Identification of the First Elementary Step in the Photocatalytic

Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Letter

Identification of the First Elementary Step in the Photocatalytic Reduction of Nitrobenzenethiols on Metallic Surface Han-Kyu Choi, Kang Sup Lee, Hyun-Hang Shin, and Zee Hwan Kim J. Phys. Chem. Lett., Just Accepted Manuscript • DOI: 10.1021/acs.jpclett.6b01852 • Publication Date (Web): 29 Sep 2016 Downloaded from http://pubs.acs.org on September 30, 2016

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.

The Journal of Physical Chemistry Letters 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 Letters

Identification of the First Elementary Step in the Photo-catalytic Reduction of Nitrobenzenethiols on Metallic Surface Han-Kyu Choi†, Kang Sup Lee†, Hyun-Hang Shin, and Zee Hwan Kim* Department of Chemistry, Seoul National University, Seoul 151-742, Korea AUTHOR INFORMATION Corresponding Author * [email protected] Author Contributions †These authors contributed equally.

1

ACS Paragon Plus Environment


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 25

Abstract Reduction of nitrobenzene is widely used for the assessment of the catalytic activities of nanoparticles, yet its mechanism is still largely unverified. Here, using the surface-enhanced Raman scattering (SERS), we have identified an intermediate of the first step in the photocatalytic reduction of nitrobenzenethiols (NBTs) on metallic surface. The formation of the intermediate is identified by a fast red-shift of the NO2 symmetric-stretching peak of the SERS spectra of reacting NBTs, prior to the slow intensity decay. Based on the laser power dependences of the rates of spectral changes, electrochemical SERS, and quantum chemical calculations, we conclude that the intermediate is the anion radical of nitrobenzenethiol that is formed by the metal-to-molecule single-electron transfer reaction. The subsequent intensity decay of the peak, which is the rate-determining step of the whole reduction reaction, corresponds

to

another

single-electron

reduction

of

the

anion

radical

into

dihydroxyaminobenzenethiol, or dianion of NBT.

TOC GRAPHICS

2


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


The reduction of nitrobenzene (NB) is the most frequently used model-reaction for assessing the catalytic activities of various semiconducting1-4 and metallic nanoparticles5-7. Despite its extensive use, however, detailed understanding of the mechanism, including the structures of intermediates and the conversion rates between them, is largely unverified. In addition, there is an indication that the photo-catalytic, electro-catalytic, and electrochemical reductions of NB may have different mechanisms. For example, under a controlled electrochemical potential, NB can be reduced to a metastable nitrobenzene anion radical8-10, yet it has not been detected as a reaction intermediate under catalytic reduction conditions. It is also suspected that the importance of cross-coupling of intermediates may be different for electroor photo-catalytic reductions11-13. Experimentally, the kinetic UV-vis absorption and surface-enhanced Raman scattering (SERS)14-15 spectra have consistently shown a slow decay of NB, and an accumulation of aniline1-7, 11-12, 16-19. With the exception of azobenzene, a long-lived intermediate with a large Raman cross-section, none of the proposed intermediates20-21 has been experimentally detected within the time-resolution of the measurement (~ 10 ms). The apparent decay rate of NB and the accumulation rate of aniline are found to be same1, 5-7. These strongly indicate that the reaction step associated with slow NB-decay is the rate-determining step for the conversion of NB into aniline. Here, we investigated the elementary steps in the photo-catalytic reduction of nitrobenzenethiols (NBTs) placed at a plasmonic junction between a silver nanoparticle (AgNP) and a gold thin-film (AuTF)20. We find that the time-resolved SERS spectra of reacting NBTs show a fast (~1 sec) red-shift of NO2-stretching of NBT, followed by a slow (~10 sec) intensity decay. We attribute that the initial red-shift correspond to the formation of NBT anion radical (NBT•‒) or its conjugate acid, induced by the single-electron transfer from 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 25

metal to NBT. Subsequent slow intensity decay corresponds to another single-electron reduction of NBT•‒ into dihydroxyaminobenzenethiol or a dianion of NBT. The result verifies the role of nitrobenzene radical anion as an intermediate of photo-catalytic reduction, and also clarifies that the apparent intensity decay of νNO-peak, which have been previously regarded to be the decay of neutral NBT, is in fact the decay of the anion radical of NBT.

Figure 1. (a) Schematic diagram of the AgNP/4-nitrobenzenethiol (NBT)/Au thin-film (AuTF) junction. The inset image (right) shows a representative SERS image monitoring νNO = 1342 cm-1, showing the spatial distribution of SERS hotspots. The inset spectrum (right) is a plasmon scattering spectrum obtained from a dark-field spectro-microscopy measurement. The laser excitation wavelength (λex) is also indicated in the spectrum. (b) Reaction steps and possible intermediates for the photo-catalytic reduction of NBT: NBT•‒ = anion radical of NBT; NBTH = conjugate acid of NBT•‒; DHABT = dihydroxyaminobenzenethiol; NBT2- = dianion of NBT; ABT = aminobenzenethiol.

The plasmonic junctions (Figure 1), defining the reaction and detection zones, are composed of Ag nanoparticles (AgNPs, diameter of 80 nm) randomly dispersed on top of a 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


monolayer of NBTs on a Au thin-film (AuTF, thickness of 10 nm). A focused laser beam (λex = 632.8 nm) locally induces photo-catalytic reduction at the junction, and also excite the SERS radiation of the NBT, intermediate, and product (Supporting Information A). A dipolar plasmon of the junction, with the resonance wavelength of ~600 nm (see inset scattering spectrum in Figure 1a), enhances the Raman scattering and also promotes the photo-catalytic reaction. The SERS from individual junction is collected with an objective lens, and is recorded by a spectrometer. As shown in the SERS image (inset of Figure 1a), the sample shows strong Raman enhancement (bright spots in the image) primarily at the AgNP/NBT/AuTF junction sites. More detailed characterization of SERS activities of NP-molecule-TF junctions, including Raman enhancement, polarization dependence, and other control measurements, can be found in Choi et al.20 and Park et al.22.

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 25

Figure 2. (a) Time-resolved SERS spectra obtained from a AgNP/NBT/AuTF junction. Vertical bars on the top denote the vibrational frequencies of NBT (blue) and DMAB (orange). The arrows in blue, red, and orange point to the time (t = 0, 1.7 and 32 sec, respectively) at which the spectra in (b) is sampled. (c) The time traces of νNO (red, NBT) and

ν3 (orange, DMAB) peak intensities, together with a fit (black). (d) Histogram of spectral shift in νNO (grey bars) and a Gaussian function fit (black curve). Also shown in vertical arrows are the average νNO peak shift measured from AuNP/NBT/AuTF junctions (orange) and NBT/AgTF (cyan). (e) Influence of temporarily blocking laser beam (laser off, 10 mins) on the temporal evolution of νNO-peak.

Figure 2a displays a time-resolved SERS spectrum obtained from a AgNP/NBT/AuTF junction, showing the progress of reduction reaction induced by laser irradiation (power density of P = 4 MW/cm2). At time-zero (t = 0 sec), the spectrum shows (Figure 2b, blue) three major peaks of NBT at ν = 1083 cm-1 (CS-stretching), 1342 cm-1 (NO2-symmetric stretching, νNO), and 1570 cm-1 (CC-stretching). As the reaction proceeds, the νNO-intensity slowly decays, and new peaks at ν = 1143, 1389, and 1430 cm-1 (named as ν1, ν2, and ν3, respectively), which are the vibrational modes of 4, 4’-dimercaptoazobenzene (DMAB) intermediate, gradually appear12-13,

20, 23-28

(see Figure 2b, orange spectrum. See also

Supporting Information B for peak assignment). At longer time scale (t > 2 min, not shown), the peaks of DMAB eventually decays away to form ABT, as confirmed by previous reports12-13, 29 (due to the close spectral overlap of peaks of ABT with those of DMAB and NBT, the formation ABT is not identifiable in our medium-resolution SERS spectra). 6


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


Intriguingly, we also find that the νNO peak redshifts by 6 cm-1 during the first ~2 seconds (Figure 2a and b), before its intensity begins to decay. During such period, the formation of DMAB is also delayed (red dotted lines in Figure 2a and c). The νNO peak-shift is consistently found in most (141 out of 147) of the SERS spectra obtained from AgNP/NBT/AuTF junctions. Microcrystalline NBTs dispersed on glass substrate, when irradiated with laser power densities ~10 times larger than the ones used in the SERS measurement, do not show any detectable νNO peak-shift or intensity decay in the Raman spectra (Supporting Information C), confirming that the metal-molecule interaction plays a role in the spectral-shift, as well as in the intensity decay, of νNO-peak. There exist significant junction-to-junction variations in SERS intensity, and in the rate of the νNO peak-shift, possibly due to the variation in local field intensities at each junctions. However, the junction-to-junction variation in the magnitude of νNO peak-shifts (∆ν) is found to be fairly small (< 2 cm-1, see Figure 2d for the histogram of ∆ν). Essentially the same magnitudes of peak-shifts are also observed from the trajectories of AuNP/NBT/AuTF junctions, and from the NBTs on rough Ag surfaces (NBT/AgTF) (see the vertical arrows in orange and cyan in Figure 2d, and Supporting Information D). Particularly, laser irradiation of the AuNP/NBT/AuTF junctions does not lead to any substantial intensity decay in νNO, or buildup of DMAB, yet we observe consistent peak-shifts (Figure S2 in Supporting Information D). Thus, while the rate of the peak-shift is influenced by the local field intensities (see below), the magnitude of peak-shift is unrelated to the local field intensities, kinds of metals, geometries of junctions, or whether or not the final reduction products (ABT and DMAB) are formed.

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 25

We find that temporarily blocking (~10 min) light during the experiment makes the νNO peak to return to its initial position (Figure 2e, t = 1.5 and t = 11 mins). Subsequent light illumination (t = 11 ~ 11.8 mins) again redshifts the νNO. This precludes the possibility that a photo-thermal, irreversible, annealing of metallic sites or molecular orientations may cause the peak-shift via site- or orientation-specific chemical-enhancement of Raman signal. Instead, the change appears to be associated with an intermediate that is relatively stable (with a lifetime of ~10 min), reversibly changing back to NBT in the absence of light (Supporting Information E).

Figure 3. (a) The zoom-in view of Figure 2a sampled around νNO and t = 0 ~ 7 sec. (b) The spectra of νNO peak sampled at t = 0.0, 0.2, 0.4, and 1.2 sec (blue, cyan, orange and red arrows in (a), respectively). (c) The spectral decomposition of νNO (grey) peak into νI (red) and νNBT (blue) components. (d) The time traces of νNBT (blue circles) and νI (red circles) components, together with fits to kinetic model (inset chemical equation). Also shown in grey

8


Page 9 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


(circles and line) is the sum of νNBT and νI components, and the associated fit. Log-log plots of (e) laser power density (P) versus k1, (f) P vs k2, and (g) k2 vs k1, which are derived from the kinetic analyses of trajectories similar to (d). Red lines and equations are the results of linear fits to the data.

Figure 3a shows a zoom-in view of the first t = 0 - 7 seconds of the νNO-spectra taken from Figure 2a. Figure 3b display four νNO-spectra sampled from Figure 3a at t = 0, 0.2, 0.4 and 1.2 seconds. They show an isosbestic point, indicating that two species with comparable Raman cross sections co-exist in the hotspot30. Analogously, a 2D-cross correlation analysis31 of the νNO-peak (Supporting Information F) not only shows the anti-correlation of the two components, but it also shows isosbestic lines (zero-correlation) located between the two. From these analyses, we identify two components at νΝΒΤ = 1342 cm-1, which is the original

νNO peak of NBT and νΙ = 1336 cm-1, which is a new peak of an intermediate (I) that is produced from a 1:1 transformation of NBT. The νNO-spectrum at each time-delay (t) is fitted to two Lorentzians centered at νNBT and νI (Figure 3c), and the amplitudes of the two are plotted as a function of time (Figure 3d). Such time-traces are again fitted to a single exponential decay function (INBT(t) = exp(-k1t) + a), modeling the irreversible 1st order decay of NBT, and a rise-and-decay function (II(t) = -exp(k1’t) + exp(-k2t)), modeling the buildup (k1’) and decay (k2) of the intermediate (see Figure 3d). Such rate equations satisfactorily fit the time-traces of NBT and intermediate. Particularly, we find that the decay rate of νNBT (k1) and the buildup rate of νI (k1’) are essentially the same (Supporting Information G), and that the a is close to zero, validating the 9



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

irreversible consecutive reaction step model, NBT I I’ (where I’ is another intermediate not detected in our measurement). We find that both the k1 and k2 scale near-linearly with laser power density (P) (Figures 3e-f), showing that both the first (k1) and second (k2) reaction steps are one-photon driven. For organic molecules without significant electronelectron correlation, one-photon absorption can only lead to single-electron transition. Therefore, we conclude that each of the reaction steps is a single-electron reduction process. We also find that the same ratio k2 / k1 = 0.29 (log10 k2 = log10 k1 -0.54) is maintained over 4 orders of magnitude variation in P (Figure 3g). With a fixed P, the rates also show a significant junction-to-junction variation (see, for example, data points in green circles obtained with a laser power density of log10 P = 3.64; See also Supporting Information H), yet the ratios of the rates remain close to k2 / k1 = 0.29 (i. e., the green data points in Figure 3g is located close to the fit line). Note that local surface heterogeneities may cause the separate variation in k1 and k2, yet it cannot lead to such linear relationship between k1 and k2. Instead, the junction-to-junction variation in plasmonic local field20 will affect in exactly the same way for the k1 and k2, providing a fixed k2 / k1 ratio. This proves that the electrontransfer is promoted by the localized plasmons of metallic gaps21, 32.

10


Page 11 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


Figure 4. (a) The electro-chemical potential-dependent SERS spectra obtained from a AgNP/NBT/AuTF junction (Supporting Information I). (b) Theoretical Raman spectra of NBT, anion radical of NBT (NBT•‒), conjugate acid of NBT (NBTH•), dianion of NBT (NBT2-), and dihydroxyaminobenzenethiol (DHABT), obtained from quantum chemical calculations (Supporting Information K) with basis sets of 6-311+G(d) (for geometry optimization), and 3-21G (frequency calculation). The peaks are convoluted with Lorentzian functions to simulate the experimental linewidth of 8 cm-1. All spectra are normalized to the peak intensity of corresponding νNO peak (red asterisks), and Raman frequencies are scaled by a constant factor of 0.952 to match the νNO of neutral NBT.

To confirm that the intermediate is the reduction product of NBT, we examine the change in SERS spectra of AgNP/NBT/AuTF junctions under a range of reductive electrochemical potential (Figure 4a). The AgNP/NBT/AuTF junction sample is immersed in 5 mM phosphate-buffered saline buffer (pH 7.4), and the working, reference, counter electrodes are placed into the solution. The SERS spectra and cyclic voltammograms (CVs) are simultaneously obtained while periodically sweeping the potential in the range of E = -0.2 V ~ -1.2 V (Supporting Information I). Application of a reductive potential can in general cause desorption of thiols from Au-substrates. For example, a recent study33 shows that a potential with E < -0.6 V causes significant desorption of methylbenzenethiols (MBTs) from Ausubstrates. The same study also reports that such desorption is locally blocked at AuNP/MBT/Au junctions, and the onset potential for desorption is shifted to E < -0.94 V. In our study, we find that the SERS νNO-peak intensities obtained from AgNP/NBT/AuTF junctions remain nearly constant (Figure S7c in Supporting Information I) during the entire 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 25

potential sweep cycles (-0.2 V~ -1.2 V), showing that the desorption of NBTs does not occur, or is locally blocked at junction sites. To minimize the photo-induced reduction, the laser power densities are kept at minimum level (P < 0.5 MW/cm2). With a fixed electro-chemical potential, the position of νNO peak was stationary. Upon applying reductive potential around 1.2 V (with reference to Ag/AgCl electrode), we start to observe a ~7 cm-1 of redshift in νNO, which is similar to the shift observed above (note that exact position of νNO, and the degree of peak-shift is slightly different from those in Figure 2, which is most likely caused by the hydration of nitro-groups34. See Supporting Information J). The original peak position was recovered through oxidative potential sweep. This strongly supports that the peak-shift arises from the reduction of NBT. We argue that the intermediate is the anion radical of nitrobenzenethiol (NBT•‒) or its conjugate acid (NBTH•) (see Figure 1b), based on the following three facts. First, the intermediate is formed through the single-electron reduction of NBT. Second, as is well known from previous studies8-10, the nitrobenzene anion radical (NB•‒) can be generated from NB by applying the reductive potential of -1.1 V, and is fairly stable under ambient condition, which is exactly what is observed in our measurement (Figures 2e and 4a). Thirdly, the redshift in νNO agrees with the general trends previously found in the IR spectra of NB•‒ and its structural analogs35-37. The LUMO of NB has a significant NO2 (π*) character, and thus the addition of an electron significantly reduces the bond-order of NO, leading to the red-shift in

νNO. To further verify our assignment, we carried out ab initio quantum chemical calculations (Supporting Information K) on the electronic ground states of NBT, NBT•‒, NBTH• (conjugated

acid

of

NBT•‒),

NBT2-

(dianion

of

NBT),

and

DHABT

(dihydroxyaminobenzenethiol), and simulated the corresponding Raman spectra (Figure 4b). 12


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


Indeed, NBT•‒ spectra show a small red-shift (~4 cm-1) in νNO with respect to that of neutral NBT, while the positions of other peaks remain the same. The NBTH•, the conjugate acid of NBT•‒, also shows similar features. On the other hand, two-electron reduction products of NBT, the NBT2- and DHABT, show νNO peaks that are blue-shifted with respect to that of NBT, and the spectra overall possess little resemblance to the spectra of the intermediate experimentally observed. The subsequent step is another single-electron reduction of NBT•‒ or NBTH• with a rate constant ~5 times smaller than that of initial step (k2 / k1 = 0.29). Currently, we cannot determine the exact product structures of the second reaction step, although the most plausible candidates are the NBT2- or DHABT. Delayed formation of DMAB (Figure 2c) arises from the “induction” time (τ ~ 1/ k1) required to form NBT•‒ (or NBTH•), and the actual buildup rate of DMAB is determined by the k2. The catalytic reduction in this work involves the electrons transferred from plasmonic structures. At this point, we are unable to tell whether or not the transferred electrons are the primary hot-electrons of plasmon excitation, or secondary, thermalized electrons. In addition, further studies are needed to determine if the anion radical intermediates will also play similar roles in the electro-catalytic reactions on metals and photo-catalytic reactions on semi-conductors. Intriguingly, we recognize similar peak-shifts in νNO from the spectra of a few previous reports38-40 on electro-catalytic reduction of NBT, although the authors did not explicitly mention, or attempt to characterize, the features. We are thus positive that the anion radical of NBT may also play a role in electro-catalytic reduction of NBT. To sum up, we have identified the anion radical of NBT as an intermediate of the first elementary step in the photo-catalytic reduction of NBT on metallic surfaces. Subsequent 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 25

single-electron reduction of the intermediate is found to be ~5 times slower than the first reduction step. Given that the intermediate and NBT have comparable Raman scattering cross-sections, apparent intensity decay of SERS νNO-peak mostly reflects the decay of anion radical, not the neutral NBT. This finding is particularly important because a majority of the kinetic SERS studies have relied on the implicit assumption that the νNO-intensity decay reflects the primary reduction step of neutral NBT, which, according to our result, is incorrect.

ASSOCIATED CONTENT Supporting Information: (a) Experimental details (b) Peak assignment for Raman and SERS spectra of NBT, ABT and DMAB (c) Raman spectra of microcrystalline NBTs on glass that are excited with HeNe laser (d) νNO peak shifts in AuNP/NBT/AuTF and NBT/AgTF sample (e) Reversible change of νNO peak position upon blocking of laser beam (f) Two-dimensional covariance analysis of νNO peak shift (g) Correlation between NBT decay rate (k1) and intermediate buildup rate (k1’) (h) Correlation between NBT decay rate (k1) and intermediate decay rate (k2). (i) Electrochemical potential-dependent SERS spectra of AgNP/NBT/AuTF junction (j) Spectral change in νNO of NBT induced by hydration (k) Detail of quantum chemical calculation method This material is available free of charge via the Internet http://pubs.acs.org.

ACKNOWLEDGMENT The work is supported by BioNano Health-Guard Research Center funded by the Ministry of Science, ICT & Future Planning (MSIP) of Korea as Global Frontier Project (H14


Page 15 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


GUARD_2013M3A6B2078947), and the Research Resettlement Fund for the New Faculty of SNU. REFERENCES (1) Tada, H.; Ishida, T.; Takao, A.; Ito, S.; Mukhopadhyay, S.; Akita, T.; Tanaka, K.; Kobayashi, H. Kinetic and DFT Studies on the Ag/TiO2-Photocatalyzed Selective Reduction of Nitrobenzene to Aniline. ChemPhysChem 2005, 6, 1537-1543. (2) Richner, G.; van Bokhoven, J. A.; Neuhold, Y.-M.; Makosch, M.; Hungerbuhler, K. In Situ Infrared Monitoring of the Solid / Liquid Catalyst Interface During the Three-Phase Hydrogenation of Nitrobenzene over Nanosized Au on TiO2. Phys. Chem. Chem. Phys. 2011, 13, 12463-12471. (3) Kimura, K.; Naya, S.-I.; Jin-nouchi, Y.; Tada, H. TiO2 Crystal Form-Dependence of the Au/TiO2 Plasmon Photocatalyst’s Activity. J. Phys. Chem. C 2012, 116, 7111-7117. (4) Naya, S.-I.; Niwa, T.; Kume, T.; Tada, H. Visible-Light-Induced Electron Transport from Small to Large Nanoparticles in Bimodal Gold Nanoparticle-Loaded Titanium(IV) Oxide. Angew. Chem. Int. Ed. 2014, 53, 7305-7309. (5) Wunder, S.; Polzer, F.; Lu, Y.; Mei, Y.; Ballauff, M. Kinetic Analysis of Catalytic Reduction of 4-Nitrophenol by Metallic Nanoparticles Immobilized in Spherical Polyelectrolyte Brushes. J. Phys. Chem. C 2010, 114, 8814-8820. (6) Zeng, J.; Zhang, Q.; Chen, J.; Xia, Y. A Comparison Study of the Catalytic Properties of Au-Based Nanocages, Nanoboxes, and Nanoparticles. Nano Lett. 2010, 10, 30-35.

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 25

(7) Herves, P.; Perez-Lorenzo, M.; Liz-Marzan, L. M.; Dzubiella, J.; Lu, Y.; Ballauff, M. Catalysis by Metallic Nanoparticles in Aqueous Solution: Model Reactions. Chem. Soc. Rev. 2012, 41, 5577-5587. (8) Geske, D. H.; Maki, A. H. Electrochemical Generation of Free Radicals and Their Study by Electron Spin Resonance Spectroscopy; the Nitrobenzene Anion Radical. J. Am. Chem. Soc. 1960, 82, 2671-2676. (9) Maki, A. H.; Geske, D. H. Electron Spin Resonance and Polarographic Investigation of Substituted Nitrobenzene Negative Ions. J. Am. Chem. Soc. 1961, 83, 1852-1860. (10) Smith, W. H.; Bard, A. J. Electrochemical Reactions of Organic Compounds in Liquid Ammonia. II. Nitrobenzene and Nitrosobenzene. J. Am. Chem. Soc. 1975, 97, 5203-5210. (11) Joseph, V.; Engelbrekt, C.; Zhang, J.; Gernert, U.; Ulstrup, J.; Kneipp, J. Characterizing the Kinetics of Nanoparticle-Catalyzed Reactions by Surface-Enhanced Raman Scattering. Angew. Chem. Int. Ed. 2012, 51, 7592-7596. (12) Xie, W.; Herrmann, C.; Kömpe, K.; Haase, M.; Schlücker, S. Synthesis of Bifunctional Au/Pt/Au Core/Shell Nanoraspberries for in Situ SERS Monitoring of Platinum-Catalyzed Reactions. J. Am. Chem. Soc. 2011, 133, 19302-19305. (13) Xie, W.; Walkenfort, B.; Schlücker, S. Label-Free SERS Monitoring of Chemical Reactions Catalyzed by Small Gold Nanoparticles Using 3D Plasmonic Superstructures. J. Am. Chem. Soc. 2013, 135, 1657-1660.

16


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


(14) Nagasawa, F.; Takase, M.; Nabika, H.; Murakoshi, K. Polarization Characteristics of Surface-Enhanced Raman Scattering from a Small Number of Molecules at the Gap of a Metal Nano-Dimer. Chem. Commun. 2011, 47, 4514-4516. (15) Sawai, Y.; Takimoto, B.; Nabika, H.; Ajito, K.; Murakoshi, K. Observation of a Small Number of Molecules at a Metal Nanogap Arrayed on a Solid Surface Using SurfaceEnhanced Raman Scattering. J. Am. Chem. Soc. 2007, 129, 1658-1662. (16) Kim, K.; Choi, J.-Y.; Shin, K. S. Photoreduction of 4-Nitrobenzenethiol on Au by Hot Electrons Plasmonically Generated from Ag Nanoparticles: Gap-Mode Surface-Enhanced Raman Scattering Observation. J. Phys. Chem. C 2015, 119, 5187-5194. (17) Shin, K. S.; Park, C. S.; Kang, W.; Kim, K. Effect of Macroscopically Smooth Silver Substrate on the Surface-enhanced Raman Scattering of 4-Nitrobenzenethiol Adsorbed on Powdered Au. Chem. Lett. 2008, 37, 180-181. (18) Han, S. W.; Lee, I.; Kim, K. Patterning of Organic Monolayers on Silver via SurfaceInduced Photoreaction. Langmuir 2002, 18, 182-187. (19) Kim, K.; Lee, Y. M.; Lee, H. B.; Park, Y.; Bae, T. Y.; Jung, Y. M.; Choi, C. H.; Shin, K. S. Visible Laser–Induced Photoreduction of Silver 4-Nitrobenzenethiolate Revealed by Raman Scattering Spectroscopy. J. Raman Spectrosc. 2010, 41, 187-192. (20) Choi, H.-K.; Park, W.-H.; Park, C.-G.; Shin, H.-H.; Lee, K. S.; Kim, Z. H. MetalCatalyzed Chemical Reaction of Single Molecules Directly Probed by Vibrational Spectroscopy. J. Am. Chem. Soc. 2016, 138, 4673-4684.

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 25

(21) Brandt, N. C.; Keller, E. L.; Frontiera, R. R. Ultrafast Surface-Enhanced Raman Probing of the Role of Hot Electrons in Plasmon-Driven Chemistry. J. Phys. Chem. Lett. 2016, 7, 3179-3185. (22) Park, W.-H.; Ahn, S.-H.; Kim, Z. H. Surface-Enhanced Raman Scattering from a Single Nanoparticle–Plane Junction. ChemPhysChem 2008, 9, 2491-2494. (23) Dong, B.; Fang, Y.; Chen, X.; Xu, H.; Sun, M. Substrate-, Wavelength-, and TimeDependent Plasmon-Assisted Surface Catalysis Reaction of 4-Nitrobenzenethiol Dimerizing to p,p′-Dimercaptoazobenzene on Au, Ag, and Cu Films. Langmuir 2011, 27, 10677-10682.

(24) Dong, B.; Fang, Y.; Xia, L.; Xu, H.; Sun, M. Is 4-Nitrobenzenethiol Converted to p,p′Dimercaptoazobenzene or 4-Aminothiophenol by Surface Photochemistry Reaction? J. Raman Spectrosc. 2011, 42, 1205-1206. (25) Sun, M.; Zhang, Z.; Zheng, H.; Xu, H. In-Situ Plasmon-Driven Chemical Reactions Revealed by High Vacuum Tip-Enhanced Raman Spectroscopy. Sci. Rep. 2012, 2, 647-650. (26) Van Schrojenstein Lantman, E. M.; Deckert-Gaudig, T.; Mank, A. J. G.; Deckert, V.; Weckhuysen, B. M. Catalytic Processes Monitored at the Nanoscale with Tip-Enhanced Raman Spectroscopy. Nat. Nanotechnol. 2012, 7, 583-586. (27) Van Schrojenstein Lantman, E. M.; de Peinder, P.; Mank, A. J. G.; Weckhuysen, B. M. Separation of Time-Resolved Phenomena in Surface-Enhanced Raman Scattering of the Photocatalytic Reduction of p-Nitrothiophenol. ChemPhysChem 2015, 16, 547-554.

18


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


(28) Harvey, C. E.; Weckhuysen, B. M. Surface- and Tip-Enhanced Raman Spectroscopy as Operando Probes for Monitoring and Understanding Heterogeneous Catalysis. Catal. Lett. 2015, 145, 40-57. (29) Kim, K.; Kim, K. L.; Shin, K. S. Photoreduction of 4,4′-Dimercaptoazobenzene on Ag Revealed by Raman Scattering Spectroscopy. Langmuir 2013, 29, 183-190. (30) Cohen, M. D.; Fischer, E. Isosbestic Points. J. Chem. Soc. 1962, 3044-3052. (31) Noda, I.; Ozaki, Y. Two-Dimensional Correlation Spectroscopy: Applications in Vibrational and Optical Spectroscopy; John Wiley & Sons, Ltd: New York, U. S. A.; 2005. (32) Sobhani, A.; Knight, M. W.; Wang, Y.; Zheng, B.; King, N. S.; Brown, L. V.; Fang, Z.; Nordlander, P.; Halas, N. J. Narrowband Photodetection in the Near-Infrared with a PlasmonInduced Hot Electron Device. Nat. Commun. 2013, 4, 1643-1648. (33) Ikeda, K.; Takeuchi, Y.; Kanamaru, K.; Suzuki, S.; Uosaki, K. Nanostructuring of Molecular Assembly Using Electrochemical Reductive Desorption of Locally Stabilized Thiol Monolayers. J. Phys. Chem. C 2016, 120, 15823-15829. (34) Schmid, E. D.; Moschallski, M.; Peticolas, W. L. Solvent Effects on the Absorption and Raman Spectra of Aromatic Nitrocompounds. Part 1. Calculation of Preresonance Raman Intensities. J. Phys. Chem. 1986, 90, 2340-2346. (35) Ezumi, K.; Miyazaki, H.; Kubota, T. Stretching Vibration of Nitro and N-oxide Groups of the Anion Radicals of 4-Nitropyridine N-oxide and Related Nitro Compounds. J. Phys. Chem. 1970, 74, 2397-2402.

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 25

(36) Ma, R.; Yuan, D.; Chen, M.; Zhou, M. Infrared Spectrum of Nitrobenzene Anion in Solid Argon. J. Phys. Chem. A 2009, 113, 1250-1254. (37) Steill, J. D.; Oomens, J. Spectroscopically Resolved Competition between Dissociation and Detachment from Nitrobenzene Radical Anion. Int. J. Mass spectrom. 2011, 308, 239252. (38) Cui, Q.; Yashchenok, A.; Li, L.; Möhwald, H.; Bargheer, M. Mechanistic Study on Reduction Reaction of Nitro Compounds Catalyzed by Gold Nanoparticles using In Situ SERS Monitoring. Colloids Surf. A Physicochem. Eng. Asp. 2015, 470, 108-113. (39) Li, J.; Wu, Y.; Sun, X.; Liu, J.; Winget, S. A.; Qin, D. A Dual Catalyst with SERS Activity for Probing Stepwise Reduction and Oxidation Reactions. ChemNanoMat 2016, 2, 786-790. (40) Zhang, J.; Winget, S. A.; Wu, Y.; Su, D.; Sun, X.; Xie, Z.-X.; Qin, D. Ag@Au Concave Cuboctahedra: A Unique Probe for Monitoring Au-Catalyzed Reduction and Oxidation Reactions by Surface-Enhanced Raman Spectroscopy. ACS Nano 2016, 10, 2607-2616.

20


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


Figure 1. (a) Schematic diagram of the AgNP/4-nitrobenzenethiol (NBT)/Au thin-film (AuTF) junction. The inset image (right) shows a representative SERS image monitoring νNO = 1342 cm-1, showing the spatial distribution of SERS hotspots. The inset spectrum (right) is a plasmon scattering spectrum obtained from a dark-field spectro-microscopy measurement. The laser excitation wavelength (λex) is also indicated in the spectrum. (b) Reaction steps and possible intermediates for the photo-catalytic reduction of NBT: NBT•- = anion radical of NBT; NBTH = conjugate acid of NBT•-; DHABT = dihydroxylaminobenzenethiol; NBT2- = dianion of NBT; ABT = aminobenzenethiol. 307x88mm (250 x 250 DPI)



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 2 (a) Time-resolved SERS spectra obtained from a AgNP/NBT/AuTF junction. Bars on the top indicates the vibrational frequencies of NBT (blue) and DMAB (orange). The arrows in blue, red, and orange point to the time (t = 0, 1.7 and 32 sec, respectively) at which the spectra in (b) is sampled. (c) The time traces of ν NO (red, NBT) and ν3 (orange, DMAB) peak intensities, together with a fit (black). (d) νNO (grey bars) and the fit to Gaussian function (black curve). Also shown in vertical arrows are the average νNO peak shift measured from AuNP/NBT/AuTF junctions (orange) and NBT/AgTF (cyan). (e) Influence of temporarily blocking the laser beam on the temporal evolution of νNO peak. 374x201mm (165 x 165 DPI)


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


Figure 3 (a) The zoom-in view of Figure 2a sampled around νNO and t = 0 ~ 7 sec. (b) The spectra of νNO peak sampled at t = 0.0, 0.2, 0.4, and 1.2 sec (blue, cyan, orange and red arrows in (a)). (c) The spectral decomposition of νNO (grey) peak into νI (red) and νNBT (blue) components. (d) The time traces of νNBT (blue circles) and νI (red circles) components, together with fits to kinetic model (inset). Also shown in grey (circles and line) is the sum of νNBT and νI components. Log-log plots of (e) laser power density (P) versus k1, (f) P vs k2, and (g) k2 vs k1 , which are derived from the kinetic analyses of trajectories similar to (d). Red lines are equations are the results of linear fits to the data. 515x261mm (125 x 125 DPI)



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 4. (a) The electro-chemical potential-dependent SERS spectra obtained from a AgNP/NBT/AuTF junction (Supporting Information I). (b) Theoretical Raman spectra of NBT, anion of NBT (NBT•-), conjugate acid of NBT (NBTH•), dianion of NBT (NBT2-), and hydroxylaminobenzenethiol (DHABT), obtained from quantum chemical calculations (Supporting Information K) with basis sets of 6-311+G(d) (for geometry optimization), and 3-21G (frequency calculation). The peaks are convoluted with Lorentzian functions to simulate the experimental linewidth of 8 cm-1. All spectra are normalized to the peak intensity of corresponding νNO peak (-red asterisks), and Raman frequencies are scaled by a constant factor of 0.952 to match the νNO of neutral NBT. 763x307mm (96 x 96 DPI)


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


Table of Contents Graphic 50x28mm (300 x 300 DPI)


Identification of the First Elementary Step in the Photocatalytic

Recommend Documents