Heterometallic antenna−reactor complexesfor photocatalysisDayne F. Swearera, Hangqi Zhaob, Linan Zhoua, Chao Zhangb, Hossein Robatjazib, John Mark P. Martirezc,Caroline M. Krauterc, Sadegh Yazdid, Michael J. McClaina, Emilie Ringea,d, Emily A. Carterc,e,f, Peter Nordlanderb,d,g,1,and Naomi J. Halasa,b,d,g,1

aDepartment of Chemistry, Rice University, Houston, TX 77005; bDepartment of Electrical and Computer Engineering, Rice University, Houston, TX 77005;cDepartment of Mechanical and Aerospace Engineering, Princeton University, Princeton, NJ 08544; dDepartment of Material Science and Nanoengineering,Rice University, Houston, TX 77005; eProgram in Applied and Computational Mathematics, Princeton University, Princeton, NJ 08544; fAndlinger Center forEnergy and the Environment, Princeton University, Princeton, NJ 08544; and gDepartment of Physics and Astronomy, Rice University, Houston, TX 77005

Contributed by Naomi J. Halas, June 20, 2016 (sent for review May 31, 2016; reviewed by Martin Moskovits and Wei David Wei)

Metallic nanoparticles with strong optically resonant propertiesbehave as nanoscale optical antennas, and have recently shownextraordinary promise as light-driven catalysts. Traditionally, how-ever, heterogeneous catalysis has relied uponweakly light-absorbingmetals such as Pd, Pt, Ru, or Rh to lower the activation energy forchemical reactions. Here we show that coupling a plasmonic nano-antenna directly to catalytic nanoparticles enables the light-inducedgeneration of hot carriers within the catalyst nanoparticles, trans-forming the entire complex into an efficient light-controlled reactivecatalyst. In Pd-decorated Al nanocrystals, photocatalytic hydrogendesorption closely follows the antenna-induced local absorptioncross-section of the Pd islands, and a supralinear power dependencestrongly suggests that hot-carrier-induced desorption occurs at thePd island surface. When acetylene is present along with hydrogen,the selectivity for photocatalytic ethylene production relative toethane is strongly enhanced, approaching 40:1. These observationsindicate that antenna−reactor complexes may greatly expand possi-bilities for developing designer photocatalytic substrates.

plasmon | photocatalysis | nanoparticle | catalysis | aluminum

Industrial processes depend extensively on heterogeneous catalystsfor chemical production and mitigation of environmental pollut-

ants. These processes often rely on metal nanoparticles dispersedinto high surface area support materials to both maximize catalyti-cally active surface area and for the most cost-effective use of ex-pensive catalysts such as Pd, Pt, Ru, or Rh (1, 2). However, catalyticprocesses utilizing transition metal nanoparticles are often energy-intensive, relying on high temperatures and pressures to maximizecatalytic activity. A transition from extreme, high-temperature con-ditions to low-temperature activation of catalytically active transitionmetal nanoparticles could have widespread impact, substantiallyreducing the current energy demands of heterogeneous catalysis.Light-driven chemical transformations offer an attractive and

ultimately sustainable alternative to traditional high-temperaturecatalytic reactions. Metallic plasmonic nanostructures are a newparadigm in photoactive heterogeneous catalysts (3–6). Plasmonicnanoparticles uniquely couple electron density with electromag-netic radiation, leading to a collective oscillation of the conductionelectrons in resonance with the frequency of incident light, knownas a localized surface plasmon resonance (LSPR). These reso-nances lead to enhanced light absorption in an area much largerthan the physical cross-section of the nanoparticle, and such opticalantenna effects result in strongly enhanced electromagnetic fieldsnear the nanoparticle surface. An LSPR can be damped throughradiative reemission of a photon, or nonradiative Landau dampingwith the creation of energetic “hot” carriers: electrons above theFermi energy of the metal and/or holes below the Fermi energy. Inthis context, “hot” refers to carriers of an energy that is a significantfraction of the plasmon energy that would not be generated ther-mally at ambient temperature. Plasmonic metal nanoparticles havebeen shown to induce chemical transformations directly on their

surfaces, through either phonon-driven or charge-carrier-drivenmechanisms in Au (7–10), Ag (11, 12), Cu (13, 14), and, recently,Al (15) nanoparticles. Although these “good” plasmonic metalsshow initial promise for plasmon-induced photocatalytic chemistry,in general they are not universally good catalytic materials despitefinding niche applications in a few industrial processes.In comparison, noncoinage transition metals have historical pre-

cedence as excellent catalysts, yet are generally considered poorplasmonic metals, because they suffer from large nonradiative damp-ing, which results in broad spectral features and weak absorption acrossthe visible region of the spectrum (16–18). Many catalytic transi-tion metal nanoparticles (Pt, Pd, Rh, Ru, etc.) possess LSPRs inthe UV, but this is disadvantageous for photocatalysis because ofpoor overlap with conventional laser sources or, alternatively, withthe solar spectrum. Increasing transition metal nanoparticle sizeredshifts optical absorption, but it increases cost and reduces sur-face area, and therefore catalytic activity. Recently, it has beenshown that plasmonic nanoparticles can be used to increase opticalabsorption in adjacent nanoparticles (19–22), for instance, enablinghydrogen detection (23, 24).Previous reports of photocatalytic transformation in plasmonic

metal nanoparticle systems rely on the metal to double as both the

Significance

Plasmon-enhanced photocatalysis holds significant promise forcontrolling chemical reaction rates and outcomes. Unfortunately,traditional plasmonic metals have limited surface chemistry, whileconventional catalysts are poor optical absorbers. By placing acatalytic reactor particle adjacent to a plasmonic antenna, thehighly efficient and tunable light-harvesting capacities of plas-monic nanoparticles can be exploited to drastically increase ab-sorption and hot-carrier generation in the reactor nanoparticles.We demonstrate this antenna−reactor concept by showing thatplasmonic aluminum nanocrystal antennas decorated with smallcatalytic palladium reactor particles exhibit dramatically increasedphotocatalytic activity over their individual components. Themodularity of this approach provides for independent control ofchemical and light-harvesting properties and paves theway for therational, predictive design of efficient plasmonic photocatalysts.

Author contributions: P.N. and N.J.H. designed research; D.F.S., H.Z., L.Z., C.Z., H.R., J.M.P.M.,C.M.K., S.Y., and M.J.M. performed research; D.F.S., H.Z., J.M.P.M., C.M.K., E.R., E.A.C.,P.N., and N.J.H. analyzed data; D.F.S., H.Z., J.M.P.M., C.M.K., E.R., E.A.C., P.N., and N.J.H.wrote the paper; D.F.S., S.Y., and E.R. performed STEM-EELS experiments; and P.N. andN.J.H. conceived the antenna−reactor concept.

Reviewers: M.M., University of California, Santa Barbara; andW.D.W., University of Florida.

The authors declare no conflict of interest.

Freely available online through the PNAS open access option.1To whom correspondence may be addressed. Email: halas@rice.edu or nordland@rice.edu.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1609769113/-/DCSupplemental.

8916–8920 | PNAS | August 9, 2016 | vol. 113 | no. 32 www.pnas.org/cgi/doi/10.1073/pnas.1609769113

Dow

nloa

ded

by g

uest

on

June

2, 2

020

light-harvesting antenna and the catalytic surface. Here we showthat the optical antenna effects of plasmonic metal nanoparticlescan be used to directly enhance light absorption and modify thecatalytic activity of directly adjacent reactive metal nanoparticlesurfaces. This “antenna−reactor” complex, with the antennaand reactor composed of two distinct materials, is illustratedschematically in Fig. 1A. We note that the experimental antenna−reactor complexes designed for this work are separated by 2- to4-nm dielectric aluminum oxide interfacial layers rather thancomplete separation as illustrated in Fig. 1A. Here we show thatantenna−reactor complexes, focusing on Al as the antenna and Pdas the reactor, can be used to photoactively drive catalytic reac-tions under mild, ambient temperature and pressure conditions.Such modular, heterometallic complexes offer greatly increaseddegrees of freedom in the design of photocatalytic complexes,expanding the possible materials that can be used as light-drivencatalysts. Manipulating the materials used for both the plasmonicantenna and catalytic reactor can theoretically lead to numerouspossibilities for controlling plasmon-assisted absorption enhance-ments and specific reactivities (SI Appendix, Figs. S1 and S2).

ResultsWe synthesized heterometallic nanoparticle complexes consisting ofaluminum nanocrystals (AlNC) (25) decorated with multiple smallerPd islands as an embodiment of the antenna−reactor concept. Pdnanoparticles function as the optically lossy catalytic material cou-pled to the plasmonic AlNC “antenna.” The AlNC core is sur-rounded by an intrinsic 2- to 4-nm self-limiting oxide (26), separatingit from Pd islands grown directly onto the outside of the Al2O3 shellusing a weak-capping growth approach (27) (SI Appendix, Fig. S3).This antenna−reactor geometry leads to internal field enhance-ments and increased optical absorption in the Pd islands decoratingthe surface. We show, via calculated near-field enhancements in a10-nm Pd island at the surface of a 110-nm diameter AlNC (Fig. 1 Band C) and, for comparison, a 110-nm Al2O3 nanosphere (Fig. 1 Dand E), the effect of coupling a plasmonic antenna to an opticallylossy transition metal. The local field enhancement in the Pd islandfor the antenna−reactor geometry is over one order of magnitudehigher than for the Al2O3 nanosphere case, due to the plasmon-enhanced near fields at the AlNC surface. This effect leads to adrastic increase in hot-carrier production in the reactive Pd islandsdecorating the surface, because hot-carrier generation is pro-portional to internal field intensity enhancement rather thanbulk material absorption (28).The wavelength-dependent absorption in Pd for the AlNC−Pd

antenna−reactor and the Al2O3−Pd control case was calculatedusing the finite difference time domain (FDTD) method and isshown in Fig. 1F. The AlNC−Pd antenna−reactor geometry (solidred curve) shows an increased absorption at ∼500 nm corre-sponding to plasmon-assisted absorption enhancements from thedipolar resonance of the AlNC (SI Appendix, Fig. S4). In contrast,the Al2O3–Pd geometry (black curve) shows only a broad, weakabsorption across the visible region of the spectrum, with slightincrease in the UV (<400 nm).We also show that the absorption enhancements in such het-

erometallic systems can be calculated directly as the product ofabsorption in the reactor metal (Fig. 1F, black curve) and thenear-field enhancements (blue curve) to yield the calculatedabsorption (dashed red curve) without the use of extensiveFDTD simulations. Interestingly, if the original Pd absorptionin an isolated geometry (black curve) is multiplied by the electricfield enhancement (blue curve) on the antenna particle alone atthe position where the Pd island would be situated, this calculatedabsorption (dashed red curve) closely matches the Pd absorptionin the coupled geometry. This provides further evidence for theantenna-induced absorption enhancement mechanism and suggeststhat the optical response of the antenna−reactor system can bedetermined by modeling separate responses from the individual

catalytic reactor and the plasmonic antenna. This insight pro-vides a path for straightforward optimization of antenna−reactorstructures, where the antenna response can be tuned indepen-dently to provide the best spectral overlap with reactor photo-catalytic activity.Experimental evidence for coupling between the Al plasmonic

nanocrystal and the Pd islands shows distinct modes correspondingto the individual antenna and reactor constituents as well as thecoupled system (Fig. 2). Using electron energy loss spectroscopy

Fig. 1. Absorption enhancements in heterometallic antenna−reactor systems.(A) Generalized schematic of a simple system containing a plasmonic antennacoupled through localized near-field enhancements to a catalytic reactormetal nanoparticle. (B) A simplified Pd−AlNC antenna−reactor model,consisting of a 20-nm Pd island on a 110-nm-diameter AlNC nanosphere.(C) Near-field enhancement in the Pd island for this antenna−reactor ge-ometry. (D) Schematic of a comparative geometry, a Pd island on a 110-nm-diameter dielectric Al2O3 nanosphere. (E) The near-field enhancement in thisgeometry is substantially reduced relative to C. (F) Absorption of Pd on Al2O3

(black) and an antenna−reactor geometry using FDTD (red solid curve) andisolated absorption multiplied by field enhancement (red dashed curve).Near-field enhancement in the Al2O3 layer of the AlNC is shown in blue.

Swearer et al. PNAS | August 9, 2016 | vol. 113 | no. 32 | 8917

APP

LIED

PHYS

ICAL

SCIENCE

S

Dow

nloa

ded

by g

uest

on

June

2, 2

020

(EELS) performed in a monochromatic, probe-corrected scanningtransmission electron microscope (STEM), distinct modes corre-sponding to Al, Pd, and the coupled Al−Pd complex are observed.Nonnegative matrix factorization was used to separate the in-formation encrypted in the EELS spectrum image into spectralcomponents (Fig. 2A) and a set of coefficients corresponding to thenanometer-scale resolution spatial distribution of the plasmon ex-citation probability, herein referred to as plasmon maps (Fig. 2 B–E) (29, 30). The two Al modes show a maximum electron lossprobability at 6 eV and 7 eV (Fig. 2A), both with a secondaryshoulder at ∼2.5 eV corresponding to the dipolar resonances ofAlNC−Pd (confirmed with UV-visible spectroscopy, SI Appendix,Fig. S4). In the plasmon maps shown in Fig. 2 B–E, an outline of thenanoparticle structure [also shown as the high-angle annular dark-field STEM (HAADF-STEM) in Fig. 2A, Inset] has been overlaid toassist in visualization relative to the specific components of thecomplex. One can see from the Al LSPR maps that prominentexcitation from Al is found at the Al−Al2O3 interface, with highestintensity in regions of the Al nanoparticle surface without Pd dec-oration (Fig. 2 B and C). This can be attributed to local damping ofthe Al LSPR through interactions and coupling with Pd interbandtransitions (31). The plasmon excitation probability is dampedwhere smaller nanoparticle aggregation is found, but is nearlycompletely eliminated in and around larger (>10 nm) Pd islands.Of the two modes attributed to Pd, only one exhibits coupling to

the Al core. This mode corresponds to plasmon-assisted en-hancements in Pd directly on the surface of the AlNC, and will bereferred to as the “antenna mode” (Fig. 2D). The second modecan be attributed to Pd interband transitions (Fig. 2E) (18). Theantenna mode reveals intensity across a broad range of energies(1.5–6 eV) that overlay the dipolar resonance contribution fromAl. The two modes contributed from Pd show how the proximityof a lossy catalytic metal to a plasmonic metal core can influencethe electronic properties of the decorating catalytic metal. Theantenna mode (red) shows clear enhancements on the surfaceof the Pd islands in areas with high coverage and large islands(Fig. 2D). Chemically synthesized AlNCs can exhibit small pro-trusions of excess Al2O3 extending from their surfaces. In thisspecific antenna−reactor nanoparticle, a larger protrusion ofAl2O3 decorated with Pd, indicated with a white arrow in theHAADF-STEM in Fig. 2A, Inset, is observed at the base of thenanoparticle. For the antenna mode of the composite nano-particle, this region shows no enhancement (Fig. 2D). However,when the Pd interband transitions are excited (Fig. 2E), a strongexcitation is observed across the entire nanoparticle, including thePd found on the Al2O3 protrusion (a secondary example and ex-perimental description of EELS plasmon mapping is provided in SIAppendix, Figs. S5 and S6).The photocatalytic properties of the AlNC−Pd antenna−reactor

complex differ dramatically from that of either individual compo-nents of the complex. We studied nonthermal, photocatalytic hy-drogen desorption, which is the rate-limiting process in the H2adsorption−desorption equilibrium on the surface of Pd. We cal-culated, from density functional theory (DFT), a ground-state energybarrier of 1.18 eV/H2 to desorb hydrogen (SI Appendix, Fig. S7 andTable S2) because of the very exothermic and barrier-free dissocia-tive adsorption of H2 on Pd(111), consistent with recent literature(32, 33). Thus, we believe that the more consequential process forPd—in contrast to earlier reports of plasmon-induced dissociation ofH2 on Au (7, 8) and Al (15)—is the interaction of the generated hotcarriers with adsorbed H and D atoms on Pd, inducing desorption ofHD. The plasmon-induced dissociation, recombination, and de-sorption processes are all charge-neutral. As the chemical reaction iscompleted, the hot electrons recombine with the holes.For the AlNC−Pd complex, the wavelength dependence of HD

production (Fig. 3A, red) closely follows the calculated absorptioncross section (black) supporting a hot-carrier mechanism (34). Whenqualitatively compared with pristine AlNCs, the wavelength de-pendence of HD production is dramatically different (green), withthe maximum HD production occurring at a photoexcitation wave-length of 800 nm, corresponding to the interband transition ofAl (15). Quantitative consumption of H2 at the dipolar LSPR ofAlNC:Pd are reported in SI Appendix, Fig. S8. Compared with ourprevious work, AlNC−Pd antenna−reactor complexes show an or-der of magnitude greater reactivity than Au/SiO2 (7) and nearly twoorders of magnitude greater reactivity than pristine AlNCs (15). Theexcitation laser power dependence of this reaction was measured at492 nm and 800 nm, corresponding to the dipolar plasmon reso-nance and Al interband transition, respectively, and showed asupralinear response at both wavelengths (Fig. 3B). Such supralinearresponses with increasing optical power density have been suggestedas a hallmark of hot-carrier-driven chemistry on nanoparticle sur-faces (3).Temperature dependence measurements between 300 K and

400 K without external illumination show an increase in HDgeneration with increasing temperature; however, the calculatedwavelength-dependent local maximum temperature increaseexpected on the nanoparticle surface is only between 2 K and 16 Kfor Al and Pd surfaces, respectively, within the experimental rangeof excitation laser power densities (SI Appendix, Fig. S9). Suchsmall local temperature increases under illumination suggest that,although photothermal heating of the Pd lattice may contributeslightly to H2 desorption, the primary cause can be attributed to

Fig. 2. EELS plasmon maps AlNC−Pd heterometallic antenna−reactor cata-lysts. (A) HAADF-STEM corresponding to the particle of interest is shown inInset. Electron energy loss spectral components corresponding to modesshown in B−E as obtained from the region indicated by the red square in Inset.A Pd-decorated Al2O3 protrusion is highlighted by the white arrow. (Scale bar,50 nm.) (B) Al LSPR mode with energy loss centered at 6 eV. (C) Al LSPR modewith energy loss centered at 7 eV. (D) Al–Pd antenna mode with broad energyloss ranging from 1.5 eV to 6 eV. (E) Pd interband transition. Spectra for modesin B–E are shown in A. Normalized EELS loss probability of spatial plasmonmaps correspond to color legend on left.

8918 | www.pnas.org/cgi/doi/10.1073/pnas.1609769113 Swearer et al.

Dow

nloa

ded

by g

uest

on

June

2, 2

020

the excitation of photoexcited hot carriers in the antenna−reactorcomplex. To estimate the importance of hot electrons versus holes forthe desorption process, we calculated densities of states and analyzedthe contributions of Pd−H bonding and antibonding interactions,from DFT (SI Appendix, Figs. S10–S13). The bonding Pd−H orbitalslie very low in energy (∼6 eV below the Fermi level), so a hot-hole-initiated process seems less likely from these initial results. In contrast,the Pd−H antibonding interactions manifest just above the Fermilevel and thus can be expected to be easily populated by hot electrons.From these DFT results, a hot-electron-initiated destabilization of thePd−H bonds leading to desorption seems more likely. However, wecaution that the band structure obtained from DFT does not neces-sarily reflect (or quantitatively capture) the correct physics of excitedcarriers on the surface. Therefore, embedded correlated wave func-tion calculations are underway that remedy the too-facile chargetransfer of DFT (35), to identify the exact mechanism of hot-carrier-mediated hydrogen desorption from Pd surfaces.Additional contributing reaction mechanisms include H2 mi-

gration through the oxide to the AlNC surface and hot-electron

transfer across the oxide into Pd. Migration of H2 followed bydissociation by plasmon- and interband-transition–induced hotcarriers was found to be the primary mechanism in pristine AlNCs(15) and certainly also contributes in the antenna−reactor com-plex. However, it is a small effect because, as shown in SI Ap-pendix, Fig. S8, the overall rate of HD production on AlNC−Pdantenna−reactors is nearly two orders of magnitude higher than inpristine AlNCs. By comparing the wavelength dependence of HDproduction on pristine AlNCs and AlNC−Pd, specifically thedrastic increase in the HD yield at the Al dipolar plasmon, we canconclude that electron transfer effects also play only a minor rolefor the overall reactivity. Thus, we conclude that the plasmon-assisted absorption enhancements in Pd at 500 nm (Fig. 1F) arethe major effect responsible for the increase in reactivity.The photocatalytic properties of the AlNC−Pd antenna–reactor

complex are translatable to other chemical reactions, such as hy-drogenation. One important and industrially relevant reaction isthe selective reduction of acetylene (36). Ethylene is a commoditychemical precursor used in the production of polyethylene-basedmaterials with widespread commercial use; however, under tradi-tional thermal conditions, ethane is also produced in a side reactionduring hydrogenation of acetylene. With AlNC−Pd antenna−reactorcomplexes, we have found a drastic increase in the selective re-duction of acetylene to ethylene under white-light illuminationcompared to traditional thermal reduction (Fig. 4). This selectivityalso shows a large increase with increased laser power density (red);representative gas chromatograms for both thermal and photo-hydrogenations are shown in SI Appendix, Fig. S14. An increase inethylene:ethane product ratio from ∼7 to ∼37 is observed for thephotohydrogenation case (red). In contrast, traditional thermalheating of the AlNC−Pd complexes showed that ethylene:ethaneselectivity leveled off at a maximum of ∼10:1 before showing a dropto ∼6:1 at 360 K (black).This selectivity enhancement seen in photohydrogenation, yet

not seen in traditional thermal hydrogenation, is likely due to theavailability of dissociated H2. In both photohydrogenation andthermal hydrogenation cases, acetylene adsorbs on the surface andundergoes the first and second hydrogenations to produce ethylene(37–39). At this point, two forward reaction pathways are possible:ethylene desorption or subsequent hydrogenation of ethylene to

25 50 75

Laser Power Density (W/cm2)10.0 12.5 15.02.5 5.0 7.5

Wavelength (nm)400 500 600 700 800 900 1000

HD

Pro

du

ctio

n (

a.u

.)

0.0

0.2

0.4

0.6

0.8

1.0

Al Interband Transition

Dipolar Surface Plasmon

Ab

sorp

tio

n C

ross

Sec

tio

n (

a.u

)

Laser Power (mW)100 125 150

HD

Rat

e (c

/s)

×106

0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

492 nm800 nm

0

AlNC-Pd

Pristine AlNC

A

B

Fig. 3. AlNC−Pd photocatalytic reactivity for the hydrogen−deuterium ex-change reaction. (A) Wavelength dependence of H2 desorption on AlNC–Pd(red) and pristine AlNC (green) and the calculated absorption cross-section ofAlNC−Pd (black). The sample was irradiated with a power density of 5 W/cm2

at all wavelengths. (B) Power dependence measurements of HD productionat the dipolar plasmon resonance (492 nm, 2.52 eV) and the Al interbandtransition (800 nm, 1.55 eV).

Fig. 4. AlNC−Pd photocatalytic reactivity toward selective acetylene hydro-genation. Selectivity of ethylene:ethane production by acetylene hydrogena-tion under illuminated (red) and thermal (black) conditions.

Swearer et al. PNAS | August 9, 2016 | vol. 113 | no. 32 | 8919

APP

LIED

PHYS

ICAL

SCIENCE

S

Dow

nloa

ded

by g

uest

on

June

2, 2

020

produce ethane. Both desorption and hydrogenation of ethylenefrom Pd(111) have similar activation barriers within the margin oferror of previous DFT calculations (37). Therefore, the availabilityof dissociated H2 dictates the branching ratio between these tworeaction pathways. In photocatalytic hydrogenations, plasmon-induced hot carriers lead to rapid desorption of H2, biasing theequilibrium toward desorption and thus limiting the availabilityof hydrogen on the surface for additional hydrogenation of eth-ylene. The hypothesis of hot-carrier-induced H-starved surfacesleading to increased selectivity is also backed up by reduced yieldsof ethylene in the photocatalytic hydrogenation case (SI Appendix,Fig. S14). With illumination, there is less surface-activated H,which also reduces the likelihood of the first and second hydro-genations of acetylene needed to produce ethylene. In thermalhydrogenations, ethylene yields are higher (T > 360 K) at theexpense of reduced selectivity, most likely due to minimal changesin dissociated H2 surface coverage, and enough kinetic energy inthe system to overcome the activation energies and favor sub-sequent hydrogenations of ethylene. The selectivity increase ob-served for the photohydrogenation of acetylene could open doorsfor developing more selective hot-carrier-driven chemistry.

ConclusionsThe direct coupling of plasmonic nanoantennas with catalyticnanoparticles into a single heterometallic complex allows for ab-sorption enhancements in poorly light-absorbing catalytic metals.With antenna−reactor complexes, hot-carrier production and

photothermal heating can be dramatically increased near cata-lytically active surfaces. This concept is a highly modular one; forexample, tuning the composition or size of the plasmonic an-tenna allows for light-induced photocatalysis at specific wave-lengths of the electromagnetic spectrum, enabling optimizationof such complexes for specific chemical reactions and reactionpathways. Likewise, by changing the “reactor” to different metals,alloys, semiconductors, or insulators, the surface chemistry andphotocatalytic activity can be highly tuned. The plasmon-inducedhydrogen desorption reported here shows that the antenna−reactor geometry increases hot-carrier production, thereby allow-ing for new, light-driven reaction pathways on the catalytic metal.Developing this antenna−reactor concept to favor specific hot-carrier-driven photocatalytic processes where control over reactionspecificities is highly desirable opens a new door for the develop-ment of precise, ultimately predictive control of catalytic chemistryusing light.

ACKNOWLEDGMENTS. This work was supported by the Air Force Officeof Science and Research under Grant FA9550-15-1-0022. N.J.H. and P.N.acknowledge support from the Robert A. Welch Foundation under GrantsC-1220 and C-1222. D.F.S. acknowledges support from the National ScienceFoundation through a Graduate Research Fellowship (0940902). C.M.K.acknowledges support by a fellowship within the Postdoc Program of theGerman Academic Exchange Service. The High Performance ComputingModernization Program of the U.S. Department of Defense and PrincetonUniversity’s Terascale Infrastructure for Groundbreaking Research in Engi-neering and Science (TIGRESS) High Performance Computing provided thecomputational resources.

1. Astruc D, Lu F, Aranzaes JR (2005) Nanoparticles as recyclable catalysts: The frontierbetween homogeneous and heterogeneous catalysis. Angew Chem Int Ed Engl 44(48):7852–7872.

2. Ross JRH (2012) Heterogeneous catalysis. Heterogeneous Catalysis (Elsevier, NewYork), pp 1–15.

3. Linic S, Aslam U, Boerigter C, Morabito M (2015) Photochemical transformations onplasmonic metal nanoparticles. Nat Mater 14(6):567–576.

4. Long R, Li Y, Song L, Xiong Y (2015) Coupling solar energy into reactions: Materialsdesign for surface plasmon-mediated catalysis. Small 11(32):3873–3889.

5. Watanabe K, Menzel D, Nilius N, Freund HJ (2006) Photochemistry on metal nano-particles. Chem Rev 106(10):4301–4320.

6. Baffou G, Quidant R (2014) Nanoplasmonics for chemistry. Chem Soc Rev 43(11):3898–3907.

7. Mukherjee S, et al. (2013) Hot electrons do the impossible: Plasmon-induced dissoci-ation of H2 on Au. Nano Lett 13(1):240–247.

8. Mukherjee S, et al. (2014) Hot-electron-induced dissociation of H2 on gold nano-particles supported on SiO2. J Am Chem Soc 136(1):64–67.

9. Wee TE, Schmidt LC, Scaiano JC (2012) Photooxidation of 9-anthraldehyde catalyzedby gold nanoparticles: Solution and single nanoparticle studies using fluorescencelifetime imaging. J Phys Chem C 116(45):24373–24379.

10. Fasciani C, Bueno Alejo CJ, Grenier M, Netto-Ferreira JC, Scaiano JC (2011) High-temperature organic reactions at room temperature using plasmon excitation: De-composition of dicumyl peroxide. Org Lett 13(2):204–207.

11. Xie W, Schlücker S (2015) Hot electron-induced reduction of small molecules onphotorecycling metal surfaces. Nat Commun 6:7570.

12. Christopher P, Xin H, Linic S (2011) Visible-light-enhanced catalytic oxidation reactionson plasmonic silver nanostructures. Nat Chem 3(6):467–472.

13. Marimuthu A, Zhang J, Linic S (2013) Tuning selectivity in propylene epoxidation byplasmon mediated photo-switching of Cu oxidation state. Science 339(6127):1590–1593.

14. Decan MR, Impellizzeri S, Marin ML, Scaiano JC (2014) Copper nanoparticle hetero-geneous catalytic ‘click’ cycloaddition confirmed by single-molecule spectroscopy. NatCommun 5:4612.

15. Zhou L, et al. (2016) Aluminum nanocrystals as a plasmonic photocatalyst for hy-drogen dissociation. Nano Lett 16(2):1478–1484.

16. Weaver J (1975) Optical properties of Rh, Pd, Ir, and Pt. Phys Rev B 11(4):1416–1425.17. Pakizeh T, Langhammer C, Zori�c I, Apell P, Käll M (2009) Intrinsic Fano interference of

localized plasmons in Pd nanoparticles. Nano Lett 9(2):882–886.18. Weaver J (1975) Low-energy interband absorption in Pd. Phys Rev B 12(2):8–9.19. Wadell C, Antosiewicz TJ, Langhammer C (2012) Optical absorption engineering in

stacked plasmonic Au-SiO₂-Pd nanoantennas. Nano Lett 12(9):4784–4790.20. Antosiewicz TJ, Apell SP, Wadell C, Langhammer C (2012) Absorption enhancement in

lossy transition metal elements of plasmonic nanosandwiches. J Phys Chem C 116(38):20522–20529.

21. Antosiewicz TJ, Wadell C, Langhammer C (2015) Plasmon-assisted indirect light ab-sorption engineering in small transition metal catalyst nanoparticles. Adv Opt Mater3(11):1591–1599.

22. Antosiewicz TJ, Apell SP (2014) Optical enhancement of plasmonic activity of catalyticmetal nanoparticles. RSC Adv 5(9):6378–6384.

23. Liu N, Tang ML, Hentschel M, Giessen H, Alivisatos AP (2011) Nanoantenna-enhancedgas sensing in a single tailored nanofocus. Nat Mater 10(8):631–636.

24. Syrenova S, et al. (2015) Hydride formation thermodynamics and hysteresis in indi-vidual Pd nanocrystals with different size and shape. Nat Mater 14(12):1236–1244.

25. McClain MJ, et al. (2015) Aluminum nanocrystals. Nano Lett 15(4):2751–2755.26. Knight MW, et al. (2014) Aluminum for plasmonics. ACS Nano 8(1):834–840.27. Zhu QL, Tsumori N, Xu Q (2015) Immobilizing extremely catalytically active palladium

nanoparticles to carbon nanospheres: A weakly-capping growth approach. J AmChem Soc 137(36):11743–11748.

28. Zheng BY, et al. (2015) Distinguishing between plasmon-induced and photoexcitedcarriers in a device geometry. Nat Commun 6:7797.

29. Ringe E, et al. (2015) Resonances of nanoparticles with poor plasmonic metal tips. SciRep 5:17431.

30. Leary RK, et al. (April 18, 2016) Structural and optical properties of discrete dendriticPt nanoparticles on colloidal Au nanoprisms. J Phys Chem, 10.1021/acs.jpcc.6b02103.

31. Pirzadeh Z, Pakizeh T, Miljkovic V, Langhammer C, Dmitriev A (2014) Plasmon–interband coupling in nickel nanoantennas. ACS Photonics 1(3):158–162.

32. Dong W, Hafner J (1997) H2 dissociative adsorption on Pd(111). Phys Rev B 56(23):15396–15403.

33. Mitsui T, Rose MK, Fomin E, Ogletree DF, Salmeron M (2003) Dissociative hydrogenadsorption on palladium requires aggregates of three or more vacancies. Nature422(6933):705–707.

34. Manjavacas A, Liu JG, Kulkarni V, Nordlander P (2014) Plasmon-induced hot carriers inmetallic nanoparticles. ACS Nano 8(8):7630–7638.

35. Libisch F, Huang C, Carter EA (2014) Embedded correlated wavefunction schemes:theory and applications. Acc Chem Res 47(9):2768–2775.

36. Tiedtke DB, et al. (2001) Chemicals influencing the activity of palladium-based catalystsfor the selective hydrogenation of acetylene to etylene in acetylene converters. Pro-ceedings of the 13th Ethylene Producers’ Conference (Am Inst Chem Eng, New York),pp 586–607.

37. Sheth PA, Neurock M, Smith CM (2003) A first-principles analysis of acetylene hy-drogenation over Pd(111). J Phys Chem B 107(9):2009–2017.

38. Sheth PA, Neurock M, Smith CM (2005) First-principles analysis of the effects of al-loying Pd with Ag for the catalytic hydrogenation of acetylene-ethylene mixtures.J Phys Chem B 109(25):12449–12466.

39. Mei D, Sheth PA, Neurock M, Smith CM (2006) First-principles-based kinetic MonteCarlo simulation of the selective hydrogenation of acetylene over Pd(111). J Catal242(1):1–15.

8920 | www.pnas.org/cgi/doi/10.1073/pnas.1609769113 Swearer et al.

Dow

nloa

ded

by g

uest

on

June

2, 2

020