Determining Plasmonic Hot Electrons and Photothermal Effects during H2 Evolution with TiN-Pt Nanohybrids

Article abstract of DOI:10.1021/acscatal.0c00343

Hydrogen storage in chemical compounds is a promising strategy to enable lightweight, high-density, and safe hydrogen technologies. However, the hydrogen release rate from these chemicals is limited by the intrinsic catalytic activity of metal catalysts, which can be enhanced by light irradiation. Here, nanohybrids including a core of plasmonic TiN and multiple Pt nanocrystal catalytic centers are assembled and show, under resonant conditions at 700 nm, hot electron-driven hydrogen evolution from ammonia borane at an apparent quantum yield of 120%. It is also demonstrated that solar irradiation enhances the activity of TiN-Pt nanohybrids by one order of magnitude through two synergistic mechanisms: hot electrons and collective-heating contributions. Using the microscopic calculation of the photo-induced temperature around a single nanocrystal, it is revealed that the collective plasmonic heating regime dominates the macroscopic temperature distribution in the system. The presented data show that plasmonic hot electrons and photothermal heating can be used in synergy to trigger hydrogen release from ammonia borane on demand, providing a general strategy for greatly enhancing the activity of metal catalysts in the dark.

Full text of DOI:10.1021/acscatal.0c00343

Subscriber access provided by JAMES COOK UNIVERSITY LIBRARY
Article
Determining Plasmonic Hot Electron and Photothermal
2
Effects during H Evolution with TiN-Pt Nanohybrids
Sourav Rej, Luca Mascaretti, Eva Yazmin Santiago, Ondrej Tomanec, Stepan Kment,
Z. Wang, Radek Zbo#il, Paolo Fornasiero, Alexander O. Govorov, and Alberto Naldoni
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.0c00343 • Publication Date (Web): 31 Mar 2020
Downloaded from pubs.acs.org on April 6, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 35
ACS Catalysis
1
2
3
4
5
6
7
8
9
Determining Plasmonic Hot Electron and
Photothermal Effects during H Evolution with
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
TiN-Pt Nanohybrids
†
†
§
†
†
Sourav Rej, Luca Mascaretti, Eva Yazmin Santiago, Ondřej Tomanec, Štěpán Kment,
‡
†
ǁ
§,‡
Zhiming Wang, Radek Zbořil, Paolo Fornasiero, Alexander O. Govorov* & Alberto
Naldoni*†
†
Regional Centre of Advanced Technologies and Materials, Faculty of Science, Palacký
University, Šlechtitelů 27, 78371 Olomouc, Czech Republic
Department of Physics and Astronomy, Nanoscale and Quantum Phenomena Institute, Ohio
§
University, Athens, Ohio 45701, United States
‡ Institute of Fundamental and Frontier Sciences, University of Electronic Science and
Technology of China, Chengdu 610054, China
ǁ Department of Chemical and Pharmaceutical Sciences, INSTM and ICCOM-CNR,
University of Trieste, via L. Giorgieri 1, Trieste 34127, Italy
Hydrogen storage in chemical compounds is a promising strategy to enable lightweight, high
density, and safe hydrogen technologies. However, the hydrogen release rate from these
chemicals is limited by the intrinsic catalytic activity of metal catalysts, which can be
enhanced by using light irradiation. Here nanohybrids including a core of plasmonic TiN and
multiple Pt nanocrystal catalytic centers are assembled and show, under resonant conditions at
700 nm, hot electrons driven hydrogen evolution from ammonia borane at an apparent
1
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
quantum yield of 120%. It is also demonstrated that solar irradiation enhances the activity of
TiN-Pt nanohybrids by one order of magnitude through two synergistic mechanisms: the hot
electrons and collective-heating contributions. Using the microscopic calculation of the
photoinduced temperature around a single nanocrystal, it is revealed that the collective
plasmonic heating regime dominates the macroscopic temperature distribution in the system.
The presented data show that plasmonic hot electrons and photothermal heating can be used
in synergy to trigger hydrogen release from ammonia borane on demand, providing a general
strategy for greatly enhancing the activity in the dark of metal catalysts.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
KEYWORDS: plasmonics, titanium nitride, ammonia-borane, hydrogen evolution,
photocatalysis
INTRODUCTION
The development of lightweight storage systems for on-board storage and controlled release
of hydrogen (H ) is a limiting factor for the widespread utilization of energy technologies
2
based on fuel cells.1–3 Transport and storage of H as a gas or liquid in highly pressurized
2
tanks may not provide the required fuel density for long transportation, while raising hazard
concerns. Thus, there is a continued interest in finding alternative means to store H in high-
2
density form from which it can be released on demand at viable rates and temperatures (below
70 °C).1
Some examples of recently investigated materials for H storage include metal hydrides and
2
2
,4
porous sorbent materials. In addition to these, a promising class of compounds is chemical
complexes wherein the H is stored in X–H bonds (X = C, B, N, or O). In this case, H can be
2
2
4
released thermally or through the aid of a catalyst. Ammonia borane (NH BH ) is a
3
3
2
ACS Paragon Plus Environment

Page 3 of 35
ACS Catalysis
1
2
3
4
5
6
7
8
9
particularly appealing candidate material owing to its high gravimetric H capacity (19.6
2
−
1
wt. %), low molecular weight (30.7 g mol ), and high stability in solution, while being
3
nonflammable and nonexplosive under standard conditions. The dehydrogenation of
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
NH BH is a downhill reaction that potentially results in three equivalents of H :
3
3
2
+
―
푁퐻 퐵퐻 +4퐻 푂→푁퐻 퐵(푂퐻) +3퐻 ↑
(1)
3
3
2
4
4
2
A typical approach to H evolution from NH BH is the use of catalytic metals, which allows
2
3
3
appreciable rates in dark conditions. Supported Pt,6–10 Rh, and Ru, based nanocatalysts
5
11
5
usually showed higher H production rates. However, the use of these catalytic metals poses
2
relevant issues in terms of costs and scalability of H storage technologies. Recently, the use
2
of plasmonic materials (noble metals and metal oxides) has emerged as a general strategy to
enhance the activity of catalytic metals for NH BH dehydrogenation by using light
3
3
irradiation.12–16
Upon light illumination, plasmonic photocatalysts support the collective oscillations of free
carriers, i.e., localized surface plasmon resonance (LSPR), which concentrates light at the
nanoscale and produce very intense electric fields.17–19 Surface plasmons may then undergo
nonradiative decay generating highly energetic electron-hole pairs (hot carriers), which have
shown the capability to drive chemical reactions20–23 with product selectivity,24–26 as well as
2
7–31
the ability to catalyze multielectron processes like CO reduction.
2
The temperature of hot carriers eventually cool down via electron–electron and
electron−phonon scattering within 100 ps, causing a rise in the lattice temperature that results
in the heating of the plasmonic material.32–34 This thermal energy can enhance the reaction
rate with an Arrhenius-like dependence and it has been the focus of recent efforts devoted to
distinguishing the photothermal contribution from the plasmonic hot electron effects.35–37
Here, we show that refractory, acid resistant, fuel cells compatible plasmonic nanohybrids
including plasmonic TiN38–42 nanocubes decorated with Pt nanocrystals efficiently drive H2
evolution from ammonia borane dehydrogenation with AQY exceeding 100% under
3
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
–2
monochromatic light at 700 nm (incident intensity: 2mW cm ). Under solar irradiation, the
activity of TiN-Pt nanohybrids is enhanced by one order of magnitude (from 32 to 346 molH2
molPt–1 min ) through the synergistic effect of plasmonic hot electrons and heating. We
–1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
demonstrate that at high solar intensity H evolution is both an electron- and thermally driven
2
process, in which hot electrons are mainly generated in TiN and then transferred to the Pt
surface, thus decreasing the energetic requirement to access the transition state of the rate
determining step, i.e., water cleavage.
EXPERIMENTAL SECTION
Synthesis of TiN–Pt nanohybrids. Before depositing Pt nanocrystals, commercial TiN
nanocubes (PlasmaChem GmbH, ~50 nm average size) were treated with concentrated HCl to
clean the TiN surface according to the following procedure. 20 mL of concentrated HCl (35%,
Penta) and 400 mg of TiN nanocubes were sonicated for 10 min at room temperature under
sealed conditions. After that, the reaction mixture was kept at 65 °C on a pre-heated water
bath for 1 h under a continuous nitrogen flow. Upon completion of the reaction, the TiN
nanocubes were then washed one time with 40 mL water/ethanol (1:1 volume ratio) mixture
and further twice with 40 mL anhydrous ethanol. TiN nanocubes were then concentrated in
5
mL anhydrous ethanol and dried in an oven at 80 °C for overnight.
The dried TiN nanocubes were crushed into a fine powder using a spatula prior to Pt
nanocrystal deposition. First, 92 mg of acid-treated TiN nanocubes were suspended in 50 mL
deionized water (DI-H O) and sonicated for 20 min. Appropriate amount of K PtCl (Sigma–
2
2
4
Aldrich) were dissolved in DI-H O and added dropwise to the aforementioned solution of TiN
2
nanocubes under vigorous stirring; afterwards, the resulting solution was stirred for additional
2
h. Subsequently, 1 mL of 3 M aqueous solution NaOH (Alfa Aesar) was added dropwise to
adjust the solution pH at ~13, followed by stirring for 1 h. 1 mL of formaldehyde (37%,
Penta) was then added dropwise to the solution with vigorous stirring and the reaction mixture
4
ACS Paragon Plus Environment

Page 5 of 35
ACS Catalysis
1
2
3
4
5
6
7
8
9
was placed in a pre-heated water bath at 85 °C for 1.5 h under a continuous nitrogen flow.
The TiN–Pt nanohybrids were washed using the same procedure as above. After washing, all
nanohybrids were then concentrated in 50 mL DI-H O. 1.8 mL of an aqueous HCl solution
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(10% vol.) were added dropwise to the nanocrystal solution and continuously stirred at 700
rpm for overnight. TiN–Pt nanohybrids were further washed and collected using the same
washing procedure described in the previous section. The as-synthesized TiN–Pt nanohybrids
were concentrated in 5 mL of anhydrous ethanol and dried in an oven at 80 °C for overnight.
The so-obtained fine powder was reduced in a furnace under a H /N flow (10%:90% vol.,
2
2
2
00 sccm) for 2 h at 200 °C. The reduced TiN–Pt nanohybrids were kept in closed glass vials
under argon atmosphere before and after their use. Inductively coupled plasma mass
spectrometry (ICP-MS) analysis confirmed that the Pt loading was 8 wt%. In order to perform
control experiments, a Pt/Al O catalyst was prepared. Pt nanocrystals were synthesized by
2
3
reducing K PtCl in the presence of NaOH (pH>12) at 145 °C for 3 h in ethylene glycol. Then
2
4
this colloidal solution was vigorously mixed with Al O followed by washing and overnight
2
3
drying at 80 °C. The so-obtained Pt/Al O catalyst was reduced at 100 °C for 1 h under H
2
2
3
flow before usage.
Characterization. Powder X-ray diffraction (XRD) patterns of material were determined by
X’Pert PRO MPD diffractometer (PANalytical) using iron-filtered Co-Kα radiation (40 kV,
3
0 mA, λ = 0.1789 nm). X-ray photoelectron spectroscopy (XPS) analysis were performed on
a PHI 5000 VersaProbe II XPS system (Physical Electronics) with monochromatized Al-Kα
source (15 kV, 50 W) and photon energy of 1486.7 eV. High-resolution spectra were scaled
using the adventitious carbon peak as a reference. The low resolution imaging of catalysts
morphology was obtained with a transmission electron microscope (TEM) JEOL equipped
with a LaB emission gun and operating at 160 kV. High-resolution micrographs and STEM-
6
elemental mapping were acquired using a FEI Titan HRTEM microscope equipped with X-
FEG electron gun operating at 80 kV. Absorbance spectra were measured by a UV-Vis
5
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
spectrophotometer SPECORDS 600 (Analytik Jena). Absorption and extinction spectra were
evaluated from transmittance and scattering measurements by a UV-vis-NIR spectrometer
(Analytik Jena Specord 250 Plus) equipped with an integrating sphere (see Section S2).
Electron energy loss imaging of free plasmons in TiN-Pt nanohybrids was recorded on a FEI-
Titan (S)TEM operated at 300kV operating in monochromated mode (with energy resolution
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0
.2 eV) and focusing on the low energy-loss regime to map the plasmonic modes of the
nanohybrids.
Opto-electronic and photothermal simulations. The simulations were carried out using
classical thermo-electrodynamics with COMSOL Multiphysics, with the objective to study
the opto-electronic and photothermal properties of our plasmonic materials. It consisted in
modeling a 40 nm TiN nanocube with a corner radius of 8 nm in water. Additionally, sixteen
Pt semispheres, with a diameter equal to 3.3 nm, were evenly distributed on each nanocube
face. These parameters were chosen to best replicate the nanoparticles utilized in the
experiment. The thermal coefficients corresponding to each material are included in the Table
S7.
The calculations assumed an average of incident wave planes over all orientations and a large
interparticle distance to avoid coupling. The system was set up to be at an ambient
temperature (Tamb) of 293.15 K before calculating its temperature increment. Using these
conditions, the calculations included the absorption, scattering and extinction cross sections,
the field enhancement factor and the temperature increment of the TiN-Pt nanohybrids.
Finally, a plain TiN nanocube was modeled to make a comparison and to observe the
contribution of the Pt nanocrystals.
To produce the photothermal calculations, the model relied on the heat transfer equation,
expressed as







T(r,t)





(r)c(r)
  k(r)T(r,t)  q(r,t)
 

t
6
ACS Paragon Plus Environment

Page 7 of 35
ACS Catalysis
1
2
3
4
5
6
7
8
9


r
where T(r,t) is the temperature increase as a function of the coordinate
and time . The
t



coefficients (r)
,
c(r)
,
k(r) are the mass density, specific heat and thermal conductivity,

1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
respectively. The function q(r,t) is the local heat source generated by the light dissipation of
the nanoparticle and it is given by




2
0
2
q(r,t)  j(r,t) E(r,t)  Im( )

NP
t
where 0 is the dielectric permittivity in vacuum, Im(NP ) is the imaginary part of
permittivity of the nanoparticle, and  is the electric field inside the nanoparticle. We are
then able to obtain a temperature difference, defined as

T  T Tamb
where Tamb is the ambient temperature.
Hydrogen evolution experiments from ammonia borane. 22 mL of DI-H O were added to
2
2
.4 mg of reduced TiN–Pt nanohybrids in a 25 mL round bottom flask. The catalyst was
dispersed by sonicating the solution for 5 min. The catalyst solution was then irradiated under
solar-simulated light (Sciencetech Light Line A4-C250 equipped with an AM 1.5G filter,
–
2
class CAA) with different intensities (from 1 to 10 Suns, i.e., from 100 mW cm to 1000 mW
–2
cm ) for 5 min under stirring to homogenize the temperature in the solution. The different
light intensities were calibrated with a thermopile detector (Standa 11UP19-H) before each
2
experiment. The light beam area impinging on the photoreactor was 4 cm . The solution
temperature was measured with a thermometer. Then, an aqueous ammonia borane (NH BH ,
3
3
Sigma–Aldrich) solution (0.28 mmol in 1 mL of water) was injected into the catalyst solution
using a micropipette. The volume of evolved hydrogen gas was measured by an inverted
water-filled burette connected to the reaction flask through inserting a rubber tube. The
volume of the evolved hydrogen gas was monitored by recording the displacement of water in
7
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
the gas burette. Completion of the reaction was confirmed when gas generation stopped. The
wavelength-dependent H evolution experiments were carried out by employing optical band-
2
–
2
pass filters (Thorlabs, FWHM = 10 nm) and using a power density of 2 mW cm at all
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
selected wavelengths.
Catalytic reduction of 4-nitrophenol. First, 8.5 mg of NH BH and 3.5 mL of 0.2 mM 4-
3
3
nitrophenol (NPh) aqueous solution were mixed in a 4 mL quartz cell equipped with a small
magnetic stirrer. Then, a 0.229 mL aqueous dispersion of TiN–Pt nanohybrids was added to
the solution under stirring condition. The molar ratio of Pt/4-NPh/NH BH was adjusted to
3
3
1
/61/24553. After adding the catalyst, the cuvette was sealed with the lid. The conversion of
4-NPh to 4-aminophenol was monitored by recording the UV-Vis spectra at different time
intervals in the range 250–550 nm. Based on the change in the intensity at λ = 400 nm as a
function of time, the rate constant of the catalytic hydrogenation of 4-NPh was determined.
The reaction was carried out under dark conditions and 5 Sun illumination. The amount of
catalysts adjusted in such a way that they did not influence the UV-Vis spectrum of 4-
nitrophenoxide anion during the time dependent study.
RESULTS AND DISCUSSION
Synthesis and characterization of TiN-Pt nanohybrids
We prepared TiN-Pt nanohybrids by developing a synthetic strategy that allows us to decorate
each TiN nanocube (average size 41±10 nm) with a multitude of Pt nanocrystals. In order to
enhance the binding of Pt nanocrystals onto the surface of the plasmonic core, commercial
TiN nanocubes were initially treated with concentrated HCl. Reacting potassium
tetrachloroplatinate in the presence of a mild reducing agent (formaldehyde) and NaOH
enabled the decoration of TiN nanocubes with well-separated Pt nanocrystals. The use of
NaOH, which neutralizes the HCl evolved during the reduction of Pt precursor, produced a
higher number of Pt nucleation centers on the surface of TiN nanocubes, which eventually led
8
ACS Paragon Plus Environment

Page 9 of 35
ACS Catalysis
1
2
3
4
5
6
7
8
9
to the uniform growth of small Pt nanocrystals. A final reduction step under H flow at 200 °C
2
for 2 h was applied to clean the Pt surface and boost the catalytic activity of plasmonic
nanohybrids before running photocatalytic experiments (See Experimental Procedure and
Section S1 in the SI for in-depth nanohybrids characterization).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1 shows the morphology of a representative reduced TiN-Pt nanohybrid. The high-
resolution transmission electron micrographs (HRTEM, Figure 1a and b) evidence that Pt
nanocrystals have an average size of 3.3±0.3 nm and form a sharp interface with the external
surface of the TiN nanocube.
The corresponding dark-field scanning TEM image and energy-dispersive X-ray spectroscopy
(EDS) elemental mapping (Figure 1c–h) highlight that Pt nanocrystals grew homogenously
around the TiN core, which retained a stoichiometric atomic distribution of Ti and N even
after the Pt decoration and reduction treatment. Notably, oxygen elemental mapping (Figure
1
g) clearly shows the presence of a 1.5 nm thick layer rich in oxygen, including a TiON/TiO2
mixture as evidenced by X-ray photoemission spectroscopy analysis (Section S1 of SI). This
passivation layer is commonly formed in TiN nanostructures upon exposure to air and/or
chemicals during the Pt nanocrystals depostion, which may act as a protective shell toward
further oxidation.
9
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 1. a, A representative TEM image of a single TiN–Pt system where Pt nanocrystals
are uniformly distributed over a TiN nanocube. b, A magnified high resolution TEM image of
a showing lattice fringes of TiN and Pt. c–h, Dark field STEM image of a representative TiN–
Pt nanohybrid and EDS elemental maps of Pt (d), Ti (e), N (f), O (g), and an overlay of Ti and
Pt (h) showing the spatial distribution of Pt nanocrystals over TiN nanocubes.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Optical properties of TiN-Pt nanohybrids
The optical properties of pristine TiN and TiN-Pt nanohybrids were studied by measuring
extinction, absorption, and scattering spectra in an aqueous solution (see details in Section S2).
Figure 2a and 2d show that the decoration of TiN nanocubes with Pt nanocrystals induced a
significant shift of the extinction peak from 664 to 720 nm as the result of the dielectric
environment change at the surface of the TiN due to the Pt deposition. Notably, TiN-Pt
nanohybrids present a remarkable absorption fraction of about 0.9, while the scattering
contribution is minimal and it is similar for both pure TiN (0.07) and for TiN-Pt (0.10).
To gain further insights on the optical properties of TiN-Pt nanohybrids, we performed finite
element method (COMSOL) simulations (see Experimental Procedure). The calculated
extinction, absorption, and scattering for a pure TiN nanocube (40 nm) and TiN nanocube
decorated with sixteen Pt nanocrystals per facet are shown in Figure 1b and 1d. Both models
include a 1.5 nm thick TiO surface layer on top of TiN. The simulated optical properties
2
agree well with the experimental measurements confirming the large absorption fraction and
low scattering. The electromagnetic simulations match very well the localized surface
plasmon resonance (LSPR) peak of TiN nanocubes, with the simulated value (670 nm) being
almost identical to the experimental one (664 nm). For the case of TiN-Pt nanohybrids, the
simulated LSPR peak is instead at ~675 nm and differs from the experimental value by only
4
5 nm. For the latter, the small difference between simulated and experimental spectra could
1
0
ACS Paragon Plus Environment

Page 11 of 35
ACS Catalysis
1
2
3
4
5
6
7
8
9
be due to the partial aggregation of TiN-Pt in solution, with Pt NPs deposition that might
favor this process if compared to the case without Pt.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. a, Experimental extinction, absorption, and scattering of TiN nanocubes. b,
Calculated extinction, absorption, and scattering of TiN nanocubes including a 1.5 nm thick
2
2
TiO surface layer. c, Electric field enhancement (
|
퐸
|
/퐸 ) distribution at 670 nm in a TiN
2

0
nanocube. d, Experimental extinction, absorption, and scattering of TiN–Pt nanohybrids
including a 1.5 nm thick TiO surface layer. e, Calculated extinction, absorption, and
2
scattering of TiN-Pt nanohybrids. f, Electric field enhancement distributions at 675 nm in a
TiN-Pt nanohybrid. g, Dark field STEM image of TiN-Pt nanohybrids on Si N membrane
3
4
(left) and corresponding electron energy loss image (right) relative to the excitation of the
1
1
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
dipolar plasmonic mode. The yellow square on the dark-field micrograph shows the focus of
the electron beam.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2c and 2f compare the E-field enhancement at the surface of a TiN nanocube and a
TiN-Pt nanohybrid induced by the resonant excitation of the dipolar LSPR. As expected, the
electric fields in the TiN nanocube localize around the corners, while for the TiN-Pt
2
nanohybrid, the optical energy is concentrated at the catalytic Pt surface. Interestingly,
|
퐸
|
/

2
퐸 at the Pt/water interface is three-times higher (40) than that generated at the corners of the
0
TiN nanocubes (13).
To map the E-field distribution we excite the free plasmon resonances in TiN-Pt nanohybrids
deposited on a Si N coated TEM grid and use dark-field scanning TEM (STEM) and electron
3
4
energy loss spectroscopy (EELS). The EELS spectrum revealed the only presence of the
dipolar plasmon resonance peaking at 2.25–2.35 eV (Supplementary Section S2).
The difference in energy position between the LSPR observed in optical measurement (1.72
eV) and the free plasmons excited with fast electrons might be due to the increased radiation
damping of polaritons (LSPR) compared to the free plasmon, which lowers the resonance
43
energy of the former (i.e., by almost 1 eV for 100 nm Ag nanoparticles). Figure 2g shows
the dark-field STEM micrograph and corresponding EELS mapping at energy of 2.25–2.35
eV showing that plasmonic electric fields propagate all over the Pt nanocrystals surface.
An ideal plasmonic antenna (optically active unit)-reactor (catalytically active unit)
nanohybrid should maximize light absorption (e.g. generation of charge carriers) while
suppressing the radiative decay channel (scattering) of LSPR.44,45 For this reason we selected
a TiN nanocube size of ~40 nm, which showed absorption as the dominant mechanism of
light-matter interaction. However, in order to assess the location of hot electron generation
and the reaction mechanism, it is also crucial to determine whether the absorption is spatially
localized in the plasmonic core (TiN) or in the catalytic material (Pt). In the first case, hot
1
2
ACS Paragon Plus Environment

Page 13 of 35
ACS Catalysis
1
2
3
4
5
6
7
8
9
electrons may be generated in the plasmonic material and then transferred to the catalytic
metal for participating to the chemical reaction. In another case, instead, hot carriers may be
directly generated onto the catalytic metal surface being readily available for catalysis.
Chavez et al. showed that core-shell nanoparticles including a core made by a Ag nanocube
and a shell composed by a catalytic metal (i.e. Pt), present peculiar optical properties, where
the plasmonic near field assists and promotes the optical transitions from the plasmonic
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
6
material to the catalytic one. This is possible in light of the high ratio (i.e. max. 15-20 at
~700 nm) between the imaginary parts of the dielectric permittivity of catalytic and

traditional plasmonic materials (Au, Ag).45,46
In order to determine the spatial location of light absorption inside our TiN-(TiO )-Pt
2
nanohybrids, we retrieved the simulated optical aborption cross section (Abs) in each
constituent component (Figure 3). The theoretical curves show that TiN accounts for 96.51%,
Pt nanoparticles for 3.26%, and TiO for 0.03% of the total absorption (Figure 3a). The lower
2
fraction of the total absorption related to the catalytic metal (Pt) in our nanohybrids is
expected since the total volume of the Pt nanoparticles is small as compared to the TiN
volume and, in addition, Pt possesses relatively small values for the imaginary part of the
dielectric permittivity at visible wavelengths.
More information about the light absorption inside the Pt nanoparticles can be obtained by
computing the absorption, abs, Pt, TiN-Pt () , for one single Pt nanoparticle inside the TiN-Pt
nanohybrid and for an isolated Pt nanoparticle, abs, Pt, isolated Pt () (Figure 3b). The ratio of the
above absorptions is expressed as follows:
abs, Pt, TiN-Pt, per one particle ()
Enh() 
(2)
(3)
abs, Pt, isolated Pt ()
where the absorption per one Pt particle inside the TiN-Pt hybrid is given by
abs, Pt, TiN-Pt ()
abs, Pt, TiN-Pt, per one particle () 
96
1
3
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
In our model, we assume that one nanohybrid contains 96 Pt semi-spheres. The ratio (2) is, in
fact, a plasmonic enhancement factor for our hybrid nanomaterial. Figure 3b shows that
abs, Pt, TiN-Pt, per one particle for one single Pt nanoparticle inside the TiN-Pt nanohybrid is strongly
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
enhanced when compared to abs, Pt, isolated Pt () of an isolated Pt nanoparticle, with a maximum
Enh ~10 at 730 nm. However, the photocatalytic experiments are done with the solar-like
spectrum and the enhancement coefficient for the absorptions should be computed
considering the extended spectrum of excitation (see SI). For the solar light, we obtain:
Enhsolar ~ 3.3
.
Figure 3. a, Simulated absorption cross sections for the complete TiN-Pt nanohybrid (blue
line), and for each single component: TiN (black line), TiO (brown line), and Pt (red line). b,
2
Simulated absorption cross section for a single, isolated Pt nanoparticle (red line) and for a
single Pt nanoparticle inside the TiN-Pt nanohybrid (blue line). The model for the Pt
1
4
ACS Paragon Plus Environment

Page 15 of 35
ACS Catalysis
1
2
3
4
5
6
7
8
9
nanoparticle is a semi-sphere. Inset: absorption cross section enhancement for a Pt
nanoparticle inside the TiN-Pt nanohybrid.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Hydrogen evolution from NH BH dehydrogenation
3
3
We describe the photocatalytic performance of TiN-Pt nanohybrids for plasmon-enhanced H2
evolution using conventional figures of merit utilized in heterogeneous catalysis and
–
1
–1
photocatalysis such as turnover frequency (TOF in mol mol min ) and apparent
H2
Pt
quantum yield (AQY% in molecules/photons). Photocatalytic performance is always
expressed as a difference between photothermal (light) and thermal (dark) reaction rates. An
exhaustive description about calculation of catalytic figures of merit is reported in the
Supporting Information Section S3.
When we tested our nanohybrids in the plasmon-enhanced H evolution from NH BH
3
2
3
dehydrogenation, we did not observe any considerable H generation rate in the absence of the
2
catalyst nor by addition of TiN nanocrystals into the aqueous NH BH solution and under
3
3
light irradiation. Conversely, we detected H evolution only when using TiN-Pt nanohybrids
2
(both in dark and light) suggesting that the presence of Pt is mandatory to activate NH BH
3 3
dehydrogenation.
Furthermore, to evaluate the photocatalytic properties of Pt nanocrystals, we synthesized 1.8
nm Pt nanocrystals supported on optically inactive Al O (see Figure S14) and tested them in
2
3
the photocatalytic hydrogen evolution from ammonia borane dehydrogenation under 5 Suns
illumination and in the dark at photothermal conditions (at final temperature generated under
5
Suns). The reaction kinetics obtained under light and dark conditions (Figure S14d) showed
an identical reaction time for producing 3 moles equivalent of H . This finding confirmed that
2
Pt nanocrystals had no significant contribution as a photocatalyst during the light-assisted
ammonia borane dehydrogenation, as instead previously observed during CO photothermal
4
7
oxidation over Pt/Al O .
2
3
1
5
ACS Paragon Plus Environment

ACS Catalysis
Page 16 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
To demonstrate that plasmonic hot electrons enhance H evolution form NH BH
3
2
3
dehydrogenation, we performed wavelength-dependent catalytic runs. We selected light with
–
2
a power density of 2 mW cm at all wavelengths, such that the solution temperature did not
deviate from room temperature (RT) along the course of reaction, i.e. no effect of
photothermal heating.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
–
1
–1
TiN–Pt nanohybrids show a TOF value of 32.1 mol mol min in dark at RT (in line with
H2
Pt
the best monometallic Pt catalysts reported in the literature, see Table S6), while in light
conditions we observed increased H production rates at all investigated wavelengths (450–
2
9
00 nm). We calculated AQYs under monochromatic light by using the reaction rate
difference between light and dark conditions. Figure 4a shows the photocatalytic AQY at
different wavelengths along with the absorption spectrum of TiN-Pt nanohybrids. The
photocatalytic AQY action spectrum closely followed the absorption of TiN-Pt, thus
confirming that AQY is related to light absorption in the reaction mixture, which can induce
both hot electron and photothermal plasmonic catalysis.24,48,49 In this case, however, we avoid
photothermal heating using low light intensity, and therefore the reported AQYs are due to the
action of plasmonic hot electrons. Notably, we obtained very high AQYs in all range of
investigated wavelengths, with AQYs exceeding 100% for experiments carried out at
1
6
ACS Paragon Plus Environment

Page 17 of 35
ACS Catalysis
1
2
3
4
5
6
7
8
9
wavelength (550, 600, 700 nm) around the LSPR of TiN-Pt nanohybrids.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. a, Apparent quantum yields (AQYs) for H evolution from NH BH
3
2
3
dehydrogenation over TiN–Pt under monochromatic light (blue dots) and TiN–Pt absorption
spectrum (red line). b, Experimental temperatures generated in the photocatalytic reactor
under different solar intensities: final temperature (FT) with TiN–Pt nanohybrids and
intermediate temperatures (IT) with TiN–Pt, TiN and water. The inset defines IT as the
temperature generated after five minutes of irradiation, whereas FT as the temperature
registered at the end of reaction. c, Turnover frequency (TOF) of H evolution from NH BH
2
3
3
dehydrogenation over TiN–Pt nanohybrids in dark at room temperature (RT), in dark at
intermediate temperature (IT) TOFtherm , and under AM 1.5G solar irradiation (Photothermal,
–
2
TOFphoto ) at different light intensities (1 Sun = 100 mW cm ). d, Hot electron generation rate
at both IT and FT under solar irradiation. e, AQYs for H evolution from NH BH
2
3
3
dehydrogenation over TiN–Pt under solar irradiation.
1
7
ACS Paragon Plus Environment

ACS Catalysis
Page 18 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The highest AQY (120%) was observed when light exactly matched the LSPR maximum at
7
00 nm. Our values are among the highest AQYs ever reported for plasmonic photocatalysis
in liquid media. For instance, plasmonic systems for liquid phase reactions have shown
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
modest AQYs reaching 20% for H evolution from methylene blue and trifluoroacetic acid
2
5
0
with Au/TiO /Ru, and 18% for the aerobic oxidation of 2-propanol to acetone with
2
5
1
AuCu/TiO . In contrast, for gas phase reactions AQYs have reached values as high as 800%
2
–
2
(using monochromatic light at 400 mW cm , i.e. 200 times more intense that our light
36
24
source), 60% for the partial oxidation of ethylene with Ag/Al O , 33.5% for ammonia
2
3
3
5
decomposition to H with Cu-Ru/MgO, and 10% for the photocatalytic oxidation of CO
2
4
4
with Ag@SiO /Pt.
2
The observed high AQYs may be due to the high absorption of TiN-Pt nanohybrids and the
modest energy barrier of the targeted chemical reaction. This allows us to make use not only
of highly energetic hot electrons but eventually also of those electrons with lower energy,17
i.e., hot electrons that already underwent inelastic scattering before reacting with NH BH
3
3
molecules.
In order to further investigate the efficiency of TiN-Pt nanohybrids in plasmon-enhanced H2
evolution, we performed NH BH dehydrogenation under AM 1.5G solar illumination using
3
3
different light intensities. In contrast to experiments carried out with monochromatic light,
solar illumination always produced an increase in temperature because of photothermal
heating (Figure 4b). To distinguish how the raise in temperature and how hot electrons affect
H evolution kinetics, we first measured the temperature increase due to light irradiation in the
2
reacting solution, performed catalytic experiments in dark conditions at each specific “photo-
thermal” temperature, and then subtracted this contribution to the H evolution rates observed
2
under light irradiation (see detailed explanation in SI, Section S3).
The correct measurement of photothermal temperature generated on the surface of plasmonic
catalysts can be challenging, especially in case of gas phase chemical reactions driven by
1
8
ACS Paragon Plus Environment

Page 19 of 35
ACS Catalysis
1
2
3
4
5
6
7
8
9
solid bed catalysts.48,49 Otherwise, in colloidal photocatalysis (as in our case), contact between
individual nanoparticles and the liquid medium ensures effective heat removal by
conduction/convection. In such a system, the difference between the nanoparticle surface
4
8
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
temperature and the bulk temperature can be reduced to an insubstantial magnitude. This
validates that our methodology to measure photothermal heating in the reacting solution by
the use of a thermocouple well approximate the measurements of the real temperature
generated at TiN-Pt nanohybrids’ surface.
In the presence of TiN-Pt, solar irradiation produced substantial heating in the reaction
solution, reaching temperatures of ~50 °C under 10 Suns. To avoid excess of heating, which
may over accelerate NH BH dehydrogenation, we defined a plasmonic intermediate
3
3
temperature (IT) and a final temperature (FT) as the collective temperatures (due to all the
nanohybrids used for catalysis) generated after five minutes of light irradiation and at the end
of reaction, respectively (Figure 4b). Figure 4b shows the evolution of temperature under
increasing light intensity for different experimental conditions and highlights that IT
accounted generally for 80–85% of FT.
To analyze the photocatalytic performance of our nanohybrids, we divide the TOF into two
parts:
TOFphoto (H)  TOFtherm (H) TOFhote (H)
(4)
where TOFphoto is the overall photothermal H evolution rate, TOFtherm is the thermal
2
contribution, and TOFhote is the contribution coming from non-thermalized energetic
electrons (hot electrons) at a given light intensity (H). Since we can measure TOFphoto and
TOFtherm , we can retrieve their difference that represents the contribution of hot electrons,
TOFhote
.
1
9
ACS Paragon Plus Environment

ACS Catalysis
Page 20 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 4c shows the TOFtherm (at IT) values under dark, and TOFphoto obtained under solar
irradiation at different light intensity. In depth kinetic study for the H generation during
2
NH BH dehydrogenation reaction catalyzed by our reduced TiN–Pt nanohybrids under
3
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
different light intensity, under different ITs and FTs in dark condition shown in Figure S16.
Notably, when we introduced solar irradiation, we observed a non-linear increase of TOFphoto
–1
–1
with light intensity. At 10 Suns illumination TOFphoto reached 346.1 mol mol min , with
H2
Pt
TiN-Pt showing therefore a ~11-fold enhancement in catalytic activity with respect to dark
conditions because the combined effect of photothermal heating and hot electrons.
Interestingly, the heating contribution considering IT (TOFtherm , Figure 4c, blue bars) is
prominent at low light irradiation, while becomes less than 50% the TOFphoto value at 10 Suns.
In other words, the hot electron effect is more prominent at increasing light intensity.24,35,44,52
The same trend is also observed when considering FT, as shown in Figure S16.
To further elucidate the role of local heating and hot electrons, we calculated TOFhote at both
4
4
ITs and FTs (see SI, Section S3) and then the hot electron generation rate per Pt nanocrystal
at different light intensity (Figure 4d). Notably, the temperature increase in the reacting
solution is dynamic during the course of reaction, therefore TOFhote values obtained using
TOFtherm at ITs (minimum photothermal heating) and FTs (maximum photothermal heating)
represent the maximum and the minimum number of usable hot electrons participating to H2
evolution, respectively.
Figure 4d shows that the hot electron generation rate during plasmon-enhanced H evolution
2
from NH BH dehydrogenation exhibits a super-linear, power-law dependence from light
3
3
intensity, a characteristic feature observed in electron-driven chemical reactions on
metals.20,52
2
0
ACS Paragon Plus Environment

Page 21 of 35
ACS Catalysis
1
2
3
4
5
6
7
8
9
Figure 4e shows the solar AQYs for photothermal and hot electron-driven rates for plasmon-
enhanced H production. The photothermal solar AQY is ~9% at 1 Sun and decreases to
2
stable values comprised between 6 and 7% at higher light intensities. In contrast, the hot
electron (either considering IT or FT) solar AQY trend shows similar values around 2–3% up
to 5 Suns illumination, while for higher light intensity it increases to ~4% at 10 Suns
illumination. Plots of AQYs highlight that at low light intensity (1–5 Suns) photothermal
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
effects plays major role in boosting the activity of TiN-Pt nanohybrids, while at higher
intensity (5–10 Suns) the contribution from hot electrons prevails over heating.35
We also evaluated the recyclability and stability of the reduced TiN–Pt nanohybrids in
NH BH dehydrogenation reaction. The catalytic performance of the nanohybrids were
3
3
comparable after three consecutive catalytic cycles (Figure S18). TEM image and photograph
of reused nanohybrids show that the size, morphology and their colloidal stability remain
unaltered after catalytic reactions confirming their high stability and recyclability (Figure
S19).
Theoretical COMSOL simulations
To further demonstrate that the reaction rates are influenced by the collective photothermal
heating53,54 rather than by thermal gradients generated nearby a single nanostructure, we
performed COMSOL simulations and calculated the raise in temperature caused by a single
TiN-Pt nanohybrids always considering a 1.5 nm surface layer of TiO . We started from the
2
case of a single nanohybrid under monochromatic excitation with a relatively high intensity
(Figure 5) and observed relatively small values for the local photoinduced temperature even
for such high intensity.
Regarding our experiments with solar-like radiation, the total temperature increase near a
single nanohybrid (NH) can be divided in two parts, as follows:
Tphoto (r)  Tcollective  Tlocal (r)
(5)
2
1
ACS Paragon Plus Environment

ACS Catalysis
Page 22 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
where Tcollective and Tlocal are the collective and local temperatures, respectively. The
contribution Tcollective comes from all neighboring nanohybrids and does not significantly
change in space in the vicinity of the considered nanohybrid. As expected, the COMSOL-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
calculated local temperature increase nearby a single TiN-Pt nanohybrid, Tlocal , is very small
–
6
and it is estimated to be 8.1  10 °C under 1 Sun illumination. Overall, the calculated local
temperature increase in our system scales linearly withe the light intensity:
Tlocal  T  I
1
sun
where Isun is the light intensity in Suns. Details of our numerical estimation for the solar local
temperature increase can be found in the Supporting Information. Notably, even under
3
–2
6
resonant illumination (λ = 675 nm) at I = 10 W cm , i.e., 10 -times higher than the
experimental light intensity, the temperature increase on the surface of TiN-Pt is only ~0.2 K
and decreases rapidly with the distance from the nanohybrid surface (Figure 5a and b).
The photo-temperature is proportional to the energy dissipation inside one nanohybrid and
can be estimated analytically as
qNH ,solar
1
Tlocal ~
(6)
4
kt,w LNH / 2
where qNH ,solar is the total heat power dissipated in the nanohybrid under solar irradiation, and
kt,w and LNH are the thermal conductivity of water and the nanohybrid size, respectively.
Supplementary Section 4 contains more details on these photothermal calculations. In a large
three-dimensional system, the collective temperature increases with the number of illuminated
5
3
nanohybrids as (Figure 5c)
2
NH
/3
Tcollective  Coeff Tlocal
N
(7)
where NNH is the total number of nanohybrids and Coeff is a coefficient dependent on the
size and concentration of the nanocrystals. In our system, the number of nanohybrids is, of
2
2
ACS Paragon Plus Environment

Page 23 of 35
ACS Catalysis
1
2
3
4
5
6
7
8
9
course, very large, and therefore we have Tcollective  Tlocal . Indeed, the measured
temperature (i.e., collective temperature) is much higher than the calculated local one.
However, a theoretical calculation of the collective temperature is beyond the scope of this
study, since it involves complex processes of heat transfer from the outside surface of the
spherical glass reactor to the air, including convection and heat diffusion.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
As expected, the enhancement coefficient owing to the collective heating is a large number:

Tlocal
Tcollective
6
Enh 
 0.610
(8)
T
where we used one-sun illuminaton and Tcollective  5C that is the temperature increase due
to the TiN-Pt nanohybrid, as determined from the experiment in Fig. 3b. This is an important
property of solution ensembles of nanohybrids. Whereas the local temperature increase is
small, the resulting collective temperature can be significant as experimentally measured
(55 °C under 10 Suns illumination). In our case, this collective photo-temperature leads to a
significant activity enhancement of thermal origin.
2
3
ACS Paragon Plus Environment

ACS Catalysis
Page 24 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 5. a, Temperature map of a single TiN-Pt nanohybrid with a 1.5 nm surface layer of
TiO and b, corresponding distance-dependent temperature profile. In a,b, the optical
2
excitation is monochromatic, at the wavelength at the plasmon peak. c, Schematic
representation of the photocatalytic reactor showing the dependence of collective heating
from the number of nanoparticles.
Hot electron-driven reaction mechanism
For the dehydrogenation of NH BH over metallic catalysts, it has been reported that the rate-
3
3
determining step (RDS) is the cleavage of O–H bond in water molecules.9,10 In order to verify
this hypothesis, we first experimentally confirmed that NH BH is not involved in the RDS of
3
3
H evolution, by observing that the reaction kinetics shows a zero-order dependence with
2
NH BH concentration (SI, Section S5).
3
3
2
4
ACS Paragon Plus Environment

Page 25 of 35
ACS Catalysis
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 6. a, Kinetics of H evolution from ammonia borane (NH BH ) dehydrogenation
2
3
3
reaction with TiN–Pt nanohybrids in H O and D O in dark conditions. b, Arrhenius plots of
2
2
apparent activation energies under light and dark conditions at plasmonic temperature. k is the
rate of H evolution. c, Schematic representation of the free energy plots showing the energy
2
barriers associated with the formation of the reaction transition state (TS) of the rate
determining step during NH BH dehydrogenation reaction under light and dark conditions.
3
3
Asterisks denote species adsorbed on the catalyst surface. d, Cartoon showing the facilitated
rupture of O–H bond of water molecule induced by plasmonic hot electrons generated on the
surface of a TiN–Pt nanohybrids.
Afterwards, we demonstrated the direct involvement of water in the RDS by performing
kinetic isotope effect (KIE) experiments with deuterated water (D O). Figure 6a shows the H
2
2
evolution kinetics for TiN-Pt nanohybrids in dark conditions using H O or D O. The KIE
2
2
(
H O rate/D O rate) was 5.1 and confirmed that H O cleavage limits the reaction rate in
2 2 2
NH BH dehydrogenation. When we run KIE experiments under 8 Suns illumination (Figure
3
3
S24) we found a higher KIE value (5.57) as compared to dark conditions (5.10), a
2
5
ACS Paragon Plus Environment

ACS Catalysis
Page 26 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
20
characteristic feature of hot electron-driven processes, thus confirming that plasmonic hot
electron play a role in accelerating the breaking of O–H bonds in water molecules during
ammonia borane dehydrogenation.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
To quantify the effect of illumination on the activation energy (E ), we used the Arrhenius
a
equation and plotted H evolution thermal rates in dark conditions at varying plasmonic
2
temperature, as well as photocatalytic rates expressed as difference between the photothermal
3
5
–1
and the thermal rates (Figure 6b and SI, Section S5). Without light, the E was 48.7 kJ mol
a
–1
(
0.5 eV), while it decreased to 29.2 kJ mol (0.3 eV) under solar light irradiation. This
suggests that hot electrons, produced upon TiN-Pt excitation, directly participate to the RDS
of H evolution (O–H bond cleavage) reaction by stabilizing the transition state (TS) by 0.2
2
eV (Figure 6c and d). Hot electrons eventually accelerate the rupture of the B–H bond, thus
finally enhancing the overall H evolution rate (Figure S25, Section S5).
2
Plasmonic tandem hydrogenation of 4-nitrophenol
We finally demonstrate the possibility of tandem hydrogenation of 4-nitrophenol using in-situ
generated H gas, evolved from plasmon-enhanced NH BH dehydrogenation by TiN–Pt
2
3
3
nanohybrids (Figure S26). Under dark conditions, the complete hydrogenation of 4-
nitrophenoxide to 4-aminophenoxide anion takes 20 min in the presence of reduced TiN–Pt
nanohybrids, whereas under 5 Suns illumination the reaction is completed in 6 min. The
hydrogenation rate constants in dark and under illumination is 0.17 and 0.42 min–1
respectively, showing that plasmonic effects can produced a 2.5-fold rate enhancement during
tandem hydrogenations reactions.
CONCLUSIONS
We reported the preparation of a plasmonic photocatalyst showing minimized scattering (0.1)
and high absorption (0.9) including acid resistant plasmonic nanocubes core (TiN) and Pt
nanocrystals. The generation of hot electrons is spatially localized mainly on the TiN core,
2
6
ACS Paragon Plus Environment

Page 27 of 35
ACS Catalysis
1
2
3
4
5
6
7
8
9
while a smaller fraction of hot carriers is directly produced in the Pt catalytic centers. The
plasmonic enhancement for the Pt-based photochemistry comes from three factors: very
strong collective photo-heating created by the plasmonic TiN component, generation of hot
electrons inside TiN (those electrons can propagate to the Pt component), and the three-fold
plasmonic enhancement of absorption inside the Pt nanoparticles (and in situ generation of hot
electrons). We demonstrated that plasmonic nanohybrids can drive photocatalytic hydrogen
evolution with 120% apparent quantum yield under resonant optical excitation at 700 nm by
harnessing hot electrons only. In contrast, when using solar simulated irradiation, TiN-Pt
nanohybrids produced both efficient local heating and hot electron generation that boosted the
TiN-Pt activity of more than one order of magnitude if compared to dark conditions, reaching
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
–1
–1
a TOF of 346 mol mol min under 10 Suns illumination. We further discerned the
H2
Pt
relative contributions of photothermal heating and hot electrons, showing that hot electrons
play the role of decreasing the energy barrier to access the transition state of rate determining
step by accelerating water splitting at Pt surface. This work provides a general strategy to
enhance the catalytic activity of transition metal catalysts for H evolution from liquid
2
feedstock through combined action of heating and hot electrons, with important implications
for H storage technologies.
2
AUTHOR INFORMATION
Corresponding Author
*
E-mail: govorov@ohio.edu; alberto.naldoni@upol.cz.
Author Contributions
S.R. synthesized nanohybrids, performed photocatalytic experiments, and wrote the draft of
the manuscript. L.M. carried out optical measurements. E.Y.S. and A.O.G. designed and
performed simulations, wrote theoretical part of the manuscript. O.T. performed electron
2
7
ACS Paragon Plus Environment

ACS Catalysis
Page 28 of 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
microscopy and EELS measurements. Š.K., Z.W., R.Z., P.F. discussed results and revised the
manuscript. A.N. conceived the concept, revised the manuscript and supervised the project.
Notes
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The authors declare no competing financial interest.
ASSOCIATED CONTENT
Supporting Information
Additional structural characterization, photocatalytic data, and theoretical plots. This material
is available free of charge via the Internet at http://pubs.acs.org.
ACKNOWLEDGMENTS
The authors gratefully acknowledge support from the Czech Science Foundation (GACR)
through the project 20-17636S. Part of this work was carried out with the support of the
CEITEC Nano Research Infrastructure (MEYS CR, 2016–2019).
REFERENCES
(
1)
Angew. Chem. Int. Ed. 2007, 46, 8116–8118.
2) Rossin, A.; Peruzzini, M. Ammonia–borane and Amine–borane Dehydrogenation
Mediated by Complex Metal Hydrides. Chem. Rev. 2016, 116, 8848–8872.
3) Staubitz, A.; Robertson, A. P. M.; Manners, I.; Ammonia–Borane and Related
Compounds as Dihydrogen Sources. Chem. Rev. 2010, 110, 4079–4124.
(4) Sutton, A. D.; Burrell, A. K.; Dixon, D. A.; Garner III, E. B.; Gordon, J. C.; Nakagawa,
Marder, T. B. Will We Soon be Fueling Our Automobiles with Ammonia–borane?
(
(
T.; Ott, K. C.; Robinson, J. P.; Vasiliu, M.; Regeneration of Ammonia–borane Spent Fuel by
Direct Reaction with Hydrazine and Liquid Ammonia, Science 2011, 331, 1426–1429.
(5)
Zhan, W.-W.; Zhu, Q.-L.; Xu, Q. Dehydrogenation of Ammonia–borane by Metal
Nanoparticle Catalysts. ACS Catal. 2016, 6, 6892–6905.
2
8
ACS Paragon Plus Environment

Page 29 of 35
ACS Catalysis
1
2
3
4
5
6
7
8
9
(
6)
Chen, W.; Ji, J.; Duan, X.; Qian, G.; Li, P.; Zhou, X.; Chen, D.; Yuan, W. Unique
Reactivity in Pt/CNT Catalyzed Hydrolytic Dehydrogenation of Ammonia–borane. Chem.
Commun. 2014, 50, 2142–2144.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(7)
Chandra, M.; Xu, Q. Room Temperature Hydrogen Generation from Aqueous
Ammonia–borane Using Noble Metal Nano-Clusters as Highly Active Catalysts. J. Power
Sources 2007, 168, 135–142.
(8)
Aijaz, A.; Karkamkar, A.; Choi, Y. J.; Tsumori, N.; Rönnebro, E.; Autrey, T.;
Shioyama, H.; Xu, Q. Immobilizing Highly Catalytically Active Pt Nanoparticles Inside the
Pores of Metal–organic Framework: A Double Solvents Approach. J. Am. Chem. Soc. 2012,
134, 13926–13929.
(9)
Li, Z.; He, T.; Matsumura, D.; Miao, S.; Wu, A.; Liu, L.; Wu, G.; Chen, P. Atomically
Dispersed Pt on the Surface of Ni Particles: Synthesis and Catalytic Function in Hydrogen
Generation from Aqueous Ammonia–borane. ACS Catal. 2017, 7, 6762–6769.
(10) Fu, F.; Wang, C.; Wang, Q.; Martinez-Villacorta, A. M.; Escobar, A.; Chong, H.;
Wang, X.; Moya, S.; Salmon, L.; Fouquet, E.; Ruiz, J.; Astruc, D. Highly Selective and Sharp
Volcano–type Synergistic Ni Pt@ZIF-8-Catalyzed Hydrogen Evolution from Ammonia–
2
borane Hydrolysis. J. Am. Chem. Soc. 2018, 140, 10034–10042.
(11) Akbayrak, S.; Tonbul, Y.; Özkar, S. Ceria Supported Rhodium Nanoparticles: Superb
Catalytic Activity in Hydrogen Generation from the Hydrolysis of Ammonia–borane. Appl.
Catal. B 2016, 198, 162–170.
(12) Rej, S.; Hsia, C.-F.; Chen, T.-Y.; Lin, F.-C.; Huang, J.-S.; Huang, M. H. Facet-
Dependent and Light-Assisted Efficient Hydrogen Evolution from Ammonia–borane Using
Gold-Palladium Core-Shell Nanocatalysts. Angew. Chem. Int. Ed. 2016, 55, 7222–7226.
(13) Fuku, K.; Hayashi, R.; Takakura, S.; Kamegawa, T.; Mori, K.; Yamashita, H. The
Synthesis of Size- and Color-Controlled Silver Nanoparticles by Using Microwave Heating
2
9
ACS Paragon Plus Environment