probing the chemistry of alumina atomic layer deposition ......atomic layer deposition (ald) relies...

Probing the Chemistry of Alumina Atomic Layer Deposition UsingOperando Surface-Enhanced Raman SpectroscopySicelo S. Masango,† Ryan A. Hackler,† Anne-Isabelle Henry,† Michael O. McAnally,† George C. Schatz,†

Peter C. Stair,†,‡ and Richard P. Van Duyne*,†

†Department of Chemistry and Center for Catalysis and Surface Science, Northwestern University, Evanston, Illinois 60208, UnitedStates‡Chemical Sciences and Engineering Division, Argonne National Laboratory, Argonne, Illinois 60439, United States

*S Supporting Information

ABSTRACT: This work demonstrates for the first time thecapability of measuring surface vibrational spectra foradsorbates during atomic layer deposition (ALD) reactionsusing operando surface-enhanced Raman spectroscopy (SERS).We use SERS to study alumina ALD growth at 55 °C on baresilver film-over nanosphere (AgFON) substrates as well asAgFONs functionalized with thiol self-assembled monolayers(SAMs). On bare AgFONs, we observe the growth of Al−Cstretches, symmetric C−H and asymmetric C−H stretchesduring the trimethylaluminum (TMA) dose half-cycle, andtheir subsequent decay after dosing with H2O. Al−C and C−H vibrational modes decay in intensity with time even without H2Oexposure providing evidence that residual H2O in the ALD chamber reacts with −CH3 groups on AgFONs. The observed Al−Cstretches are attributed to TMA dimeric species on the AgFON surface in agreement with density functional theory (DFT)studies. We observe Al−C stretches and no thiol vibrational frequency shifts after dosing TMA on AgFONs functionalized withtoluenethiol and benzenethiol SAMs. Conversely, we observe thiol vibrational frequency shifts and no Al−C stretches forAgFONs functionalized with 4-mercaptobenzoic acid and 4-mercaptophenol SAMs. Lack of observed Al−C stretches forCOOH- and OH-terminated SAMs is explained by the spacing of Al−(CH3)x groups from the SERS substrate. TMA penetratesthrough SAMs and reacts directly with Ag for benzenethiol and toluenethiol SAMs and selectively reacts with the −COOH and−OH groups for 4-mercaptobenzoic acid and 4-mercaptophenol SAMs, respectively. The high sensitivity and chemical specificityof SERS provides valuable information about the location of ALD deposits with respect to the enhancing substrate. Thisinformation can be used to evaluate the efficacy of SAMs in blocking or allowing ALD deposition on metal surfaces. The abilityto probe ALD reactions using SERS under realistic reaction conditions will lead to a better understanding of the mechanisms ofALD reactions.

■ INTRODUCTIONAtomic layer deposition (ALD) relies on sequential self-limitingbinary reactions between gaseous precursor molecules and asubstrate to deposit thin films in a layer-by-layer fashion.1−4

ALD provides angstrom-level precision over film thickness.Unlike films deposited by line-of-sight techniques, ALD filmsare highly uniform and can be deposited on most substratesregardless of whether they are flat, porous, or have complexthree-dimensional topologies. A wide range of materials can bedeposited by ALD including metal oxides, metal nitrides, metalsulfides, pure metal thin films, and others. Recent applicationsof ALD extend beyond thin films and include deposition ofmetal nanoparticles such as Pd,5−8 Pt,9−11 Ag,12 Ru,13 Ru−Pt,14and Pt−Pd15,16 alloys for applications in catalysis.The nucleation and growth of ALD materials is highly

dependent on the surface chemistry involved.17,18 A deepunderstanding of ALD reactions at the molecular level is crucialto optimize ALD processes and grow materials efficiently.Operando spectroscopic techniques, in which reactions are

studied in situ under temperatures and pressures employed inALD reactions,19,20 can provide valuable insight towardunderstanding mechanistic details of ALD processes. Most insitu characterization techniques that have been used toinvestigate the mechanisms of ALD reactions either do notprovide detailed chemical structural information or areperformed under conditions that differ from realistic ALDconditions.18 Examples include quartz crystal microbalance(QCM),21 quadrupole mass spectrometry (QMS),22 Fouriertransform infrared (FTIR) spectroscopy,23 and spectroscopicellipsometry.24 QCM studies are used to monitor mass gains onthe surface during ALD deposition and determine the growthrate per cycle and film thickness as a function of the number ofcycles. A major limitation of QCM measurements is that theydo not provide any chemically specific information.18 QMS is

Received: November 24, 2015Revised: January 28, 2016

Article

pubs.acs.org/JPCC

© XXXX American Chemical Society A DOI: 10.1021/acs.jpcc.5b11487J. Phys. Chem. C XXXX, XXX, XXX−XXX

pubs.acs.org/JPCChttp://dx.doi.org/10.1021/acs.jpcc.5b11487

used to detect products that desorb into the gas phase duringALD reactions, which helps to identify possible mechanisms ofALD reactions. However, QMS does not provide directmolecular-level information about adsorbed surface species. Inaddition, some of the gas-phase species generated during ALDprocesses may not be stable, and many undergo furtherconversion in the gas phase, upon collision with the walls of thereactor, or in the ionizer of the mass spectrometer beforedetection.18 Spectroscopic ellipsometry provides informationabout the thickness, growth rate, optical properties, crystallinephase, and material composition of ALD films but does notdirectly probe the presence of chemical bonds at surfaces.24 X-ray photoelectron spectroscopy (XPS) is another techniqueoften used to study ALD reactions but is usually performed inultrahigh vacuum (UHV), an environment that is different fromtypical conditions at which ALD reactions are performed.25,26

As a result, the data collected may not be representative of theproper surface species during the ALD process. Operando XPSsetups typically require the use of a specially constructedelectrostatic lens system and a differential pumping stagebetween the sample cell and the electron energy analyzer tominimize the scattering of photoelectrons by gas-phasemolecules.27 Additionally, a synchrotron radiation source thatcan deliver a much higher photon flux than a conventional X-ray source is often used to increase the total photoelectronsignal, although Tao and co-workers recently designed and builtan operando XPS setup that uses a benchtop Al Kα X-raysource.28−31 Compared to a synchrotron-based setup, alaboratory-based operando XPS instrument has limited spectralresolution and surface sensitivity because of its X-ray sourceintensity and fixed photon energy, which affects the accuracy ofthe quantitative analysis.32 FTIR provides molecular-levelinformation about adsorbed surface species and does notneed a UHV environment.23,33 FTIR readily detects vibrationalmodes in the ∼1000−3000 cm−1 region but has difficultymeasuring low frequency vibrations (∼100−1000 cm−1).34Raman spectroscopy is a complementary technique to FTIR

that can probe both low frequency and high frequencymolecular vibrations. However, Raman scattering has inherentlylow sensitivity because of small Raman scattering cross sections(∼10−30 cm2 steradian−1 molecule−1).35 Surface-enhancedRaman spectroscopy (SERS) overcomes the sensitivitylimitations of normal Raman scattering.36−38 The enhancementof Raman signals is attributed to the excitation of localizedsurface plasmon resonances (LSPRs) that result in enhancedelectromagnetic fields around noble metal nanostructures suchas Ag, Au, and Cu.39,40 The high sensitivity and distancedependence of SERS38 should make it possible to evaluate thelocation of ALD deposits with respect to the enhancingsubstrate. To the best of our knowledge, in situ studies that useSERS to monitor ALD reactions do not exist in the literature.Most of the studies published so far use SERS to monitorcatalytic reactions occurring on thermally robust SERSsubstrates.41,42 In contrast, our study focuses on demonstratingthat operando SERS can be used to monitor adsorbed surfacespecies during the ALD process itself, which results in animproved understanding of the surface chemistry involved.Using in situ QCM and ex situ LSPR spectroscopy, it was

demonstrated that ALD Al2O3 grows on bare Ag surfaces with agrowth rate of ∼1.1 Å/cycle and induces a redshift in the LSPRpeak position due to a change in refractive index.43 Thedeposition of Al2O3 by ALD on hydroxylated surfaces is themodel ALD process and has been the subject of extensive

investigations.2,3,44,45 A few studies have investigated thegrowth of ALD Al2O3 on metallic substrates.

46,47 ALD Al2O3thin films have been shown to protect metal nanoparticles fromsintering at elevated temperatures.48,49 This offers both apromising route to stabilize metal nanoparticles, which haveapplications in catalysis, and in situ monitoring of catalyticreactions using operando surface-enhanced optical spectros-copies.In this work, we demonstrate, for the first time, that SERS

can be used to probe ALD reactions in situ with high sensitivity.We use SERS to monitor the surface chemistry of Al2O3 ALDat 55 °C on Ar plasma-cleaned silver-film-over nanosphere(AgFON) substrates. We also use SERS to monitor ALD Al2O3growth on AgFONs functionalized with thiol self-assembledmonolayers (SAMs). Thiol SAMs are ideal molecular systemsto use because they form well-defined and well-orderedstructures on metal surfaces that have been extensively studiedin the literature.50 SAMs provide a well-characterized startingsurface that can be used as an internal standard against whichspectral changes after ALD growth can be measured. Further,thiol SAMs on AgFONs are known to displace carboncontamination and offer an alternative way to clean SERSsubstrates.51,52 They also preserve the enhancing nature ofSERS substrates and eliminate the need for plasma cleaning,which reduces the SERS signal.51 Using operando SERS, weinvestigate the nucleation and growth of ALD Al2O3 aroundSAMs possessing different terminal groups. The use of SAMspossessing different terminal functional groups provides a wayto tailor the surface and direct ALD Al2O3 growth to achievearea-selective ALD.53,54

■ EXPERIMENTAL SECTIONFabrication of AgFON SERS Substrates. AgFONs were

fabricated on polished 25 mm silicon wafers according to astandard procedure described in previous publications.55 Siliconwafers were cleaned by immersion in piranha solution (3:1 byvolume H2SO4/30% H2O2) for 1 h. Caution: Piranha isextremely corrosive, and appropriate personal protectiveequipment should be used during handling. Clean siliconwafers were thoroughly rinsed with deionized (DI) water. Thewafers were then sonicated for 1 h in 5:1:1 by volume H2O/NH4OH/30% H2O2 followed by rinsing with DI water. Silicananospheres (390 nm, Bangs Laboratories) were diluted to 5%silica by volume. The solvent was replaced twice with Millipore(Milli-Q, 18.2 MΩ·cm−1) H2O by a conventional centrifuga-tion/supernatant removal procedure, followed by sonication for1 h. The solvent-replaced nanosphere solution (10−12 μL) wasdrop-coated and distributed homogeneously across theprepared silicon wafer surface. The solvent was then allowedto evaporate under ambient conditions and spheres assembledin a hexagonal close-packed array as verified by SEMmeasurements. Ag films (200 nm) were deposited at a rate of2 Å/s under vacuum (6 × 10−6 Torr) over the nanospheresusing a home-built thermal vapor deposition system. Thesubstrates were spun during deposition while the metalthickness and deposition rate were measured by a 6 MHzgold-plated QCM (Sigma Instruments, Fort Collins, CO).AgFONs, which were not plasma-cleaned, were incubated in 1mM ethanolic solutions of benzenethiol, toluenethiol, 4-mercaptobenzoic acid, or 4-mercaptophenol (Sigma-Aldrich)for a minimum of 4 h. The extinction spectra of AgFONs weremeasured using a fiber-optic coupled halogen light source(World Precision Instruments) and UV/vis spectrometer (SD

The Journal of Physical Chemistry C Article

DOI: 10.1021/acs.jpcc.5b11487J. Phys. Chem. C XXXX, XXX, XXX−XXX

B

http://dx.doi.org/10.1021/acs.jpcc.5b11487

2000, Ocean Optics) in specular reflectance mode. A Ag mirrorwas used as a spectral reference and was fabricated bydepositing 200 nm of Ag film on cleaned glass using thermalvapor deposition.Surface-Enhanced Raman Spectroscopy. A 532 nm

continuous wave (CW) laser (Innovative Photonic Solutions)was used for all spectroscopic measurements. Laser light wasdirected using protected silver mirrors to a 3 mm right-angleprism and then focused using a visible achromatic doublet lens(1 in. diameter, 4 in. focal length) through a quartz window to aplasmonic substrate placed inside the ALD reactor. The spotsize radius measured at the sample was ∼124 μm and ∼57 μmfor the 532 and 633 nm lasers, respectively, using a scanningknife-edge technique. Raman scattered light was collected in a180° backscattering geometry and focused onto a 0.3 mimaging spectrograph (Acton SpectraPro 2300i) using a visibleachromatic doublet lens (1 in. diameter, 4 in. focal length).Scattered light was dispersed (1200 grooves/mm grating, 500nm blaze) onto a liquid N2-cooled CCD detector (PrincetonInstruments, Model 7509-0001). SER spectra were collectedwith 1−7 mW of laser power (Paq), 1−10 s of acquisition time(taq), and 1−10 accumulations each, depending on the systeminvestigated. No background contribution or SERS signalattenuation was observed from the quartz window (Figure S1).Computational Modeling. Electronic structure calcula-

tions presented in this work have been performed with theamsterdam density functional (ADF) computational chemistrypackage.56 Full geometry optimization, frequency, and polar-izability calculations for isolated monomer and dimer TMAcomplexes were completed using the Becke−Perdew (BP86)generalized gradient approximation (GGA) exchange correla-tion functional and a triple-ζ polarized (TZP) Slater orbitalbasis set.Static Raman polarizabilities (ω = 0) were calculated in the

RESPONSE package by two-point numerical differentiationusing the RAMANRANGE keyword. Raman scatteringintensities were determined by the scattering factor 45α̅j′ +7γj̅′, where α̅j′ and γj̅′ are the isotropic and anisotropicpolarizability tensors with respect to the jth vibrational mode.The Raman intensity for each vibrational mode were broadenedto a Lorentzian line shape with full-width at half-maximum(fwhm) of 10 cm−1 for comparison to experimental data.Atomic Layer Deposition. ALD was performed in a home-

built viscous flow reactor that has been described previouslyand is shown in Figure 1.12,21 SERS substrates were mountedon a movable sample holder, placed inside the ALD chamberunder vacuum (∼0.05 Torr), and heated to ∼55 °C to preservetheir fine nanostructure. SER spectra were taken before andafter dosing 60 sccm of trimethylaluminum (TMA) and 60sccm H2O using ultrahigh purity (UHP) N2 as the carrier gas.For one-half-cycle of Al2O3, TMA was dosed on the SERSsubstrate for 10 or 30 s unless otherwise stated, and SERspectra were acquired while purging with N2. The second half-cycle involved dosing H2O for 60 s and thereafter acquiringSER spectra during N2 purging. Thermocouples were placedabove and below the sample compartment to ensure uniformheating around the sample. Variacs were used to manuallyadjust the voltage supplied to heating tapes to keep thetemperature at 55 ± 2 °C throughout experiments.

■ RESULTSSEM and LSPR Characterization of AgFONs. Reactive

ion etching (RIE) using Ar plasma was used to remove carbon

contamination on AgFONs, decreasing the background of SERspectra prior to operando monitoring of ALD reactions, andultimately improve the sensitivity of the AgFONs (Figure S2).To ensure that the AgFON structure was not altered duringRIE, scanning electron microscopy (SEM) was used to observethe Ag film morphology. Figure 2 shows SEM images ofAgFONs after RIE exposure times of 25 s (Figure 2A,B) and300 s (Figure 2C,D). The top-down views (Figures 2A,C)clearly show that the Ag film undergoes dramatic structuralchanges upon long (300 s) Ar plasma exposure, with theappearance of a vitreous-like, smoothened film (Figure 2C). Asimilar trend was observed on the side observation of thesubstrates (Figures 2B,D). It is also important to notice that themetal pillars that act as electromagnetic hot spots55,57 in bareAgFONs are maintained after 25 s of Ar plasma exposure. Thisis consistent with the high intensity of the SER spectra (FigureS2C). On the basis of these observations, 25 s of Ar plasmacleaning was chosen because it had no adverse effects on thepillar-like structure of the AgFONs while still removingbackground signal due to carbonaceous species. Additionally,Figure S2D shows that the LSPR of the AgFONs wasoptimized for 532 nm laser excitation, with a reflectanceminimum indicative of the LSPR at 534 nm.

Reaction of TMA with Bare AgFON. Figure 3 shows SERspectra acquired before and after two cycles of ALD Al2O3growth on an Ar plasma-cleaned AgFON at 55 °C. In Figure3A, only peaks below 900 cm−1 are shown because Ar plasmaremoves most of the surface contaminants observed in thatregion. Two peaks at 583 and 671 cm−1 are observed afterdosing TMA for 10 s. TMA dosing for 5−10 s was long enoughto saturate the surface as shown in Figure S3. Figure 3B showsSER difference spectra for the first two ALD cycles. The 583and 671 cm−1 peaks decay after dosing H2O for 60 s andreappear after 10 s of TMA dosing in the second cycle. The 583and 671 cm−1 peaks are assigned to the symmetric andasymmetric Al-CH3 stretching modes,

58 respectively (see Table1 and Table S1).Figure 4 shows both the recorded SER (A) and difference

spectra (B) in the C−H stretching region. A peak at 2892 cm−1appears after 30 s of TMA exposure. This peak decays after 60 sof H2O dosing and reappears after the second TMA half-cycle.The difference spectra in Figure 4B clearly show the spectralchanges before and after ALD Al2O3 growth. Additionalpositive peaks at 2822 and 2921 cm−1 are seen after the first

Figure 1. Experimental schematic for operando ALD-SERS reactor. IF= interference filter, M = mirror, FM = flip mirror, ND = neutraldensity filter.



C

http://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://dx.doi.org/10.1021/acs.jpcc.5b11487

TMA half-cycle. All positive peaks observed after TMAexposure (red curves, Figures 3B and 4B) switch and becomenegative after dosing H2O, indicative of loss of surface-boundAl−(CH3)x species (blue curves, Figures 3B and 4B). TMAdosing in the second cycle restores the positive peaks, however,with reduced signal intensity because of the distance depend-ence of the electromagnetic enhancement in SERS (red curves,Figures 3B, 4B). The 2822 and 2892 cm−1 peaks are assigned totwo different symmetric C−H stretches of Al−(CH3)x surfacespecies.72 The 2921 cm−1 peak is assigned to the asymmetricC−H stretch.59−62 The observation of surface-bound speciesfrom ALD precursors shows that operando SERS is capable ofmonitoring ALD half-reactions and can provide valuable insight

into the mechanistic details of the surface chemistry of ALDprocesses.

Reaction of TMA with Toluenethiol and Benzenethiol.The nucleation and growth of ALD Al2O3 on AgFONsfunctionalized with thiol SAMs was also investigated at 55 °C.Figure 5 shows the first ALD Al2O3 cycle performed on atoluenethiol (TT)-functionalized AgFON. TMA was dosed onthe surface for 10 min to ensure complete saturation. Tenminutes of TMA exposure was long enough to allow TMA tocompletely react with Ag (Figure S9). In Figure 5A, a peak at585 cm−1 (symmetric Al−CH3 stretch) is observed after thefirst TMA half-cycle. A slight reduction in TT peak intensitiesafter TMA exposure is also seen. The 585 cm−1 peak disappearsafter dosing H2O (see Figure S10 for difference spectra). On abenzenethiol (BT)-functionalized AgFON, peaks at 583 cm−1

(symmetric Al−CH3 stretch) and 1211 cm−1 (symmetric Al−CH3 stretch + CH3 bend)

23 appear and decay after TMA andH2O exposures, respectively (see Figures S11 and S12). Figure5B−D and Table S2 show the absence of vibrational frequencyshifts of TT and BT SAMs after dosing TMA on the surface.Figure 5B−D and Table S3 shows that the 622 (CC bend),1081 (CCC in-plane bend + C−S stretch), and 1381 cm−1 (CCstretch)67,73 peaks of TT decrease in intensity after TMAexposure. The fwhm of the peaks is roughly the same beforeand after TMA exposure (see Table S3).

Reaction of TMA with Mercaptobenzoic Acid andMercaptophenol. The reaction of TMA with 4-mercapto-benzoic acid (MBA) and 4-mercaptophenol (MP) on AgFONswas also investigated because the −COOH and −OH groupscould react directly with TMA.74,75 Figure 6 shows the firstALD cycle of Al2O3 performed on MBA-functionalized AgFONat 55 °C. After 10 min of TMA exposure, thiol peak shifts from845 to 849 cm−1 (COO− out-of-plane bend)63 and 1080−1082cm−1 (CCC ring deformation and C−S stretch),65 and theappearance of a broad shoulder peak at ∼1397 cm−1 (COO−stretch)71 is seen. On MP-functionalized AgFON, thiol peakshifts from 1074 to 1075 cm−1 (CCC ring deformation and C−S stretch),65 1290−1276 cm−1 (C−OH stretch + ringbreathing),66 and 1577−1584 cm−1 (CC stretch and OHbend)65,66 are observed (Figure S13). In addition to thiol peakshifts, the peak intensities generally decreased after TMAexposure (except for the 845 cm−1 peak of MBA) for bothMBA and MP SAMs (Table S3). Table S3 also shows dramaticchanges in the fwhm of MBA peaks after TMA exposure; thefwhm of the 845 cm−1 peak increases from 33 to 52 cm−1. ForMP, the fwhm of the 1290 cm−1 peak increases from 86 to 113cm−1. The fwhm of the 1577 cm−1 peak decreases from 45 to37 cm−1 after TMA exposure.

C−H Stretching Region of Functionalized AgFONs.Figure 7 displays SER and difference spectra of TT- and MBA-functionalized AgFONs in the C−H stretching region. On aTT-functionalized AgFON, the 2822 and 2892 cm−1 peaks,assigned to two different symmetric C−H stretches of Al−CH3surface species, are observed after TMA exposure (Figure7A,B). They decay after H2O exposure and reappear after TMAexposure in the second cycle (Figure 7B). On an MBA-functionalized AgFON, the symmetric C−H stretch of TMAappears at 2898 cm−1 after TMA exposure (Figures 7C,D).Similarly to a bare AgFON, the TMA symmetric C−H stretchpeak decays after H2O exposure and reappears after TMAexposure in the second cycle (Figure 7D).

Figure 2. SEM images of AgFONs after 25 s (A, B) and 300 s (C, D)of Ar plasma cleaning. Top-down view is shown in A and C and sideview is displayed in B and D.



D

http://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://dx.doi.org/10.1021/acs.jpcc.5b11487

■ DISCUSSIONReaction of TMA with Bare AgFON. The 583 (symmetric

Al−CH3 stretch) and 671 cm−1 (asymmetric Al−CH3 stretch)modes shown in Figure 3 and Table 1 match well with thevibrational frequencies of dimeric TMA calculated by DFT (seeFigure S5 and Table 1). In the first ALD cycle, we propose thatTMA reacts directly with Ag and decomposes into Al atomsand Al−(CH3)x surface species. The Al−CH3 stretches

observed at 583 and 671 cm−1 (Figure 3B) become negativeafter H2O exposure because H2O replaces Al−(CH3)x surfacespecies with Al−(OH)x species. They reappear after TMAdosing because TMA reacts with Al−(OH)x species andreplaces them with Al−(CH3)x species. Previous DFT studiessuggest that TMA reacts directly with metals such as Pd, Pt,and Ir by dissociative chemisorption to form Al−CH3 and Alsurface species, which transform to Al−(OH)x surface species

Figure 3. (A) SER spectra of a 25 s Ar plasma-cleaned AgFON before ALD (a) and after 10 s of TMA (1st cycle) (b), 60 s of H2O (1st cycle) (c),10 s of TMA (2nd cycle) (d), and 60 s of H2O (2nd cycle) (e). (B) Difference SER spectra for the first 2 ALD cycles. SER spectra were acquiredwith λexc = 532 nm, Pexc = 2 mW, taq = 5 s and 10 accumulations each. Data are shifted on y-axis for clarity.

Table 1. Vibrational Modes and Corresponding Assignments*

peak positions (cm−1)

TMA thiol SAMs

expt calcd mon calcd dim TT BT MBA MP assignment23,58−71

583 572 ν(Al−CH3)symm622 γ(CC)

671 671 ν(Al−CH3)asymm691 β(CCC) + ν(C−S)

845 γ(COO−)1000 β(CCC)1022 β(C−H)ring

1081 1072 1080 1074 β(CCC) + ν(C−S)1211 1190 1190 ν(Al−C)asymm, bending CH3

1290 ν(C−OH) + ring breathing1381 ν(CC)

1397 ν(COO−)1432 ν(CC)

1577 ν(CC) + β(OH)1599 ν(CC)

2822 ν(C−H)symm2865 ν(C−H)symm (CH3 group)

2892 2905 ν(CbridgeH)symm2917 ν(C−H)asymm (CH3 group)

2919 2919 ν(C−H)symm2921 2996, 3022 2960, 3008, 3032 ν(C−H)asymm

3051 3058 3055 3055 ν(C−H)ring*Experimental and DFT-calculated vibrational frequencies and peak assignments. TMA = trimethylaluminum, mon = monomer, dim = dimer, TT =toluenethiol, MBA = 4-mercaptobenzoic acid, BT = benzenethiol, MP = 4-mercaptophenol. γ, β, and ν indicate the out-of-plane bending, in-planebending, and stretching modes, respectively.



E

http://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://dx.doi.org/10.1021/acs.jpcc.5b11487

after H2O exposure.46 Nucleation of TMA could also occur on

hydroxyl (−OH) groups adsorbed on a very thin layer of oxideon the AgFON surface.43 The percentage of oxygen atoms inthe volume sampled by XPS was calculated to be ∼8% for both0 and 25 s Ar plasma-cleaned AgFONs (Figure S14). If

nucleation of TMA occurs on −OH groups, then Al−Ostretches could be observed by SERS but they were notobserved in our studies even with an optimized AgFON for thelow frequency Al−O stretching region (Figure S2D). Ramanpeaks from Al2O3 compounds are typically observed in the

Figure 4. SER spectra (A) of a 25 s Ar plasma-cleaned AgFON depicting the C−H stretching region before ALD (a) and after 30 s of TMA (1stcycle) (b), 60 s of H2O (1st cycle) (c), 30 s of TMA (2nd cycle) (d), and 60 s of H2O (2nd cycle) (e). Difference spectra (B) showing C−Hstretching region after 2 ALD Al2O3 cycles. SER spectra were acquired with λexc = 532 nm, Pexc = 5 mW, taq = 10 s and 10 accumulations each.

Figure 5. (A) SER spectra of a AgFON functionalized with toluenethiol (TT) before and after 1 ALD Al2O3 cycle. (B−D) SER and differencespectra showing no thiol peak shifts before and after ALD Al2O3. The red curves in B−D represent Lorentzian peak fits. SER spectra were acquiredwith λexc = 532 nm, Pexc = 1 mW, taq = 2 s, and 10 accumulations each.



F

http://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://dx.doi.org/10.1021/acs.jpcc.5b11487

∼700−1000 cm−1 region.76 The scarcity of SERS studies thatreport the observation of Al−O vibrations is attributed to thelow Raman cross sections of amorphous Al2O3 as reported byStair and co-workers.76

The TMA C−H stretches positioned at 2822, 2892, and2921 cm−1 in Figure 4B become negative after H2O exposurebecause H2O replaces (CH3)x surface species with (OH)xspecies. In the second cycle, TMA replaces (OH)x specieswith (CH3)x species. Figure S6 shows gradual decay over timeof the TMA symmetric C−H stretch at 2892 cm−1 duringconstant 5 mW of laser irradiation after TMA exposure. Thepeak intensity decreases to 50 and 15% of the initial intensityafter 240 and 780 s, respectively. A control experiment shownin Figure S7 provides evidence that the decay of the C−Hstretches over time was not caused by laser irradiation but wasdue to the reaction of residual H2O inside the ALD chamberwith adsorbed Al−(CH3)x surface species. A power dependencestudy of the 2892 cm−1 peak is displayed in Figure S8 andshows increasing SERS intensity when the laser power wasincreased from 0.2 to 7 mW. The study was performed veryquickly for each laser power using 3 s acquisition time and 10accumulations to minimize the reaction of TMA with residualH2O.Reaction of TMA with Toluenethiol and Benzenethiol.

The toluenethiol SAM does not prevent TMA from reactingwith the AgFON since the symmetric Al−CH3 stretch of TMA(585 cm−1) is observed (Figure 5A). We propose that TMA

penetrates through the monolayer and reacts with Ag as shownin Scheme 1. It is also possible that TMA displaces somesurface toluenethiol molecules and reacts with the AgFON asindicated by the slight reduction in toluenethiol SERS signalafter TMA exposure (Figures 5A−D and Table S2). Anotherpossible explanation for the decrease of the toluenethiol signalis that TMA changes the spatial orientation of toluenethiolmolecules. The toluenethiol molecules are initially orientedalmost perpendicular to the AgFON surface.77 When TMAreacts with the AgFON surface, it changes the orientation ofthe toluenethiol molecules to a more tilted position. Thechange in molecular orientation lowers the Raman scatteringcross-section and affects the SERS signal as seen previously forother molecular systems.78

As with the reaction of TMA with an unfunctionalizedAgFON, the Al−CH3 peak at 585 cm−1 (Figure 5A) and TMAC−H symmetric stretches at 2822 and 2892 cm−1 (Figures 7A,7B) decay after H2O exposure because H2O reacts with Al−(CH3)x groups to produce Al−(OH)x species. Figure 5B−Dand Table S2 show that toluenethiol vibrational modes do notshift after TMA exposure. Lack of vibrational frequency shiftsindicate that toluenethiol molecules are not significantlyperturbed by TMA and this supports the claim that TMAreacts directly with the Ag surface. In addition, the fwhm oftoluenethiol peaks does not change significantly after TMAexposure (Table S2), which might further indicate that TMAdoes not react directly with toluenethiol but with the Ag

Figure 6. (A) SER spectra of the first ALD Al2O3 cycle deposited on a AgFON functionalized with 4-mercaptobenzoic acid (MBA). (B−D) SER anddifference spectra of MBA showing thiol peak shifts observed after 10 min of TMA exposure. The red curves in B−D are Lorentzian peak fits.Spectra were acquired with λexc = 532 nm, Pexc = 1 mW, taq = 2 s, and 10 accumulations each.



G

http://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://dx.doi.org/10.1021/acs.jpcc.5b11487

surface. We observed that the fwhm of vibrational modes ofSAMs with nonreactive groups to TMA did not changesignificantly after TMA exposure unlike SAMs with reactivegroups as we discuss in the following section. In similar fashion,the symmetric Al−CH3 stretch at 583 cm−1 and the Al−CH3deformation mode at 1211 cm−1 were observed after TMAexposure on benzenethiol-functionalized AgFON and no thiolvibrational frequency shifts were observed (Figures S11, S12and Table S2). We believe that TMA also reacts with Ag ratherthan the benzenethiol molecules similar to the TT-function-alized AgFON.Reaction of TMA with Mercaptobenzoic Acid and

Mercaptophenol. On MBA-functionalized AgFON, vibra-tional frequency shifts are seen after TMA exposure as depictedin Figure 6B−D. The 4 cm−1 shift of the COO− bendingvibration63 at 845 cm−1 is due to complexation with TMA asillustrated in Scheme 1. A shoulder also appears at 1397 cm−1,which is attributed to the replacement of −COO−H groupswith −COO−Al− species after TMA exposure. This assign-ment is consistent with literature studies that reported strongerCOO− stretches of adsorbed MBA at ∼1400 cm−1 in its

deprotonated form than the protonated form.71 Chabal and co-workers observed a 1476 cm−1 band after TMA exposure onCOOH-terminated SAMs on Si and assigned it to the COstretch mode found in acid salt structures. They also observedcomplete disappearance of the CO stretch mode at 1718cm−1 after TMA exposure.80 Unfortunately, CO stretches ofMBA are usually very weak in SERS.63 The 2 cm−1 peak shift ofthe 1080 cm−1 peak (CCC ring deformation mode and C−Sstretch)65 (Figure 6C) and the shifted position of the TMA C−H stretch at 2898 cm−1 (Figure 7C,D) are explained by thedirect reaction of TMA with the −COOH groups of MBA,which perturbs MBA molecules and changes their vibrationalfrequencies. Similarly to a bare AgFON and TT-functionalizedAgFON, the TMA C−H stretch at 2898 cm−1 (Figure 7D)decays after H2O exposure because H2O reacts with Al−(CH3)x surface groups and replaces them with Al−(OH)xgroups. TMA regenerates Al−(CH3)x species in the secondTMA half-cycle.For the MP-functionalized AgFON, vibrational frequency

shifts of MP peaks were seen after TMA exposure (FiguresS13A−D and Table S3). The CCC ring deformation and C−S

Figure 7. SER spectra of a TT-functionalized AgFON (A) and MBA-functionalized AgFON (C) showing the evolution of C−H peaks before (a)and after 10 min of TMA exposure (first cycle) (b), 60 s of H2O (first cycle) (c), 10 min of TMA (second cycle) (d), and 60 s of H2O (secondcycle) (e). Difference spectra of the first 2 ALD Al2O3 cycles on a TT-functionalized AgFON (B) and MBA-functionalized AgFON (D). Spectra in Awere obtained with λexc = 532 nm, Pexc = 1 mW, taq = 10 s and 10 accumulations. Spectra in C were acquired with λexc = 532 nm, Pexc = 5 mW, taq = 4s and 10 accumulations each.



H

http://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://dx.doi.org/10.1021/acs.jpcc.5b11487

stretch mode shifted from 1074 to 1075 cm−1.65 The C−Ostretch64−66 shifted from 1290 to 1276 cm−1 after the reactionof TMA with the −COH group. The CC stretch and OHbending mode65,66 shifted from 1577 to 1584 cm−1 as a resultof TMA reacting with the −OH group (Figures S13B−D).Vibrational frequency shifts of SAMs suggest that TMA

selectively reacts with the −COOH and −OH groups of SAMson AgFON. The fwhm of MBA and MP peaks associated withthe −COOH and −OH groups increases significantly (TableS3) after TMA exposure, which suggests that TMA reactsdirectly with the −COOH and −OH groups of the SAMs.Changes in the peak intensities after TMA exposure did notreveal significant differences in the reactivity of TMA with thetypes of SAMs investigated. The observation of the symmetricAl-CH3 stretch on TT- and BT-functionalized and not onMBA- and MP-functionalized AgFONs is consistent with thedistance dependence of SERS in which vibrational modes ofmolecules closer to the plasmonic substrate show strongersignals than those spaced further away.38 Furthermore, theTMA C−H stretches for MBA-functionalized AgFONs (Figure7C,D) are ∼7× weaker than TMA C−H stretches for TT-functionalized AgFONs (Figures 7A,B). This could beexplained by the larger distance of the TMA−CH3 groupsfrom the AgFON surface for MBA-functionalized AgFONsthan TT-functionalized AgFONs as illustrated in Scheme 1. It isalso possible that there is a difference in the amount of TMAdeposited on MBA- and TT-functionalized AgFONs. Forexample, if the amount of TMA deposited on MBA-functionalized SAMs is higher than on TT-functionalizedAgFONs, then the TMA −CH3 groups will be much fartheraway from the surface and would appear much weaker forMBA-functionalized AgFONs. At present, we assume that theamount of TMA deposited on MBA- and TT-functionalizedAgFONs is the same (∼1.1 Å/cycle).The S:Ag atomic ratio, determined by XPS, provides

information about the relative packing density of SAMs onAgFONs. Figure S15 shows S 2p and Ag 3d peaks for all SAMs;the calculated S:Ag ratio increases from 0.0575 to 0.0824 in thefollowing order: TT < BT < MP < MBA. This indicates a lowerpacking density of TT and BT SAMs on AgFONs than MP and

MBA SAMs. This is consistent with the work of Lee and co-workers who observed a lower packing density of BT SAMscompared to MP and MBA SAMs on Au.66 These studiessupport our claim that TMA is able to penetrate through lessdensely packed, unreactive SAMs and react directly with theAgFON surface.

Comparison with Other Studies. Several studies haveused SAMs as enablers of area-selective ALD, a process inwhich ALD material is deposited only where desired. On SiO2surfaces, Chen et al. showed that to completely block HfO2ALD, long (>12 carbon units), linear, densely packed,hydrophobic alkyltrichlorosilane SAMs are necessary.53,81

Using in situ IR spectroscopy, Chabal and co-workers observeddisappearance of CO stretches and no perturbation of alkylchain modes and concluded that COOH-terminated SAMs onSi were effective in blocking TMA from reacting with Si.80 Tothe best of our knowledge, studies that investigate ALDblocking mechanisms using SAMs have been extensivelystudied on Au but not as thoroughly on Ag. Preiner et al.showed that SAMs with hydrophobic tail groups such as 1-dodecanethiol on Au act as ALD resists and block Al2O3 growthwhile SAMs with hydrophilic groups such as MBA allow fastergrowth of Al2O3.

82 Our study on CH3-terminated SAMs isconsistent with Hooper et al., who studied the interaction ofvapor-deposited Al atoms with CH3-terminated SAMs on Auand observed Al−S species by XPS to indicate penetration of Alatoms through the SAM to the Au/SAM interface.83

Furthermore, lack of major perturbations in the SAM IRspectra indicated that no chemical interaction occurredbetween Al and the SAM.83 On COOH-terminated SAMs onAu, Fisher et al. concluded that the −COOH group reacts withAl to prevent Al atoms from penetrating into the monolayer.XPS data indicated that Al atoms bind with both oxygen atomsof the SAM to form metal−organic species. IR spectra showeda 14 cm−1 shift and 25% drop in intensity of the SAM’s COstretch after Al deposition to indicate direct reaction of Alatoms with the −COOH group.74 Similarly, they observedsignificant perturbation of the C−O stretch (a decrease inintensity and 35 cm−1 shift) of OH-terminated SAMs on Au

Scheme 1. Proposed Scheme for the Reaction of TMA with AgFONs Functionalized with Toluenethiol and 4-MercaptobenzoicAcid SAMsa

aThe structure of dimeric TMA contains bridging pentacoordinated carbon atoms as determined by gas-phase electron diffraction.69,79.



I

http://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfhttp://dx.doi.org/10.1021/acs.jpcc.5b11487

using IR spectroscopy and concluded that deposited Alselectively reacts with −OH groups to form −OAl groups.75These observations are consistent with our studies in which

we observe vibrational frequency shifts of OH- and COOH-terminated SAMs after TMA exposure. Finding the ideal SAMon AgFON to completely block ALD growth is beyond thescope of this investigation. Rather, the focus of this study is todemonstrate that operando SERS provides unique informationabout where ALD deposition occurs on SAM-functionalizedsurfaces, information that is difficult to obtain from techniquessuch as FTIR, QCM, QMS, and spectroscopic ellipsometry.SERS provides valuable information about surface-boundmolecular species and on where the ALD deposition occurson the surface under realistic ALD working conditions. Ourstudy reveals that TMA forms dimeric species on the AgFONsurface and this information is difficult to obtain from othertechniques. The rich structural information provided by SERScan be used to provide insight about the surface chemistry ofALD reactions.

■ CONCLUSIONSThis work demonstrates the capability of operando SERS tomeasure surface vibrational spectra of adsorbates during ALDreactions. On bare AgFONs, Al−C, symmetric C−H, andasymmetric C−H stretches were observed after TMA exposureat 55 °C, and they decayed after dosing H2O. Residual H2O inthe ALD chamber reacted with −CH3 groups on AgFONs asevidenced by the decay of the Al−C and C−H stretches evenbefore H2O exposure. DFT studies revealed that the observedAl−C stretches were consistent with TMA dimeric species onthe AgFON surface. On thiol SAM-functionalized AgFONs,Al−C stretches and no thiol vibrational frequency shifts wereseen after TMA exposure in the case of toluenethiol andbenzenethiol SAMs. Thiol vibrational frequency shifts and noAl−C stretches were observed after TMA exposure onAgFONs functionalized with 4-mercaptobenzoic acid and 4-mercaptophenol SAMs. For COOH- and OH-terminatedSAMs, the absence of observable Al−C stretches was due totheir larger distance away from the enhancing SERS substrate.Operando SERS revealed that TMA selectively reacted with theAgFON surface for benzenethiol and toluenethiol SAMs andwith the −COOH and −OH groups for 4-mercaptobenzoicacid and 4-mercaptophenol SAMs, respectively. This shows thatSAMs can be used to tailor the surface and direct ALD Al2O3growth to achieve area-selective ALD. The surface sensitivityand rich structural information provided by operando SERS willallow for the probing of the deposition mechanisms of otherALD reactions such as metal ALD processes under realisticreaction conditions. Knowledge of the surface chemistry ofALD reactions will help in the selection of better precursorsand in the optimization of ALD processes.

■ ASSOCIATED CONTENT*S Supporting InformationThe Supporting Information is available free of charge on theACS Publications website at DOI: 10.1021/acs.jpcc.5b11487.

Extinction spectra of AgFONs, SER spectra acquiredwith and without the quartz window, SER spectra ofAgFONs before and after Ar plasma cleaning, DFT-calculated Raman spectra of TMA, SER spectra showingreaction of surface species with residual H2O, laser powerdependence study, SER spectra of benzenethiol and 4-

mercaptophenol before and after ALD, and XP spectra(PDF)

■ AUTHOR INFORMATIONCorresponding Author*E-mail: [email protected] ContributionsThe manuscript was written through contributions of allauthors. All authors have given approval to the final version ofthe manuscript.

NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSWe acknowledge Dr. Jon Dieringer, Dr. Bogdan Negru, Dr.Neil Schweitzer, Dr. Dragos Seghete, Dr. Nathan Greeneltch,Stephanie Zaleski, and Naihao Chiang for experimental help,data analysis, and valuable discussions. We gratefully acknowl-edge financial support from the Northwestern UniversityInstitute for Catalysis in Energy Processes (ICEP). ICEP isfunded through the US Department of Energy, Office of BasicEnergy Science (Award No. DE-FG02-03-ER15457). M.O.M.acknowledges support from the National Science FoundationGraduate Fellowship Research Program under Grant No. DGE-0824162. This work made use of the EPIC facility (NUANCECenter-Northwestern University), which has received supportfrom the MRSEC program (NSF DMR-1121262) at theMaterials Research Center, the International Institute forNanotechnology (IIN), and the State of Illinois, through theIIN.

■ REFERENCES(1) George, S. M. Atomic Layer Deposition: An Overview. Chem.Rev. 2010, 110, 111−131.(2) Puurunen, R. L. Surface Chemistry of Atomic Layer Deposition:A Case Study for the Trimethylaluminum/Water Process. J. Appl. Phys.2005, 97, 121301.(3) Miikkulainen, V.; Leskela,̈ M.; Ritala, M.; Puurunen, R. L.Crystallinity of Inorganic Films Grown by Atomic Layer Deposition:Overview and General Trends. J. Appl. Phys. 2013, 113, 021301.(4) Suntola, T. Atomic Layer Epitaxy. Thin Solid Films 1992, 216,84−89.(5) Lu, J.; Stair, P. C. Nano/Subnanometer Pd Nanoparticles onOxide Supports Synthesized by AB-type and Low-Temperature ABC-type Atomic Layer Deposition: Growth and Morphology. Langmuir2010, 26, 16486−16495.(6) Lu, J.; Stair, P. C. Low-Temperature ABC-type Atomic LayerDeposition: Synthesis of Highly Uniform Ultrafine Supported MetalNanoparticles. Angew. Chem., Int. Ed. 2010, 49, 2547−2551.(7) Lu, J.; Fu, B.; Kung, M. C.; Xiao, G.; Elam, J. W.; Kung, H. H.;Stair, P. C. Coking- and Sintering-Resistant Palladium CatalystsAchieved Through Atomic Layer Deposition. Science 2012, 335,1205−1208.(8) Feng, H.; Libera, J. A.; Stair, P. C.; Miller, J. T.; Elam, J. W.Subnanometer Palladium Particles Synthesized by Atomic LayerDeposition. ACS Catal. 2011, 1, 665−673.(9) Christensen, S. T.; Elam, J. W.; Rabuffetti, F. A.; Ma, Q.;Weigand, S. J.; Lee, B.; Seifert, S.; Stair, P. C.; Poeppelmeier, K. R.;Hersam, M. C.; et al. Controlled Growth of Platinum Nanoparticles onStrontium Titanate Nanocubes by Atomic Layer Deposition. Small2009, 5, 750−757.(10) Baker, L.; Cavanagh, A. S.; Seghete, D.; George, S. M.; Mackus,A. J. M.; Kessels, W. M. M.; Liu, Z. Y.; Wagner, F. T. Nucleation andGrowth of Pt Atomic Layer Deposition on Al2O3 Substrates Using



J

http://pubs.acs.orghttp://pubs.acs.org/doi/abs/10.1021/acs.jpcc.5b11487http://pubs.acs.org/doi/suppl/10.1021/acs.jpcc.5b11487/suppl_file/jp5b11487_si_001.pdfmailto:[email protected]://dx.doi.org/10.1021/acs.jpcc.5b11487

(Methylcyclopentadienyl)-Trimethyl Platinum and O2 Plasma. J. Appl.Phys. 2011, 109, 084333.(11) Li, J.; Liang, X.; King, D. M.; Jiang, Y.-B.; Weimer, A. W. HighlyDispersed Pt Nanoparticle Catalyst Prepared by Atomic LayerDeposition. Appl. Catal., B 2010, 97, 220−226.(12) Masango, S. S.; Peng, L.; Marks, L. D.; Van Duyne, R. P.; Stair,P. C. Nucleation and Growth of Silver Nanoparticles by AB and ABC-Type Atomic Layer Deposition. J. Phys. Chem. C 2014, 118, 17655−17661.(13) Biener, J.; Baumann, T. F.; Wang, Y.; Nelson, E. J.; Kucheyev, S.O.; Hamza, A. V.; Kemell, M.; Ritala, M.; Leskela,̈ M. Ruthenium/Aerogel Nanocomposites via Atomic Layer Deposition. Nanotechnol-ogy 2007, 18, 055303.(14) Christensen, S. T.; Feng, H.; Libera, J. A.; Guo, N.; Miller, J. T.;Stair, P. C.; Elam, J. W. Supported Ru-Pt Bimetallic NanoparticleCatalysts Prepared by Atomic Layer Deposition. Nano Lett. 2010, 10,3047−3051.(15) Weber, M. J.; Mackus, A. J. M.; Verheijen, M. A.; van der Marel,C.; Kessels, W. M. M. Supported Core/Shell Bimetallic NanoparticlesSynthesis by Atomic Layer Deposition. Chem. Mater. 2012, 24, 2973−2977.(16) Lei, Y.; Liu, B.; Lu, J.; Lobo-Lapidus, R. J.; Wu, T.; Feng, H.;Xia, X.; Mane, A. U.; Libera, J. A.; Greeley, J.; et al. Synthesis of Pt-PdCore-Shell Nanostructures by Atomic Layer Deposition: Applicationin Propane Oxidative Dehydrogenation to Propylene. Chem. Mater.2012, 24, 3525−3533.(17) Zaera, F. The Surface Chemistry of Thin Film Atomic LayerDeposition (ALD) Processes for Electronic Device Manufacturing. J.Mater. Chem. 2008, 18, 3521−3526.(18) Zaera, F. The Surface Chemistry of Atomic Layer Depositionsof Solid Thin Films. J. Phys. Chem. Lett. 2012, 3, 1301−1309.(19) Banares, M. A. Operando Methodology: Combination of In SituSpectroscopy and Simultaneous Activity Measurements UnderCatalytic Reaction Conditions. Catal. Today 2005, 100, 71−77.(20) Guerrero-Perez, M. O.; Banares, M. A. From Conventional insitu to Operando Studies in Raman Spectroscopy. Catal. Today 2006,113, 48−57.(21) Elam, J. W.; Groner, M. D.; George, S. M. Viscous Flow Reactorwith Quartz Crystal Microbalance for Thin Film Growth by AtomicLayer Deposition. Rev. Sci. Instrum. 2002, 73, 2981.(22) Matero, R.; Rahtu, A.; Ritala, M. In Situ Quadrupole MassSpectrometry and Quartz Crystal Microbalance Studies on the AtomicLayer Deposition of Titanium Dioxide from Titanium Tetrachlorideand Water. Chem. Mater. 2001, 13, 4506−4511.(23) Goldstein, D. N.; McCormick, J. A.; George, S. M. Al2O3Atomic Layer Deposition with Trimethylaluminum and OzoneStudied by in Situ Transmission FTIR Spectroscopy and QuadrupoleMass Spectrometry. J. Phys. Chem. C 2008, 112, 19530−19539.(24) Langereis, E.; Heil, S. B. S.; Knoops, H. C. M.; Keuning, W.; vande Sanden, M. C. M.; Kessels, W. M. M. In Situ SpectroscopicEllipsometry as a Versatile Tool for Studying Atomic LayerDeposition. J. Phys. D: Appl. Phys. 2009, 42, 073001.(25) Bayer, T. J. M.; Wachau, A.; Fuchs, A.; Deuermeier, J.; Klein, A.Atomic Layer Deposition of Al2O3 onto Sn-Doped In2O3: Absence ofSelf-Limited Adsorption during Initial Growth by Oxygen Diffusionfrom the Substrate and Band Offset Modification by Fermi LevelPinning in Al2O3. Chem. Mater. 2012, 24, 4503−4510.(26) Johnson, A. M.; Stair, P. C. Alternative Low-Pressure SurfaceChemistry of Titanium Tetraisopropoxide on Oxidized Molybdenum.J. Phys. Chem. C 2014, 118, 29361−29369.(27) Starr, D. E.; Liu, Z.; Havecker, M.; Knop-Gericke, A.; Bluhm, H.Investigation of Solid/Vapor Interfaces Using Ambient Pressure X-rayPhotoelectron Spectroscopy. Chem. Soc. Rev. 2013, 42, 5833.(28) Schnadt, J.; Knudsen, J.; Andersen, J. N.; Siegbahn, H.; Pietzsch,A.; Hennies, F.; Johansson, N.; Martensson, N.; Ohrwall, G.; Bahr, S.;et al. The New Ambient-Pressure X-ray Photoelectron SpectroscopyInstrument at MAX-Lab. J. Synchrotron Radiat. 2012, 19, 701−704.(29) Knudsen, J.; Andersen, J. N.; Schnadt, J. A Versatile Instrumentfor Ambient Pressure X-ray Photoelectron Spectroscopy: The Lund

Cell Approach. Surf. Sci. 2016, 646, 160−169, http://dx.doi.org/10.1016/j.susc.2015.10.038,.(30) Tao, F. Design of an In-house Ambient Pressure AP-XPS Using aBench-top X-ray Source and the Surface Chemistry of Ceria UnderReaction Conditions. Chem. Commun. 2012, 48, 3812−3814.(31) Tao, F. Operando Studies of Catalyst Surfaces During Catalysisand Under Reaction Conditions: Ambient Pressure X-Ray Photo-electron Spectroscopy with a Flow-Cell Reactor. ChemCatChem 2012,4, 583−590.(32) Jurgensen, A.; Heutz, N.; Raschke, H.; Merz, K.; Hergenroder,R. Behavior of Supported Palladium Oxide Nanoparticles UnderReaction Conditions, Studied with Near Ambient Pressure XPS. Anal.Chem. 2015, 87, 7848−7856.(33) Biswas, M.; Libera, J. A.; Darling, S. B.; Elam, J. W. New Insightinto the Mechanism of Sequential Infiltration Synthesis from InfraredSpectroscopy. Chem. Mater. 2014, 26, 6135−6141.(34) Kim, H.; Kosuda, K. M.; Van Duyne, R. P.; Stair, P. C.Resonance Raman and Surface- and Tip-Enhanced Raman Spectros-copy Methods to Study Solid Catalysts and Heterogeneous CatalyticReactions. Chem. Soc. Rev. 2010, 39, 4820.(35) McCreery, R. L. Raman Spectroscopy for Chemical Analysis; JohnWiley: New York, 2000; Vol. 157.(36) Jeanmaire, D. L.; Van Duyne, R. P. Surface RamanSpectroelectrochemistry. Part I. Heterocyclic, Aromatic, and AliphaticAmines Adsorbed on the Anodized Silver Electrode. J. Electroanal.Chem. Interfacial Electrochem. 1977, 84, 1−20.(37) Dieringer, J. A.; Lettan, R. B.; Scheidt, K. A.; Van Duyne, R. P. AFrequency Domain Existence Proof of Single-Molecule Surface-Enhanced Raman Spectroscopy. J. Am. Chem. Soc. 2007, 129,16249−16256.(38) Stiles, P. L.; Dieringer, J. A.; Shah, N. C.; Van Duyne, R. P.Surface-Enhanced Raman Spectroscopy. Annu. Rev. Anal. Chem. 2008,1, 601−626.(39) Schatz, G. C.; Van Duyne, R. P., Electromagnetic Mechanism ofSurface-Enhanced Spectroscopy. In Handbook of Vibrational Spectros-copy; Chalmers, J. M., Griffiths, P. R., Eds.; Wiley: New York, 2002; pp759−774.(40) Willets, K. A.; Van Duyne, R. P. Localized Surface PlasmonResonance Spectroscopy and Sensing. Annu. Rev. Phys. Chem. 2007,58, 267−297.(41) John, J. F.; Mahurin, S. M.; Dai, S.; Sepaniak, M. J. Use ofAtomic Layer Deposition to Improve the Stability of Silver Substratesfor In Situ High-Temperature SERS Measurements. J. RamanSpectrosc. 2010, 41, 4−11.(42) Formo, E. V.; Wu, Z.; Mahurin, S. M.; Dai, M. In Situ HighTemperature Surface Enhanced Raman Spectroscopy for the Study ofInterface Phenomena: Probing a Solid Acid on Alumina. J. Phys. Chem.C 2011, 115, 9068−9073.(43) Whitney, A. V.; Elam, J. W.; Zou, S.; Zinovev, A. V.; Stair, P. C.;Schatz, G. C.; Van Duyne, R. P. Localized Surface Plasmon ResonanceNanosensor: A High-Resolution Distance-Dependence Study UsingAtomic Layer Deposition. J. Phys. Chem. B 2005, 109, 20522−20528.(44) Rahtu, A.; Alaranta, T.; Ritala, M. In Situ Quartz CrystalMicrobalance and Quadrupole Mass Spectrometry Studies of AtomicLayer Deposition of Aluminum Oxide from Trimethylaluminum andWater. Langmuir 2001, 17, 6506−6509.(45) Groner, M. D.; Fabreguette, F. H.; Elam, J. W.; George, S. M.Low-Temperature Al2O3 Atomic Layer Deposition. Chem. Mater.2004, 16, 639−645.(46) Lu, J.; Liu, B.; Guisinger, N. P.; Stair, P. C.; Greeley, J. P.; Elam,J. W. First-Principles Predictions and In Situ Experimental Validationof Alumina Atomic Layer Deposition on Metal Surfaces. Chem. Mater.2014, 26, 6752−6761.(47) Lu, J.; Liu, B.; Greeley, J. P.; Feng, Z.; Libera, J. A.; Lei, Y.;Bedzyk, M. J.; Stair, P. C.; Elam, J. W. Porous Alumina ProtectiveCoatings on Palladium Nanoparticles by Self-Poisoned Atomic LayerDeposition. Chem. Mater. 2012, 24, 2047−2055.



K

http://dx.doi.org/10.1016/j.susc.2015.10.038http://dx.doi.org/10.1016/j.susc.2015.10.038http://dx.doi.org/10.1021/acs.jpcc.5b11487

(48) Feng, H.; Lu, J.; Stair, P. C.; Elam, J. W. Alumina Over-coatingon Pd Nanoparticle Catalysts by Atomic Layer Deposition: EnhancedStability and Reactivity. Catal. Lett. 2011, 141, 512−517.(49) Whitney, A. V.; Elam, J. W.; Stair, P. C.; Van Duyne, R. P.Toward a Thermally Robust Operando Surface-Enhanced RamanSpectroscopy Substrate. J. Phys. Chem. C 2007, 111, 16827−16832.(50) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.;Whitesides, G. M. Self-Assembled Monolayers of Thiolates on Metalsas a Form of Nanotechnology. Chem. Rev. 2005, 105, 1103−1170.(51) Norrod, K. L.; Rowlen, K. L. Removal of CarbonaceousContamination from SERS-Active Silver by Self-Assembly ofDecanethiol. Anal. Chem. 1998, 70, 4218−4221.(52) Lin, X.-M.; Cui, Y.; Xu, Y.-H.; Ren, B.; Tian, Z.-Q. Surface-Enhanced Raman Spectroscopy: Substrate-Related Issues. Anal.Bioanal. Chem. 2009, 394, 1729−1745.(53) Chen, R.; Kim, H.; McIntyre, P. C.; Porter, D. W.; Bent, S. F.Achieving Area-Selective Atomic Layer Deposition on PatternedSubstrates by Selective Surface Modification. Appl. Phys. Lett. 2005, 86,191910.(54) Fang, M.; Ho, J. C. Area-Selective Atomic Layer Deposition:Conformal Coating, Subnanometer Thickness Control, and SmartPositioning. ACS Nano 2015, 9, 8651−8654.(55) Greeneltch, N. G.; Blaber, M. G.; Henry, A.-I.; Schatz, G. C.;Van Duyne, R. P. Immobilized Nanorod Assemblies: Fabrication andUnderstanding of Large Area Surface-Enhanced Raman SpectroscopySubstrates. Anal. Chem. 2013, 85, 2297−2303.(56) Baerends, E. J. ADF 2014; SCM: Amsterdam, 2015.(57) Kleinman, S. L.; Frontiera, R. R.; Henry, A.-I.; Dieringer, J. A.;Van Duyne, R. P. Creating, Characterizing, and Controlling Chemistrywith SERS Hot Spots. Phys. Chem. Chem. Phys. 2013, 15, 21−36.(58) Ogawa, T. Vibrational Assignments and Normal Vibrations ofTrimethylaluminium. Spectrochim. Acta 1968, 24, 15−20.(59) Ferguson, J. D.; Weimer, A. W.; George, S. M. Atomic LayerDeposition of Al2O3 Films on Polyethylene Particles. Chem. Mater.2004, 16, 5602−5609.(60) Soto, C.; Tysoe, W. T. The Reaction Pathway for the Growth ofAlumina on High Surface Area Alumina and in Ultrahigh Vacuum by aReaction Between Trimethyl Aluminum and Water. J. Vac. Sci.Technol., A 1991, 9, 2686−2695.(61) George, S. M.; Ott, A. W.; Klaus, J. W. Surface Chemistry forAtomic Layer Growth. J. Phys. Chem. 1996, 100, 13121−13131.(62) Goldstein, D. N.; George, S. M. Enhancing the Nucleation ofPalladium Atomic Layer Deposition on Al2O3 Using Trimethylalumi-num to Prevent Surface Poisoning by Reaction Products. Appl. Phys.Lett. 2009, 95, 143106.(63) Lee, S. B.; Kim, K.; Kim, M. S. Surface-Enhanced RamanScattering of o-Mercaptobenzoic Acid in Silver Sol. J. Raman Spectrosc.1991, 22, 811−817.(64) Francioso, O.; Sanchez-Cortes, S.; Tugnoli, V.; Ciavatta, C.;Sitti, L.; Gessa, C. Infrared, Raman, and Nuclear Magnetic Resonance(1H, 13C, and 31P) Spectroscopy in the Study of Fractions of PeatHumic Acids. Appl. Spectrosc. 1996, 50, 1165−1174.(65) Cabalo, J.; Guicheteau, J. A.; Christesen, S. Toward Under-standing the Influence of Intermolecular Interactions and MolecularOrientation on the Chemical Enhancement of SERS. J. Phys. Chem. A2013, 117, 9028−9038.(66) Barriet, D.; Yam, C. M.; Shmakova, O. E.; Jamison, A. C.; Lee,R. T. 4-Mercaptophenylboronic Acid SAMs on Gold: Comparisonwith SAMs Derived from Thiophenol, 4-Mercaptophenol, and 4-Mercaptobenzoic Acid. Langmuir 2007, 23, 8866−8875.(67) Nagabalasubramanian, P. B.; Periandy, S.; Mohan, S.;Govindarajan, M. FTIR and Raman Spectra, Vibrational Assignments,Ab Initio, DFT and Normal Coordinate Analysis of α,α Dichlor-otoluene. Spectrochim. Acta, Part A 2009, 73, 277−280.(68) Sebek, J.; Knaanie, R.; Albee, B.; Potma, E. O.; Gerber, R. B.Spectroscopy of the C-H Stretching Vibrational Band in SelectedOrganic Molecules. J. Phys. Chem. A 2013, 117, 7442−7452.(69) Berthomieu, D.; Bacquet, Y.; Pedocchi, L.; Goursot, A.Trimethylaluminum Dimer Structure and its Monomer Radical

Cation: A Density Functional Study. J. Phys. Chem. A 1998, 102,7821−7827.(70) Willis, B. G.; Jensen, K. F. An Evaluation of Density FunctionalTheory and ab Initio Predictions for Bridge-Bonded AluminumCompounds. J. Phys. Chem. A 1998, 102, 2613−2623.(71) Osinkina, L.; Lohmuller, T.; Jackel, F.; Feldmann, J. Synthesis ofGold Nanostar Arrays as Reliable, Large-Scale, HomogeneousSubstrates for Surface-Enhanced Raman Scattering Imaging andSpectroscopy. J. Phys. Chem. C 2013, 117, 22198−22202.(72) Sano, H.; Mizutani, G.; Ushioda, S. Absolute Raman ScatteringCross Sections of Trimethylgallium and Trimethylaluminum in theVapor Phase. Chem. Phys. Lett. 1992, 194, 398−402.(73) Lee, T. G.; Kim, K.; Kim, M. S. Raman Scattering of α-Toluenethiol Adsorbed on Silver Sol. J. Raman Spectrosc. 1991, 22,339−344.(74) Fisher, G. L.; Hooper, A. E.; Opila, R. L.; Allara, D. L.;Winograd, N. The Interaction of Vapor-Deposited Al Atoms withCO2H Groups at the Surface of a Self-Assembled AlkanethiolateMonolayer on Gold. J. Phys. Chem. B 2000, 104, 3267−3273.(75) Fisher, G. L.; Walker, A. V.; Hooper, A. E.; Tighe, T. B.; Bahnck,K. B.; Skriba, H. T.; Reinard, M. D.; Haynie, B. C.; Opila, R. L.;Winograd, N.; et al. Bond Insertion, Complexation, and PenetrationPathways of Vapor-Deposited Aluminum Atoms with HO- and CH3O-Terminated Organic Monolayers. J. Am. Chem. Soc. 2002, 124, 5528−5541.(76) Xiong, G.; Elam, J. W.; Feng, H.; Han, C. Y.; Wang, H.-H.; Iton,L. E.; Curtiss, L. A.; Pellin, M. J.; Kung, M.; Kung, H.; et al. Effect ofAtomic Layer Deposition Coatings on the Surface Structure of AnodicAluminum Oxide Membranes. J. Phys. Chem. B 2005, 109, 14059−14063.(77) Carron, K. T.; Hurley, L. G. Axial and Azimuthal AngleDetermination with Surface-Enhanced Raman Spectroscopy: Thio-phenol on Copper, Silver, and Gold Metal Surfaces. J. Phys. Chem.1991, 95, 9979−9984.(78) Allen, C. S.; Van Duyne, R. P. Orientational Specificity ofRaman Scattering from Molecules Adsorbed on Silver Electrodes.Chem. Phys. Lett. 1979, 63, 455−459.(79) Almenningen, A.; Halvorsen, S.; Haaland, A. A Gas PhaseElectron Diffraction Investigation of the Molecular Structures ofTrimethylaluminium Monomer and Dimer. Acta Chem. Scand. 1971,25, 1937−1945.(80) Li, M.; Dai, M.; Chabal, Y. J. Atomic Layer Deposition ofAluminum Oxide on Carboxylic Acid-Terminated Self-AssembledMonolayers. Langmuir 2009, 25, 1911−1914.(81) Chen, R.; Kim, H.; McIntyre, P. C.; Bent, S. F. Investigation ofSelf-Assembled Monolayer Resists for Hafnium Dioxide Atomic LayerDeposition. Chem. Mater. 2005, 17, 536−544.(82) Preiner, M. J.; Melosh, N. A. Creating Large Area MolecularElectronic Junctions Using Atomic Layer Deposition. Appl. Phys. Lett.2008, 92, 213301.(83) Hooper, A.; Fisher, G. L.; Konstadinidis, K.; Jung, D.; Nguyen,H.; Opila, R.; Collins, R. W.; Winograd, N.; Allara, D. L. ChemicalEffects of Methyl and Methyl Ester Groups on the Nucleation andGrowth of Vapor-Deposited Aluminum Films. J. Am. Chem. Soc. 1999,121, 8052−8064.



L
http://dx.doi.org/10.1021/acs.jpcc.5b11487

probing the chemistry of alumina atomic layer deposition ......atomic layer deposition (ald) relies...

Documents