Mechanism of Surface Alkylation of a Gold Aerogel with Tetra-n

Letter pubs.acs.org/JPCL

Mechanism of Surface Alkylation of a Gold Aerogel with Tetra‑n‑butylstannane‑d36: Identification of Byproducts

Monika Benkovičová,† Dan Wen,¶ Jan Plutnar,† Martina Č ížková,† Alexander Eychmüller,¶ and Josef Michl*,†,§ †

Institute of Organic Chemistry and Biochemistry, Academy of Sciences of the Czech Republic, Flemingovo nám. 2, 16610 Prague 6, Czech Republic ¶ TU Dresden, Bergstrasse 66b, 01062 Dresden, Germany § Department of Chemistry and Biochemistry, University of Colorado, Boulder, Colorado 80309-0215, United States S Supporting Information *

ABSTRACT: The formation of self-assembled monolayers on surfaces is often likely to be accompanied by the formation of byproducts, whose identification holds clues to the reaction mechanism but is difficult due to the minute amounts produced. We now report a successful identification of self-assembly byproducts using gold aerogel with a large specific surface area, a procedure likely to be applicable generally. Like a thin gold layer on a flat substrate, the aerogel surface is alkylated with n-butyl-d9 groups upon treatment with a solution of tetra-n-butylstannane-d36 under ambient conditions. The reaction byproducts accumulate in the mother liquor in amounts sufficient for GC-MS analysis. In chloroform solvent, they are butene-d8, butane-d10, octane-d18, and tributylchlorostannane-d27. In hexane, they are the same except that tributylchlorostannane-d27 is replaced with hexabutyldistannane-d54. The results are compatible with an initial homolytic dissociation of a C−Sn bond on the gold surface, followed by known radical processes.

T

detailed nature of the attachment. The initial assumption was that an R3Sn−CH2R′ bond is cleaved, the R3Sn moiety is attached to a surface Au atom or atoms, and the R′CH2− alkyl group is either lost to solution5 or attached to the Au surface as well.8 More recent experiments with deuterated alkylstannanes showed that under the ambient conditions used this cannot be entirely correct. Although tin is co-deposited with the alkyls, it does not appear as a trialkylstannyl group, whose deuterated form would be detectable in IR spectra, but as a tin oxide, detected by XPS.6 In the one case in which the alkylation reaction order with respect to stannane concentration in bulk solution was determined,6 it was found to be 0.5. It was suggested that two butyls may be transferred simultaneously in a way reminiscent of reductive elimination and that the resulting stannylene R2Sn or Sn(0) atom is then oxidized by atmospheric oxygen, leading ultimately to the formation of tin oxides. This would be analogous to Au surface alkylation by RHgX, which co-deposits elemental mercury.9,10 An investigation of the mechanism of benzylation with benzyltrimethylstannane under ultrahigh vacuum conditions led to the conclusion that the benzylic bond dissociates and both the C6H5CH2 and (CH3)3Sn groups are attached to the surface,11 in contrast to the findings for alkylstannanes in ambient

he attachment of organic molecules to a gold surface has been of considerable interest for a long time.1−3 Despite much effort, many mechanistic questions remain, and direct identification of byproducts of the attachment reactions would help to solve them. In view of their minute amounts, identification of the byproducts is hard. To our knowledge, only a pH decrease upon treatment of gold nanoparticles with three thiols and thus a proton release have been demonstrated,4 but detection of molecular byproducts has not been accomplished. We now report successful use of gold aerogel with a large surface area for the purpose. The monolayer self-assembly reaction studied is the alkylation of a gold surface with an alkylstannane solution under ambient conditions. This treatment is known to coat a flat surface of gold, mostly Au(111), with an alkyl-containing monolayer.5 It is significant for at least two reasons: (i) the resulting layers are distinct from those commonly produced by treatment with alkanethiols in their permeability and thermal and chemical stability,6 and (ii) direct carbon-to-gold attachment has much higher electrical conductivity than attachment through a sulfur atom.7,8 While methyl groups are not transferred from a tin atom to the Au surface and ethyl groups are transferred only very slowly, propyl and longer alkyl groups are transferred efficiently.6 Similar solution alkylation processes do not proceed under ambient conditions with alkylsilanes,5 but they do take place with the organomercurials RHgX.9,10 The mechanism of gold surface alkylation with alkylstannane solutions under ambient conditions is not known nor is the © XXXX American Chemical Society

Received: February 7, 2017 Accepted: April 25, 2017 Published: May 1, 2017 2339

DOI: 10.1021/acs.jpclett.7b00296 J. Phys. Chem. Lett. 2017, 8, 2339−2343

Letter

The Journal of Physical Chemistry Letters solution where they are not, but the mechanisms could well be different in the two environments. Although additional kinetic studies are clearly desirable, they alone are not likely to elucidate the mechanism of gold surface alkylation with alkylstannanes, and a determination of the nature of reaction byproducts would appear very useful. Although we are aware of the risk that the reaction mechanisms on different facets of a gold surface need not be the same, we believe that at least some initial guidance could be obtained from work on gold that has a large surface area. We were unable to find anhydrous conditions for the formation of small (2−3 nm diameter) gold nanoparticles protected by ligands well enough to prevent coagulation without also inhibiting the alkylation with a stannane (it appears that hydroxylic solvents interfere with the alkylation process). The recent development of procedures for the preparation of noble metal aerogels12 offers an easier route. We have now investigated the reaction of tetra-n-butylstannane-d36 (1) with a gold aerogel whose specific surface area is on the order of 50 m2/g.13 The use of a perdeuterated reagent allows identification of reaction products in the presence of contaminants present in the aerogel or derived from the ambient atmosphere and the solvent. The aerogel13 and the deuterated stannane 16 were prepared as described in the literature. The solvents and reagents were commercial. Aerogel infrared (IR) spectra in KBr pellets and polarization-modulated IR reflection/absorption spectra (PM IRRAS) of monolayers on a flat gold surface were taken using a FT-IR Nicolet 6700 spectrometer. Transmission electron microscopy (TEM) used a JEOL JEM-1011 transmission electron microscope at 60 kV. X-ray photoelectron spectra (XPS) were acquired with an ESCAProbeP spectrometer by Omicron Nanotechnology Ltd. Gas chromatography/mass spectrometry (GCMS) analysis was performed using an Agilent 5975B MSD quadrupole mass spectrometer coupled to a 6890N gas chromatograph equipped with a 30 m HP-5MS MePhSiO column. We used 1 μL injection, 1.0 mL/min He flow, a temperature ramp of 250−320 °C in 40 min, and 25− 600 m/z MS detection. High-resolution mass spectra (HRMS) were obtained with a Waters GCT Premier instrument using electron impact (EI) ionization. When produced as reported, Au aerogel contains significant amounts of impurities, primarily chlorides, borates, and dopamine.12 We found that immersion of a cleaned flat gold surface into a 10, 100, or 1000 μM aqueous solution of any one of these contaminants for 2 h followed by drying does not interfere with a subsequent 24 h surface alkylation reaction with 1 in THF, as judged by the appearance of a C−D stretching vibration peak in PM IRRAS. When 10 mg of Au aerogel12 is treated with 2 mL of a 44 mM solution of 1 in n-hexane for 24 h and the solid is then twice stirred for 1 h with 2 mL of fresh hexane and dried, its IR spectrum (Figure 1) shows a distinct broad peak at 2213−2217 cm−1 in the C−D stretching region that did not display any peaks before the reaction with 1. The peak is not removed by subsequent extraction with 300 mL of hexane in a Soxhlet apparatus for 3 days. The frequency of its maximum, 2215 cm−1, is similar to the values of 2219−2220 cm−1 reported6 for deuterated n-butyl groups on a flat gold surface treated with a solution of 1. The TEM of the aerogel appears identical before and after treatment with 1 (Figure 2) and suggests that treatment with 1 did not change the morphology and the surface area. XPS reveals the presence of tin in the form of an

Figure 1. FTIR spectra of Au aerogel in a KBr pellet before (black) and after (red) reaction with 1. The arrow indicates a C−D stretching vibration.

oxide (a peak for Sn 3d5/2 at 486.15 ± 0.4 eV), as in experiments with flat gold surfaces, where the value 486.5 ± 0.1 eV was reported.6 We conclude that deuteriated butyl groups are chemisorbed on the gold aerogel surface and are transferred there from the stannane, as in earlier experiments with flat gold surfaces. The GC traces of n-hexane and hexane-d14 remain unchanged by treatment with aerogel under ambient conditions (this is not true of THF, which suffers partial oxidation). The presence of pentanes and hexanes as impurities masks much of the region of retention times tr (all in min) shorter than 2 min. In the GC trace of the mother liquor recovered from treatment of the aerogel with 1 in n-hexane or hexane-d14, new peaks appeared at retention times of tr = 4.18 (n-octane-d18, 2), 19.46 (1), and 26.37 (hexakis(n-butyl-d9)distannane, 3). With one batch of the aerogel, a peak at tr = 3.82 also appeared and its MS showed a strong signal for C7H7+ (with hexane-d14, C7D7+), suggesting that the peak might be due to toluene (and toluene-d8). The intensity of this peak was not increased noticeably when n-heptane was used as the solvent or when it was added to the n-hexane solvent, and we have not examined it further. When chloroform instead of n-hexane is used as the solvent under otherwise identical reaction conditions, the peak at tr = 26.37 does not appear, but a new peak at tr = 19.30 (tri-nbutylchlorostannane-d27) emerges. The previously obscured region of short retention times now reveals a weak peak at at tr = 1.37 due to a mixture of butane-d10 (4) and but-1-ene-d8 (5), which have been independently determined to have the same retention time. The peaks were identified using HRMS and a comparison of tr with values for authentic samples. Examples of the results are shown in Figure 3 (GC trace of mother liquor in chloroform and solvent alone), Figure 4 (MS of peaks at tr = 4.18 and 26.37), and Figure 5 (MS of peak at tr = 1.37). Additional GC traces and MS are shown in the Supporting Information. Because the new products 2−5 only form in the simultaneous presence of both 1 and the aerogel in hexane and not in the presence of the aerogel alone, we believe that they form during the surface alkylation reaction and have arisen from further transformations of initial reaction byproducts. All four can be accounted for readily as resulting from dissociation of a Sn−C bond of 1 physisorbed on the gold surface to produce a radical pair, presumably at least initially also physisorbed on the surface (indicated by the / sign). 2340


Letter

The Journal of Physical Chemistry Letters

Figure 2. Transmission electron micrograph of Au aerogel before (A) and after (B) alkylation.

Figure 3. GC traces: (Top) Mother liquor after reaction of Au aerogel with 1 in chloroform; (Bottom) virgin solvent.

(n‐C4 D9)4 Sn/Au(s) (1) → n‐C4 D9• /Au(s) (6) + (n‐C4 D9)3 Sn•/Au(s) (7) (1)

This would be in keeping with the original mechanistic suggestions.5,8 Neither radical is apt to accumulate on the surface or in the solution above the surface, and their most likely modes of disappearance are bimolecular: the reversal of reaction 1 or disproportionation (2) and dimerization (3) of the surface-bound or free butyl radical 6 2n‐C4 D9• /Au(s) (6) → n‐C4 D10 (4) + n‐C4 D8 (5)

(2)

2n‐C4 D9• /Au(s) (6) → n‐C8D18 (2)

(3)

Figure 4. Mass spectra of new peaks after reaction of Au aerogel with 1 in n-hexane. (A) tr = 4.18, 2; (B) tr = 26.37, 3.

attached alkyl, such as a loss of one or more deuterium atoms. In contrast, against the original guesses,5,7,8 previous work6 suggests that reaction 6 hardly ever occurs because deuterated trimethylstannyl groups are not found by IR spectroscopy on a gold surface treated with alkyltrimethylstannanes deuteriated in the methyl groups. It is easy to imagine alternative fates for the physisorbed radical 7, such as reductive elimination of dibutylstannylene-d18 (8) with transfer of an alkyl to the surface (reaction 7), which offers an alternative pathway to the observed surface alkylation

and dimerization (eq 4) of the surface-bound or free stannyl radical 7 2(n‐C4 D9)3 Sn•/Au(s) (7) → (n‐C4 D9 )3 Sn−Sn(n‐C4 D9)3 (3) (4)

The processes in eqs 2−4 would be expected to compete with monomolecular reactions in which a physisorbed radical attaches covalently to the gold surface, possibly keeping its simple alkyl structure but possibly not n‐C4 D9• /Au(s) (6) → n‐C4 D9Au n

(5)

(n‐C4 D9)3 Sn•/Au(s) (7) → (n‐C4 D9)3 SnAu n

(6)

(n‐C4 D9)3 Sn•/Au(s) (7) → (n‐C4 D9)2 Sn/Au(s) (8) + n‐C4 D9Au n

(7)

14

The expected polymerization of the stannylene 8 would preclude its observation in the GCMS experiment, but other decay paths can be imagined, such as another possible surface

Reaction 5 could be the main route to the observed alkylated gold surface and might involve structural changes in the 2341


Letter

The Journal of Physical Chemistry Letters

to be a tortuous route toward a detailed mechanistic understanding of this surface alkylation process. Both byproduct and kinetic studies need to be extended to additional stannanes before a more complete picture can emerge.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpclett.7b00296. GC traces and MS (PDF)

■

AUTHOR INFORMATION

ORCID

Josef Michl: 0000-0002-4707-8230 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We are grateful to Dr. Vladimiŕ Vrkoslav for help with the GCMS measurements and to Dr. Jiři ́ Kaleta for valuable discussions and help with the preparation of figures. This work was supported by the Grant Agency of the Czech Republic (1402337S), the Institute of Organic Chemistry and Biochemistry (RVO: 61388963), and the ERC (AdG 2013 AEROCAT).

■

REFERENCES

(1) Ulman, A. Formation and structure of self-assembled monolayers. Chem. Rev. 1996, 96, 1533−1554. (2) Schreiber, F. Structure and growth of self-assembling monolayers. Prog. Surf. Sci. 2000, 65, 151−257. (3) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Self-assembled monolayers of thiolates on metals as a form of nanotechnology. Chem. Rev. 2005, 105, 1103−1170. (4) Ansar, S. M.; Perera, G. S.; Jiang, D.; Holler, R. A.; Zhang, D. Organothiols self-assembled onto gold: Evidence for deprotonation of the sulfur-bound hydrogen and charge transfer from thiolate. J. Phys. Chem. C 2013, 117, 8793−8798. (5) Khobragade, D.; Stensrud, E. S.; Mucha, M.; Smith, J. R.; Pohl, R.; Stibor, I.; Michl, J. Preparation of covalent long-chain trialkylstannyl and trialkylsilyl salts and an examination of their adsorption on gold. Langmuir 2010, 26, 8483−8490. (6) Kaletová, E.; Kohutová, A.; Hajduch, J.; Kaleta, J.; Bastl, Z.; Pospíšil, L.; Stibor, I.; Magnera, T. F.; Michl, J. The scope of direct alkylation of gold surface with solutions of C1-C4 n-Alkylstannanes. J. Am. Chem. Soc. 2015, 137, 12086−12099. (7) Cheng, Z.-L.; Skouta, R.; Vazquez, H.; Widawsky, J. R.; Schneebeli, S.; Chen, W.; Hybertsen, M. S.; Breslow, R.; Venkataraman, L. In situ formation of highly conducting covalent Au-C contacts for single-molecule junctions. Nat. Nanotechnol. 2011, 6, 353−357. (8) Chen, W.; Widawsky, J. R.; Vázquez, H.; Schneebeli, S. T.; Hybertsen, M. S.; Breslow, R.; Venkataraman, L. Highly conducting πconjugated molecular junctions covalently bonded to gold electrodes. J. Am. Chem. Soc. 2011, 133, 17160−17163. (9) Mucha, M.; Kaletová, E.; Kohutová, A.; Scholz, F.; Stensrud, E. S.; Stibor, I.; Pospíšil, L.; von Wrochem, F.; Michl, J. Alkylation of gold surface by treatment with C18H37HgOTs and anodic Hg stripping. J. Am. Chem. Soc. 2013, 135, 5669−5677. (10) Scholz, F.; Kaletová, E.; Stensrud, E. S.; Ford, W. E.; Kohutová, A.; Mucha, M.; Stibor, I.; Michl, J.; von Wrochem, F. Formation of nalkyl monolayers by organomercury deposition on gold. J. Phys. Chem. Lett. 2013, 4, 2624−2629. (11) Batra, A.; Kladnik, G.; Gorjizadeh, N.; Meisner, J.; Steigerwald, M.; Nuckolls, C.; Quek, S. Y.; Cvetko, D.; Morgante, A.;

Figure 5. Mass spectrum of the peak at tr = 1.37 in the GC of mother liquor after reaction of the Au aerogel with 1 in chloroform. (A) Observed MS; (B) superimposed mass spectra of n-butane-d10 (4, blue) and n-but-1-ene-d8 (5, red) deduced from published MS of nbutane and n-but-1-ene by converting the mass of H to D, assuming no isotope effects on fragmentation. I = relative intensity.

alkylation pathway, double alkyl transfer to the surface with the formation of a Sn(0) atom. This atom, the stannylene 8, or the radical 6 would be a source of reactivity with oxygen, which needs to be invoked ultimately to explain the formation of tin oxides. The radical 7 would be expected to abstract a chlorine atom from chloroform, and it is thus reasonable that the dimer 3 is not observed among the products when the alkylation reaction is performed in this solvent, whereas tri-n-butylchlorostannaned27 is observed. The set of reactions 1−7 accounts for the observed byproducts 2−5 and is compatible with other observations. For instance, it is easy to believe that the smallest radicals, especially methyl, would dimerize the fastest (reaction 2) and would be less likely to become covalently attached to the gold surface. As for the kinetic observations that have been made on di-n-butyldimethylstannane and showed an order of 0.5 with respect to bulk stannane concentration, we prefer not to speculate until byproduct analysis and kinetic analysis are both available for a larger set of stannanes. The use of gold aerogel is likely to be generally useful for the identification of byproducts of self-assembly reactions on gold surfaces. Our results for alkylation with tetra-n-butylstannaned36 (1) are compatible with the proposal that the dissociation of a C−Sn bond upon contact with gold is the first step. However, this observation merely represents a beginning in what is likely 2342


Letter

The Journal of Physical Chemistry Letters Venkataraman, L. Trimethyltin-mediated covalent gold-carbon bond formation. J. Am. Chem. Soc. 2014, 136, 12556−12559. (12) Liu, W.; Herrmann, A.-K.; Bigall, N. C.; Rodriguez, P.; Wen, D.; Oezaslan, M.; Schmidt, T. J.; Gaponik, N.; Eychmüller, A. Noble metal aerogels: synthesis, characterization, and application as electrocatalysts. Acc. Chem. Res. 2015, 48, 154−162. (13) Wen, D.; Liu, W.; Haubold, D.; Zhu, C.; Oschatz, M.; Holzschuh, M.; Wolf, A.; Simon, F.; Kaskel, S.; Eychmüller, A. Gold aerogels: three-dimensional assembly of nanoparticles and their use as electrocatalytic interfaces. ACS Nano 2016, 10, 2559−2567. (14) Neumann, W. P. Germylenes and stannylenes. Chem. Rev. 1991, 91, 311−334.

2343


Mechanism of Surface Alkylation of a Gold Aerogel with Tetra-n

Recommend Documents