Controlling Hydrogenation Selectivity by Size: 3-Hexyne on Supported

Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE

C: Surfaces, Interfaces, Porous Materials, and Catalysis

Controlling Hydrogenation Selectivity by Size: 3-Hexyne on Supported Pt Clusters Marian D Rötzer, Andrew S Crampton, Maximilian Krause, Katharina Thanner, Florian F Schweinberger, and Ueli Heiz J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.8b12151 • Publication Date (Web): 15 Feb 2019 Downloaded from http://pubs.acs.org on February 16, 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 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

The Journal of Physical Chemistry

Controlling Hydrogenation Selectivity by Size: 3-Hexyne on Supported Pt Clusters M. D. Rötzer,† A. S. Crampton,†,‡ M. Krause,† K. Thanner,†,¶ F. F. Schweinberger,†,§ and U. Heiz∗,† †Technische Universität München, Lehrstuhl für Physikalische Chemie, Zentralinstitut für Katalyseforschung und Fakultät für Chemie, Lichtenbergstr. 4, 85748 Garching, Germany ‡Current Address: L3 Open Water Power, 28 Dane St., Somerville, MA 02143, United States of America ¶Current Address: Helmholtz Institut Ulm (HIU), Helmholtzstr. 11, 89081 Ulm, Germany §Current Address: Roche Diagnostics GmbH, Staffelseestr. 2-8, 81477 Munich, Germany E-mail: [email protected]

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

Abstract Size-selected metal clusters supported on metal oxides have recently gained significant scientific attention due to their potential to investigate hydrogenation reactions on a fundamental level. To expand previous studies on ethylene hydrogenation, we report on the selective hydrogenation of 3-hexyne using the same model systems of Pt clusters supported on MgO, but introducing the additional parameter of reaction selectivity. Isotopically labelled temperature-programmed reaction (TPR) experiments show that the surface chemistry of 3-hexyne is dependent on cluster size and characterized by desorption of several reaction products. The latter include formation of molecules involving dehydrogenation as well as hydrogenation steps. By comparison between hydrogenation of hexyne and ethylene an atomic window is found for Pt9 , where activation barriers favor triple over double bond hydrogenation effectively leading to enhanced selectivity. The favored hydrogenation of the triple bond is caused by cluster morphology and correlated adsorption sites available for the alkyne. The interplay between cluster and adsorbed 3-hexyne leads to enhanced activation of hydrogen not observed for the bare metal clusters. This is the first experimental evidence of the potential use of supported, size-selected metal clusters for selective hydrogenation reactions.

Keywords: Platinum, Cluster, TPR, Alkyne, (De-)Hydrogenation, Dehydrocyclization, Selectivity.

2


Page 2 of 25

Page 3 of 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


Introduction Cluster-based materials with control over atomic size have been intensively studied under ultra-high vacuum (UHV) conditions due to their unique and unpredictable catalytic properties evolving by addition of single atoms . 1–4 Typically, these investigations dealt either with CO oxidation or acetylene cyclotrimerization catalyzed by late transition metals, in particular Au, Pd and Pt, supported on various metal oxides including MgO , 5–8 SiO2 /Si , 9–11 Al2 O3 12,13 and TiO2 . 14–18 In recent years the focus of cluster research has shifted towards (de)-hydrogenation reactions of unsaturated hydrocarbons . 19–23 Our group has shown that supported Pt clusters in the size regime of seven to twenty atoms show distinguishable catalytic activity as a function of size for ethylene hydrogenation . 19 For instance, Pt13 supported on MgO(100) shows higher hydrogenation activity stemming from increased intrinsic resistance towards undesired dehydrogenation channels leading to poisoning of the metal surface. This finding for cluster-based materials results in structure sensitivity, which is not observed for larger nanoparticles and bulk materials . 24 The proclivity of monodisperse clusters for (de)-hydrogenation reactions was shown to be affected by local electron densities defining unique catalytic active sites. The local electron density is not only controlled by cluster size, 19,20 but also by support material 25 and stoichiometry 26 offering additional possibilities to steer chemical reactivity. Having established an advanced understanding of ethylene hydrogenation on supported Pt clusters on both, an experimental and theoretical basis, the question arises whether these gained insights can be expanded to more complex systems involving other functional groups and alkyl-substituents. Particularly, the replacement of the double bond by a triple bond introduces the supplemental parameter of selectivity in addition to overall hydrogenation activity. In order to gain insights into these systems, the hydrogenation of 3-hexyne has been chosen as a model system for long-chained alkynes keeping cluster component (Ptn , n between 3 and 13) and MgO(100) as support material constant. This approach allows direct comparison between an unstudied reaction and well-characterized catalyst systems. 3



Page 4 of 25

The tendency to form different reaction products defining cluster selectivity has been tested by isotopically labelled temperature-programmed reaction (TPR) experiments and are compared to results of a Pt(111) single crystal.

Experimental All experiments were performed in a stainless steel ultrahigh vacuum (UHV) chamber operating at a base pressure of 2 · 10−10 mbar described in detail elsewhere. 27 A Mo(100) single crystal (MaTecK, Germany, 0.785 cm2 ) single was cleaned by electron bombardement (2000 K, 30 s) and oxidation (900 K, p(O2 )= 4 · 10−7 mbar, 10 min).

Film prepa-

ration was conducted by evaporization of magnesium in an oxygen background (p(O2 )= 5 · 10−7 mbar) at 600 K onto the clean Mo(100) surface. After film growth the sample was annealed at 800 K for 10 min in order to minimize defect sites. The resulting film has a thickness of 8-10 ML (monolayers) and has been shown to be defect-poor . 19 The cleanliness of the single crystal and the synthesized film is controlled via Auger electron spectroscopy and ultraviolet photoelectron spectroscopy (UPS). Further details about film preparation and characterization can be found in the literature . 28 The size-selected Pt clusters were synthesized using a laser ablation source. 27 In short, the second harmonic of a Nd:YAG (100 Hz, Innolas Spitlight DPSS, Germany) is focused onto a rotating Pt target (99.95 %, Alfa-Aesar) inducing a metal plasma, which is subsequently cooled by a delayed He-pulse (6.0 purity, Westfalen, Germany) and supersonic expansion into vacuum. The resulting cluster ions are guided by ion optics and an octupole towards a quadrupole bender, where only positively charged clusters are transmitted into the analysis chamber. Mass-selection down to one specific cluster size is achieved by an additional quadrupole mass filter (QMF, 16.000 amu, Extrel, USA) before deposition under soft-landing conditions (kinetic energy < 1 eV per cluster atom). The neutralization current on the 0.785 cm2 single crystal was used to determine the

4


Page 5 of 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


effective coverage assuming unit charge per cluster. The coverage was varied for better comparability of the metal loading on the support. An overview of the used coverages, the amount of particles and the total number of Pt atoms is given in table 1. After deposition the temperature of the samples was shortly increased to 300 K leading to desorption of hydrogen adsorbed from the background. In order to compare activities a Pt(111) single crystal (MaTecK, Germany) was cleaned by repeated cycles of Ar+ sputtering (≈ 10 µA, 1.5 kV, 900 K, p(Ar)= 4 · 10−6 mbar, 45 min), followed by oxidation (650 K, p(O2 )= 4 · 10−7 mbar, 10 min) and subsequent annealing in vacuo (1300 K, 1 min). Dosage of gases was performed using a calibrated molecular beam doser originally designed by Yates et al. at 100 K assuming unity sticking probability. 29,30 The purity of D2 was 100.00 % (Westfalen, Germany). 3-Hexyne was purchased from Sigma-Aldrich (purity > 99 %) and the absence of benzene and hexenes was additionally checked by GC-MS, where only additional stabilizers were found. Prior to use the alkyne was further purified by repeated freeze-pump-thaw cycles and the cleanliness was monitored in situ with mass spectroscopy. The reported dosages are given per surface atom (SA) of the MgO support assuming 2.25 · 1015 atoms/cm2 . For all experiments 0.4 /SA D2 was dosed first followed by 0.22 /SA 3-hexyne except for the dehydrogenation of 3-hexyne on supported Pt atoms, where only 0.44 /SA 3-hexyne was dosed without deuterium.

5



Table 1: The cluster coverage on the 0.785 cm2 single crystal used for different experiments: TPR experiments were performed dosing 0.4/SA D2 followed by 0.22/SA 3-hexyne, whereas TPD on Pt atoms solely used 0.44/SA 3-hexyne. Cluster Size 13

10 9 8 7 6 5 4 3 1

Particles 8.4 · 1012 4.5 · 1012 9.0 · 1012 9.0 · 1012 9.0 · 1012 9.0 · 1012 9.0 · 1012 9.0 · 1012 7.5 · 1012 1.1 · 1013 1.8 · 1013 9.0 · 1013

Total Pt Atoms 1.1 · 1014 5.9 · 1013 1.2 · 1014 9.0 · 1013 8.1 · 1013 7.2 · 1013 6.3 · 1013 5.4 · 1013 3.8 · 1013 4.5 · 1013 5.4 · 1013 9.0 · 1013

Experiment TPR CO IRRAS after deposition CO IRRAS after TPR to 300 and 600 K TPR TPR TPR TPR TPR TPR TPR TPR TPD

A quadrupole mass spectrometer (QMS, Balzers QMG 430, Liechtenstein) was used for TPR measurements, which were performed using a heating ramp of 2 K/s. The cracking patterns of 3-hexyne, 3-(cis/trans)-hexene, hexane, benzene were controlled in situ by backfilling the chamber and are in good agreement with the NIST database. The areas determined by TPR experiments were corrected using QMS sensitivity factors determined for each molecule . 19,20 In short the area of the CO monolayer on Pt(111) was determined using the TPD technique. Afterwards identical fluxes through the nozzle of the QMS were installed by applying a background pressure in the chamber and taking into account the ion gauge sensitivity. Comparison between the signal of CO and the gas of interest led to QMS sensitivity factors, which included the influence of QMS transmission and cracking patterns. Deuterated molecules were supposed to have identical cracking patterns compared to their non deuterated counterpart, which might cause a systematic error. However, the size of this error is negligible and the discussion of this work is unaffected since no quantitative comparisons are made. The following m/z ratios were used for identification: 86 for d2 -hexene, 90 for d4 -hexane, 78 for benzene and 69 for H2 -hexene (correlating to the ion CH3 CH2 CH−CHCH2+ caused 6


Page 6 of 25

Page 7 of 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


by fragmentation of hexene). The benzene trace has been baseline-corrected for the overlapping cracking pattern of physisorbed 3-hexyne at low temperatures. Since this work mainly focused on the surface chemistry of C6 molecules, the m/z ratios have been chosen to exclude overlapping cracking patterns with HD-exchanged molecules. No other traces have been measured and consequently no statement can be made about the extent of additional exchange reactions and hydrogen desorption. The presented error bars were determined from multiple measurements on a single cluster size. Infrared reflection absorption spectroscopy (IRRAS, Thermo Electron Corp. Nicolet FT6700) was performed on the Pt13 clusters after defined temperature steps. After background measurements at 100 K, approximately 10 Langmuir CO were dosed at the same temperature and the spectra were recorded. This was done for clusters as deposited and for clusters after 3-hexyne and D2 TPR to 300 and 600 K, respectively. For clusters as deposited a total of 4.5· 1012 particles was used. After TPR experiments the initial coverage was doubled to 9.0 · 1012 clusters to increase sensitivity. The experiments were conducted in single reflection mode with an external MCT-detector (Thermo Electron Corp., MCTA-TRS) using a resolution of 4 cm-1 and averaging 256 scans.

Results TPR spectra of 3-hexyne and deuterium on supported, size-selected Ptn clusters, bare MgO(100) and Pt(111) are shown in figure 1 and 2. Figure 1 is devoted to hydrogenation products from reaction with coadsorbed D2 , whereas figure 2 contains species involving either direct or indirect dehydrogenation steps. The two possible reaction products of 3-hexyne and deuterium are deuterated hexene and hexane, which can be distinguished using their molecular mass as shown in figure 1. No molecular desorption of d2 -hexene at m/z=86 is observed for cluster sizes containing between three and seven atoms and the bare MgO(100) support within the sensitivity of the

7



experiment. A small onset of hexene desorption is detected for Pt8 around 150 to 300 K. Significant hexene desorption are found for Pt9, 10 and Pt13 having broad desorption features and maxima around 250 K. For Pt(111), two major desorption signals around 150 and 250 K can be seen in figure 1. No molecular desorption of d4 -hexane at m/z= 90 is detected for all cluster sizes and only minor desorption at 184 K for the platinum single crystal. b) m/z=90, d4-Hexane

a) m/z=86, d 2-Hexene

Pt(111) Pt13 ion signal /a.u.

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

Pt10 Pt9 Pt8 Pt7 Pt6 Pt5 Pt4 Pt3

MgO 200 300 400 500 temperature /K

200 300 400 500 temperature /K

Figure 1: TPR results of 3-hexyne and D2 on size-selected Pt clusters supported on MgO(100) tracing m/z= 86 (deuterated hexene, a) and m/z= 90 (deuterated hexane, b). Desorption signals are highlighted by grey background. The results obtained for the bare MgO(100) support and a Pt(111) single crystal are shown for comparison. The two black arrows of Pt13 indicate the temperature, after which CO IRRA spectra are recorded. Besides the formation of deuterated products, additional reaction pathways including dehydrogenation steps have to be taken into account for the surface chemistry of 3-hexyne as shown in figure 2. The m/z ratio of 69 was used to determine the amount of selfhydrogenation leading to the formation of H2 -hexene. In contrast to deuteration the hydrogen atoms stem from dehydrogenation of 3-hexyne making atomic hydrogen available for intermolecular hydrogenation of 3-hexyne. No self-hydrogenation is observed for cluster sizes below eight atoms. The signal below 200 K is caused by physisorbed 3-hexyne and can be

8


Page 8 of 25

Page 9 of 25

neglected. Cluster sizes between 8 and 13 atoms show broad signals ranging from 200 up to 400 K, and hexene desorption occurs additionally below 200 K on Pt(111). At temperatures above 400 K, benzene desorption is observed for all cluster sizes and Pt(111). b) m/z=78, Benzene

a) m/z=69, H2-Hexene

Pt(111) Pt13 ion signal /a.u.

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


Pt10 Pt9 Pt8 Pt7 Pt6

Pt5 Pt4 Pt3

MgO 200 300 400 500 temperature /K

200 300 400 500 temperature /K

Figure 2: TPR results of 3-hexyne and D2 on size-selected Pt clusters and Pt(111) showing m/z= 69 (H2 -hexene, a) and m/z= 78 (benzene formation, b). Desorption signals of H2 hexene (grey) are caused by self-hydrogenation and the feature below 160 K originates from the cracking pattern of 3-hexyne physisorbed on the support. The desorption of benzene occurs above 400 K for all cluster sizes.

Integration and normalization of the TPR curves to the amount of Pt atoms per sample are shown in figure 3. Four desorbing species have been identified including hydrogenated hexene, self-hydrogenated hexene, hexane and benzene. The only observable reaction pathway of 3-hexyne and deuterium on Pt clusters containing between 3 and 7 atoms is the formation of benzene. Pt8 is the minimum cluster size where hexene desorption occurs in both, hydrogenation and self-hydrogenation pathways. Larger cluster sizes (9, 10 and 13) show higher activity than the Pt(111) surface for all products. To further investigate carbonaceous deposits on the metal clusters, IRRA spectra of supported Pt13 clusters are shown in figure 4 using CO as a probe molecule. Here, clean

9



d2-Hexene d4-Hexane H2-Hexene Benzene

normalized TPR areas

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 10 of 25

3

4

5

6

7

8

9 10

13

Pt(111)

number of atoms per cluster

Figure 3: The normalized TPR areas from figures 1 and 2 (highlighted in grey) are displayed as a function of cluster size: For each molecule, the determined area is first corrected by the QMS sensitivity factor and afterwards normalized to the total amount of Pt atoms given by coverage and cluster size. clusters exhibit one on-top CO stretch frequency at 2075 cm-1 . After the 3-hexyne and D2 TPR to 300 K, no signal is detected, whereas after TPR to 600 K one single band at 2069 cm-1 is observed. These two temperature steps are shown in figure 1 by black arrows.

Discussion The discussion is divided into two parts. First, reactions not necessarily requiring the presence of deuterium are dealt with, which include formation of H2 -hexene and benzene. In the second part, reactions of 3-hexyne explicitly including deuterium and their impact on selective hydrogenation reactions are discussed.

10


Page 11 of 25

clean

2075 after TPR to 300 K

transmittance /%

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


after TPR to 600 K

0.2 2069

2150

2100

2050

wavenumber /cm

2000

-1

Figure 4: IRRA spectra of the CO frequency on Pt13 clusters supported on MgO: three spectra are shown for clusters as deposited (denoted as clean), and after 3-hexyne and D2 TPR to 300 and 600 K. All spectra are recorded at 100 K. It should be noted that only half the particles were used for the spectrum of clean clusters.

Dehydrogenation pathways The thermal chemistry of unsaturated hydrocarbons on clean transition metals under UHV conditions is known to be characterized by dehydrogenation leading to the formation of strongly adsorbed carbonaceous species on the metal surface . 31 This process releases atomic hydrogen, which in turn can either recombine to form molecular hydrogen or hydrogenate additional molecules on the surface. The intermolecular hydrogenation was monitored in this work by desorption of H2 -hexene. As expected, no production of H2 -hexene was observed for the blank MgO(100) support, but also platinum metal clusters up to seven atoms showed no significant intermolecular hydrogenation. This observation can be rationalized by the number of adsorption sites available under the applied reaction conditions. Smaller clusters exhibit only one adsorption site of 3-hexyne effectively suppressing intermolecular interactions between adsorbed hexynes. This can be illustrated on Pt3 clusters, which are assumed to have a trigonal planar structure on the surface. Alkynes adsorb in a di-σ/π con11



figuration on platinum , 32 which requires three surface atoms, and consequently all cluster atoms are in direct contact with one single hexyne molecule and not available for interactions with additional molecules. For larger clusters, e.g. Pt7 , additional effects have to be taken into account. The alkyl-substituents of 3-hexyne might sterically hinder co-adsorption or even weaken co-adsorption due to intermolecular interactions, which in turn lead to desorption before dehydrogenation. Besides these interactions, the transition between Pt7 and Pt8 might be accompanied by significanct changes in cluster morphology, which was proposed between these sizes on TiO2 (110) . 15,17 As soon as the cluster reaches a critical size, co-adsorption and self-hydrogenation takes place. The broad desorption feature, exhibiting at least two maxima for Pt9 and Pt10 , reflects the multi-step nature of the dehydrogenation. First, significant dehydrogenation takes place below room temperature leading to hexene desorption below 300 K for larger clusters and the metal surface. The desorption temperature of 1-hexene on Pt(111) is found between 275 to 284 K depending on coverage , 33,34 which may be indicative of repulsive interactions between co-adsorbed species of hexene and hexyne tested here. The occurance of hexene desorption above room temperature is clearly reaction-rate limited by hydrogen release during dehydrogenation. One reaction pathway of exclusively dehydrogenation is the formation of benzene shown in figure 2 b). Benzene desorption is observed for all cluster sizes investigated in this work. This reaction occurs via intra-molecular dehydrocyclization, since no evidence for CC-bond scission was found for both, cluster samples and single crystal. Our previous work has studied the thermal chemistry of 3- and 1-hexyne on Pt(111) in detail, concluding that cyclization follows several dehydrogenation steps for both molecules similar to a copper platinum alloy . 35,36 Furthermore, this dehydrocyclization path is also open for the according hexenes regardless of the initial stereochemistry. The dehydrogenation mechanism followed by ringclosure seems to be applicable to platinum clusters, since dehydrogenation, as observed by H2 -hexene desorption in figure 2, has vanished below 400 K, which is significantly lower than

12


Page 12 of 25

Page 13 of 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


benzene desorption. The desorption of benzene was shown to emerge only at higher coverages for Pt(111), whereas low coverages lead to complete coking of the surface without molecular desorption . 37 Consequently, the amount of formed benzene in figure 3 could be underestimated due to limited desorption. One evidence of limited desorption and formation of highly dehydrogenated species remaining on the clusters surface stems from CO IRRA spectra shown in figure 4. After desorption of hydrogenated products at 300 K the surface of Pt13 clusters is still blocked by adsorbed hydrocarbons. A distinct red-shift of 6 cm-1 in CO frequency and a loss in intensity are observed after 600 K and completed benzene desorption. This shift is typically associated with the presence of co-adsorbed and dehydrogenated carbonaceous species . 38,39 Three metal atoms within one cluster are sufficient for dehydrocyclization, which is explained by the adsorption mode as described above. Further evidence for this adsorption induced reaction behavior arises from dehydrogenation experiments of 3-hexyne on single Pt atoms, where no benzene formation has been observed as shown in figure 5. Here, the signals below 200 K are caused by the cracking contribution of 3-hexyne physisorbed on the inert support. No signals are observed for dehydrogenation (m/z= 2), self-hydrogenation (m/z= 69, 57), and dehydrocyclization (m/z= 78). It should be noted, that benzene formation from 3-hexyne reported in this work has distinct differences to the well-established cyclotrimerization of acetylene on Pd clusters, since the latter requires coadsorption of three molecules and represents an intermolecular reaction . 7

Hydrogenation pathways Figure 1 demonstrates, that d4 -hexane is only observed as a reaction product from Pt(111) in a small peak at 190 K. This desorption temperature is slightly lower than that observed for hexane in submonolayer quantities on clean Pt(111) 40 indicating repulsive intermolecular interactions. No hexane desorption is detected for all cluster sizes, which might be explained 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

ion signal /a.u.


Page 14 of 25

m/z=2 (x0.01) m/z=78 (x0.1) m/z=69 m/z=57

200

300

400

500

temperature /K

Figure 5: TPD of 0.44/SA 3-hexyne on single Pt atoms supported on MgO(100) (amount of Pt atoms 9.0 · 1013). either by intrinsic high selectivity for the alkene or quantities below the detection limit. In order to gain insight into cluster selectivity, it is convenient at this point to compare formation of d2 -hexene representative of triple bond hydrogenation to a model system of double bond hydrogenation under the same experimental conditions, e.g. ethylene. This comparison between different unsaturated functional groups for identical cluster sizes is shown in figure 6. As determined by integration and normalization of the ion current in figure 1, cluster sizes below eight atoms show no significant hydrogenation of 3-hexyne, whereas an on-off effect is observed between Pt9 and Pt10 for ethane formation . 19 The absence of hydrogenation products in TPR experiments does not exclude the clusters capability for hydrogenation under steady-state conditions, but general consistency is found between TPR and pulsed, isothermal experiments . 20 For cluster sizes without hydrogenation products in TPR experiments, the turn-over frequencies determined under isothermal conditions typically do not exceed the one determined for the extended surface.

14


Page 15 of 25

Enhanced Selectivity

d2-Hexene from 3-Hexyne Ethane from Ethylene

integrated ion signal /a.u.

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


3

4

5

6

7

8

9 10

13

Pt(111)

number of atoms per cluster

Figure 6: Comparison between TPR results obtained for 3-hexyne hydrogenation in this work (black circle) and ethylene hydrogenation taken from reference 19 (grey triangle). Clusters containing less than eight atoms show no hydrogenation products for 3-hexyne indicative of unfavorable hydrogenation reaction barriers, whereas Pt10 and Pt13 clusters show reactivity towards both reactions. Only Pt9 exhibits high activity towards a triple bond and low activity for a double bond resulting in increased selectivity.

Cluster sizes having increased hydrogenation activity compared to the extended surface in figure 6, in particular Pt9 , Pt10 and Pt13 , all show similar desorption temperatures below 300 K in figure 1. This leads to the conclusion that activation barriers for hydrogenation as well as mechanistic pathway leading to hydrogenation are quite similar for these cluster sizes. In contrast to the platinum clusters, two distinct distinct signals are observed for the single crystal. We tentatively attribute the low temperature signal at 180 K to hydrogenation of a weakly bound, physisorbed hexyne molecule, whereas the high temperature signals originates from hydrogenation of hexyne chemisorbed on the metal surface. The interaction between metal and weakly bound species is decreased by preadsorbed hydrogen acting as an isolation layer, an effect which has also been observed for 1,3-butadiene adsorbed on hydrogen covered Pt(111) . 41 Given similar activation energies, but different activities and selectivities, the question arises of the origin of the latter. 15



A requirement for increased activity might stem from either more favorable hydrogen or alkyne adsorption. In order to test hydrogen adsorption, we have performed HD scrambling experiments on the clean cluster sizes showing increased reactivity towards hexyne. Here, Pt9 showed limited activity towards HD-exchange in comparison to Pt10 and Pt13 . 19,20 Assuming a three-dimensional prismatic structure for Pt9 (one low-energy structure found by firstprinciples calculations), the limited adsorption and dissociation of hydrogen measured by TPR experiments are caused by different interactions between hydrogen and Pt atoms of the bottom layer and top layer, respectively. Hydrogen dissociatively adsorbs on Pt atoms of the bottom layer in contrast to the top layer. This dissociation difference is only strictly valid for a prismatic structure of Pt9 . Recent calculations on the same system by Kolsbjerg et al. expanded the possible low-energy structures of Pt9 employing a neural-network-enhanced evolutionary algorithm 42 . They found asymmetric configurations of Pt9 , which are slightly more stable than the prismatic one by 0.18 eV. The impact of this finding on inter-structural conversions and hydrogen adsorption at elevated temperatures remains to be answered. However, co-adsorption of deuterium and 3-hexyne on Pt9 investigated in this work here, unequivocally demonstrates that Pt9 is among the most active cluster sizes available for hexyne hydrogenation. Consequently the clusters capability to adsorb and dissociate hydrogen as part of the reaction network can be ruled out to be a limiting factor for both, activity and selectivity. On the contrary, co-adsorption of 3-hexyne on the D2 precovered Pt9 clusters has to facilitate hydrogen activation by a synergistic effect in order to achieve high hydrogenation activity as evidenced by the experiments of this work. This adsorbate-adsorbate interaction is decisively controlled by the adsorption mode and number of available sites of 3-hexyne on the metal clusters. These parameters are influenced by several factors making direct correlation rather difficult. First, it was shown for the activation of ethylene, that the arrangement and in particular the local electron density of single metal atoms within the cluster define unique adsorption sites leading to distinguishable activities . 19 A direct measurement of the clusters local electron densities, via e.g. photoelectron

16


Page 16 of 25

Page 17 of 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


spectroscopies , 13,18 leads to overlapping signals concealing the underlying local nature. An indirect measurement by using the CO frequency in IR spectroscopy was also not able to appreciably resolve the unique atomic sites . 25 Further hampering experimental evidence, the local electron density governed by cluster morphology cannot be regarded as a constant, but rather changes over the course of the experiment due to fluxionality . 43,44 This cluster relaxation is here mainly controlled by temperature and adsorption of 3-hexyne. During TPR experiments inter-isomer structural conversion occurs with more isomers being thermally accessible at higher temperatures . 23 At the same time, the adsorption energy of an alkyne is higher than the one found for alkenes and typically around ≈ 2 eV , 45 which is comparable to the metal binding energies within the cluster . 46 Therefore adsorption of 3-hexyne is expected to have a major influence on cluster morphology by reaction-induced fluctionality. These interactions also include the activation of co-adsorbed deuterium, which is not observed for Pt9 clusters without adsorbed alkynes. A complex interplay between these factors leads to favorable hydrogenation conditions of 3-hexyne on Pt9 , Pt10 and Pt13 , whereas only for Pt9 high selectivity can be achieved by suppressed alkene hydrogenation.

Conclusions In this paper the surface chemistry of 3-hexyne was addressed on size-selected Pt clusters supported on MgO(100). The selectivity for different reaction channels was tested by isotopically labelled TPR experiments with the conclusions that (i) all cluster sizes between 3 and 13 atoms showed intramolecular dehydrocyclization of 3-hexyne to form benzene comparable to the Pt(111) single crystal. This finding is correlated to the di-σ/π adsorption mode only requiring the presence of three metal atoms. (ii) Co-adsorption of several 3-hexyne molecules is hindered for cluster sizes smaller than 7 atoms, as evidenced by desorption of H2 -hexene, which stems from intermolecular, repulsive interactions caused by steric hindrance. (iii) Small

17



cluster sizes show no hydrogenation products, whereas Pt9 , Pt10 and Pt13 are more reactive than the single crystal on a per atom basis. By comparison to ethylene hydrogenation, a small atomic window is found for Pt9 , which shows high triple bond, but low double bond hydrogenation activity. The resulting enhanced selectivity can be attributed to a change of cluster morphology induced by the higher alkyne adsorption energy in comparison to ethylene. This fluxionality of the clusters is also responsible for activation of hydrogen required for increased hydrogenation activity. The results presented here demonstrate the potential usage of cluster-based materials for selective hydrogenation reactions, but also the difficulties encountered when designing a catalyst on an atom basis. Not only changes as a function of size are observed, but the investigations of more complex molecules are also convoluted by additional reaction pathways, which become accessible as the size is changed. Identification and understanding these pathways constitute a major challenge in order to steer catalytic selectivity and this work provides foundation for further investigations of selective hydrogenation reactions of unsaturated hydrocarbons on atomically controlled catalysts.

Acknowledgement This project was supported by Clariant in the framework of MuniCat- Munich Catalysis Alliance of Clariant and Technical University Munich. We also acknowledge financial support of the Deutsche Forschungsgemeinschaft (DFG, project He3454/23-1). M. K. acknowledges financial support of the Studienstiftung des Deutschen Volkes. M. D. R. is grateful to Richard Fischer for helpful discussions and Normen Szesni for stimulating the project.

Supporting Information Available The following files are available free of charge. The Supporting Information is available free of charge on the ACS Publications website 18


Page 18 of 25

Page 19 of 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


at DOI: Blank TPD experiments of H2 and D2 on MgO.

References (1) Yoon, B.; Häkkinen, H.; Landman, U.; Wörz, A. S.; Antonietti, J.-M.; Abbet, S.; Judai, K.; Heiz, U. Charging Effects on Bonding and Catalyzed Oxidation of CO on Au8 Clusters on MgO. Science 2005, 307, 403–407. (2) Tyo, E. C.; Vajda, S. Catalysis by Clusters with Precise Numbers of Atoms. Nat. Nanotechnol. 2015, 10, 577–588. (3) Watanabe, Y. Atomically Precise Cluster Catalysis Towards Quantum Controlled Catalysts. Sci. Technol. Adv. Mater. 2014, 15, 063501. (4) Schweinberger, F. In Encyclopedia of Interfacial Chemistry; Wandelt, K., Ed.; Elsevier: Oxford, 2018; pp 625–637. (5) Yoon, B.; Landman, U.; Habibpour, V.; Harding, C.; Kunz, S.; Heiz, U.; Moseler, M.; Walter, M. Oxidation of Magnesia-Supported Pd30 Nanoclusters and Catalyzed CO Combustion: Size-Selected Experiments and First-Principles Theory. J. Phys. Chem. C 2012, 116, 9594–9607. (6) Kunz, S.; Schweinberger, F. F.; Habibpour, V.; Röttgen, M.; Harding, C.; Arenz, M.; Heiz, U. Temperature Dependent CO Oxidation Mechanisms on Size-Selected Clusters. J. Phys. Chem. C 2010, 114, 1651–1654. (7) Abbet, S.; Sanchez, A.; Heiz, U.; Schneider, W.-D. Tuning the Selectivity of Acetylene Polymerization Atom by Atom. J. Catal. 2001, 198, 122–127. (8) Abbet, S.; Sanchez, A.; Heiz, U.; Schneider, W. D.; Ferrari, A. M.; Pacchioni, G.;

19



Rösch, N. Size-effects in the Acetylene Cyclotrimerization on Supported Size-Selected Pdn Clusters (1≤n≤30). Surf. Sci. 2000, 454, 984–989. (9) Dai, Y.; Gorey, T. J.; Anderson, S. L.; Lee, S.; Lee, S.; Seifert, S.; Winans, R. E. Inherent Size Effects on XANES of Nanometer Metal Clusters: Size-Selected Platinum Clusters on Silica. J. Phys. Chem. C 2017, 121, 361–374. (10) Heiz, U.; Sherwood, R.; Cox, D. M.; Kaldor, A.; Yates, J. T. CO Chemisorption on Monodispersed Platinum Clusters on SiO2 : Detection of CO Chemisorption on Single Platinum Atoms. J. Phys. Chem. 1995, 99, 8730–8735. (11) Yasumatsu, H.; Fukui, N. Steady-State Reaction Kinetics of CO Oxidation Catalyzed by Uni-Sized Pt30 Clusters Directly Bound to Si Surface. Cat. Sci. Tec. 2016, 6, 6910– 6915. (12) Yin, C.; Negreiros, F. R.; Barcaro, G.; Beniya, A.; Sementa, L.; Tyo, E. C.; Bartling, S.; Meiwes-Broer, K.-H.; Seifert, S.; Hirata, H. et al. Alumina-Supported Sub-Nanometer Pt10 Clusters: Amorphization and Role of the Support Material in a Highly Active CO Oxidation Catalyst. J. Mater. Chem. A 2017, 5, 4923–4931. (13) Kane, M. D.; Roberts, F. S.; Anderson, S. L. Effects of Alumina Thickness on CO Oxidation Activity over Pd20 /Alumina/Re(0001): Correlated Effects of Alumina Electronic Properties and Pd20 Geometry on Activity. J. Phys. Chem. C 2015, 119, 1359–1375. (14) Reber, A. C.; Khanna, S. N.; Roberts, F. S.; Anderson, S. L. Effect of O2 and CO Exposure on the Photoelectron Spectroscopy of Size-Selected Pdn Clusters Supported on TiO2 (110). J. Phys. Chem. C 2016, 120, 2126–2138. (15) Watanabe, Y.; Wu, X.; Hirata, H.; Isomura, N. Size-Dependent Catalytic Activity and Geometries of Size-Selected Pt Clusters on TiO2 (110) Surfaces. Cat. Sci. Tec. 2011, 1, 1490–1495.

20


Page 20 of 25

Page 21 of 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


(16) Isomura, N.; Wu, X.; Hirata, H.; Watanabe, Y. Cluster Size Dependence of Pt CoreLevel Shifts for Mass-Selected Pt Clusters on TiO2 (110) Surfaces. J. Vac. Sci. Technol. A 2010, 28, 1141–1144. (17) Isomura, N.; Wu, X.; Watanabe, Y. Atomic-Resolution Imaging of Size-Selected Platinum Clusters on TiO2 (110) Surfaces. J. Chem. Phys. 2009, 131, 164707. (18) Kaden, W. E.; Wu, T.; Kunkel, W. A.; Anderson, S. L. Electronic Structure Controls Reactivity of Size-Selected Pd Clusters Adsorbed on TiO2 Surfaces. Science 2009, 326, 826–829. (19) Crampton, A. S.; Rötzer, M. D.; Ridge, C. J.; Schweinberger, F. F.; Heiz, U.; Yoon, B.; Landman, U. Structure Sensitivity in the Nonscalable Regime Explored via Catalysed Ethylene Hydrogenation on Supported Platinum Nanoclusters. Nat. Commun. 2016, 7, 10389. (20) Crampton, A. S.; Rötzer, M. D.; Ridge, C. J.; Yoon, B.; Schweinberger, F. F.; Landman, U.; Heiz, U. Assessing the Concept of Structure Sensitivity or Insensitivity for Sub-Nanometer Catalyst Materials. Surf. Sci. 2016, 652, 7–19. (21) Imaoka, T.; Akanuma, Y.; Haruta, N.; Tsuchiya, S.; Ishihara, K.; Okayasu, T.; Chun, W.-J.; Takahashi, M.; Yamamoto, K. Platinum Clusters with Precise Numbers of Atoms for Preparative-Scale Catalysis. Nat. Commun. 2017, 8, 688. (22) Ha, M.-A.; Baxter, E. T.; Cass, A.; Anderson, S. L.; Alexandrova, A. N. Boron Switch for Selectivity of Catalytic Dehydrogenation on Size-Selected Pt clusters on Al2 O3 . J. Am. Chem. Soc. 2017, 139, 11568–11575. (23) Baxter, E. T.; Ha, M.-A.; Cass, A. C.; Alexandrova, A. N.; Anderson, S. L. Ethylene Dehydrogenation on Pt4,7,8 Clusters on Al2 O3 : Strong Cluster Size Dependence Linked to Preferred Catalyst Morphologies. ACS Catal. 2017, 3322–3335.

21



(24) Rioux, R. M.; Song, H.; Hoefelmeyer, J. D.; Yang, P.; Somorjai, G. A. High-SurfaceArea Catalyst Design: Synthesis, Characterization, and Reaction Studies of Platinum Nanoparticles in Mesoporous SBA-15 Silica. J. Phys. Chem. B 2005, 109, 2192–2202. (25) Crampton, A. S.; Rötzer, M. D.; Landman, U.; Heiz, U. Can Support Acidity Predict Sub-Nanometer Catalyst Activity Trends? ACS Catal. 2017, 6738–6744. (26) Crampton, A. S.; Rötzer, M. D.; Schweinberger, F. F.; Yoon, B.; Landman, U.; Heiz, U. Controlling Ethylene Hydrogenation Reactivity on Pt13 Clusters by Varying the Stoichiometry of the Amorphous Silica Support. Angew. Chem. Int. Ed. 2016, 55, 8953– 8957. (27) Heiz, U.; Vanolli, F.; Trento, L.; Schneider, W.-D. Chemical Reactivity of Size-Selected Supported Clusters: An Experimental Setup. Rev. Sci. Instrum. 1997, 68, 1986–1994. (28) Schweinberger, F. F.; Crampton, A. S.; Zimmermann, T.; Kwon, G.; Ridge, C. J.; Günther, S.; Heiz, U. Submonolayer Sensitive Adsorption Study of Trichloroethene on Single Crystal Surfaces by Means of MIES, UPS and TPD. Surf. Sci. 2013, 609, 18–29. (29) Winkler, A.; Yates Jr., J. T. Capillary Array Dosing and Angular Desorption Distribution Measurements: A General Formalism. J. Vac. Sci. Technol. A 1988, 6, 2929–2932. (30) Bozack, M. J.; Muehlhoff, L.; Russell Jr., J. N.; Choyke, W. J.; Yates Jr., J. T. Methods in Semiconductor Surface Chemistry. J. Vac. Sci. Technol. A 1987, 5, 1–8. (31) Bond, G. Metal-Catalysed Reactions of Hydrocarbons; Springer US, 2006. (32) Koestner, R. J.; Frost, J. C.; Stair, P. C.; Van Hove, M. A.; Somorjai, G. A. Evidence for the Formation of Stable Alkylidyne Structures from C3 and C4 Unsaturated Hydrocarbons Adsorbed on the Pt(111) Single Crystal Surface. Surf. Sci. 1982, 116, 85–103.

22


Page 22 of 25

Page 23 of 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


(33) Zhao, H.; Koel, B. E. Probing Selectivity over Pt-Sn Catalysts in Reactions of n-C6 Hydrocarbons: Adsorption and Reactivity of n-Hexane, 1-Hexene, and 1,5-Hexadiene on Pt(111) and Sn/Pt(111) Surface Alloys. J. Phys. Chem. C 2009, 113, 18152–18162. (34) Murillo, L. E.; Khan, N. A.; Chen, J. G. The Effect of Hydrocarbon Structure and Chain Length on the Low-Temperature Hydrogenation Activity on Ni/Pt(111) Bimetallic Surfaces. Surf. Sci. 2005, 594, 27–42. (35) Rötzer, M. D.; Krause, M.; Crampton, A. S.; Mitterreiter, E.; Heenen, H. H.; Schweinberger, F. F.; Reuter, K.; Heiz, U. Surface Chemistry of 1- and 3-Hexyne on Pt(111): Desorption, Decomposition, and Dehydrocyclization. J. Phys. Chem. C 2018, 122, 4428–4436. (36) Teplyakov, A. V.; Bent, B. E. Mechanism of Dehydrocyclization of 1-Hexene to Benzene on Cu3 Pt(111). J. Phys. Chem. B 1997, 101, 9052–9059. (37) Campbell, J. M.; Seimanides, S.; Campbell, C. T. Probing Ensemble Effects in Surface Reactions. 2. Benzene Adsorption on Clean and Bismuth-Covered Platinum(111). J. Phys. Chem. 1989, 93, 815–826. (38) Lundwall, M. J.; McClure, S. M.; Goodman, D. W. Probing Terrace and Step Sites on Pt Nanoparticles Using CO and Ethylene. J. Phys. Chem. C 2010, 114, 7904–7912. (39) Rioux, R. M.; Hoefelmeyer, J. D.; Grass, M.; Song, H.; Niesz, K.; Yang, P.; Somorjai, G. A. Adsorption and Co-adsorption of Ethylene and Carbon Monoxide on SilicaSupported Monodisperse Pt Nanoparticles: Volumetric Adsorption and Infrared Spectroscopy Studies. Langmuir 2008, 24, 198–207. (40) Tait, S. L.; Dohnálek, Z.; Campbell, C. T.; Kay, B. D. n-Alkanes on Pt(111) and on C(0001)/Pt(111): Chain Length Dependence of Kinetic Desorption Parameters. J. Chem. Phys. 2006, 125, 234308.

23



(41) Zhao, H.; Welch, L. A.; Koel, B. E. Site-Blocking Effects of Preadsorbed H on Pt(111) Probed by 1,3-Butadiene Adsorption and Reaction. Surf. Sci. 2009, 603, 3355–3360. (42) Kolsbjerg, E. L.; Peterson, A. A.; Hammer, B. Neural-Network-Enhanced Evolutionary Algorithm Applied to Supported Metal Nanoparticles. Phys. Rev. B 2018, 97, 195424, PRB. (43) Landman, U.; Yoon, B.; Zhang, C.; Heiz, U.; Arenz, M. Factors in Gold Nanocatalysis: Oxidation of CO in the Non-Scalable Size Regime. Top. Catal. 2007, 44, 145–158. (44) Häkkinen, H.; Abbet, S.; Sanchez, A.; Heiz, U.; Landman, U. Structural, Electronic, and Impurity-Doping Effects in Nanoscale Chemistry: Supported Gold Nanoclusters. Angew. Chem. Int. Ed. 2003, 42, 1297–1300. (45) Mittendorfer, F.; Thomazeau, C.; Raybaud, P.; Toulhoat, H. Adsorption of Unsaturated Hydrocarbons on Pd(111) and Pt(111): A DFT Study. J. Phys. Chem. B 2003, 107, 12287–12295. (46) Chaves, A. S.; Rondina, G. G.; Piotrowski, M. J.; Tereshchuk, P.; Da Silva, J. L. F. The Role of Charge States in the Atomic Structure of Cun and Ptn (n = 2-14 atoms) Clusters: A DFT Investigation. J. Phys. Chem. A 2014, 118, 10813–10821.

24


Page 24 of 25

Page 25 of 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


Graphical TOC Entry Size and Selectivity?

Pt9

+ H2

25


Pt3

Controlling Hydrogenation Selectivity by Size: 3-Hexyne on Supported

Recommend Documents