Active Site and Electronic Structure Elucidation of ... - ACS Publications

Subscriber access provided by University of Newcastle, Australia

Article

Active Site and Electronic Structure Elucidation of Pt Nanoparticles Supported on Phase-pure Molybdenum Carbide Nanotubes Shuai Tan, Lucun Wang, Shibely Saha, Rebecca Fushimi, and Dongmei (Katie) Li ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b01217 • Publication Date (Web): 06 Mar 2017 Downloaded from http://pubs.acs.org on March 6, 2017

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 free 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 accessible to all readers and 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.

ACS Applied Materials & Interfaces 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 26

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

ACS Applied Materials & Interfaces

Active Site and Electronic Structure Elucidation of Pt Nanoparticles Supported on Phase-pure Molybdenum Carbide Nanotubes Shuai Tan a,c, Lucun Wang b,c, Shibely Saha a, Rebecca R Fushimi b,c, Dongmei Li a,c,* a

Department of Chemical Engineering, University of Wyoming, Laramie, WY 82071, USA

b

Biological and Chemical Processing Department, Energy and Environment Science and

Technology, Idaho National Laboratory, Idaho Falls, ID, USA 83415 c

Center for Advanced Energy Studies, Idaho Falls, ID, USA 83401.

ABSTRACT: We recently showed that phase-pure molybdenum carbide nanotubes can be durable supports for platinum (Pt) nanoparticles in hydrogen evolution reaction (HER). In this paper we further characterize surface properties of the same Pt/β-Mo2C catalyst platform using carbon monoxide (CO)-Pt and CO-Mo2C bond strength of different Pt particle sizes in the < 3 nm range. Results from diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) and temporal analysis of products (TAP) revealed the existence of different active sites as Pt particle size increases. Correlation between the resultant catalyst activity and deposited Pt particle size was further investigated using water-gas-shift (WGS) as a probe reaction, suggesting that precise control of particle diameter and thickness is needed for optimized catalytic activity.

ACS Paragon Plus Environment

1


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 2 of 26

KEYWORDS: platinum, molybdenum carbide, atomic layer deposition, strain-ligand effect, active sites, electronic structure 1. INTRODUCTION Early work of Levy and Boudart1 indicated the platinum (Pt)-like properties of transition metal carbides (TMCs), resulting from their similarity to Pt in geometric and electronic structures. Recent work has focused on replacing Pt with TMCs in electrochemical devices, such as PEMFC and eletrolyzers.2–5 Catalytic properties of TMCs, especially molybdenum carbide (Mo2C) and Pt/Mo2C, for low temperature water gas shift (WGS) have also been investigated.6–11 Schweitzer et al. reported the superiority of Pt/Mo2C catalyst over the oxide supported catalysts and predicted that the WGS reaction occurred near the Pt/Mo2C interface, followed by computational simulation.11 Using both ex situ and in situ x-ray absorption (XAS), Sabins et al found that the synergy between Pt-Mo alloy nanoparticle and Mo2C was the cause for the enhanced Pt/Mo2C activity.8,9 Despite the comprehensive investigation on the origin of the enhanced catalytic activity of supported noble metal catalysts relative to bare Mo2C support, there still lacks study on how the size of supported noble metal, such as Pt, affects resultant metal/carbide catalyst activity when: 1) both the TMC support and deposited metal particles are in the nano-scale regime; 2) the carbide support has pristine phase purity. Chemical properties of supported metals are attributed to two effects: ligand and strain effect.12,13 Ligand effect results from orbital interaction between a deposited metal and its support, inducing electronic charge transfer at the interface, while strain effect results from lattice mismatch. These effects often result in d-band center shift and, consequently, changes in metal-adsorbate bond strength and catalytic reactivity of resultant catalysts.14–16 Here we report a unique Pt/Mo2C catalyst platform where phase-pure Mo2C nanotube was used as a support, with


2

Page 3 of 26

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


Pt deposited onto the support in a rotary atomic layer deposition (ALD) reactor, providing atomic level size control.2 This level size control has only been achieved on oxide supports17–19 until recently.20 In contrast to the mesoporous Mo2C structures used in open literature,8 which makes detailed imaging study difficult. In contrast, the phase-pure β-Mo2C nanotube in our study provides much-needed detail of surface characterizations, such as Pt lattice change as Pt particle size grows from atomic to 2-3 nm.20 Specifically, 15, 50 and 100 cycles of Pt deposition were chosen based on pioneering work by Hsu et al,21 with resultant samples being referred to as 15, 50 and 100 Pt/Mo2C, respectively. However, we like to point out that 15 Pt/Mo2C samples generated noisy and non-repeatable results.20 We are currently investigating surface properties of this sample via X-ray absorption fine structure spectroscopy (XAFS)22 combined with DRIFTS23 to further our understanding of this sample. As such, this paper focuses on properties of 50 and 100 Pt/Mo2C samples. By characterizing CO adsorption onto and desorption from the Pt/Mo2C nanotube surface via infrared Fourier transform spectroscopy (DRIFTS) and temporal analysis of products (TAP), we report our experimental findings that active sites transitioned as Pt particle size increased from 2 to 2.7 nm, which resulted in different working temperature ranges for WGS reaction. These surface properties further shed light on the synergetic effect between Pt and transition metal carbides. 2. EXPERIMENTAL SECTION 2.1. Materials Synthesis. Mo2C nanotubes were synthesized via a salt flux reaction as reported by our previous publication.20 Sodium fluoride, multi-walled carbon nanotube (MWNT) and molybdenum powder were all purchased from Sigma-Aldrich. Synthesized Mo2C powder was dried and stored in vials and left in air for days before each use. To deposit Pt onto Mo2C nanotubes,

a

rotating

ALD

reactor

was

used,24

with

Platinum

(IV)

Trimethyl


3


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 4 of 26

(methylcyclopentadienyl) (MeCpPtMe3) as and oxygen as precursors. The deposition temperature was maintained at 200oC. 2.2. Materials Characterizations. Pt mass loading was measured using PerkinElmar atomic absorption spectroscopy (AAS) instrument and inductively coupled plasma optical emission spectrometry (ICP-OES) techniques when it is detectable. X-ray diffraction (XRD) patterns of Pt/Mo2C catalysts were obtained using a Rigaku Model RU300 instrument with a rotating anode generator and a monochromatic detector. Cu Kα radiation was used, with a power setting of 40 kV and 40 mA. X-ray photoelectron spectroscopy (XPS) pattern of Pt/Mo2C catalysts was collected from Kratos Axis Ultra Delay Line Detector (DLD) equipped with a monochromatic Al-Kα x-ray source, where both a hemispherical analyzer and spherical mirror analyzer were used. Pt particle morphology and size distribution on Mo2C was characterized under bright field (BF) mode with an FEI Tecnai F2 G20 S-Twin TEM with an accelerating voltage of 200 kV. Brunauer, Emmett and Teller (BET) surface area measurement of Pt/Mo2C mesopores (2-50 nm) was carried out using single point N2 adsorption/desorption cycles at 75.4 K (Micromeritics ASAP 2020). Additionally, Dubinin-Radushkevich (DR) method was used to characterize micropore surface area (< 2 nm) by CO2 on Pt/Mo2C at 273 K. Pt dispersion on Mo2C nanotube was measured under hydrogen (H2) chemisorption process. It should be noted that bare Mo2C was characterized under the same condition and no H2 adsorption was observed from bare Mo2C surface, confirming that Pt atoms are the only H2 absorption sites. 2.3. Characterization of CO Bond Strength on Pt. Nicolet iS50 FT-IR spectrometer (Thermo Scientific) with iS50 ABX automated beamsplitter exchanger (DRIFTS) were used to investigate CO bond strength with Pt/Mo2C catalyst surfaces. The samples were loaded into the high temperature reaction chamber with ZnSe windows. Typically, 256 scans were collected


4

Page 5 of 26

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


with 4 cm-1 step size in the wavelength range of 1500-2500 cm-1. The samples were pretreated in the furnace by flowing 30 ml/min 15% CH4/H2 over 4 hours at 590 oC and then cooled down to 25 oC while purging N2 with flow rate of 30 ml/min. The pretreated samples were sequentially mixed with potassium bromide (KBr, non-absorbing matrix) to dilute samples to 1% samples (by mass). After loading samples into the DRIFTS chamber, the chamber temperature was maintained at 100 oC for over 30 minutes to remove the moisture in the samples. Sequentially, the chamber temperature was cooled down to room temperature in order to collect background spectrum. 20% CO (balanced in N2) was then introduced into the chamber at the flow rate of 30 ml/min for over 30 minutes to ensure that CO gas molecules were saturated on catalyst surfaces. Next, helium (He) was purged through the chamber at flow rate of 30 ml/min to remove the gas phase CO till no signals were detected in the downstream by mass spectrometry. The chamber was maintained under He for extended duration. Finally, the temperature was quickly ramped to each desired temperature under He flow. FT-IR spectra were then collected at 1, 5, 10, 15, 20 and 25 minutes till the spectra were stabilized for each temperature. 2.4. Temporal Analysis of Products (TAP). The pulse experiments were carried out in a commercial TAP-2 reactor.25 In a typical pulse response experiment a narrow pulse of reactant gas (ca. 10-8 mol/pulse, pulse width 0.135 ms) was injected into a packed-bed microreactor using a high-speed, magnetically controlled pulse valve. The microreactor was a stainless steel tubular reactor with a length of 43.7 mm and a diameter of 6.35 mm, housing an internal thermocouple. The microreactor sat on top of a high-throughput vacuum chamber (10-8 Torr) containing a RGA200 mass spectrometer, which was used to detect the effluent from the reactor. The catalyst was sandwiched in a thin-zone configuration in the middle of the microreactor packed among inert silica particles with diameters between 250 and 300 µm. Pulse sizes were chosen to ensure


5


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 6 of 26

Knudsen flow throughout the reactor. Following reduction in 15% CH4/H2 flow at 590oC, the samples were exposed to ambient conditions and, sequentially, loaded in a thin-zone configuration26 in the TAP reactor and the pressure was reduced to 10-8 torr. The freshly pretreated sample (Mo2C, 50 ALD Pt/Mo2C, and 100 ALD Pt/Mo2C) was heated from room temperature (25°C) to 400 °C at a ramping rate of 20 °C/min under vacuum and then cooled down to room temperature. This experiment was repeated to ensure that the surface is clean. Then, the catalyst was exposed to 5000 pulses of CO/Ar (CO:Ar =1:1) at 35.0 psi and room temperature to saturate the catalyst surface with adsorbed CO. 2.5. Water-Gas-Shift (WGS) Measurement. Catalysts were loaded into a quartz tube (0.9 mm OD, 0.7 mm ID and 20 cm long), supported on a quartz wool plug. The Pt/Mo2C samples were pretreated by flowing a gas mixture of 15% CH4 in H2 at a flow rate of 150 ml/min and temperature of 590 oC for 4 hours to reduce the Mo2C surface that might be oxidized in air. Following the pretreatment, the catalysts were exposed to the feed gas (12.5% CO-37.5 H2O50% N2) at 15000 h-1 GHSV under atmospheric pressure. Deionized water was fed to the system using a peristaltic pump and mixed with the dry gas stream in a vaporizer maintained at 150°C. The CO, CO2, and N2 concentrations of the effluent stream were analyzed using an in-line Cirrus 2 MKS mass spectroscopy (MS) that is equipped with a CO sensor. Water steam in the outlet was cooled by flowing through a condenser filled with ice water and then collected in a cylindrical desiccator filled with silica gel. 3. RESULTS AND DISCUSSION Table.1 summarizes Pt mass loading, dispersion, BET surface area and Pt NP sizes. We note that mass loading of Pt in 15 Pt/Mo2C sample was too small to be detected using ICP-OES. Pt particle size measured by H2 chemisorption increased from 0.2-0.3 nm (15 Pt/Mo2C) to ~2 nm


6

Page 7 of 26

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


(50 Pt/Mo2C), and to ~2.7 nm (100 Pt/Mo2C) as the number of ALD cycles increased from 15 to 50 and 100, respectively, which is consistent with size measurement using transmission electron microscopy (TEM) images (Fig.1 and Fig.S1). It should be noted that no H2 adsorption was observed from bare Mo2C samples, confirming that Pt atoms/clusters are the only H2 absorption sites. Different from the raft-like Pt particles supported on Mo2C reported by Schweitzer et al,11 both TEM images (Fig.1 and Fig.S1) and Pt dispersion (Table.1) suggest that NPs of 50 Pt/Mo2C have smaller diameter (2 nm) and are much thinner (~1 nm) than those in 100 Pt/Mo2C (2.7 nm, thickness: ~2 nm). In other words, as the Pt particles grow in diameter with increasing ALD cycles, their thickness also increases. While TEM images qualitatively revealed this trend, the higher dispersion in 50 Pt/Mo2C (57%) than 100 Pt/Mo2C (41%) measured via H2 chemisorption quantitatively proved this observation (Table.1). Table 1. Characterization of Mo2C and Pt/Mo2C Catalyst

Pt Loading (wt%) a

Dispersion b (%)

Pt Particle Size (nm) c

N2 surface area (m2/g)

CO2 surface area (m2/g)

Mo2C

0

0

~

30

109

15 Pt/Mo2C

~

~

~0.2-0.3

29

108

50 Pt/Mo2C

1.07

57

~2

27

101

100 Pt/Mo2C

4.4

41

~2.7

26

66

a

Measured by ICP-MS. Measured and calculated from H2 chemisorption. c Measured from H2 chemisorption. b


7


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 8 of 26

Figure 1. HRTEM images and corresponding size distribution of Pt particles in: (a, b) 15 Pt/Mo2C, (c, d) 50 Pt/Mo2C and (e, f) 100 Pt/Mo2C. Our previous XPS study of the 15, 50 and 100 Pt/Mo2C samples demonstrated different binding energy (BE) shifts from Pt0 due to electron transfer between Pt atom and its support (ligand effect),12 with the volcano profile of BE shift vs. the number of ALD cycles indicating that 50 Pt/Mo2C samples have the largest BE shift.20 We note that there is no noticeable BE shift from local charging based on C1s signals, further confirming that the BE shift in all three samples resulted from the interaction between Pt and Mo2C (Fig.S2). On the other hand, strain effect from the lattice mismatch between supported metal catalyst particles and its support was also reported using lattice spacing analysis,20 showing that Pt (111) surface facet in smaller Pt particles for 50 Pt/Mo2C were stretched evenly along [110] and [200] directions (Fig.2).27 The distortion of Pt (111) surface facet caused by strong interaction between Pt and Mo2C support is


8

Page 9 of 26

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


consistent with findings reported by Daio et al.28 It was also noted that bigger particles in 100 Pt/Mo2C exhibiting uneven lattice expansion from the center to the edge.20 We postulate that particles with smaller diameters in 50 Pt/Mo2C are also thinner, resulting in more Pt atoms to be either in direct contact with or in close proximity with Mo2C lattice. On the other hand, thicker particles from 100 Pt/Mo2C samples possess enough Pt atoms on the top or in the proximity of top atoms that experience the least of strain-ligand effect, resulting in the smallest BE shift and least lattice spacing expansion at the top center. The strong coupling interaction between Pt particles and Mo2C nanotube support revealed by XPS and TEM results is consistent with what was reported in open literature.3,29–31 It should be noted that there is noticeable, but noisy Pt (111) peaks in 50 and 100 Pt/Mo2C samples, relevant to bare Mo2C XRD patterns (Fig.S3), due to the low Pt loading (3 nm.34


10

Page 11 of 26

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


Figure 3. DRIFTS results of: a) bare Mo2C, b) 50 Pt/Mo2C and c) 100 Pt/Mo2C under different helium purging time and temperatures. We note that there is a lack of CO-Ptδ+ signals from 100 Pt/Mo2C, compared to 50 Pt/Mo2C, indicating that thicker Pt particles have substantially less Pt atoms impacted by strain-ligand effect, since Ptδ+ sites are the results of strong ligand-stain effect. In other words, the higher signal intensity from CO-Ptδ+ than CO-Pt0 in 50 Pt/Mo2C indicate that CO-Ptδ+ sites dominated the electronic nature of Pt NPs, as depicted in Fig.4a (left), which is consistent with the fact that 50 Pt/Mo2C possesses the highest Pt dispersion (57%) summarized in Table.1 and largest BE shift (0.70 eV) among all the Pt/Mo2C samples.20 In contrast, dominating amount of Pt sites in 100 Pt/Mo2C samples were not affected significantly by strain-ligand effect as the particles grow thicker, except for those along the peripheral interface and in the proximity with Mo2C substrate (Fig.4a, right). Consequently, CO-Pt0 signals were much stronger for 100 Pt/Mo2C than for 50 Pt/Mo2C, consistent with the much lower BE shift from Pt0 (0.22 eV) and significantly lower dispersion (41%).

Figure 4. Schematic of CO desorption process for: 250 oC (a), and above 300 oC (b), for 50 and 100 Pt/Mo2C. CO bond strength to Mo2C sites provides another crucial indicator of surface properties. For 100 Pt/Mo2C (Fig.3c), the intensities of CO-Mo2C peaks decreased with longer He purging time


11


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 12 of 26

at 150oC, indicating the removal of CO molecules from Mo2C surface. This peak remained constant for over 20 minutes for both 50 and 100 Pt/Mo2C till temperature increases to 200oC. However, these two samples responded differently with temperature increase. For 100 Pt/Mo2C, CO molecules adsorbed on Mo2C sites were removed substantially at 200oC, evidenced by the disappearance of CO-Mo2C peaks (Fig.3c). The intensity of CO-Mo2C peaks in 50 Pt/Mo2C sample (Fig.3b) only had a slight decrease when temperature increased to 200-250oC and did not disappear until temperature increased to 300oC. We postulate that the strong strain-ligand effect in the smaller and thinner 50 Pt/Mo2C NPs may have affected the electronic structure Mo2C sites, possibly forming Pt-Mo alloy.8 Similar trend with temperature was also observed for CO-Pt0 peaks. For 100 Pt/Mo2C, intensity of the linear CO-Pt0 peaks remained constant at 200oC, but significantly decreased when temperature increased to 250oC and above (Fig.3c), indicating desorption of CO molecules from Pt0 sites, as depicted in Fig.4a (right). It was also worth noting that these two peaks redshifted to lower frequency with He purging time and temperature increase (dashed lines in Fig. 3c). This was attributed to dipole-dipole coupling change between CO and Pt atom as CO coverage decreased.19,35 The signal of CO-Pt0 for 50 Pt/Mo2C is fairly weak, following a similar trend to 100 Pt/Mo2C with He purging time. When temperature increased to 200oC, the signal further weakened to the noise level of DRIFTS. However, the dominating CO-Ptδ+ signal from 50 Pt/Mo2C remained at 300oC and became sharper (Fig.3c), suggesting that CO desorption temperature from Ptδ+ sites is above 300oC (Fig.4b, left). We like to point out that our DRIFTS observation on the size effect on CO-Pt bond strength is consistent with what Lentz et al reported with their DFT calculation.34 The differences in the commercial catalyst (EuroPt-1) for their DRIFTS study and our Pt/Mo2C platform are twofold:


12

Page 13 of 26

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


1) EuroPt-1 has wider size distribution (0.9-3.5 nm, centered at 1.8 nm), while our two Pt/Mo2C samples have much narrower size distribution with 50 Pt/Mo2C ranging between 1-2.4 nm (centered at 2 nm), 100 Pt/Mo2C varying between 2-3.2 nm (centered at 2.7 nm); 2) EuroPt-1 uses inert SiO2 as support. Evidently, the narrower size distribution range of our samples with different median sizes provides additional experimental evidence to what was calculated in their DFT work. The electronically conductive Mo2C support, resulting in CO-Ptδ+ signal in our DRIFTS data, contributes to the existing catalyst platform repertoire in the nanoscale range for future theoretical study and experimental validation. To correlate the CO-Pt0 and CO-Ptδ+ bond strength with catalytic activity, temporal analysis of products (TAP) reactor system25 was used to carefully examine the effect of CO adsorption/desorption on CO oxidation step (CO->CO2) by temperature programmed CO pulse response experiments. One advantage of transient experiments in vacuum is the higher time resolution that enables observation of subtle processes that could be missed in DRIFTS experiment. Following the removal of surface species during pretreatment process, the samples were exposed to CO pulses at room temperature first.

Figure 5. CO retention and CO2 fractional yield when CO is pulsed at room temperature in vacuum.


13


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 14 of 26

High CO retention, shown in Fig.5, was detected for all samples with the greatest per pulse reaching 90% for the 50 Pt/Mo2C sample. The CO molecules must be strongly bound to the surface because no gas phase products were detected. This enhanced CO interaction supports the strain-ligand effect hypothesis and DRIFTS results (Fig.3). CO also interacts with the pristine Mo2C and 100 Pt/Mo2C sample at room temperature. However, for these materials CO2 production was observed (Fig.5). Particularly for the bare Mo2C material, a CO2 fractional yield near 40% was detected. The carbon balance shows significant CO retention on the surface for all materials. The pronounced CO2 production at room temperature from Mo2C and 100 Pt/Mo2C can be attributed to the forward Boudouard reaction (2CO → C + CO2). Pulse response experiments using CO isotopes are needed to definitively rule out participation of the surface species. However, the participation of surface species is considered unlikely since sustained CO2 production, as shown in Fig.6, was observed for all samples as pulses continued with temperature being slowly increased at the same time. When observing pulse response shape such as those in Fig.6, the most obvious point of reference is the peak height. Although the CO2 production rapidly drops for the Mo2C sample (Fig.6a) in the initial pulses, indicating adsorption site poisoning through carbon accumulation, it remains sustained throughout the experiment. The carbon mass balance suggests a significant amount of the pulsed CO is not detected in the gas phase products and remains on the surface.


14

Page 15 of 26

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


Figure 6. CO2 product pulse response curves with CO being pulsed while temperature slowly (1 °C/min) increasing for different samples: (a) Mo2C, (b) 50 Pt/Mo2C, (c) 100 Pt/Mo2C. Only every 10th pulse is shown for clarity. An alternative presentation of the CO and CO2 pulse response data is presented in Fig.7, with the CO retention in each pulse and the CO2 yield being indicated. All samples indicate an initial drop in CO retention when temperature increased from room temperature to 50oC, which could be attributed to site poisoning by C accumulation from the forward Boudouard reaction. As the temperature increases, however, there is a turning point (Fig.7a) where CO retention begins to increase (near 50oC for Mo2C and 50 Pt/Mo2C, near 100oC for 100 Pt/Mo2C). The CO2 production for the Mo2C sample continuously drops as the temperature is raised and plateaued around 150oC. The 50 Pt/Mo2C sample maintains CO retention (Fig. 7a) but does not indicate CO2 production until temperatures greater than 150oC are reached (Fig.7b). At 200oC, the CO retention is still high while CO2 production is low indicating that CO is strongly bound to the surface in corroboration with the CO-Ptδ+ peak for 50 Pt/Mo2C in DRIFTS data (Fig. 3b). The


15


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 26

fact that bare Mo2C support and the 50 Pt/Mo2C catalyst does not behave the same for CO2 production suggests that the ALD deposited Pt NPs in 50 Pt/Mo2C may have changed the electronic structure of the Mo2C surface. The 100 Pt/Mo2C sample indicates unique behavior for temperature dependent CO2 production with a low temperature regime (100-275oC) transitioning to a higher temperature regime (>275oC) where CO2 yield is pronounced (Fig.6c and Fig.7b). This indicates the presence of distinct routes to CO2 formation through different active sites, consistent with different BE shift in 50 and 100 Pt/Mo2C and DRIFTS results (Fig.3).

Figure 7. CO retention (a), and CO2 fractional yield (b), for bare Mo2C, 50 and 100 Pt/Mo2C samples. The particle size effect on catalytic activity was also investigated under WGS conditions using catalysts loading of 400 (Fig.8a) and 100 mg (Fig.S5). Both 50 and 100 Pt/Mo2C showed significantly enhanced CO conversion, compared to bare Mo2C nanotubes. 50 Pt/Mo2C generally required higher working temperature, which is very pronounced in the temperature range of 200350oC and reflected by a slightly higher apparent activation energy (Ea) (55.8 kJ/mol) than that of 100 Pt/Mo2C (53.6 kJ/mol) (Fig.8c). We attribute this to the fact that 50 Pt/Mo2C has more substantial amount of Pt atoms affected by strain-ligand effect than 100 Pt/Mo2C. The stronger bonding of CO-Ptδ+ required higher temperature to activate the Ptδ+ sites, which resulted in lower WGS rate in 50 Pt/Mo2C than in 100 Pt/Mo2C for low temperature range (Fig.8b). However,


16

Page 17 of 26

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


when temperature increased above 300oC, 50 Pt/Mo2C showed ~2.5 times higher WGS rate (0.82 mol CO/molPt·s) than 100 Pt/Mo2C (0.35 mol CO/molPt·s) (Fig.8b). For the same reaction conditions, similar WGS rate trend was observed for lower catalyst loading (100 mg, Fig.S5).

Figure 8. The CO conversion (a), WGS rate (mol CO/molPt·s) (b), and apparent activation energy (∆Eapp, kJ/mol) (c), of Pt/Mo2C catalysts for WGS reaction. Feed gas: 12.5% CO-37.5% H2O-50% N2, 15000 h-1 GHSV). 4. CONCLUSION To summarize, we demonstrated that Pt NPs deposited on phase-pure Mo2C nanotubes can be used as an experimental platform for elucidating active sites and electronic structure of both Pt


17


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 18 of 26

and Mo2C. The Pt particle size change in the range of < 3 nm not only resulted in different BE shift from Pt0, but may also have changed the electronic structure of Mo2C sites, suggested by different CO-Mo2C bond strengths between 50 and 100 Pt/Mo2C materials. The biggest BE shift from Pt0 for 50 Pt/Mo2C materials was further corroborated by evident CO-Ptδ+ signal in DRIFTS and much higher CO2 production temperature in TAP results. WGS rates suggest that there may be an optimal particle perimeter and thickness that balances working temperature and activity. Future study with X-ray absorption fine structure spectroscopy (XAFS)22 combined with DRIFTS23 will facilitate further probing of the interface between Pt NP and TMC support with the reported Pt/Mo2C catalyst platform.

ASSOCIATED CONTENT Supporting Information The detailed catalytic characterizations, calculations for WGS reaction and XPS patterns were included in the Electronic Supporting Information (ESI). This material is available free of charge via the Internet at http://pubs.acs.org. AUTHOR INFORMATION Corresponding Author Dongmei Li, Department of Chemical Engineering, University of Wyoming, Laramie, WY, 82072. [email protected]. ACKNOWLEDGMENT


18

Page 19 of 26

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


We are grateful to Professors Joseph Holles, Maohong Fan, Brian Leonard and Dr. Erwin Sabio for their assistance with BET/chemisorption, DRIFTS, Mo2C synthesis and HRTEM/XPS experiments. We thank the Office of Research and Economic Development, School of Energy Research (SER) and NASA EPSCoR at the University of Wyoming for their generous financial support. We acknowledge of support of the U.S. Department of Energy, under DOE Idaho Operations Office Contract DE-AC07-05ID14517. Accordingly, the U.S. Government retains a nonexclusive, royalty-free license to publish or reproduce the published form of this contribution, or allow others to do so, for U.S. Government purposes.


19


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 20 of 26

REFERENCES (1)

Levy, R. B.; Boudart, M. Platinum-Like Behavior of Tungsten Carbide in Surface Catalysis. Science. 1973, 181 (4099), 547–549.

(2)

Wan, C.; Regmi, Y. N.; Leonard, B. M. Multiple Phases of Molybdenum Carbide as Electrocatalysts for the Hydrogen Evolution Reaction. Angew. Chemie Int. Ed. 2014, 53 (25), 6407–6410.

(3)

Zhang, K.; Yang, W.; Ma, C.; Wang, Y.; Sun, C.; Chen, Y.; Duchesne, P.; Zhou, J.; Wang, J.; Hu, Y.; Banis, M. N.; Zhang, P.; Li, F.; Li, J.; Chen, L. A Highly Active, Stable and Synergistic Pt nanoparticles/Mo2C Nanotube Catalyst for Methanol ElectroOxidation. NPG Asia Mater. 2015, 7 (1), e153.

(4)

Liu, Y.; Kelly, T. G.; Chen, J. G.; Mustain, W. E. Metal Carbides as Alternative Electrocatalyst Supports. ACS Catal. 2013, 3 (6), 1184–1194.

(5)

Hunt, S. T.; Milina, M.; Alba-rubio, A. C.; Hendon, C. H.; Dumesic, J. A.; Románleshkov, Y. Self-Assembly of Noble Metal Monolayers on Transition Metal Carbide Nanoparticle Catalysts. Science. 2016, 352 (6288), 974–978.

(6)

Ma, Y.; Guan, G.; Hao, X.; Zuo, Z.; Huang, W.; Phanthong, P.; Li, X.; Kusakabe, K.; Abudula, A. Embedded Structure Catalyst: A New Perspective from Noble Metal Supported on Molybdenum Carbide. RSC Adv. 2015, 5 (20), 15002–15005.

(7)

Patt, J.; Moon, D. J.; Phillips, C.; Thompson, L. Molybdenum Carbide Catalysts for Water – Gas Shift. Catal. Letters 2000, 65, 193–195.

(8)

Sabnis, K. D.; Akatay, M. C.; Cui, Y.; Sollberger, F. G.; Stach, E. A.; Miller, J. T.;


20

Page 21 of 26

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


Delgass, W. N.; Ribeiro, F. H. Probing the Active Sites for Water–gas Shift over Pt/molybdenum Carbide Using Multi-Walled Carbon Nanotubes. J. Catal. 2015, 330, 442–451. (9)

Sabnis, K. D.; Cui, Y.; Akatay, M. C.; Shekhar, M.; Lee, W.-S.; Miller, J. T.; Delgass, W. N.; Ribeiro, F. H. Water–gas Shift Catalysis over Transition Metals Supported on Molybdenum Carbide. J. Catal. 2015, 331, 162–171.

(10)

Schaidle, J. a.; Lausche, A. C.; Thompson, L. T. Effects of Sulfur on Mo2C and Pt/Mo2C Catalysts: Water Gas Shift Reaction. J. Catal. 2010, 272 (2), 235–245.

(11)

Schweitzer, N. M.; Schaidle, J. a; Ezekoye, O. K.; Pan, X.; Linic, S.; Thompson, L. T. High Activity Carbide Supported Catalysts for Water Gas Shift. J. Am. Chem. Soc. 2011, 133, 2378–2381.

(12)

Kitchin, J. R.; Nørskov, J. K.; Barteau, M. a.; Chen, J. G. Role of Strain and Ligand Effects in the Modification of the Electronic and Chemical Properties of Bimetallic Surfaces. Phys. Rev. Lett. 2004, 93 (15), 4–7.

(13)

Strasser, P.; Koh, S.; Anniyev, T.; Greeley, J.; More, K.; Yu, C.; Liu, Z.; Kaya, S.; Nordlund, D.; Ogasawara, H.; Toney, M. F.; Nilsson, A. Lattice-Strain Control of the Activity in Dealloyed Core–shell Fuel Cell Catalysts. Nat. Chem. 2010, 2 (6), 454–460.

(14)

Jia, Q.; Caldwell, K.; Strickland, K.; Ziegelbauer, J. M.; Liu, Z.; Yu, Z.; Ramaker, D. E.; Mukerjee, S. Improved Oxygen Reduction Activity and Durability of Dealloyed PtCo X Catalysts for Proton Exchange Membrane Fuel Cells: Strain, Ligand, and Particle Size Effects. ACS Catal. 2015, 5 (111), 176.


21


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

(15)

Page 22 of 26

Friebel, D.; Viswanathan, V.; Miller, D. J.; Anniyev, T.; Ogasawara, H.; Larsen, A. H.; Ogrady, C. P.; Nørskov, J. K.; Nilsson, A. Balance of Nanostructure and Bimetallic Interactions in Pt Model Fuel Cell Catalysts: In Situ XAS and DFT Study. J. Am. Chem. Soc. 2012, 134 (23), 9664–9671.

(16)

Xie, S.; Choi, S. Il; Lu, N.; Roling, L. T.; Herron, J. a.; Zhang, L.; Park, J.; Wang, J.; Kim, M. J.; Xie, Z.; Mavrikakis, M.; Xia, Y. Atomic Layer-by-Layer Deposition of Pt on Pd Nanocubes for Catalysts with Enhanced Activity and Durability toward Oxygen Reduction. Nano Lett. 2014, 14 (6), 3570–3576.

(17)

Peterson, E. J.; DeLaRiva, A. T.; Lin, S.; Johnson, R. S.; Guo, H.; Miller, J. T.; Hun Kwak, J.; Peden, C. H. F.; Kiefer, B.; Allard, L. F.; Ribeiro, F. H.; Datye, A. K. LowTemperature Carbon Monoxide Oxidation Catalysed by Regenerable Atomically Dispersed Palladium on Alumina. Nat. Commun. 2014, p 4885.

(18)

Setthapun, W.; Williams, W. D.; Kim, S. M.; Feng, H.; Elam, J. W.; Rabuffetti, F. a.; Poeppelmeier, K. R.; Stair, P. C.; Stach, E. a.; Ribeiro, F. H.; Miller, J. T.; Marshall, C. L. Genesis and Evolution of Surface Species during Pt Atomic Layer Deposition on Oxide Supports Characterized by in Situ XAFS Analysis and Water-Gas Shift Reaction. J. Phys. Chem. C 2010, 114 (21), 9758–9771.

(19)

Ding, K.; Gulec, A.; Johnson, A. M.; Schweitzer, N. M.; Stucky, G. D.; Marks, L. D.; Stair, P. C. Identification of Active Sites in CO Oxidation and Water-Gas Shift over Supported Pt Catalysts. Science. 2015, 350 (6257), 1688–1690.

(20)

Saha, S.; Martin, B.; Leonard, B.; Li, D. Probing Synergetic Effects between Platinum Nanoparticles Deposited via Atomic Layer Deposition and a Molybdenum Carbide


22

Page 23 of 26

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


Nanotube Support. J. Mater. Chem. A Mater. energy Sustain. 2016, 4, 9253–9265. (21)

Hsu, I. J.; Hansgen, D. A.; Mccandless, B. E.; Willis, B. G.; Chen, J. G. Atomic Layer Deposition of Pt on Tungsten Monocarbide (WC) for the Oxygen Reduction Reaction. J. Phys. Chem. C 2011, 115 (9), 3709–3715.

(22)

Becknell, N.; Kang, Y.; Chen, C.; Resasco, J.; Kornienko, N.; Guo, J.; Markovic, N. M.; Somorjai, G. A.; Stamenkovic, V. R.; Yang, P. Atomic Structure of Pt3Ni Nanoframe Electrocatalysts by in Situ X ‑ Ray Absorption Spectroscopy. J. Am. Chem. Soc. 2015, 137, 15817–15824.

(23)

Brieger, C.; Melke, J.; Bosch, N. Van Der; Reinholz, U.; Riesemeier, H.; Guilherme, A.; Krishna, M.; Derr, I.; Schökel, A.; Roth, C. A Combined in-Situ XAS – DRIFTS Study Unraveling Adsorbate Induced Changes on the Pt Nanoparticle Structure. J. Catal. 2016, 339, 57–67.

(24)

McCormick, J. A.; Cloutier, B. L.; Weimer, A. W.; George, S. M. Rotary Reactor for Atomic Layer Deposition on Large Quantities of Nanoparticles. J. Vac. Sci. Technol. A Vacuum, Surfaces, Film. 2007, 25 (1), 67.

(25)

Gleavesa, J. T.; Yablonskii, G. S.; Phanawadee, P.; Schuurmanb, Y. TAP-2 : An Interrogative Kinetics Approach. Appl. Catal. A Gen. 1997, 160, 55–88.

(26)

Shekhtman, S. O.; Yablonsky, G. S.; Chen, S.; Gleaves, J. T. Thin-Zone TAP-ReactorTheory and Application. Chem. Eng. Sci. 1999, 54, 4371–4378.

(27)

Zhong, X.; Feng, Y.; Lieberwirth, I.; Wolfgang, K. Facile Synthesis of MorphologyControlled Platinum Nanocrystals. Chem. Mater. 2006, 18 (10), 2468–2471.


23


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

(28)

Page 24 of 26

Daio, T.; Staykov, A.; Guo, L.; Liu, J.; Tanaka, M.; Matthew Lyth, S.; Sasaki, K. Lattice Strain Mapping of Platinum Nanoparticles on Carbon and SnO2 Supports. Sci. Rep. 2015, 5, 13126.

(29)

He, C.; Tao, J.; He, G.; Shen, P. K.; Qiu, Y. Small Size Mo2C Nanocrystal Coupled with Reduced Graphene Oxide Enhance the Electrochemical Activity of Palladium Nanoparticles towards Methanol Oxidation Reaction. Catal. Sci. Technol. 2016, 6 (19), 7316–7322.

(30)

He, C.; Tao, J.; Ke, Y.; Qiu, Y. Graphene-Supported Small Tungsten Carbide Nanocrystals Promoting a Pd Catalyst towards Formic Acid Oxidation. Rsc Adv. 2015, 5 (82), 66695–66703.

(31)

Yan, Z.; Wang, H.; Zhang, M.; Jiang, Z.; Jiang, T.; Xie, J. Pt Supported on Mo2C Particles with Synergistic Effect and Strong Interaction Force for Methanol Electro-Oxidation. Electrochim. Acta 2013, 95, 218–224.

(32)

Green, I. X.; Tang, W.; Neurock, M.; Yates Jr., J. T. Spectroscopic Observation of Dual Catalytic Sites During Oxidation of CO on a Au/TiO2 Catalyst. Science. 2011, 333, 736– 740.

(33)

Wang, J.; Kispersky, V. F.; Nicholas Delgass, W.; Ribeiro, F. H. Determination of the Au Active Site and Surface Active Species via Operando Transmission FTIR and Isotopic Transient Experiments on 2.3wt.% Au/TiO2 for the WGS Reaction. J. Catal. 2012, 289, 171–178.

(34)

Lentz, C.; Jand, S. P.; Melke, J.; Roth, C.; Kaghazchi, P. DRIFTS Study of CO Adsorption on Pt Nanoparticles Supported by DFT Calculations. J. Mol. Catal. A Chem.


24

Page 25 of 26

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


2017, 426, 1–9. (35)

Crossley, A.; King, D. A. Infrared Spectra for CO Isotopes Chemisorbed on Pt {111}: Evidence for Strong Absorbate Coupling Interactions. Surf. Coatings Technol. 1977, 68, 528–538.


25


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 26 of 26

For Table of Contents (TOC) Only


26

Active Site and Electronic Structure Elucidation of ... - ACS Publications

Recommend Documents