Plasmon-Enhanced Autocatalytic N-Demethylation - ACS Publications

Subscriber access provided by READING UNIV

C: Plasmonics; Optical, Magnetic, and Hybrid Materials

Plasmon-Enhanced Autocatalytic N-Demethylation Tefera Entele Tesema, Christopher J. Annesley, and Terefe G. Habteyes J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.8b07078 • Publication Date (Web): 12 Aug 2018 Downloaded from http://pubs.acs.org on August 13, 2018

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 33 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

Plasmon-Enhanced Autocatalytic N-Demethylation Tefera E. Tesema1, Christopher Annesley2 and Terefe G. Habteyes1* 1

Departments of Chemistry and Chemical Biology, and Center for High Technology Materials,

University of New Mexico, Albuquerque, New Mexico 87131, USA 2

Air Force Research Lab, Kirtland AFB, New Mexico 87117, USA

ABSTRACT. Increasing experimental results indicate that optically excited plasmonic metal nanoparticles can drive photochemical reactions at photon flux comparable to that of solar radiation. However, experimental evidence that provides insight into the mechanism of the reactions on plasmonic surfaces has been limited. Here, using plasmon-enhanced Ndemethylation (PEND) of methylene blue (MB) as model reaction, we report mechanistic analysis of photochemical reactions on plasmonic gold nanoparticles under different adsorption and atmospheric conditions using surface enhanced Raman scattering (SERS) as operando spectroscopy to monitor the reaction as a function of exposure time to the light source. We found that in air and oxygen atmospheres and in the presence of co-adsorbed water molecules, MB undergoes photochemical N-demethylation to yield thionine (complete N-demethylation product) and other partial N-demethylation products that have distinct vibrational signatures. The product signals are negligible when the MB-particle system is illuminated in nitrogen atmosphere. Consistent with the well studied mechanism in solution, the PEND reaction appears to be initiated by singlet oxygen generated via energy transfer from the excited state of MB to oxygen

ACS Paragon Plus Environment

1

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

Page 2 of 33

molecule, and therefore the reaction may tentatively be described as an autocatalytic photochemical process. The results of this study provide an important insight that electronic excitations of adsorbates pumped by the localized surface plasmon field can lead to selective reaction pathways.

1. INTRODUCTION Localized surface plasmon resonances of metal nanoparticles can drive photochemical reactions1 and enhance spectroscopic signals2 simultaneously, combining surface chemistry and operando spectroscopy to obtain fundamental understanding of surface-molecule interaction and heterogeneous catalysis3-6 by probing the light-driven chemical and physical processes in-situ. Experimental evidence for photochemical reactions on optically excited plasmonic metal nanoparticles is increasing.7-19 The mechanism of the reactions on the plasmonic nanoparticles can vary depending on many factors. Excitation of surface plasmon resonances generates hot charge carriers20 that can initiate chemical reactions21,

22

and at the same time the plasmon

resonances can cause local heating23-26 that may assist the surface reaction.27-29 Surface bound ligands that are inherent to colloidal metal nanoparticles can play critical role in mediating charge transfer and reaction pathways.30 In cases where the plasmon resonance frequency overlaps with the electronic transition energy of the adsorbate, the intense plasmon local field can enhance the rate of photoexcitation of adsorbate states19, 31, 32 that can lead to generation of reactive species such as singlet oxygen,33-36 which is known to initiate chemical transformation of organic molecules.37,

38

It has also been reported that photoexcitation of small metal

nanoparticles produces singlet oxygen directly as well.39,

40

In addition, the involvement of

activated oxygen in plasmon driven photochemical reactions has been reported.11, 41


2

Page 3 of 33 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 mechanisms of photochemical reactions on plasmonic metal nanoparticles are mainly discussed in terms of hot electron transfer to adsorbates as detailed in several recent reviews.1, 21, 42, 43

Particle-molecule charge transfer processes have been studied intensively in the mechanistic

studies of surface enhanced Raman scattering (SERS), focusing on molecules that are chemically linked to the surface through specific atoms such as nitrogen, sulfur or oxygen.44-47 On the other hand, the involvement of hot electrons in photochemical reactions is broadly implied even when there is no certainty in the adsorption geometry. For example, resonant charge transfer excitation to unoccupied electronic state of methylene blue (MB) adsorbed on silver nanocube has been claimed based on an assumption that MB is chemically linked to the surface through the N-atom in the thiazine ring.48, 49 However, theoretical calculations show that MB adsorbs on gold and silver through weak dispersion forces, where the molecule orients flat and parallel to the surface.50-52 In addition, comparing the SERS spectra of MB at 532 nm and 785 nm, resonant charge transfer excitation to adsorbate states has been implied at the longer wavelength (785 nm) but not at shorter wavelength (532 nm) that has more overlap with the plasmon resonance.48, 49 On the other hand, it has recently been shown that MB adsorbed on plasmonic nanoparticles undergoes wavelength dependent N-demethylation,16 producing thionine that has a strong vibrational peak at ~479 cm-1 and moderately strong peak at ~804 cm-1 that do not exist in MB SERS spectrum.16, 53 In this regard, the vibration bands in the spectra obtained at 532 nm in references 48 and 49 appears to be mainly due to conversion of MB to thionine product during the SERS experiment. This is why the peak values obtained at 532 and 785 nm excitation wavelengths provided in Table 1 of reference 48 do not agree at all. Since resonant charge transfer excitation to adsorbate states is claimed at 785 nm excitation wavelength, and the SERS result at this excitation wavelength is characteristic of MB,48 one can conclude that the charge


3


Page 4 of 33

transfer does not necessarily result in chemical transformation, suggesting the critical importance of other factors and processes such as near-field enhanced electronic transition of adsorbates. In fact, it has recently been demonstrated that the dissociation of (CH3S)2 on gold and silver nanoparticles is initiated by plasmon pumped electronic transition of the adsorbate.19 In this work, we use plasmon enhanced N-demethylation of MB (PEND-MB) as model reaction to investigate the mechanism of photochemical reactions on resonantly excited plasmonic gold nanoparticles. In our first report, it has been shown that PEND-MB on gold nanoparticles takes place when the excitation energy overlaps with both the electronic transition energy of the molecule and the plasmon resonance of the nanoparticles.16 However, the reaction was studied in open air, and therefore systematic mechanistic study was difficult as it was impossible to confirm the involvement of molecules from the gas phase. Here, the mechanism of the photochemical N-demethylation on resonantly excited gold nanoparticles is investigated systematically by performing the reaction in reactive (air and oxygen) and inert (nitrogen) atmospheres at different adsorption conditions. We find that in air and oxygen atmospheres and in the presence of co-adsorbed water molecules, PEND-MB takes place, producing thionine and other intermediate N-demethylation products that exhibit distinct vibrational bands in the SERS spectra. The reaction does not take place in the absence of oxygen as evidenced by the negligible PEND product signals when the sample is exposed to laser in nitrogen atmosphere. Consistent with the N-demethylation initiated by thermally generated singlet oxygen in solution phase chemistry,38, 54, 55 the mechanism of the PEND-MB on gold nanoparticles may be initiated by singlet oxygen that can be generated via energy transfer from triplet excited state of MB to triplet ground state oxygen. Since the proposed mechanism involves generation of singlet oxygen by the MB reactant itself, it may be described as an autocatalytic chemical process. Only N-


4

Page 5 of 33 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


demethylation products are detected in our experiment, indicating that plasmon pumped electronic excitations of adsorbates can lead to selective reactivity by breaking specific chemical bonds. 2. EXPERIMENTAL METHODS Sample preparation. An aqueous solution of gold nanorods (AuNRs, ~40 nm diameter and ~85 nm length) with cetyltrimethylammonium bromide (CTAB) stabilizing surfactant was obtained from Nanopartz, Inc. Solid powders of methylene blue (MB), azure B, azure A and thionine were obtained from Sigma-Aldrich. After the excess CTAB surfactant was removed through one round of centrifugation, the AuNRs were re-suspended in 15 µM aqueous solution of either MB, azure B, azure A or thionine for at least 3 hours to ensure adsorption of the molecules on the AuNRs. Unabsorbed excess molecules were removed through one round of centrifugation. The AuNRs with the adsorbates were then re-suspended in ultrapure water and drop-casted on coverslip and dried at ambient condition. To understand whether co-adsorbed water molecules are involved in the photochemical reaction, adsorption of MB on AuNRs was also carried out in non-aqueous medium as follows. After the AuNRs were settled from the original solution through centrifugation and the supernatant was removed, they were re-suspended in freshly prepared solution of MB in anhydrous ethanol, and incubated for at least 3 hours. The solution was then centrifuged again to remove unadsorbed MB by discarding the supernatant. The solid residue was then dissolved in ethanol and aggregates of MB-AuNR were prepared on coverslip by drop-casting the solution and allowing drying at ambient condition. This procedure is not expected to completely remove water but it is sufficient to suppress the N-demethylation to demonstrate that whether water molecules are involved in the reaction or not.


5


Page 6 of 33

To directly adsorb MB on gold surface without the interference of surface ligands and investigate the effect of surface-molecule proximity, bare gold nanostructures were prepared by depositing gold of thickness less than 10 nm on glass substrate using electron-beam evaporation. At this thickness, the evaporation procedure inherently produces plasmonic nanostructures with broad resonances.56-58 The effect of surface-molecule proximity is tested by adsorbing MB directly on the bare gold nanostructures and on poly(sodium 4-styrenesulfonate) (PSS) monolayer by incubating the substrate inside 15 µM MB/ethanol solution overnight. The PSS monolayer, which is about 1 nm thick, was created by immersing the gold/substrate inside 2 mg/ml solution of PSS that is prepared by dissolving in 0.1 M aqueous solution of NaCl for five minutes. The film was then pulled out and rinsed thoroughly with ultrapure water and blown dry with nitrogen gas. Photochemical reaction and SERS measurement under controlled atmospheric condition. The photochemistry of MB on plasmonic surfaces is studied at room temperature in a closed chamber (Warner Instruments Inc., RC-21B,) that is sealed by bare coverslip from the top side and sample containing coverslip (with the sample side facing upward) from bottom side of the chamber as illustrated by the schematics in Figure 1a. The coverslips are placed on the recessed grooves to which vacuum grease is applied to create a tight seal of the coverslips. SERS measurements are performed at ambient atmosphere, or under continuous flow of pure O2 or N2 gases. At the chamber output, a length of tube is used as a back pressure control to ensure smooth flow of gases. The importance of water molecules in the gas phase is studied by expanding the N2 or O2 gases through a water bubbler filled with ultrapure water. The photochemical reaction is initiated using 633 nm excitation wavelength that overlaps with the plasmon resonances of the gold nanoparticles and electronic transition energy of MB. The


6

Page 7 of 33 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


reaction is monitored using SERS signal of the reactant and product species by collecting the scattered light through the coverslip using 0.7 NA objective. About 90% of the signal collected from the sample is directed to a spectrometer (IsoPlane Spectrograph of Princeton Instruments) that uses thermoelectrically cooled (−75 °C) and back-illuminated deep depletion CCD camera, while the remaining 10% of the signal is directed to the Olympus UC30 camera attached to the GX51 microscope for monitoring the focus and inspecting the sample. The comparison of Stokes and anti-Stokes Raman intensities was carried out using our atomic force /near-field microscope that has been described elsewhere59, 60 but the same spectrometer mentioned above is used for detection. 3. RESULTS As mentioned above, the photochemical reaction is initiated using 633 nm excitation wavelength that overlaps with the resonances of the nanoparticles and molecules as illustrated in Figure S1 in the Supporting Information (SI). It is known that adsorption of molecules on metal surfaces causes red-shift and broadening of electronic absorption bands with respect to that in solution and isolated molecules.47, 61, 62 These adsorption effects can result in more significant overlap of the excitation wavelength with the absorption band of thionine than shown in Figure S1. The overlap of the excitation wavelength with the MB absorption band is also expected to remain significant upon adsorption56, 63 due to the significant broadening.47, 62 As a result, the MB reactant and thionine product species can be monitored with high sensitivity by taking advantage of the large SERS cross-section that results from resonance Raman and electromagnetic enhancement effects.64 We note that the strongest vibrational band of thionine at 479 cm-1 is completely absent in the MB spectrum as shown in Figure 1b, providing high


7


contrast of reactant and product signals. In addition, the vibrational spectra of azure B and azure A, display characteristic features (see Figure 1b) that can be used to identify these species if they MB adsorbed on gold nanostructures

(a)

coverslip

Gas in

Gas out

coverslip Excitation light (λ = 633 nm)

Raman scattering

(b) Thionine Powder SERS

Azure A Powder SERS

Normalized intensity

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 33

Azure B Powder SERS

Methylene blue Powder SERS

600

800

1000

1200

1400

1600

-1

Raman shift (cm )

Figure 1. (a) Schematic showing the experimental setup of the photochemical reaction under the flow of different gases over the solid sample (MB adsorbed on gold nanostructures). (b) Normal Raman spectra of solid powders (black lines) and SERS spectra (red lines) of methylene blue, azure B, azure A and thionine as labeled. All the spectra are acquired using 633 nm excitation wavelength, one after the other at the same optical settings so that comparison of peak positions is reliable. The vertical dashed lines are included to facilitate the comparison of vibrations frequencies for the different molecules.


8

Page 9 of 33 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


are formed due to partial N-demethylation of MB. These spectral contrasts between reactant and product species are used to investigate the effects of atmospheric and adsorption conditions as follows. We note that, in addition to the vibrational signals, the background intensity changes with exposure time as illustrated in Figure S2. In the data analysis, the background is subtracted as described in the SI. Atmospheric condition: The temporal evolution of the spectra obtained under ambient condition as well as the flow of oxygen and nitrogen gases are compared in Figure 2. The vibrational frequencies at 479 cm-1 (due to skeletal deformation mode of thionine) and 804 cm-1 (due to NH2 rocking vibration of thionine) at which prominent peaks are expected due to conversion of MB to thionine through PEND reaction are indicated by the red asterisks in Figures 2a-c. We will be using these frequencies to identify the peaks throughout this analysis. Under ambient air and oxygen atmosphere, continuous exposure of the sample results in the appearance and growth of the vibrational bands at 479 cm-1 and 804 cm-1 (Figures 2a and b), while their intensities are within the background level in nitrogen atmosphere (Figure 2c). These observations indicate that oxygen is involved in the PEND-MB reaction. In addition to the appearance of prominent new peaks at 479 cm-1 and 804 cm-1, the vibrational band that initially peaks at 1435 cm-1 appears to red-shift with exposure time, and the shift is accompanied by relative intensity increase depending on the atmospheric condition as shown in the insets in Figures 2a-c. Interestingly, although the vibrational signatures of complete N-demethylation appears to be absent in N2 atmosphere, the frequency shift for the 1435 cm-1 band is appreciable (see Figure S3), suggesting that the peak position for this band is sensitive to partial N-demethylation reaction that may take place due to trace amount of oxygen molecules present as adsorbates. This frequency shift may also arise from light-induced surface-molecule


9


Page 10 of 33

interaction but this effect may not be significant based on the excellent agreement of peak positions in the SERS and normal Raman spectra as shown in Figure 1b. We note that the 479 cm-1 and 804 cm-1 bands can be observed only after an NH2 functional group is formed at least at one end of MB,16 which involves removal of two methyl groups. The integrated intensities of representative vibrational bands are plotted as a function of time as shown in Figures 2d-f. The 1392 cm-1 band has similar intensity in both MB and thionine spectra, and can be used as a reference to observe the rise of vibrational band intensities that are unique to thionine product. As can be seen in Figures 2d-e, the intensity of thionine product signal increases with time and becomes higher than that of the 1392 cm-1 band after about 40 s of exposure in air and oxygen. In nitrogen atmosphere, the relative intensity at 479 cm-1, where the strongest vibrational band of thionine is expected, remains within the background level during the continuous illumination of the sample (Figure 2f), confirming that conversion of MB to thionine is negligible in the absence of oxygen. Additionally, the rise of the thionine product signal is correlated with the decline of the MB reactant signal as a function of exposure time in different atmospheres and is shown in Figure S4. As mentioned above, the relative intensity of the 1435 cm-1 band increases during the first 30 seconds of exposure to the incident laser and then decreases slowly or remains constant depending on the atmospheric conditions (see the plot with enlarged scale in Figure S5). This vibrational mode has recently been assigned to the in-plane NCH bending vibration of MB,65 and therefore, should disappear upon complete N-demethylation of MB (thionine formation). Hence, as suggested above, the change in frequency and intensity of this mode with exposure time can be attributed to partial N-demethylation products that include azure B, azure A and azure C.


10

Page 11 of 33

40

Integrated relative intensity of representative vibrational bands (arb. unit)

(a) Air 0.5 s 30 s 100 s 200 s

*

1400

Relative SERS intensity (arbitrary unit)

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


1440

1480

*

*

(b) Oxygen 0.5 s 30 s 100 s 200 s

*

1400

1440

1480

*

*

(c) Nitrogen 0.5 s 30 s 100 s 200 s 1400

1440

1480

*

(d) Air -1

479 cm

35 30 25

1392 cm

20

-1

-1

1435 cm 15

479 cm

(e) Oxygen

25

-1

20 1392 cm 15

-1

-1

1435 cm

10 (f) Nitrogen

1392 cm

-1

1435 cm

-1

20

15

* * 600

800

479 cm

1000

1200

1400

1600

-1

Raman shift (cm )

0

40

80

120

160

-1

200

Time (s)

Figure 2 Temporal evolution of the SERS signal of MB adsorbed on gold nanorods and illuminated with 0.4 mW of 633 nm CW laser (focused with 0.7 NA objective) in Air (a, d), oxygen (b, e) and nitrogen (c, f) atmospheres. The SERS spectra are acquired continuously with 0.5 s acquisition time for over 200 s of continuous illumination, and the black, green, red and blue lines in (a) – (c) represent average spectra of at least ten different spots on the same sample at different exposure times as labeled. The integrated intensities extracted from 400 spectra acquired within 207 s are plotted in (d) – (f) for representative vibrational bands. In order to obtain further experimental evidence for the formation of partial Ndemethylation products, the photochemical reactions in different atmospheres have been repeated at lower incident laser power (0.2 mW) and shorter acquisition time (0.1 s) than used to obtain the data in Figure 2 (0.4 mW and 0.5 s), recording the spectra continuously for 120 s. The results obtained from these measurements are summarized in Figure 3. In air and oxygen


11


atmospheres, the relative intensity of the 1435 cm-1 band increases with exposure time, while that of the other bands remain nearly constant in air and decrease slightly in oxygen as shown in Figures 3a-b. Again, the intensity increase for the band that initially peaks at ~1435 cm-1 unlike any of the other bands is accompanied by more significant frequency red-shift depending on the atmospheric condition as shown in Figures (3d-f). For the 1435 cm-1 band, about ~5 cm-1 redshift is observed in air and oxygen, compared to about 3 cm-1 overall red-shift in nitrogen atmosphere within the 120 s exposure time at the given incident laser power. The magnitudes of the shifts increase with incident laser power and exposure time. For example, at 0.4 mW incident laser power, the 1435 cm-1 band red-shifts by ~7 cm-1 within 200 s of illumination as can be seen in Figure S3.

Normalized intensity

1.4

(a) Air

1435 cm

(b) Oxygen

-1

(c) Nitrogen

1.2 1392 cm

-1

1620 cm

-1

1.0

0.8

(d) Air

1620 cm

-1

(e) Oxygen

(f) Nitrogen

0 -1

∆ν (cm )

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 33

1392 cm

-1

1435 cm

-1

-2

-4

-6 0

40

80

Time (s)

120

0

40

80

120

Time (s)

0

40

80

120

Time (s)

Figure 3. (a) – (c) Integrated and normalized relative intensities of representative vibrational bands plotted as a function of exposure time in (a) air, (b) oxygen and (c) nitrogen atmospheres. (d) – (f) Vibrational frequency shift of the different bands as labeled based on the initial peak frequencies. The vibrational frequencies are determined by fitting a Gaussian function to the spectra. The integrated intensities and frequencies are extracted from 1000 spectra each recorded with 0.1 s acquisition time within 120 seconds of continuous exposure time using 0.2 mW of 633 nm CW laser that is focused with 0.7 NA objective.


12

Page 13 of 33 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


In nitrogen atmosphere, the relative intensities of all the bands decrease gradually with exposure time as shown in Figure 3c. However, it is interesting to note that the intensity decline for the 1435 cm-1 band is the slowest, while the decline rates for 1392 cm-1 and 1620 cm-1 bands are reversed in nitrogen atmosphere from that in air and oxygen. The reversal of the intensity trend for the 1392 cm-1 and 1620 cm-1 is also observed in the frequency shift going from air/oxygen to nitrogen. The 1620 cm-1 vibration mode (stretching vibrations of the CC and CN bonds of the fused aromatic ring)65 slightly blue-shifts in air, remains nearly constant in oxygen, and red-shifts in nitrogen atmosphere. In addition, the 1392 cm-1 mode (the CN stretching vibration of the –N(CH3)2 functional groups coupled to the CN and CC stretching vibration of the fused aromatic ring)65 red-shifts in air and oxygen compared to slight blue-shift in nitrogen atmosphere. That is, in nitrogen atmosphere, the 1392 cm-1 and the 1620 cm-1 bands shifts in opposite direction by about the same magnitude, albeit the shifts are very small, as shown in Figure 3f. This trend is reproducible at low laser intensity in nitrogen atmosphere (see results in Figure S6 for a different sample), and it may be attributed to light-induced surface molecule interaction45 and stark effects.66 The effect disappears when the incident laser power is increased (Figure S3) possibly due to the overwhelming effect of the photochemical N-demethylation. For the 1435 cm-1 band, partial N-demethylation and light-induced surface molecule interaction may shift the peak position in the same direction. However, as mentioned earlier, based on the agreement between the peak positions in the SERS and normal Raman spectra of solid powders (Figure 1b), the frequency shift due to light-induced surface molecule interaction may not be significant. Therefore, the substantial frequency shift observed in Figure 3f for the band that initially peak at 1435 cm-1 can be attributed to partial N-demethylation due to oxygen molecules that can be present as adsorbates. If there was strictly no demethylation reaction occurring, the


13


Page 14 of 33

shifts for the bands that initially peak at 1392 cm-1 and 1620 cm-1 might be more significant than observed in Figure 3f. To confirm that the observed relative intensity increase and frequency shift is due to partial N-demethylation, the SERS spectra of MB and azure B (that can be formed when one CH3 group of MB is replaced by H atom) recorded in N2 atmosphere are compared in Figure 4a. As illustrated on the figure, replacing one of the methyl groups of MB with hydrogen results in the peak frequency red-shifting for the bands that appear at 1392 cm-1 and 1435 cm-1 for MB, which is in good agreement with the frequency shifts observed due to photochemical conversion of MB in Figures 3 and S3 in air and oxygen. For the 1435 cm-1 band, the frequency downshift going from MB to azure B is accompanied by relative intensity increase, again in agreement with the observations in Figure 3. Interestingly, the very small blue shift observed as a function of exposure time in Figure 3d for the 1620 cm-1 band is discernible in Figure 4a comparing the red and black spectra. Further N-demethylation does not appear to change the peak positions significantly but it increases the relative intensity of the 1426 cm-1 band and broadens the 1620 cm-1 band as can be seen comparing the Raman spectra of azure B and azure A presented in Figure 1b. In general, the temporal evolution of the bands in the 1360 – 1450 cm-1 region can be attributed to partial N-demethylation with possible contribution from surface-molecule interaction. Further insights into the origin of the frequency shift and intensity rise can be obtained by investigating the adsorption conditions.


14

(a) -1

1426 cm

Azure B

-1

1435 cm

MB 1400

1500

1600 -1

Raman shift (cm )

(b) Intensity ratio (1426/1392)

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


Intensity (arb. unit)

Page 15 of 33

1.0

MB

AZ B

AZ A

0

1

2

Thionine

0.5 3

4

# CH3 replaced with H Figure 4 (a) SERS spectra of methylene blue (black line) and azure B (red line) recorded at 0.4 mW of 633 nm CW laser and 0.5 s acquisition time in nitrogen atmosphere. The spectra are recorded one after another at the same optical settings so that exactly the same reference point is used in determining the Raman shifts for the two compounds. (b) Ratio of the intensity at 1426 cm-1 to the peak intensity at ~1392 cm-1 (MB) or ~1386 cm-1 (azure B, azure A and thionine). The spectra for azure A (AZ A) and thionine are given in Figure 1B. Adsorption condition: surface-molecule proximity. In all the experimental results presented so far, MB is adsorbed on gold nanorods coated with CTAB surface ligand. To directly adsorb MB on gold surface and evaluate the effect of proximity to the surface, bare gold nanostructures are prepared using electron-beam evaporation to deposit a targeted thickness of less than 10 nm. At this thickness, it is well known that the evaporation procedure inherently produces plasmonic nanostructures with broad resonances56-58 as shown in Figures S8 and S9. In Figure 5, the results obtained for MB adsorbed directly on bare gold nanostructures (MB-AuNS)


15


Page 16 of 33

is compared with that obtained by adsorbing on similar gold nanostructures but coated with monolayer of poly(sodium 4-styrenesulfonate) (MB-PSS-AuNS). For MB-AuNS sample, two peaks at 1421 cm-1 and 1435 cm-1 are observed, and the relative intensity of the former increases with exposure time. For the MB-PSS-AuNS sample, the 1421 cm-1 has shifted to 1426 cm-1, and its relative intensity increases with exposure time. Based on the results presented in Figures 4a, the 1426 cm-1 peak position can be assigned to azure B or C, and the red-shift to 1421 cm-1 for the MB-AuNS sample can be attributed to surface-molecule interaction because of direct coupling of the molecule to the gold surface. As shown in Figure 5b, the 1421 cm-1 to 1392 cm-1 peak intensity ratio for the MB-AuNS sample (open circles) is higher than the corresponding peak intensity ratio for the MB-PSS-AuNS sample (solid circles). On the other hand, the 479 cm1

to 446 cm-1 peak intensity ratio is higher for the MB-PSS-AuNS sample (solid triangles) than

for the MB-AuNS sample (open triangles), indicating that conversion of MB to thionine is favored in the presence of spacer layer. The fact that the 1421/1426 cm-1 and 1435 cm-1 peaks are resolved in Figure 5 indicates that the apparent red-shift from the initial 1435 cm-1 peak position (observed in Figures 2 and 3) is in fact due to the appearance of another peak at ~1426 cm-1 due to the formation of intermediate Ndemethylation products. In addition, the 1426 cm-1 peak position for the MB-PSS-AuNS sample agrees with the corresponding peak position for the MB adsorbed on AuNRs, for which CTAB surface ligand serves as natural spacer.


16

Page 17 of 33

(a) Intensity (arb. unit)

MB-PSS-AuNS 207 s 0.5 s p1

p2

p3

p4

MB-AuNS 207 s 0.5 s

440

480

1360

1400

1440

1480

-1

Raman shift (cm ) p2/p1 (MB-PSS-AuNS)

(b) p4/p3 (MB-AuNS)

1.2

Intensity ratio

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


p2/p1 (MB-AuNS) 0.8 p4/p3 (MB-PSS-AuNS)

0.4 0

50

100

150

200

Time (s)

Figure 5 Spacer layer effect. (a) Representative spectra showing the relative intensities of the peaks labeled p1 – p4 at the beginning (black line) and end (red line) of exposure for 207 seconds in the presence (upper two spectra) and absence (lower two spectra) of poly(sodium 4styrenesulfonate (PSS) coating on e-beam evaporated gold nanostructures (AuNS) as labeled. Each spectrum represents average of at least 10 different spots on the same sample that is illuminated at 633 nm excitation wavelength and 0.4 mW incident laser power in air. (b) Peak intensity ratios extracted from the 400 spectra acquired during the 207 seconds of continuous illumination. Notice that the relative intensity of thionine signal p2 with respect to p1 is higher in the presence of PSS spacer layer, while the opposite is true for the intermediate signal p4 with respect to p3. Adsorption condition: hydration effect. The results presented so far are for MB adsorbed on the colloidal gold nanorods and evaporated gold nanostructures in aqueous solution, which could result in the adsorption of hydrated MB in the dried sample. Adsorbed water may be involved in the photochemical N-demethylation reaction. To obtain experimental evidence for


17


Page 18 of 33

the involvement of water molecules in the reaction, the water molecules in the surface-molecule complex have been minimized by suspending both the molecule and the gold nanorods in ethanol during the solution processes and the results are summarized in Figure 6. Comparing the relative intensity of the bands at 479 cm-1 and 804 cm-1 in Figure 6a (obtained under a flow of oxygen over dehydrated surface-molecule complex) to the corresponding intensities in Figure 2b (obtained under the flow of oxygen over hydrated surfacemolecule complex), it can be seen that the N-demethylation reaction is suppressed significantly when the surface-molecule complex is dehydrated. In addition, the frequency shift of the 1435 cm-1 band is negligible as shown by the intensity map in the inset of Figure 6a. To further understand the hydration effect, we repeat the experiment after introducing water molecules in the gas phase by expanding oxygen gas through a water bubbler, and representative spectra obtained under this condition are plotted in Figure 6b. Like in Figure 6a, the vibrational signatures of thionine remain very weak. The peak intensities plotted in Figure 6c-d show that the relative intensity of the 479 cm-1 band remains significantly lower than the intensity of the 1392 cm-1 band throughout the exposure time. This trend is in contrast to that observed in Figures 2d & e, where the relative intensity of the 479 cm-1 band becomes larger than that of the 1392 cm-1 band after some time of exposure. In addition, the frequency shifts for the bands in the 1360 – 1450 cm-1 region is negligible as can be seen in the insets in Figures 6a-b, suggesting that partial N-demethylation is also suppressed in the absence of water. The fact that vibrational signatures of N-demethylation remain weak even in the presence of water molecules in the gas phase suggests that it is the hydrated form of MB that undergoes N-demethylation reaction. In hydrated MB, water molecules are likely to be in the proximity of the –N(CH3)2 functional group


18

Page 19 of 33

because of a favorable electrostatic interaction, and can be the source of hydrogen atoms that replace the methyl groups. (a) O2 gas

(c) O2 gas 2.0

0.5 s 30 s 100 s

-1

1392 cm 1400

1450

1.5

1500

-1

Relative peak intensity

1435 cm

Relative SERS intensity

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


(b) O2/H2O gas 0.5 s 30 s 100 s

1400

1450

1.0

0.5 -1

479 cm 1.5

(d) O2/H2O gas -1

1392 cm

1500

1.0 -1

479 cm

0.5

-1

1435 cm 600

800

1000

1200

1400

1600

0

-1

Raman shift (cm )

50

100

150

200

Time (s)

Figure 6. Temporal evolution of the SERS signal for dehydrated MB-AuNR complex. The amount of adsorbed water is minimized by using ethanol solution of MB and AuNRs during the sample preparation. Representative spectra acquired under the flow of (a) dry O2 gas, and (b) O2/H2O gas formed by passing the O2 gas through a water bubbler. (c, d) Relative peak intensities extracted from the spectra acquired under the flow of (c) dry O2 gas, and (d) O2/H2O gas.

4. DISCUSSION From the results presented in Figures 2 and 6, we conclude that O2 and H2O molecules are involved in the N-demethylation of MB. The results in Figures 3 and 4 indicate signatures of partial N-demethylation products. The results in Figure 5 suggests that adsorption of MB directly on bare plasmonic gold surface favors formation of partial N-demethylation over complete Ndemethylation of MB. In addition, it has been shown that conversion of MB to thionine on gold


19


Page 20 of 33

nanoparticles takes place at 633 nm excitation wavelengths but not at 532 nm and 808 nm excitation wavelengths, where only the 633 nm excitation overlaps with the MB electronic transition.16 Considering that MB is a well known photosensitizer,67 this observation may suggest the involvement of singlet oxygen in the photochemical N-demethylation. Recent experimental observations indicate that plasmonic nanoparticles enhance the efficiency of photosensitizers to generate singlet oxygen.34-36 On the other hand, it has been shown that thermally generated singlet oxygen (1O2) drives N-demethylation reaction.38, 54, 55 Together, these results show that it is likely that PEND-MB follows a similar mechanism. As illustrated in Figure 7, photoexcitation promotes MB from its singlet ground state (S0) to singlet excited state (S1) that undergoes intersystem crossing to its excited triplet state (T1). The energy transfer from MB (T1) to oxygen in its triplet ground state (3O2) can generate the reactive 1O2. That is, the plasmon field enhances the photochemical N-demethylation by pumping the electronic excitation of the MB adsorbate that leads to enhanced generation of 1O2. This suggestion is supported by the experimental observation presented in Figure 5 that shows enhanced thionine product signal in the presence of PSS spacer. This follows the work of Nitzan and Brus, where the PSS spacer creates optimal surface-molecule proximity to minimize competing radiative and non-radiative processes.31 In addition, intersystem crossing to triplet states, which have relatively long lifetime, can reduce the competition of nonradiative processes as suggested by Wolkow and Moskovits.32 As a result, plasmon-enhanced electronic excitation of MB adsorbate can lead to enhanced singlet oxygen generation that can interact with MB to form a charge-transfer complex (exciplex).68, 69 H-atom transfer within the complex leads to formation of a radical species, which in the presence of water leads to N-demethylation and other byproducts as shown in Reaction 1 in-line with the solution phase mechanism proposed by Lapi et al.55


20

Page 21 of 33

(1) where only one of the methyl groups is shown for simplicity. Z+ represents the rest of the cationic structural constituent of MB including the fused ring system and the two methyl groups attached to the N atom on the other end of the molecule.

Energy

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


Metal nanoparticle

Hot electrons Energy transfer Plasmon field

Adsorbate states

Singlet (S1)

Triplet (T1) 1O 2

EF

3

O2

Singlet (S0)

Figure 7 Schematic showing possible photophysical processes for MB on gold nanoparticle when the excitation energy is in resonance with the particle plasmon resonance and MB adsorbate electronic transition. The plasmon near-field of the particle pumps the S0 → S1 electronic transition of the MB adsorbate (blue lines). S1 to T1 intersystem crossing (purple arrow) can populate the MB T1 state, from which energy transfer (green arrows) can promote oxygen from its triplet ground state (3O2) to singlet excited state (1O2). The black arrows indicate energy transfer to the metal surface, and red arrows indicate hot electron transfer to the unoccupied adsorbate states.

In addition to enhanced local field, the surface provides steric hindrances that may assist the photochemical N-demethylation. In the N-demethylation of α-methyl substituted Nmethylpiperidines initiated by thermally generated singlet oxygen in acetonitrile, it has been


21


observed that the ratio of secondary amine to N-formyl derivatives increase almost linearly with increasing number of α-methyl groups, underscoring the importance of the steric effect for the reaction.55 In our experiment, no vibrational signatures of N-formyl derivatives have been detected, which suggests N-demethylation on plasmonic surfaces is selective. For MB adsorbed on surfaces, N-demethylation minimizes the steric effect by replacing the relatively bulky methyl groups with H-atoms and improves surface-molecule interaction. The calculated adsorption geometry of MB and its N-demethylated derivatives on Au (111) shows that the molecules orient parallel to the surface with the separation between the center of mass of the molecules and the surface decreasing from 3.31 Å (MB) to 3.24 Å (thionine) upon N-demethylation.52 In addition, the calculated adsorption energies indicate that N-demethylation of MB adsorbate is thermodynamically feasible. That is, every replacement of CH3 group with H atom results in more favorable surface-molecule interaction, leading to a cascade of photochemical reactions until all the methyl groups are replaced by hydrogen atoms as depicted in Figure 8, which is based on the results of the theoretical calculations of Zhou et al.52

0

∆Eadsorption (J/g)

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 22 of 33

MB Az B

-20 Az A -40

Az C

-60

Thionine 0

1

2

3

4

# CH3 group replaced by H

Figure 8 Relative adsorption energy per gram for methylene blue (MB), azure B (AZ B), azure A (AZ A), azure C (AZ C) and thionine based on the theoretical results in reference 50.


22

Page 23 of 33 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


Other possible mechanisms may include hot electron transfer to adsorbed species including oxygen and MB. Transient anionic oxygen created via hot plasmon electron injection has been implicated in the surface photochemistry of para-aminothiophenol41, 70, 71 as well as in ethylene epoxidation.11 For PEND-MB on gold nanoparticles, this possibility can be ruled out based on the absence of product signal at 532 nm and 808 nm excitation wavelengths.16 Other studies have implicated direct charge excitation in MB-silver complex has been proposed based on the unusually large anti-Stokes to Stokes Raman intensity ratio at 785 nm excitation wavelength, although N-demethylation of MB has not been observed at this excitation wavelength.48, 49 In our experiment on MB-gold system at 633 nm excitation wavelength, the anti-Stokes intensity has always been much weaker than that of the Stokes as shown in Figure S7. In addition to the excitation wavelength difference, this discrepancy can be attributed to different adsorption properties of MB on gold and silver surfaces with different surface ligands. In general, charge transfer processes are important when the adsorbate is chemically linked to the surface.45, 47 In the case of MB, the adsorption can be described as physisorption based on the agreement between the vibrational peak positions in the SERS and normal Raman spectra (see Figure 1B), which is consistent with the theoretical predictions of Zhou et al.52 The presence of additional spacer layers such as surface ligands is expected to reduce the product signal drastically if the reaction is driven by charge transfer processes.72 In contrast to this expectation, the results presented in Figure 5 shows that the thionine product signal increases in the presence of PSS spacer layer, which strongly supports that the N-demethylation reaction is driven by near-field enhanced electronic excitation of MB in-line with the earlier proposals of plasmon enhanced photochemistry.31, 32, 73


23


Page 24 of 33

However, signatures of light-induced surface molecule interaction and partial charge transfer processes are apparent in the SERS spectra acquired as a function of exposure time. Specifically, the fact that, in N2 atmosphere (Figure 3f), the 1392 cm-1 and 1620 cm-1 vibration modes shift in opposite direction to what is expected from N-demethylation chemical conversion effect indicates the importance of light-induced surface-molecule interaction. The interaction may cause delocalization of the π-electron density of the fused aromatic ring. This charge delocalization can strengthen the 1392 cm-1 vibration mode, which is due to the stretching vibration of the CN attached to the methyl groups coupled to the stretching vibration of the fused aromatic ring. In contrast, the 1620 cm-1 ring vibration mode can be softened as the electron density is pulled from the aromatic system.74 This light-induced partial charging of the molecule may assist the photochemical reactions by activating the surface-molecule complex to readily undergo chemical transformation. In addition, the fact that the PEND-MB reaction is slow may indicate the importance of local heating, though it may be small,25 due to dissipation of energy from the resonant excitation of the molecule and particle resonances. 5. CONCLUSION The mechanisms of a plasmon enhanced photochemical reaction, N-demethylation of methylene blue, is investigated under different atmospheric and adsorption conditions using SERS as operando spectroscopy. We found that, in the presence of oxygen in the atmosphere and water molecules as adsorbates, MB undergoes photochemical N-demethylation to produce thionine and other intermediate species that have distinct vibrational signatures. The mechanism of the reaction appears to involve singlet oxygen generated via energy transfer from MB excited state to oxygen molecule. The localized plasmon field enhances the generation of singlet oxygen by pumping the electronic transition of methylene blue. Singlet oxygen can react with MB to


24

Page 25 of 33 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


form an exciplex that can transform MB to thionine and partial N-demethylation products in the presence of water. Unlike random degradation of organic molecules on traditional semiconductor catalysts, the photochemical reaction observed in this work is selective to N-demethylation of MB indicating that electronic excitations of adsorbates pumped by localized surface plasmon field can lead to selective reaction pathways. ASSOCIATED CONTENT Supporting Information The Supporting Information is available free of charge on the ACS Publications website. Data analysis procedures and nine supporting figures with captions are available. AUTHOR INFORMATION Corresponding Author *[email protected] Notes The authors declare no competing financial interests. ACKNOWLEDGMENT This research has been supported by the U.S. Air Force Office of Scientific Research, Grant No. FA9550-15-1-0305 and task number 18RVCOR121.

REFERENCES 1. Zhang, Y.; He, S.; Guo, W.; Hu, Y.; Huang, J.; Mulcahy, J. R.; Wei, W. D., SurfacePlasmon-Driven Hot Electron Photochemistry. Chem. Rev. 2018, 118, 2927-2954. 2. Zrimsek, A. B.; Chiang, N. H.; Mattei, M.; Zaleski, S.; McAnally, M. O.; Chapman, C. T.; Henry, A. I.; Schatz, G. C.; Van Duyne, R. P., Single-Molecule Chemistry with Surface- and Tip-Enhanced Raman Spectroscopy. Chem. Rev. 2017, 117, 7583-7613. 3. Bjerneld, E. J.; Svedberg, F.; Johansson, P.; Kall, M., Direct Observation of Heterogeneous Photochemistry on Aggregated Ag Nanocrystals Using Raman Spectroscopy: The Case of


25


Page 26 of 33

Photoinduced Degradation of Aromatic Amino Acids. J. Phys. Chem. A 2004, 108, 41874193. 4. Joseph, V.; Engelbrekt, C.; Zhang, J. D.; Gernert, U.; Ulstrup, J.; Kneipp, J., Characterizing the Kinetics of Nanoparticle-Catalyzed Reactions by Surface-Enhanced Raman Scattering. Angew. Chem.-Int. Edit. 2012, 51, 7592-7596. 5. 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. 6. Choi, H.-K.; Park, W.-H.; Park, C.-G.; Shin, H.-H.; Lee, K. S.; Kim, Z. H., Metal-Catalyzed Chemical Reaction of Single Molecules Directly Probed by Vibrational Spectroscopy. J. Am. Chem. Soc. 2016, 138, 4673-4684. 7. Goncher, G. M.; Parsons, C. A.; Harris, C. B., Photochemistry on Rough Metal-Surfaces. J. Phys. Chem. 1984, 88, 4200-4209. 8. Jeong, D. H.; Jang, N. H.; Suh, J. S.; Moskovits, M., Photodecomposition of Diazanaphthalenes Adsorbed on Silver Colloid Surfaces. J. Phys. Chem. B 2000, 104, 35943600. 9. Redmond, P. L.; Wu, X. M.; Brus, L., Photovoltage and Photocatalyzed Growth in CitrateStabilized Colloidal Silver Nanocrystals. J. Phys. Chem. C 2007, 111, 8942-8947. 10. Huang, Y.-F.; Zhu, H.-P.; Liu, G.-K.; Wu, D.-Y.; Ren, B.; Tian, Z.-Q., When the Signal Is Not from the Original Molecule to Be Detected: Chemical Transformation of ParaAminothiophenol on Ag During the SERS Measurement. J. Am. Chem. Soc. 2010, 132, 9244-9246. 11. Christopher, P.; Xin, H. L.; Linic, S., Visible-Light-Enhanced Catalytic Oxidation Reactions on Plasmonic Silver Nanostructures. Nat. Chem. 2011, 3, 467-472. 12. Christopher, P.; Xin, H. L.; Marimuthu, A.; Linic, S., Singular Characteristics and Unique Chemical Bond Activation Mechanisms of Photocatalytic Reactions on Plasmonic Nanostructures. Nat. Mater. 2012, 11, 1044-1050. 13. Mukherjee, S.; Libisch, F.; Large, N.; Neumann, O.; Brown, L. V.; Cheng, J.; Lassiter, J. B.; Carter, E. A.; Nordlander, P.; Halas, N. J., Hot Electrons Do the Impossible: PlasmonInduced Dissociation of H2 on Au. Nano Lett. 2013, 13, 240-247.


26

Page 27 of 33 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. Zhang, Y.; Xiao, Q.; Bao, Y.; Zhang, Y.; Bottle, S.; Sarina, S.; Zhaorigetu, B.; Zhu, H., Direct Photocatalytic Conversion of Aldehydes to Esters Using Supported Gold Nanoparticles under Visible Light Irradiation at Room Temperature. J. Phys. Chem. C 2014, 118, 19062-19069. 15. Zhao, J.; Nguyen, S. C.; Ye, R.; Ye, B. H.; Weller, H.; Somorjai, G. A.; Alivisatos, A. P.; Toste, F. D., A Comparison of Photocatalytic Activities of Gold Nanoparticles Following Plasmonic and Interband Excitation and a Strategy for Harnessing Interband Hot Carriers for Solution Phase Photocatalysis. ACS Central Sci. 2017, 3, 482-488. 16. Tesema, T. E.; Kafle, B.; Tadesse, M. G.; Habteyes, T. G., Plasmon-Enhanced Resonant Excitation and Demethylation of Methylene Blue. J. Phys. Chem. C 2017, 121, 7421-7428. 17. Cao, E.; Guo, X.; Zhang, L. Q.; Shi, Y.; Lin, W. H.; Liu, X. C.; Fang, Y. R.; Zhou, L. Y.; Sun, Y. H.; Song, Y. Z.; Liang, W. J.; Sun, M. T., Electrooptical Synergy on PlasmonExciton-Codriven Surface Reduction Reactions. Adv. Mater. Interfaces 2017, 4, 1700869. 18. Yu, S. J.; Wilson, A. J.; Heo, J.; Jain, P. K., Plasmonic Control of Multi-Electron Transfer and C-C Coupling in Visible-Light-Driven CO2 Reduction on Au Nanoparticles. Nano Lett. 2018, 18, 2189-2194. 19. Kazuma, E.; Jung, J.; Ueba, H.; Trenary, M.; Kim, Y., Real-Space and Real-Time Observation of a Plasmon-Induced Chemical Reaction of a Single Molecule. Science 2018, 360, 521-525. 20. Hartland, G. V.; Besteiro, L. V.; Johns, P.; Govorov, A. O., What's So Hot About Electrons in Metal Nanoparticles? ACS Energy Lett. 2017, 2, 1641-1653. 21. Linic, S.; Aslam, U.; Boerigter, C.; Morabito, M., Photochemical Transformations on Plasmonic Metal Nanoparticles. Nat. Mater. 2015, 14, 567-576. 22. Brus, L., Noble Metal Nanocrystals: Plasmon Electron Transfer Photochemistry and SingleMolecule Raman Spectroscopy. Acc. Chem. Res. 2008, 41, 1742-1749. 23. Pozzi, E. A.; Zrimsek, A. B.; Lethiec, C. M.; Schatz, G. C.; Hersam, M. C.; Van Duyne, R. P., Evaluating Single-Molecule Stokes and Anti-Stokes SERS for Nanoscale Thermometry. J. Phys. Chem. C 2015, 119, 21116-21124. 24. Nguyen, S. C.; Zhang, Q.; Manthiram, K.; Ye, X. C.; Lomont, J. P.; Harris, C. B.; Weller, H.; Alivisatos, A. P., Study of Heat Transfer Dynamics from Gold Nanorods to the Environment Via Time-Resolved Infrared Spectroscopy. ACS Nano 2016, 10, 2144-2151.


27


Page 28 of 33

25. Keller, E. L.; Frontiera, R. R., Ultrafast Nanoscale Raman Thermometry Proves Heating Is Not a Primary Mechanism for Plasmon-Driven Photocatalysis. ACS Nano 2018, 12, 5848– 5855. 26. Carattino, A.; Caldarola, M.; Orrit, M., Gold Nanoparticles as Absolute Nanothermometers. Nano Lett. 2018, 18, 874-880. 27. Adleman, J. R.; Boyd, D. A.; Goodwin, D. G.; Psaltis, D., Heterogenous Catalysis Mediated by Plasmon Heating. Nano Lett. 2009, 9, 4417-4423. 28. Golubev, A. A.; Khlebtsov, B. N.; Rodriguez, R. D.; Chen, Y.; Zahn, D. R. T., Plasmonic Heating Plays a Dominant Role in the Plasmon-Induced Photocatalytic Reduction of 4Nitrobenzenethiol. J. Phys. Chem. C 2018, 122, 5657-5663. 29. Zhang, X.; Li, X. Q.; Reish, M. E.; Zhang, D.; Su, N. Q.; Gutierrez, Y.; Moreno, F.; Yang, W. T.; Everitt, H. O.; Liu, J., Plasmon-Enhanced Catalysis: Distinguishing Thermal and Nonthermal Effects. Nano Lett. 2018, 18, 1714-1723. 30. Kafle, B.; Poveda, M.; Habteyes, T. G., Surface Ligand-Mediated Plasmon-Driven Photochemical Reactions. J. Phys. Chem. Lett. 2017, 8, 890-894. 31. Nitzan, A.; Brus, L. E., Theoretical-Model for Enhanced Photochemistry on Rough Surfaces. J. Chem. Phys. 1981, 75, 2205-2214. 32. Wolkow, R. A.; Moskovits, M., Enhanced Photochemistry on Silver Surfaces. J. Chem. Phys. 1987, 87, 5858-5869. 33. Zhang, Y. X.; Aslan, K.; Previte, M. J. R.; Geddes, C. D., Metal-Enhanced Singlet Oxygen Generation: A Consequence of Plasmon Enhanced Triplet Yields. J. Fluoresc. 2007, 17, 345349. 34. Zhang, Y.; Aslan, K.; Previte, M. J. R.; Geddes, C. D., Plasmonic Engineering of Singlet Oxygen Generation. Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 1798-1802. 35. Planas, O.; Macia, N.; Agut, M.; Nonell, S.; Heyne, B., Distance-Dependent PlasmonEnhanced Singlet Oxygen Production and Emission for Bacterial Inactivation. J. Am. Chem. Soc. 2016, 138, 2762-2768. 36. Lv, J. L.; Wu, G. Q.; He, Y.; Zhang, L. J.; Yi, Y. H., Methylene Blue-Loaded Gold Nanobipyramids @SiO2 Enhanced Singlet Oxygen Generation for Phototherapy of Cancer Cells. Opt. Mater. Express 2017, 7, 409-414.


28

Page 29 of 33 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


37. Li, H. R.; Wu, L. Z.; Tung, C. H., Reactions of Singlet Oxygen with Olefins and Sterically Hindered Amine in Mixed Surfactant Vesicles. J. Am. Chem. Soc. 2000, 122, 2446-2451. 38. Baciocchi, E.; Del Giacco, T.; Lanzalunga, O.; Lapi, A., Singlet Oxygen Promoted CarbonHeteroatom Bond Cleavage in Dibenzyl Sulfides and Tertiary Dibenzylamines. Structural Effects and the Role of Exciplexes. J. Org. Chem. 2007, 72, 9582-9589. 39. Vankayala, R.; Sagadevan, A.; Vijayaraghavan, P.; Kuo, C. L.; Hwang, K. C., Metal Nanoparticles Sensitize the Formation of Singlet Oxygen. Angew. Chem.-Int. Edit. 2011, 50, 10640-10644. 40. Ho-Wu, R.; Yau, S. H.; Goodson, T., Efficient Singlet Oxygen Generation in Metal Nanoclusters for Two-Photon Photodynamic Therapy Applications. J. Phys. Chem. B 2017, 121, 10073-10080. 41. Huang, Y. F.; Zhang, M.; Zhao, L. B.; Feng, J. M.; Wu, D. Y.; Ren, B.; Tian, Z. Q., Activation of Oxygen on Gold and Silver Nanoparticles Assisted by Surface Plasmon Resonances. Angew. Chem.-Int. Edit. 2014, 53, 2353-2357. 42. Zhang, Z. L.; Xu, P.; Yang, X. Z.; Liang, W. J.; Sun, M. T., Surface Plasmon-Driven Photocatalysis in Ambient, Aqueous and High-Vacuum Monitored by SERS and TERS. J. Photochem. Photobiol. C-Photochem. Rev. 2016, 27, 100-112. 43. Kim, M.; Lin, M.; Son, J.; Xu, H. X.; Nam, J. M., Hot-Electron-Mediated Photochemical Reactions: Principles, Recent Advances, and Challenges. Adv. Opt. Mater. 2017, 5, 1700004. 44. Lombardi, J. R.; Birke, R. L.; Lu, T. H.; Xu, J., Charge-Transfer Theory of Surface Enhanced Raman-Spectroscopy - Herzberg-Teller Contributions. J. Chem. Phys. 1986, 84, 4174-4180. 45. Morton, S. M.; Jensen, L., Understanding the Molecule-Surface Chemical Coupling in SERS. J. Am. Chem. Soc. 2009, 131, 4090-4098. 46. Valley, N.; Greeneltch, N.; Van Duyne, R. P.; Schatz, G. C., A Look at the Origin and Magnitude of the Chemical Contribution to the Enhancement Mechanism of SurfaceEnhanced Raman Spectroscopy (SERS): Theory and Experiment. J. Phys. Chem. Lett. 2013, 4, 2599-2604. 47. Birke, R. L.; Lombardi, J. R.; Saidi, W. A.; Norman, P., Surface-Enhanced Raman Scattering Due to Charge-Transfer Resonances: A Time-Dependent Density Functional Theory Study of Ag13-4-Mercaptopyridine. J. Phys. Chem. C 2016, 120, 20721-20735.


29


Page 30 of 33

48. Boerigter, C.; Campana, R.; Morabito, M.; Linic, S., Evidence and Implications of Direct Charge Excitation as the Dominant Mechanism in Plasmon-Mediated Photocatalysis. Nat. Commun. 2016, 7, 10545. 49. Boerigter, C.; Aslam, U.; Linic, S., Mechanism of Charge Transfer from Plasmonic Nanostructures to Chemically Attached Materials. ACS Nano 2016, 10, 6108-6115. 50. Anderson, S., Orientation of Methylene Blue Molecules Adsorbed on Solids. J. Opt. Soc. Am. 1949, 39, 49-56. 51. Greathouse, J. A.; Geatches, D. L.; Pike, D. Q.; Greenwell, H. C.; Johnston, C. T.; Wilcox, J.; Cygan, R. T., Methylene Blue Adsorption on the Basal Surfaces of Kaolinite: Structure and Thermodynamics from Quantum and Classical Molecular Simulation. Clay Clay Min. 2015, 63, 185-198. 52. Zhou, L.; Johnson, R.; Habteyes, T.; Guo, H., Adsorption of Methylene Blue and Its NDemethylated Derivatives on the (111) Face of Coinage Metals: The Importance of Dispersion Interactions. J. Chem. Phys. 2017, 146, 164701. 53. Hutchinson, K.; Hester, R. E.; Albery, W. J.; Hillman, A. R., Raman-Spectroscopic Studies of a Thionine-Modified Electrode. J. Chem. Soc.-Faraday Trans. I 1984, 80, 2053-2071. 54. Baciocchi, E.; Del Giacco, T.; Lapi, A., Oxygenation of Benzyldimethylamine by Singlet Oxygen. Products and Mechanism. Org. Lett. 2004, 6, 4791-4794. 55. Baciocchi, E.; Del Giacco, T.; Lapi, A., Quenching of Singlet Oxygen by Tertiary Aliphatic Amines. Structural Effects on Rates and Products. Helv. Chim. Acta 2006, 89, 2273-2280. 56. Rai, V. N.; Srivastava, A. K.; Mukherjee, C.; Deb, S. K., Surface Enhanced Absorption and Transmission from Dye Coated Gold Nanoparticles in Thin Films. Appl. Optics 2012, 51, 2606-2615. 57. Leosson, K.; Ingason, A. S.; Agnarsson, B.; Kossoy, A.; Olafsson, S.; Gather, M. C., UltraThin Gold Films on Transparent Polymers. Nanophotonics 2013, 2, 3-11. 58. Kossoy, A.; Merk, V.; Simakov, D.; Leosson, K.; Kena-Cohen, S.; Maier, S. A., Optical and Structural Properties of Ultra-Thin Gold Films. Adv. Opt. Mater. 2015, 3, 71-77. 59. Habteyes, T. G.; Dhuey, S.; Kiesow, K. I.; Vold, A., Probe-Sample Optical Interaction: Size and Wavelength Dependence in Localized Plasmon Near-Field Imaging Opt. Express 2013, 21, 21607.


30

Page 31 of 33 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


60. Kafle, B.; Tesema, T. E.; Kazemi, A.; Habteyes, T. G., Stripping and Transforming Alloyed Semiconductor Quantum Dots via Atomic Interdiffusion. J. Phys. Chem. C 2016, 120, 12850-12859. 61. Craighead, H. G.; Glass, A. M., Optical-Absorption of Small Metal Particles with Adsorbed Dye Coats. Opt. Lett. 1981, 6, 248-250. 62. Darby, B. L.; Auguie, B.; Meyer, M.; Pantoja, A. E.; Le Ru, E. C., Modified Optical Absorption of Molecules on Metallic Nanoparticles at Sub-Monolayer Coverage. Nat. Photonics 2016, 10, 40-45. 63. Narband, N.; Uppal, M.; Dunnill, C. W.; Hyett, G.; Wilson, M.; Parkin, I. P., The Interaction between Gold Nanoparticles and Cationic and Anionic Dyes: Enhanced UV-Visible Absorption. Phys. Chem. Chem. Phys. 2009, 11, 10513-10518. 64. Lombardi, J. R.; Birke, R. L., A Unified Approach to Surface-Enhanced Raman Spectroscopy. J. Phys. Chem. C 2008, 112, 5605-5617. 65. Roy, S. D.; Ghosh, M.; Chowdhury, J., Adsorptive Parameters and Influence of Hot Geometries on the SER(R)S Spectra of Methylene Blue Molecules Adsorbed on Gold Nanocolloidal Particles. J. Raman Spectrosc. 2015, 46, 451-461. 66. Li, Y. J.; Zolotavin, P.; Doak, P.; Kronik, L.; Neaton, J. B.; Natelson, D., Interplay of BiasDriven Charging and the Vibrational Stark Effect in Molecular Junctions. Nano Lett. 2016, 16, 1104-1109. 67. Tardivo, J. P.; Del Giglio, A.; de Oliveira, C. S.; Gabrielli, D. S.; Junqueira, H. C.; Tada, D. B.; Severino, D.; Turchiello, R. D. F.; Baptista, M. S., Methylene Blue in Photodynamic Therapy: From Basic Mechanisms to Clinical Applications. Photodiagnosis Photodyn. Ther. 2005, 2, 175-191. 68. Gorman, A. A.; Gould, I. R.; Hamblett, I.; Standen, M. C., Reversible Exciplex Formation between Singlet Oxygen, 1∆g, and Vitamin-E. Solvent and Temperature Effects. J. Am. Chem. Soc. 1984, 106, 6956-6959. 69. Gould, I. R.; Young, R. H.; Mueller, L. J.; Farid, S., Mechanisms of Exciplex Formation. Roles of Superexchange, Solvent Polarity, and Driving-Force for Electron-Transfer. J. Am. Chem. Soc. 1994, 116, 8176-8187.


31


Page 32 of 33

70. Xu, P.; Kang, L. L.; Mack, N. H.; Schanze, K. S.; Han, X. J.; Wang, H. L., Mechanistic Understanding of Surface Plasmon Assisted Catalysis on a Single Particle: Cyclic Redox of 4-Aminothiophenol. Sci Rep 2013, 3, 2997. 71. Zhao, L. B.; Liu, X. X.; Zhang, M.; Liu, Z. F.; Wu, D. Y.; Tian, Z. Q., Surface Plasmon Catalytic Aerobic Oxidation of Aromatic Amines in Metal/Molecule/Metal Junctions. J. Phys. Chem. C 2016, 120, 944-955. 72. Kim, Y.; Wilson, A. J.; Jain, P. K., The Nature of Plasmonically Assisted Hot-Electron Transfer in a Donor-Bridge-Acceptor Complex. ACS Catal. 2017, 7, 4360-4365. 73. Nitzan, A.; Brus, L. E., Can Photochemistry Be Enhanced on Rough Surfaces. J. Chem. Phys. 1981, 74, 5321-5322. 74. Phan, H. T.; Haes, A. J., Impacts of pH and Intermolecular Interactions on Surface-Enhanced Raman Scattering Chemical Enhancements. J. Phys. Chem. C 2018, 122, 14846-14856.


32

Page 33 of 33

TOC graphic

Reactant PEND-MB

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


Product

Intermediate

Exposure time


33

Plasmon-Enhanced Autocatalytic N-Demethylation - ACS Publications

Recommend Documents