oriented

86

randomly on the support surface, possible anisotropies in the NP shape and structure could not be

87

resolved. In the present study, we have used similarly synthesized ensembles of well-defined metal

88

NPs but have deposited them on planar single crystals to unveil, in the absence of adsorbates (in a 3 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

89

UHV measurement environment), the role of the support. Furthermore, by employing X-ray

90

polarization-dependent XAFS measurements, we are now able to extract information on the

91

differences in the metal-metal bonds within NPs along different directions (parallel and

92

perpendicular to the support surface), and also to single out the contribution of the bonds between

93

the NPs and the support.46,47 These data allowed us to estimate the most favorable structural

94

configuration for our NPs deposited on a strongly interacting support (TiO2) versus a weak one

95

(SiO2).

96

The interpretation of in situ polarized XAFS data using conventional approaches is challenging,

97

especially in cases like the present one, where the low metal loading results in limited EXAFS data

98

quality. Reliable multiple scattering EXAFS analysis of distant coordination shell contributions

99

that is necessary to resolve NPs shape is not possible in this case. Instead, we have applied recently

100

developed advanced XAFS analysis methods based on the neural network (NN) approach to

101

extract the NP structural information from the experimental XANES data, which have much better

102

signal to noise ratio.48-50 CNs extracted from XANES are then used to construct a 3D structural

103

model. Its agreement with available EXAFS data is further validated using reverse Monte Carlo

104

(RMC) simulations.48,51 RMC-EXAFS analysis allowed us also to extract the entire bond-length

105

distribution function (radial distribution function (RDF)) that encodes information about the

106

variations in bond lengths due to static and dynamic thermal disorder effects.49,52 Here we employ

107

for the first time the RMC analysis for the interpretation of polarization-dependent EXAFS data.

108

By combing polarization-dependent XAFS measurements of NPs on single crystal supports with

109

the RMC-EXAFS simulation method, we extracted information about the possible anisotropy of

110

RDFs, and obtained accurate interatomic distance distributions. In particular, the RMC simulations

111

yielded non-Gaussian, support-dependent shapes for the RDF peaks that indicate the presence of

112

strong particle surface- and support-induced disorder and/or essentially anharmonic atomic

113

dynamics,48,53,54 which could have an impact on the catalytic properties of these NPs. In addition,

114

we were able to analyze separately the bond length distributions for the bonds that are parallel and

115

perpendicular to the substrate surface. Therefore, we were able to probe directly the effect of

116

particle-support interactions on the anisotropy of the particle structure and dynamics. Such detailed

117

insight into the interactions between nanometer-sized particles and the underlying support,

118

including insight into buried interfaces, was enabled by a synergistic combination of scanning-

119

probe microscopy and spectroscopic techniques and advanced data analysis. 4 ACS Paragon Plus Environment

Page 4 of 31

Page 5 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

120 121 122

Methods A. Sample preparation and experimental details

123

Size-selected Pt NPs were synthesized using inverse micelle encapsulation.55,56 A commercial

124

diblock copolymer poly(styrene)-block-poly(2-vinylpyredine)[PS(26000)-P2VP(4800)] was

125

dissolved in toluene to form the micelles. Subsequently, H2PtCl6∙6H2O with a metal/P2VP ratio of

126

0.18 was added to the solution to create the metal NPs. A monolayer thick film of NPs was

127

achieved after dip coating the TiO2(110) crystal and natively oxidized SiO2/Si (111) substrate into

128

the micellar solution. In situ O2-plasma treatment at 5×10-6 mbar for 100 min was performed to

129

remove the polymers. To stabilize the Pt NPs on TiO2, the sample was stepwise annealed in

130

vacuum from 300 °C up to 900 °C in 100 °C intervals.

131

To get information about the morphology of the as prepared samples, AFM images were acquired

132

using a Veeco multimode microscope (Digital instrument, Nanoscope IIIa). All STM images were

133

acquired at room temperature using an Aarhus 150 HT STM microscope (SPECS GmbH). An

134

electrochemically etched tungsten tip was used for the STM measurements. Due to tip-convolution

135

effects, the NP height (and not the diameter) was used as the representative NP size descriptor. In

136

order to confirm the complete removal of the ligands, X-ray photoelectron spectroscopy (XPS)

137

measurements were conducted using a monochromatic Al Kα X-ray source (1486.6 eV) and a

138

Phoibos (SPECS) electron energy analyzer.

139

X-ray absorption fine-structure spectroscopy (XAFS) measurements were performed at the 12-

140

BM beamline of the Argonne National Laboratory, which is equipped with a UHV chamber

141

compatible with XAFS measurements. The experiments were conducted in fluorescence mode at

142

the Pt L3-edge (11562 eV) at room temperature. The ex situ synthesized samples were first

143

transferred into UHV and annealed in O2 at 450°C for 30 min in order to remove adventitious

144

carbon from the NPs. Subsequently the samples were reduced in vacuum at 600°C. To get the

145

information about the anisotropic structure of the NPs, in the case of Pt/TiO2(110), EXAFS

146

measurements were carried out with the X-ray polarization in three different orientations, from

147

which two were parallel to the substrate surface and one was perpendicular to the substrate. The

148

two parallel measurements were done along (i) TiO2(110)[001] (denoted as ║-1), and (ii) 5 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

149

TiO2(110)[ 110] (denoted as ║-2) directions. Since no well-defined crystalline orientations were

150

available at the native SiO2/Si(111) interface, on this substrate no polarization-dependent

151

measurements were performed, and we collected XAFS data with one X-ray polarization only.

152

The Athena and FEFFIT programs were used for XAFS data normalization, background

153

subtraction and conventional non-linear least-square fitting of the experimental EXAFS data.57

154

Fitting of the first-shell EXAFS data was done over a range of 3-10.5 Å-1 in k- space and 1.0-3.4

155

Å in R-space. EXAFS data from Pt/TiO2 were fitted with three components corresponding to Pt—

156

Pt (~2.7 Å), Pt—Ti (~2.7Å) and Pt—O (~2.0 Å) bonds, while for Pt/SiO2 only Pt-Pt and Pt-O were

157

considered. Due to limited k-range and R-range, available for the analysis and the large number of

158

fitting parameters, multiple constraints were employed: correction to photoelectron reference

159

energy ΔE0=1.8±0.7 eV was fixed to be the same for all the samples (this constrain is motivated

160

by the similarity of the obtained XANES spectra for different samples), Debye-Waller factors σ2

161

and interatomic distances R for Pt—Pt, Pt—O and Pt—Ti bonds were fixed to be the same for all

162

three spectra, corresponding to Pt/TiO2 system. The passive electron reduction factor (𝑆20) was

163

estimated to be 0.87 by fitting EXAFS spectrum of platinum foil. For calculations of the theoretical

164

scattering amplitudes and phase shifts used in the fits, the FEFF8 ab initio code has been

165

employed,38,58 using fcc Pt structure to simulate the Pt—Pt scattering path, PtO2 structure to

166

simulate the Pt—O scattering path, and TiO2 structure with embedded Pt atom to simulate the Pt—

167

Ti scattering path.

168

B. XAFS data interpretation

169

The sub-mono-layer sample coverage with widely spaced small NPs employed in this study

170

resulted in low metal concentration in the samples, thus XAFS data quality is not adequate to

171

conduct reliable conventional multiple scattering (MS) EXAFS analysis of contributions beyond

172

the first coordination shell. Additional information is therefore necessary to resolve the NP shape.

173

Such additional information can be extracted by considering the polarization of the X-rays in

174

EXAFS measurements. We employ here the fact that the contribution of different interatomic

175

bonds to the total spectrum depends on the relative orientation of the bond with respect to the X-

176

ray polarization vector.59 For K-absorption edges, one can introduce a simple relation between the

177

effective coordination number (ECN), i.e., the number of bonds per absorbing atom that contribute

6 ACS Paragon Plus Environment

Page 6 of 31

Page 7 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

178

to the X-ray absorption for a specific polarization direction, and the angle between the bond

179

direction and the X-ray polarization:60,61

180

𝑁=

( )∑ 3 𝑁𝑎𝑡

𝑐𝑜𝑠2𝜃𝑗,#(1)

𝑗

181

where 𝑁𝑎𝑡 is the number of absorbing atoms in the particle, θj is the angle between the jth bond

182

direction and the electric field vector of the incident X-ray. This simple relation, however, is not

183

accurate for L2 and L3 absorption edges.62 The effect of X-ray polarization on ECN for these

184

absorption edges can be assessed numerically, using codes like FEFF,63 which treat polarization

185

effects explicitly. As demonstrated in the Supporting Information, Fig.S1, for the Pt L3-edge, the

186

shape of the ECN(θ) dependence qualitatively still resembles a cos2(θ) function, but the effect of

187

the polarization is less pronounced than for the K-edge. This means that we can still use Eq.(1) at

188

least for qualitative analysis of L3-edge XAFS data. As demonstrated in Fig. S1, the error in ECN

189

that we introduce in this case is around 15%. In this study, the ECNs extracted from polarized

190

EXAFS data were compared to those obtained from particle structure modeling to extract

191

information on the shape of Pt NPs.

192

For unambiguous determination of the NP shape, information about CNs in more distant

193

coordination shells is necessary.39,42,64 For this purpose we employ here NN-based XANES

194

analysis, which yields CNs from distant coordination shells even in the cases when the low data

195

quality does not allow us to use MS EXAFS analysis. Our approach for the analysis of XANES

196

data using neural networks is explained in detail in Ref.50 and we use here the same NN. Briefly,

197

a large set of ab-initio calculated XANES spectra is constructed for particle models with different

198

sizes and shapes using FEFF63 and FDMNES65 codes. For each of the used structure models we

199

calculate also the corresponding CNs for the first few coordination shells. The constructed

200

database of calculated XANES spectra and corresponding CNs values is then used to train a NN,

201

which maps the relationship between XANES features and CNs. The trained NN then can be used

202

to determine CNs from experimental XANES spectra. Importantly, our NN was trained on

203

structure models of completely metallic particles only, thus, it cannot recognize and quantify the

204

interactions of the particle with the support (bonding, charge transfer, etc). For example, it cannot

205

give any answers regarding Pt—O and Pt—Ti bonds for Pt particles on the TiO2 support.

206

Nevertheless, our previous studies have shown that if the contribution of these bonds is small, as 7 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

207

it is expected at least for the Pt/SiO2 system, our NN can still provide reasonably reliable results

208

about the Pt—Pt CNs that can be used complimentary with CNs values extracted from the

209

traditional EXAFS data fitting.

210

The knowledge of CNs from EXAFS and NN-XANES analysis allows us to determine the NP size

211

and shape. In addition to that, EXAFS data contain also the details about the distribution of bond

212

lengths, which can be characterized by RDF, and contains key information about the static

213

structure strain and atomic dynamics.48,51 To extract this information, and also to additionally

214

validate the NP structural models obtained from conventional EXAFS and NN-XANES analysis,

215

we employ the RMC method. In RMC simulations, the aforementioned structure model (which

216

includes Pt NP model as well as support atoms, when applicable) is further optimized, by

217

proposing a small and random displacements to all atoms in the model to account for disorder

218

effects. Ab-initio EXAFS calculations are then carried out, and the obtained spectrum is compared

219

with the experimental data. For comparison of theoretical and calculated spectra, we use wavelet

220

transform (WT).66,67 Depending on the agreement, the proposed structure modification is either

221

accepted or discarded. The process is repeated, until after several thousands of iterations a good

222

agreement between experimental and theoretical EXAFS spectra is achieved. Importantly, since

223

only small atomic displacements are allowed, the overall size and shape of the NP do not change

224

during the RMC optimization.

225

In this work, for RMC-EXAFS analysis we rely on the EvAX code.68 Calculations for Pt NPs on

226

SiO2 are carried out as in Ref.50 To benefit from the sensitivity of EXAFS spectra to polarization

227

effects, and to obtain a more unambiguous structure model for Pt NPs on TiO2, we have modified

228

the EvAX code, so that polarization effects on EXAFS spectra can now be modeled explicitly.

229

Note that the calculations of polarized EXAFS spectra are carried out accurately in the EvAX

230

program (which uses FEFF code for EXAFS calculations). In this particular case, at each RMC

231

𝑡𝑒𝑜𝑟 𝑡𝑒𝑜𝑟 iteration three theoretical EXAFS spectra are calculated (𝜒𝑡𝑒𝑜𝑟 1 (𝑘), 𝜒2 (𝑘) and 𝜒3 (𝑘)),

232

corresponding to the same structure model, but three different X-ray polarizations (perpendicular

233

to substrate surface and parallel to TiO2(110)[001] (denoted further as ║-1) and TiO2(110)[110]

234

(denoted further as ║-2) directions), and all three are compared independently with the

235

𝑒𝑥𝑝 𝑒𝑥𝑝 corresponding experimental EXAFS spectra (𝜒𝑒𝑥𝑝 1 (𝑘), 𝜒2 (𝑘) and 𝜒3 (𝑘)). The residual that

236

characterizes the goodness of current RMC atomic configuration is then calculated as 8 ACS Paragon Plus Environment

Page 8 of 31

Page 9 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

3

‖𝑊𝑇[𝜒𝑡ℎ𝑒𝑜𝑟 (𝑘)] ― 𝑊𝑇[𝜒𝑒𝑥𝑝 𝑖 𝑖 (𝑘)]‖ .#(2) ∑ ‖𝑊𝑇[𝜒𝑒𝑥𝑝 𝑖 (𝑘)]‖ 𝑖=1

1 𝜉= 3

237 238 239 240

Here WT[..] denotes Morlet wavelet transform and ||..|| - Euclidean norm. Thus, at the end of the

241

RMC run, we get a single atomic configuration, which agrees with the EXAFS data collected for

242

the three different X-ray polarizations. Other technical details of the RMC simulations are given

243

in Ref.50

244 245 246

Results and Discussion A. Microscopic Morphological Characterization (AFM/STM)

247

AFM images of as-prepared micellar Pt NPs supported on SiO2 and TiO2(110) are shown in Fig.

248

1(a) and (b), respectively. According to the AFM data, the average NP heights in these samples

249

are ~1.8 ± 0.8 nm and ~1.9 ± 0.6 nm, respectively (the apparent difference in the NP diameter seen

250

in the images is due to the fact that both samples were measured with tips of different end radius).

251

Height histograms are shown in Supporting Information, Fig. S2. The NPs are distributed with a

252

hexagonal ordering and homogenous interparticle distance (~ 40 ±10 nm) on both substrates. An

253

STM image of the Pt NPs supported on TiO2(110) acquired after annealing in vacuum at a

254

temperature of 1100°C is shown in Figure 1(c). Note that since the SiO2 substrate has a wide band

255

gap, STM measurements are not feasible for this sample unless it is highly doped, which was not

256

the present case. A homogenous interparticle distance has been observed after the high temperature

257

annealing, indicating the high stability and low mobility of the NPs on TiO2. The high resolution

258

STM images in Fig. 1(d),(e) reveal the flat NP top, which can be attributed to truncated octahedron

259

NPs with (100) top and interfacial facets. The corresponding line profiles are shown in Fig. 1(f).

260

It should be noted that due to the tip convolution effect, the lateral size of the NPs is overestimated

261

in STM and AFM measurements. However, the height of the NPs (Fig. S2) is less affected by the

262

tip apex artifact and can be used as a realiable measure of the particle size .

9 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

263 264

Figure 1. (1×1 μm) AFM images of as-prepared Pt NPs supported on (a) SiO2/Si(111) and (b)

265

TiO2(110). (c, d, e) STM images of polymer-free Pt NPs on TiO2(110) acquired at 25°C after

266

annealing in vacuum at 1100°C. The inset in (c) and the high resolution images in (d,e) display

267

truncated octahedron Pt NPs on TiO2(110) with (100) top and interfacial facets. (f) Line profiles

268

of the Pt NPs are shown in (d,e).

269 270

B. Spectroscopic Structural Characterization (EXAFS/XANES)

271

1. XANES data

272

Experimental XANES spectra of our Pt NPs measured at room temperature in UHV on SiO2 and

273

TiO2(110) are shown in Fig. 2. In all cases the spectra resemble strongly the spectrum of the Pt

274

foil (also shown in Fig. 2), confirming that Pt is mostly metallic in these samples, and no significant

275

oxidation takes place under our UHV measurement environment. One can note that XANES

276

oscillations are broadened and have lower amplitudes for all NPs samples, in comparison to those

277

of bulk Pt, which can be attributed to particle size-effects.50 Also, a minor shift of spectra to higher

278

energies can be detected for all spectra for Pt NPs, which can be due to the interaction with the

279

support. The differences between the spectra for NPs on different supports are rather subtle,

280

suggesting that particle size and structure are similar in samples on both supports. XANES spectra 10 ACS Paragon Plus Environment

Page 10 of 31

Page 11 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

281

for Pt on TiO2, acquired with different X-ray polarizations, are also quite close. For Pt NPs on

282

SiO2, we did not perform polarization-dependent measurements because no well-defined

283

crystalline orientations were available at the native SiO2/Si(111) interface. For Pt NPs on TiO2,

284

spectra, acquired with both substrate-parallel X-ray polarizations, overlap completely. The

285

intensity of white line in the spectrum, acquired with X-ray polarization, perpendicular to sample

286

surface, is slightly lower, suggesting some anisotropy in particle structure and/or particle-support

287

interactions. This difference, however, is much smaller than the one observed in a similar study of

288

raft-like (and, hence, very anisotropic) Pt clusters on α-Al2O3.46

289 290

Figure 2. Experimental Pt L3-edge XANES spectra measured at 25°C in UHV corresponding to

291

Pt NPs supported on SiO2 and TiO2(110). The data on TiO2 were acquired in three different X-ray

292

polarizations, two parallel to the support surface and one perpendicular. Data from Pt foil are also

293

shown for reference.

294 295

2. Analysis of EXAFS data (non-linear least square fitting)

296

Fig.3 shows k2-weighted EXAFS spectra in k-space (a) and their Fourier-transforms (FT) (b-d),

297

extracted from the analysis of Pt L3-edge data for NPs on two different supports. Clear differences

298

can be observed. In particular, amplitude of EXAFS oscillations is slightly larger for Pt on SiO2

299

support. In addition, the shoulder in FT-EXAFS data between 1 and 2 Å that can be attributed to

300

a contribution from low-Z elements is absent in the data for Pt NPs on SiO2. Note that since the 11 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

301

same NP solution was used to synthesize the NPs deposited on TiO2 and SiO2, the resulting NPs

302

are expected to have the same average number of atoms in both samples. Therefore, the observed

303

differences can be attributed to the differences in particle shape and/or bonding with the support.

304

Polarization effects in EXAFS data for Pt NPs on TiO2 are also shown in Fig. 3. First of all, note

305

the significant difference between spectra for substrate-perpendicular X-ray polarization and X-

306

ray polarization along [001] direction (ǁ-1 direction). The FT-EXAFS data for the former one

307

contain a pronounced shoulder between 1 and 2 Å due to a contribution from low-Z elements,

308

which can interpreted as an evidence of Pt bonding with the oxygen atoms in the support. Due to

309

the dependence of contributions of different bonds on the angle between bond orientation and X-

310

ray polarization direction (Eq.1), these bonds are not expected to contribute significantly to the

311

spectra with substrate-parallel X-ray polarization, which explains the lack of this contribution in

312

ǁ-1 spectrum.

313

However, in contrast to the ǁ-1 spectrum the contribution of low-Z elements is clearly visible in

314

EXAFS spectrum, collected with X-ray polarization in another substrate-parallel direction ([110]

315

or ǁ-2 direction). Since substrate-perpendicular Pt—O bonds cannot contribute significantly to this

316

spectrum, this observation suggests the presence of some oxygen atoms also on the side-surfaces

317

of the Pt nanoparticle, which can be interpreted as a partial encapsulation of Pt particle by TiO2

318

substrate and/or presence of adsorbed oxygen on side facets of Pt particle. Note that encapsulation

319

of Pt NPs in TiO2 has been observed, for example, through STM studies.25,31 The difference

320

between results for ǁ-1 and ǁ-2 directions in this case can be explained by an anisotropy of

321

TiO2(110) surface. This anisotropy and its effect on interactions with Pt clusters have also been

322

acknowledged previously.69, 24,31

323

For quantitative analysis, we perform non-linear least-square EXAFS fitting. The results of such

324

fitting are shown in R-space in Fig. 3, while k-space data are shown in Supporting Information,

325

Fig. S3. The contributions of the individual scattering components are also plotted in R-space

326

(Fourier-transformed) in Fig.3. Corresponding values of structure parameters are reported in Table

327

1 and Table S1 in the Supporting Information. Our strongly constrained fitting model that accounts

328

for the presence of Pt—Pt, Pt—O and Pt—Ti bonds can fit experimental data quite accurately, and

329

all the values of interatomic distances and Debye-Waller factors seem to be physically reasonable,

330

giving us confidence in the obtained values of structure parameters. It is worth mentioning that an 12 ACS Paragon Plus Environment

Page 12 of 31

Page 13 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

331

alternative fitting model with an extra Pt—O bond (ca. 2.7 Å) instead of Pt—Ti bond also resulted

332

in a reasonable fit. However, the obtained Pt—Pt structure parameters were similar in both cases.

333

In the case of Pt NPs on SiO2, higher Pt—Pt and lower Pt—O CNs (as compared to those for NPs

334

on the TiO2 substrate) are obtained. Since the number of atoms per particle should be close in both

335

samples, the higher Pt—Pt coordination numbers obtained for Pt/SiO2 can be interpreted as an

336

evidence for more spherical particle shape, which is in an agreement with our earlier TEM study

337

showing nearly spherical Pt NPs supported on SiO2 after annealing at 727°C.55 Low Pt—O CN for

338

Pt on SiO2 is in agreement with the results of visual examination of EXAFS data above, and

339

suggests relatively weak particle-support interaction for this system.

340

Pt—Pt coordination numbers obtained for Pt NPs on TiO2 with different X-ray polarizations are

341

relatively close, suggesting that NPs on this support are also rather three-dimensional (as opposed

342

to flat, raft-like structures, reported elsewhere for small Pt NPs with very strong particle-support

343

interactions).4,16,34 Similarity of Pt—Pt coordination numbers in ǁ-1 and ǁ-2 direction also suggests

344

the lack of noticeable elongation of Pt NPs in TiO2(110) plane, reported in the literature24,31 for

345

PVD NPs. This discrepancy can be explained based on the differences in the synthesis methods

346

employed and in the size of the NPs used. In our study, the NPs were prepared by colloidal

347

chemistry and therefore originally had a spherical shape, while in the other studies (e.g., by Dulub

348

et al, Ref. 31) the NPs were directly grown on the substrate. The shape of the NPs evaporated on a

349

substrate is impacted by the growth kinetics such as possible ad-atom diffusion asymmetries. In

350

addition, while our NPs were smaller than 2 nm in size, the PVD-grown NPs in the mentioned

351

study had a lateral size of around 20 nm. Since the strain and the support interactions depend on

352

the NP size, the stress build-up in a larger NP (such as PVD-grown NPs1,70) may facilitate an

353

asymmetrical growth of the NPs.

354

The main difference between XAFS results for different X-ray polarizations thus lies in the

355

difference in Pt—O and Pt—Ti coordination numbers. In particular, contribution of Pt—O bonds

356

is significantly lower for ǁ-1 spectrum than for other two spectra, confirming the presence of

357

anisotropic particle-support interactions in this system.

13 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

358 359

Figure 3. k-space k2-weighted Pt L3-ege EXAFS data (a) and Fourier-transformed EXAFS data

360

(b) for Pt NPs on SiO2 and TiO2(110). Fitted spectra considering Pt—Pt and Pt—O bonds for NPs

361

on SiO2/Si(111), and Pt-Pt, Pt−O, and Pt−Ti bonds for NPs on TiO2(110) are also shown. For

362

clarity, spectra are shifted vertically in (b).

363 364 365 366 367 368 369

14 ACS Paragon Plus Environment

Page 14 of 31

Page 15 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Sample/X-ray polarization

NPt-Pt

NPt-O

NPt-Ti

Pt/TiO2 (110) -║-1

7.3 (9)

0.1(2)

0.5(3)

Pt/TiO2 (110) -║-2

6.4(8)

0.6(2)

1.0(4)

Pt/TiO2 (110) -┴

7.3(9)

0.6(2)

0.3(2)

Pt/SiO2

8.3 (3)

0.2 (1)

---

370

Table 1. ECNs for Pt NPs on TiO2(110) obtained for different X-ray polarizations with respect to

371

the support surface (two in-plane orientations parallel to the TiO2(110)[001] (║-1) and TiO2(110)[1

372

10] (║-2) support directions and one perpendicular), obtained from the analysis of EXAFS data.

373

CNs from EXAFS data for Pt on SiO2 are also shown. Uncertainties in the last digit are given in

374

parentheses.

375

3. Neural network-based analysis of XANES data

376

Quantitative interpretation of EXAFS spectra is always complicated by the correlations between

377

different structural parameters. These correlations especially hinder the analysis of strongly

378

overlapping contributions beyond the first coordination shell for data with limited k-range and

379

poor signal-to-noise ratio. Therefore, here we complement EXAFS analysis with the NN-based

380

analysis of XANES data. Experimental XANES data for Pt NPs are shown in Fig. 2. These spectra

381

were used as input for NN analysis,50 and the obtained CNs for the first four coordination shells

382

are given in Table 2.

383

According to the EXAFS analysis, bonding with the support is the weakest for the sample on SiO2,

384

thus in this case the results, obtained by NN that is trained on metallic particles only, should be the

385

most reliable. Indeed, the 1st shell CN obtained from XANES analysis for this sample (7.9±0.4)

386

is in an excellent agreement with the results of conventional EXAFS data fitting (8.3±0.3). This

387

agreement shows the power of NN-based analysis of XANES data, which could be a very useful

388

method in the cases, where high quality EXAFS spectra cannot be achieved (e.g., for diluted

15 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 31

389

samples, time dependent studies, measurements in liquid or high-pressure gas environments due

390

to attenuation of beam).

391

The applicability of NN-XANES method for the samples on TiO2 is less certain due to a larger

392

contribution of bonds with the support. Interpretation of the CNs values, obtained from NN-

393

XANES analysis, should be done with caution in this case. However, one can note that at least for

394

the first coordination shell the results from NN-XANES analysis (Table 2) agree reasonably with

395

the independently obtained results of EXAFS analysis (Table 1), giving us some confidence that

396

contribution of Pt—O and Pt—Ti bonds is sufficiently low in this case for NN-XANES method to

397

work, and that also the CNs for the 2nd, 3rd and 4th coordination shells are determined reliably. In

398

addition, perhaps, the most important conclusion from NN-XANES analysis, which should not be

399

affected by the presence of any systematic errors, is that all four coordination numbers for three

400

different polarization directions are similar, suggesting, again, a rather three-dimensional shape of

401

Pt NPs, and the lack of significant elongations along any specific direction. Sample/X-ray

N1

N2

N3

N4

Pt/TiO2 (110) -║-1

6.2(3)

3.5(3)

5(2)

4(2)

Pt/TiO2 (110) -║-2

6.0(4)

3.7(3)

7(1)

4(1)

Pt/TiO2 (110) -┴

6.1(4)

3.6(2)

5(2)

3(1)

Pt/SiO2

7.9 (4)

4.0(4)

7(2)

5(1)

polarization

402

Table 2. Pt—Pt effective coordination numbers (ECNs) for the first, second, third and fourth

403

coordination shells (N1, N2, N3 and N4, correspondingly) extracted by NN analysis from

404

experimental XANES data. Uncertainties in the last digit are given in parentheses.

405 406 407 16 ACS Paragon Plus Environment

Page 17 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

408

4. Resolving the structure of Pt NPs

409

A MATLAB code was written to generate a database of Pt NP structure models to match the CNs

410

extracted through the analysis of EXAFS and NN-based analysis of XANES. Different NP shapes

411

were obtained by truncating a bulk Pt fcc structure along three different planes (111), (110) and

412

(100) to generate NPs with sizes ranging from 0.8 to 3 nm. This database includes a variety of NP

413

shapes such as cuboctahedron, truncated octahedron and hexahedron. For each shape, the

414

corresponding ECNs were calculated by considering different X-ray polarization directions

415

(Eq.(1)). Subsequently, the ECNs and particle sizes for the NPs models were compared with the

416

experimental results extracted from NN-based analysis of XANES data, the 1st shell conventional

417

EXAFS data analysis and AFM and STM measurements. Fig. 4(a) shows the particle model for

418

the Pt/SiO2 system with the best agreement with the experimental results (AFM-determined

419

average NP height and CNs from XAFS analysis): all four CNs, obtained from the experimental

420

XAFS data, agree well with those for a cuboctahedral particle with 55 atoms (corresponding CNs

421

for the model are reported in Table 3). The size and shape of this particle model also agree well

422

with the available TEM data for analogously synthesized NPs.55

423

In the case of Pt/TiO2, the determination of particle shape is possible only by combining insights

424

from different techniques, since by its own neither EXAFS, nor XANES, nor microscopy result is

425

conclusive for this complex system, and can be biased by significant systematic errors. To

426

determine the most likely NPs shape, we use the first shell coordination numbers from EXAFS

427

analysis, and the first four CNs from XANES analysis for each polarization direction. We use also

428

insights from STM and AFM data (Fig.1) regarding possible particle size and shape. Furthermore,

429

we exploit the observation that the total number of atoms in the particles on TiO2 should be close

430

to the number of atoms in the particle on SiO2 (55). The NP model that gives the best agreement

431

with these data is shown in Figs. 4(b) and (c): a truncated octahedron with Pt(100) top and

432

interfacial facet and with 50 atoms. Corresponding CNs for this model particle are reported in

433

Table 3 and agree well with the results of XAFS data analysis. It is worth mentioning that the 1st

434

CN (N1) of ca. 7 has been estimated to correspond to a NP with ca. 1 nm in diameter.39,71 Similar

435

shapes have been observed in our earlier STM study of larger Pt NPs (