Reactivity of hydrogen with solid-state films of alkylamine- and tetraoctylammonium bromide-stabilized Pd, PdAg, and PdAu nanoparticles for sensing and catalysis applications

Article abstract of DOI:10.1021/ja076000h

Hydrogen gas spontaneously adsorbs to Pd metal as atomic hydrogen and diffuses into the lattice to form PdH_x. We previously showed that films of hexanethiolate-coated Pd monolayer-protected clusters (MPCs) do not readily react with H₂ due to the strong chemical bonding of the thiolate to the Pd, which inhibits the reaction. Consequently, these films require ozone or heat treatment for reactivity to occur, which is inconvenient for sensing or catalysis applications. In this report, we describe the reactivity between H₂ and solid-state films of alkylamine-coated Pd, PdAg (10:1), and PdAu (10.1) MPCs and films of tetraoctylammonium bromide (TOABr)-stabilized Pd and PdAg (10:1) nanoparticles as determined by changes in film conductivity. Our data show that Pd nanoparticles coated with these more weakly coordinated amine or ammonium groups readily react with H₂ without any need for ozone or heat treatment. The conductivity of films of octylamine (C8NH₂)- or dodecylamine (C12NH₂)-coated Pd, PdAg, and PdAu MPCs increases irreversibly upon initial exposure to 100% H ₂ to varying degrees and with different reaction kinetics and then exhibits stable, reversible changes in the presence of H₂ concentrations ranging from 9.6 to 0.08%. The behavior upon initial exposure to H₂ (conditioning) and the direction and magnitude of the reversible conductivity changes depend on the alkyl chainlength and alloy composition. Films of TOABr-coated Pd and PdAg nanoparticles show stable, reversible increases in conductivity in the presence of H₂ concentrations from 9.6 down to 0.11% without conditioning. Surface FTIR spectroscopy and atomic force microscopy (AFM) provide information about the organic monolayer and film morphology, respectively, following reactivity with H₂. This work demonstrates a simple approach toward preparing films of chemically synthesized Pd-containing nanoparticles with controlled reactivity to H₂ for sensing and catalysis applications.

Full text of DOI:10.1021/ja076000h

Published on Web 12/20/2007
Reactivity of Hydrogen with Solid-State Films of Alkylamine-
and Tetraoctylammonium Bromide-Stabilized Pd, PdAg, and
PdAu Nanoparticles for Sensing and Catalysis Applications
Francisco J. Iba n˜ ez and Francis P. Zamborini*
Department of Chemistry, UniVersity of LouisVille, LouisVille, Kentucky 40292
Received August 9, 2007; E-mail: f.zamborini@louisville.edu
Abstract: Hydrogen gas spontaneously adsorbs to Pd metal as atomic hydrogen and diffuses into the
lattice to form PdH
clusters (MPCs) do not readily react with H
x
. We previously showed that films of hexanethiolate-coated Pd monolayer-protected
due to the strong chemical bonding of the thiolate to the Pd,
2
which inhibits the reaction. Consequently, these films require ozone or heat treatment for reactivity to occur,
which is inconvenient for sensing or catalysis applications. In this report, we describe the reactivity between
H and solid-state films of alkylamine-coated Pd, PdAg (10:1), and PdAu (10:1) MPCs and films of
2
tetraoctylammonium bromide (TOABr)-stabilized Pd and PdAg (10:1) nanoparticles as determined by
changes in film conductivity. Our data show that Pd nanoparticles coated with these more weakly coordinated
amine or ammonium groups readily react with H
conductivity of films of octylamine (C8NH )- or dodecylamine (C12NH
increases irreversibly upon initial exposure to 100% H to varying degrees and with different reaction kinetics
and then exhibits stable, reversible changes in the presence of H concentrations ranging from 9.6 to 0.08%.
The behavior upon initial exposure to H (conditioning) and the direction and magnitude of the reversible
conductivity changes depend on the alkyl chainlength and alloy composition. Films of TOABr-coated Pd
and PdAg nanoparticles show stable, reversible increases in conductivity in the presence of H concentrations
from 9.6 down to 0.11% without conditioning. Surface FTIR spectroscopy and atomic force microscopy
AFM) provide information about the organic monolayer and film morphology, respectively, following reactivity
with H . This work demonstrates a simple approach toward preparing films of chemically synthesized Pd-
containing nanoparticles with controlled reactivity to H
2
without any need for ozone or heat treatment. The
2
2
)-coated Pd, PdAg, and PdAu MPCs
2
2
2
2
(
2
2
for sensing and catalysis applications.
Introduction
of the Pd, which has been heavily exploited for H sensing.
2
Palladium’s ability to break and adsorb H has also led to its
use for catalyzing various reactions.
The reaction between hydrogen and Pd or Pd-containing
alloys is of great interest because of its relevance in H2 sensing
and heterogeneous and homogeneous catalysis. It is well known
that H2 spontaneously adsorbs to Pd as atomic H and diffuses
into the lattice to form PdHx. The initial R-phase Pd becomes
â-phase PdHx through an R-â phase transition. The Pd lattice
H2 is a useful energy source that has the potential to reduce
the need for fossil fuels in the future. A great deal of effort has
1
(
13) Dwivedi, D.; Dwivedi, R.; Srivastava, S. K. Sens. Actuators, B 2000, 71,
161-168.
spacing changes throughout these phase changes, depending on
the H2 concentration1,2 in the surrounding atmosphere. The phase
(14) Favier, F.; Walter, E. C.; Zach, M. P.; Benter, T.; Penner, R. M. Science
2001, 293, 2227-2231.
(15) Huang, L.; Gong, H.; Peng, D.; Meng, G. Thin Solid Films 1999, 345,
transitions and changes in lattice spacing lead to measurable
217-221.
changes in the optical properties,3
-11
12-32
2,8
(16) Hughes, R. C.; Schubert, W. K. J. Appl. Phys. 1992, 71, 542-544.
resistance,
and mass
(17) Hughes, R. C.; Schubert, W. K.; Zipperian, T. E.; Rodriguez, J. L.; Plut,
T. A. J. Appl. Phys. 1987, 62, 1074-1083.
(
1) Lewis, F. A. The Palladium/Hydrogen System; Academic Press Inc.:
(18) Kaltenpoth, G.; Schnabel, P.; Menke, E.; Walter, E. C.; Grunze, M.; Penner,
R. M. Anal. Chem. 2003, 75, 4756-4765.
London, 1967.
(
2) Christofides, C.; Mandelis, A. J. Appl. Phys. 1990, 68, 1-30.
3) B e´ venot, X.; Trouillet, A.; Veillas, C.; Gagnaire, H.; Cl e´ ment, M. Sens.
Actuators, B 2000, 67, 57-67.
(19) Kang, W. P.; G u¨ rb u¨ z, Y. J. Appl. Phys. 1994, 75, 8175-8181.
(20) Kolmakov, A.; Klenov, D. O.; Lilach, Y.; Stemmer, S.; Moskovits, M.
Nano Lett. 2005, 5, 667-673.
(
(4) Garcia, J. A.; Mandelis, A. ReV. Sci. Instrum. 1996, 67, 3981-3983.
5) Kalli, K.; Othonos, A.; Christofides, C. ReV. Sci. Instrum. 1998, 69, 3331-
(21) Kong, J.; Chapline, M. G.; Dai, H. AdV. Mater. 2001, 13, 1384-1386.
(22) Luongo, K.; Sine, A.; Bhansali, S. Sens. Actuators, B 2005, 111-112, 125-
129.
(
3
338.
(
6) Kalli, K.; Othonos, A.; Christofides, C. J. Appl. Phys. 2002, 91, 3829-
(23) Lutz, B. J.; Fan, Z. H. Anal. Chem. 2005, 77, 4969-4975.
(24) Mizsei, J.; Voutilainen, J.; Saukko, S.; Lantto, V. Thin Solid Films 2001,
391, 209-215.
3
840.
(
(
(
7) Lin, H.; Gao, T.; Fantini, J.; Sailor, M. J. Langmuir 2004, 20, 5104-5108.
8) Smith, A. L.; Shirazi, H. M. Thermochim. Acta 2005, 432, 202-211.
9) Zhao, Z.; Carpenter, M. A. J. Appl. Phys. 2005, 97, 124301.
(25) Morris, J. E.; Kiesow, A.; Hong, M.; Wu, F. Int. J. Electron. 1996, 81,
441-447.
(
(
(
10) Zhao, Z.; Carpenter, M. A.; Xia, H.; Welch, D. Sens. Actuators, B 2006,
13, 532-538.
11) Zhao, Z.; Sevryugina, Y.; Carpenter, M. A.; Welch, D.; Xia, H. Anal. Chem.
004, 76, 6321-6326.
12) Dankert, O.; Pundt, A. Appl. Phys. Lett. 2002, 81, 1618-1620.
(26) Sakamoto, Y.; Takai, K.; Takashima, I.; Imada, M. J. Phys.: Condens.
Matter 1996, 8, 3399-3411.
1
(27) Sayago, I.; Terrado, E.; Lafuente, E.; Horrillo, M. C.; Maser, W. K.; Benito,
A. M.; Navarro, R.; Urriolabeitia, E. P.; Martinez, M. T.; Gutierrez, J. Synth.
Met. 2005, 148, 15-19.
2
6
22 J. AM. CHEM. SOC. 2008, 130, 622-633
9
10.1021/ja076000h CCC: $40.75 © 2008 American Chemical Society

Nanoparticles for Sensing and Catalysis Applications
A R T I C L E S
been put forth to develop H2-fueled motor vehicles to fulfill
increasing energy demands for transportation. Also, H2 is present
as a common reagent in industry and used as an O2 scavenger
in metallurgy, in hydrocracking for refined fuels, and in
degradation of synthetic materials.33 Operating with H2 can be
Early chemiresistive H2 sensors reported in the literature were
based on conductive Pd films whose resistance increased in the
presence of hydrogen due to the formation of the more resistive
1,26
PdHx. More recent reports on this type of sensing mechanism
demonstrated improved response times, higher sensitivity, and
lower detection limits by using nanostructured materials (Pd
2
dangerous because it is explosive in air above 4%. Accordingly,
34
31
one of the aims in fuel cell research and other applications is
nanotubes). Another type of sensing mechanism involves the
3
5,36
14,28
12,30
to safely store
and release H2 in a controlled manner. For
use of Pd nanowires
or films of nanoparticles
that contain
these reasons, it is important to develop simple, reliable, low-
cost sensors for the detection of H2 over a range of concentra-
tions.
disconnected, high resistance metal-metal junctions. These
materials exhibit a decrease in resistance in the presence of H2
due to the formation of PdHx, which expands in volume and
forms a more connected, lower resistance structure. Penner and
co-workers demonstrated this behavior with Pd mesowires
electrochemically synthesized by step-edge decoration of highly
In the area of catalysis, there is an increasing interest in the
use of chemically synthesized, reusable metal nanoparticles to
catalyze reactions with large turnover rates and high selectivity.
Organic stabilizers have the critical role of providing size and
shape control and reduced size dispersity during synthesis. They
also provide stability by preventing nanoparticle aggregation
and size and shape changes during the catalytic reaction. While
serving these functions, the stabilizer should also prevent
poisoning of the catalyst and not deter the reactivity of the metal
nanoparticle itself. Chemically synthesized catalyst particles can
often be easily separated from the reaction products and reused.
Crooks and co-workers recently demonstrated the use of
dendrimer-encapsulated Pd nanoparticles for catalyzing the
1
4,28,32
oriented pyrolytic graphite (HOPG)
and lithographically
18,44
fabricated Pd wires.
Others have described similar behavior
for discontinuous films of Pd, which contain Pd nanoparticles
evaporated or sputtered so that they are below the percolation
1
2,22,25,30
threshold for conductivity.
onstrated the importance of the functionality of the substrate
Zach and co-workers dem-
3
0
that the Pd is deposited on. In general, these types of sensors
exhibit very fast, reversible, and sensitive responses with
detection limits as low as 0.05 ppm and response times on the
1
4,30
order of milliseconds.
37
We previously described the reactivity of hexanethiolate-
coated Pd monolayer-protected clusters (MPCs) to H2 for
hydrogenation of olefins. Others catalyzed the transformation
of aromatic nitro and azide compounds to their corresponding
amines by the reaction with Pd nanoparticles attached to ferrite
nanoparticles through amine groups, where the catalyst could
4
5
chemiresistive sensing applications. This was the first report
on films of chemically synthesized Pd nanoparticles for H2
sensing, which has several potential benefits over evaporated
or sputtered Pd films in terms of simplicity, cost, reproducibility,
and control over the electronic properties and sensing behavior.
Although the electronic properties of Pd MPCs can be tailored
by the surrounding monolayer, the presence of the strongly
chemisorbed thiolate group prevented the reaction between Pd
and H2. This required ozone or heat treatment to desorb thiolates
from the surface and promote H2 reactivity, which is inconve-
nient and more complicated. It would be beneficial to synthesize
Pd nanoparticles that could be deposited as a film and directly
utilized for H2 sensing and catalysis without treatment. Here
we report the reactivity of solid-state films containing alky-
3
8
be easily separated magnetically via the ferrite nanoparticles.
39
Other examples include thioether-, tetraoctylammonium bro-
mide (TOABr)-, and chalcogenide-stabilized Pd nanoparticles
in the presence of thiols41 for the hydrogenation of 6-bromo-
40
1
-hexane, styrene, and various alkynes, respectively. Thiol-
42
stabilized Pt nanoparticles and binol-functionalized Au nano-
particles43 have also served as catalysts for the hydrogenation
of allyl alcohol and benzaldehyde, respectively. There have been
no previous studies to our knowledge on the reactivity of
alkylamine-stabilized Pd nanoparticles with H2 for sensing or
catalysis and no study on TOABr-coated Pd nanoparticles for
H2 sensing.
4
6
lamine- and tetraoctylammonium bromide (TOABr)-coated
40
Pd, PdAg, and PdAu alloys with H2 as determined by changes
in conductivity as well as surface reflectance FTIR spectroscopy
and atomic force microscopy (AFM) measurements on the films.
Importantly, we find that the alkylamine and ammonium-
containing molecules stabilize the Pd nanoparticles but do not
inhibit their reactivity with H2, as in the case of hexanethiolates.
We describe the details of the reactivity as a function of alkyl
chainlength, type of ligand (amine or ammonium), and metal
composition.
(
(
(
(
(
(
(
28) Walter, E. C.; Favier, F.; Penner, R. M. Anal. Chem. 2002, 74, 1546-
1
553.
29) Wolfe, D. B.; Love, J. C.; Paul, K. E.; Chabinyc, M. L.; Whitesides, G.
M. Appl. Phys. Lett. 2002, 80, 2222-2224.
30) Xu, T.; Zach, M. P.; Xiao, Z. L.; Rosenmann, D.; Welp, U.; Kwok, W. K.;
Crabtree, G. W. Appl. Phys. Lett. 2005, 86, 203104.
31) Yu, S.; Welp, U.; Hua, L. Z.; Rydh, A.; Kwok, W. K.; Wang, H. H. Chem.
Mater. 2005, 17, 3445-3450.
32) Yun, M.; Myung, N. V.; Vasquez, R. P.; Lee, C.; Menke, E.; Penner, R.
M. Nano Lett. 2004, 4, 419-422.
33) Ramachandran, R.; Menon, R. K. Int. J. Hydrogen Energy 1998, 23, 593-
5
98.
34) Che, G.; Lakshmi, B. B.; Fisher, E. R.; Martin, C. R. Nature 1998, 393,
3
46-349.
(
35) Schalpbach, L.; Zuttel, A. Nature 2001, 414, 353-358.
36) Horonouchi, S.; Yamanoi, Y.; Yonezawa, T.; Mouri, T.; Nishihara, H.
Langmuir 2006, 22, 1880-1884.
Experimental Section
(
(37) Niu, Y.; Yeung, L. K.; Crooks, R. M. J. Am. Chem. Soc. 2001, 123, 6840-
Chemicals. Sodium borohydride (99%), tetraoctylammonium bro-
mide (99%), toluene (99.9%), 2-propanol (99.9%), ethanol (200 proof),
dichloromethane (99%), and acetonitrile (99%) were purchased from
6
846.
(
38) Guin, D.; Baruwati, B.; Manorama, S. V. Org. Lett. 2007, 9, 1419-
1
421.
(
(
39) Ganesan, M.; Freemantle, R. G.; Obare, S. O. Chem. Mater. 2007.
40) Hwang, C.-B.; Fu, Y.-S.; Lu, Y.-L.; Jang, S.-W.; Chou, P.-T.; Wang, C.
R. C.; Yu, S.-J. J. Catal. 2000, 195, 336-341.
8 2
VWR Scientific Products. Hexanethiol (C6S), octylamine (C H25NH ),
(
41) Ananikov, V. P.; Orlov, N. V.; Beletskaya, I. P.; Khrustalev, V. N.; Antipin,
M. Y.; Timofeeva, T. V. J. Am. Chem. Soc. 2007, 129, 7252-7253.
42) Eklund, S. E.; Cliffel, D. E. Langmuir 2004, 20, 6012-6018.
43) Marubayashi, K.; Takizawa, S.; Kawakuzu, T.; Arai, T.; Sasai, H. Org.
Lett. 2003, 5, 4409-4412.
(44) Im, Y.; Lee, C.; Vazquez, R. P.; Bangar, M. A.; Myung, N. V.; Menke, E.
J.; Penner, R. M.; Yun, M. Small 2006, 2, 356-358.
(45) Iba n˜ ez, F. J.; Zamborini, F. P. Langmuir 2006, 22, 9789-9796.
(46) Kumar, A.; Mandal, S.; Selvakannan, P. R.; Pasricha, R.; Mandale, A. B.;
Sastry, M. Langmuir 2003, 19, 6277-6282.
(
(
J. AM. CHEM. SOC.
9
VOL. 130, NO. 2, 2008 623

A R T I C L E S
Iba n˜ ez and Zamborini
dodecylamine (C12
H
25NH
2
), K
2
PdCl
4
, and AgC
2
F
3
O
2
were purchased
) containing
) were purchased from
2
O was synthesized from metallic Au.
the C6S Pd MPCs, except the hexanethiol was not added. These
nanoparticles are insoluble when isolated as a powder.5 Thus, after
reduction and separation from the water phase, the toluene solution
was reduced to ∼50 mL by rotary evaporation, filtered with a microdisc
filter (acrodisc, 2 µm PTFE membrane) to remove insolubles, and then
used from solution. TOABr PdAg nanoparticles were synthesized by
0
from Aldrich Chemical Co. 1-hexadecylamine (C16
and 1-octadecylamine (C18
Alfa Aesar Co. HAuCl · 3 H
H36NH
2
C
18
H37NH
2
H37NH
2
4
Barnstead Nanopure water (17.8 MΩ-cm) was employed for all aqueous
solutions.
Synthesis of Hexanethiolate-Coated Pd MPCs. Hexanethiolate-
adding AgC
performing the rest of the synthesis the same as that for pure Pd. The
10-fold excess of NaBH was with respect to the total metal. The
average diameter of TOABr Pd nanoparticles was 7.0 ( 0.5 nm as
determined by AFM measurements, and UV-vis spectra were consis-
tent with the formation of metal nanoparticles.
2 3 2
F O to the solution in a final 10:1 Pd:Ag molar ratio and
coated (C6S) Pd monolayer-protected clusters (MPCs) were synthesized
4
7
45,48
according to a modified Brust reaction as reported previously.
Briefly, 0.40 g of K PdCl was dissolved in 25 mL of water, and
.00 g of tetraoctylammonium bromide (TOABr) was dissolved in
50 mL of toluene. The two solutions were combined and stirred until
4
2
4
1
1
2-
all of the PdCl
4
transferred into the toluene phase. The toluene phase
Sensor Device. Two Au electrodes separated by 23 µm were
fabricated in a clean room by standard photolithography, sputtering,
was separated, and 90 µL of hexanethiol, corresponding to a 1:2 thiol/
Pd ratio, was added to the toluene and stirred. The solution was cooled
and lift-off procedures on a Si/SiO substrate. The 100 nm thick Au
x
to ∼0 °C using an ice bath and a 10-fold excess of NaBH
4
(0.46 g in
electrodes were sputtered over a 10 Å thick Cr adhesion layer during
the process. Wire leads were attached to the Au electrodes with Ag
epoxy (cured 12 h, 80 °C), which was further insulated with an
overlayer of Torr-seal epoxy (cured 12 h, 80 °C). The electrode was
cleaned by rinsing in acetonitrile, dichloromethane, acetone, ethanol,
1
0 mL of water) with respect to Pd was added to the toluene solution
with stirring. The solution turned black within a few seconds, indicating
the formation of metallic Pd MPCs. Ten milliliters of additional water
was added, and the solution was stirred overnight. The toluene layer
was separated and removed by rotary evaporation. The remaining black
solid was suspended in 200 mL of acetonitrile and collected by filtration
on a glass fritted B u¨ chner funnel. The black solid product was washed
with an additional 250 mL of acetonitrile and thoroughly dried before
collecting. The average diameter of Pd MPCs prepared this way is
and 2-propanol before drying under N . The device was then placed in
2
a UVO ozone cleaner (Jelight Company Inc., Irvine, CA) for 10 min
before depositing the films of nanoparticles over the electrodes. Films
of alkylamine-coated MPCs were drop-cast deposited using 1-3 drops
of a 4-70 mg/mL toluene solution, which leads to many multilayers
of nanoparticles with film thicknesses well above the thickness of the
electrodes (>100 nm). Previous transmission electron microscopy
(TEM) studies show that monolayer films are hexagonally close-packed
4
8
3
.0 nm according to literature.
Synthesis of Alkylamine-Coated Pd, PdAg, and PdAu MPCs.
Alkylamine-coated Pd MPCs were synthesized similar to previously
reported alkylamine-coated Au MPCs.49 Solutions of 0.50 g (1.59 mmol)
51
with the alkyl chains interdigitated and detailed conductivity studies
of K
2
PdCl
4
dissolved in 20 mL of water and 1.92 g (3.51 mmol) of
model the films as being well-ordered and hexagonally close-packed
or packed in a cubic lattice with reasonable results.52 The concentration
and number of drops were chosen to obtain current through the film
above the baseline current. Films of TOABr-coated nanoparticles were
drop-cast deposited using 2 drops of an approximately 3.0 mg/mL
toluene solution, which is the concentration in terms of Pd when
reducing the volume to 50 mL by rotary evaporation. The concentration
used to prepare the films generally increased with increasing chainlength
on the Pd MPCs. The range of concentrations used leads to different
film thicknesses; however, the amount deposited in all cases is enough
to completely cover the two electrodes and data from a few selected
TOABr dissolved in 100 mL of toluene were combined and stirred
until all PdCl
alkylamine ligand was then added in a 12:1 alkylamine/Pd ratio and
2
-
4
transferred into the toluene phase. The appropriate
rapidly stirred for ∼2 h. This ratio corresponds to 0.019 mol of
2
alkylamine, which is 3.10 mL for octylamine (C8NH ), 3.39 g for
dodecylamine (C12NH
-octadecylamine (C16-18NH
white upon addition of the alkylamine due to complexation between
2
), and 4.42 g for 1-hexadecylamine containing
1
2
). The aqueous layer became cloudy
2
- 49
the protonated amine and the PdCl
4
4
. Next, 0.84 g of NaBH (0.022
mol) dissolved in 10 mL of water was added to the two-phase solution
while stirring. The solution turned black within a few seconds, indicating
the formation of metallic Pd MPCs. Ten milliliters of additional water
was added, and the solution was stirred overnight. The toluene layer
was separated and removed by rotary evaporation. The remaining black
solid was suspended in 200 mL of acetonitrile, collected by filtration
on a glass fritted B u¨ chner funnel, and washed with an additional 150
mL of acetonitrile and 150 mL of ethanol before thoroughly drying
2
samples (C6S Pd and C8NH PdAg) showed that the sensing response
was not statistically different over a small thickness range.
Gas-Sensing Experiments. Gas-sensing experiments were per-
formed with a CH Instruments 660A (Austin, TX) electrochemical
workstation operating in chronoamperometry mode. The current was
monitored with time while a -0.3 V potential was applied between
the two electrodes and the sample was exposed to alternating flow of
and collecting. The average diameter of C8NH
.8 nm as determined by atomic force microscopy (AFM) measure-
ments.
2
Pd MPCs was 3.0 (
pure N
2
2 2
or air and different concentrations of H in the N or air carrier
0
gas. A range of H
2
concentrations was obtained using a set of flow
meters (Cole Parmer, 2% error at full scale) located between the sample
and gas cylinders and operated by a 3-way valve, which allows a
constant H flow during mixing with N and avoids overpressures and
Alkylamine-coated PdAg and PdAu alloy MPCs were synthesized
in a 10:1 Pd:Ag or Pd:Au ratio by adding the appropriate amount of
2
2
K
2
PdCl
4
and AgC
2
F
3
O
2
or HAuCl
4
‚3H
2
O salts. AgC
2
F
3
O
2
was directly
artificial current spikes during sensing. The different concentrations
-
soluble in the toluene phase whereas AuCl
4
was phase transferred
was
2:1 and 15:1 with respect to total metal content. Nuclear magnetic
of H and the total flow rates (H + N /air) used are as follows: 9.6 (
2
2
2
-
1
-1
into toluene with TOABr. The amount of alkylamine and NaBH
1
4
0.3% (3.1 ( 0.1 L.min ), 6.2 ( 0.2% (4.9 ( 0.1 L.min ), 3.2 (
-
1
-1
0.1% (4.7 ( 0.1 L.min ), 1.0 ( 0.1% (4.7 ( 0.1 L.min ), 0.50 (
-
1
-1
resonance (NMR) and UV-vis spectroscopy measurements on Pd,
PdAg, and PdAu MPCs were consistent with the successful synthesis
of pure MPCs.
0.02% (4.6 ( 0.1 L.min ), 0.21 ( 0.02% (4.6 ( 0.1 L.min ), 0.11
-
1
-1
( 0.02% (4.6 ( 0.1 L.min ), 0.08 ( 0.02% (4.6 ( 0.1 L.min ).
Characterization. Films of alkylamine- and TOABr-coated Pd and
Pd alloy nanoparticles were drop-cast deposited onto Si(100)/TiW(50
Synthesis of TOABr-Coated Pd and PdAg Nanoparticles. TOABr
coated Pd nanoparticles were synthesized in an identical manner as
(
50) Isaacs, S. R.; Culter, E. C.; Park, J.-S.; Lee, T. R.; Shon, Y.-S. Langmuir
(
(
(
47) Brust, M.; Walker, M.; Bethell, D.; Schiffrin, D. J.; Whyman, R. Chem.
2005, 21, 5689-5692.
Comm. 1994, 7, 801-802.
(51) Wuelfing, W. P.; Green, S. J.; Pietron, J. J.; Cliffel, D. E.; Murray, R. W.
J. Am. Chem. Soc. 2000, 122, 11465-11472.
(52) Zamborini, F. P.; Smart, L.; Leopold, M. C.; Murray, R. W. Anal. Chim.
Acta 2003, 496, 3-16.
48) Zamborini, F. P.; Gross, S. M.; Murray, R. W. Langmuir 2001, 17, 481-
4
87.
49) Leff, D. V.; Brandt, L.; Heath, J. R. Langmuir 1996, 12, 4723-4730.
6
24 J. AM. CHEM. SOC. VOL. 130, NO. 2, 2008
9

Nanoparticles for Sensing and Catalysis Applications
A R T I C L E S
Figure 1. Optical microscope images of the electrode device (A) before
and (B) after drop-cast deposition of a film of C8NH2 Pd MPCs between
the 23 µm electrode gap. The scheme of the nanoparticle shows the different
metal compositions and different ligand stabilizers used in our previous
work (C6S Pd MPCs) and this report (alkylamine- and TOABr-coated Pd
and Pd alloy nanoparticles).
Å)/Au(2000 Å) and electrode devices for surface reflectance Fourier
Transform Infrared (FTIR) spectroscopy and AFM experiments,
respectively. FTIR data were acquired using a Digilab FTS 7000
spectrometer (Varian, Cambridge, MA) in reflectance mode with a
2
liquid N -cooled MCT detector. AFM images were acquired with a
Veeco Digital Instruments Nanoscope 3A Multimode Scanning Probe
Microscope (Santa Barbara, CA) using a Si tip operating in tapping
mode. H Proton NMR and UV-vis spectroscopy were obtained with
Figure 2. (A) Cyclic voltammograms (CVs) of drop-cast deposited films
of C8NH2, C12NH2, C16-18NH2, and TOABr Pd nanoparticles obtained
in air from +0.3 to -0.3 V (started at 0.0 V) at a sweep rate of 100 mV/s.
1
an INOVA 500 MHz and a Varian Cary 50 spectrometer, respectively.
(B) Expanded CVs of the background (no film) and films of C12NH2 Pd
and C16-18NH2 Pd as indicated by the dashed rectangle in (A). (C)
Chronoamperometry (CA) plots of the same drop-cast films from (A)
measured in air at -0.3 V for 1000 s.
Results and Discussion
Sensor Device. Figure 1 shows the device (see Experimental
Section) used to test the reactivity of hydrogen with various
solid-state films of alkylamine- and tetraoctylammonium bromide-
coated Pd, PdAg, and PdAu nanoparticles by monitoring
changes in conductivity in the presence of H2 with N2 or air as
the carrier gas. Frame A shows the device before and Frame B
shows the device after drop-cast deposition of a film of
octylamine (C8NH2)-coated Pd MPCs. The figure also illustrates
the various metal compositions (Pd, PdAg, and PdAu) and
protecting ligands used in our previous work and in this study,
which includes hexanethiolate (C6S), octylamine (C8NH2),
dodecylamine (C12NH2), hexadecylamine-octadecylamine (C16-
medium surrounding the clusters, and cluster size.5
1,53-56
Note
that there is almost no current through the film of C16-18 NH2
Pd MPCs because the cluster edge-to-edge distance is too large
for significant electron hopping to occur, which is expected to
be greater than 20 Å considering that the distance is estimated
to be ∼1.2 times the chainlength of one ligand due to
interdigitation of the monolayers from adjacent clusters.51
The CV of the film of TOABr-coated Pd nanoparticles is
markedly different compared to that of C8NH2 Pd MPCs. The
current is on the same order of magnitude, but the plot does
not show ohmic behavior. Instead, there is large hysteresis on
the forward and reverse scan, which is consistent with the current
1
8NH2), and tetraoctylammonium bromide (TOABr).
+
Electronic Properties of Alkylamine- and TOABr-Coated
being dominated by ions, likely due to the large excess of TOA
and Br ions in the film, since these nanoparticles could not be
-
Pd Nanoparticles. Figure 2A shows cyclic voltammograms
CVs) of solid-state, drop-cast deposited films of C8NH2,
(
purified as a solid.
C12NH2, C16-18NH2, and TOABr-coated Pd nanoparticles.
Figure 2B shows the films of C12NH2 Pd, C16-18NH2 Pd, and
the background expanded (see dashed box in Figure 2A). The
current through the film of C8NH2 Pd MPCs is linear with
potential, showing ohmic behavior. Current through the film
of C12NH2 is also ohmic but shows hysteresis because the
current is close to the background (with no MPCs deposited).
The CV of the film of C16-C18NH2 Pd MPCs is the same as
the background, indicating no conductivity through this film.
The shape of the curves and the fact that the conductivity, which
is proportional to the slope of the CV, decreases with increasing
chainlength is consistent with an electron hopping conductivity
mechanism through the films, which depends on the cluster
edge-to-edge distance (exponentially), the dielectric of the
Figure 2C shows chronoamperometry (CA) plots in air of
films of C8NH2, C12NH2, C16-18NH2, and TOABr-coated Pd
nanoparticles measured for 1000 s at -0.3 V. For the films of
alkylamine-coated Pd MPCs, the current is fairly constant over
time, consistent with ohmic behavior. The magnitude of current
-9
-10
-11
is 5.5 × 10 , 1.0 × 10 , and 2.3 × 10
A for films of
(
53) Terrill, R. H.; Postlethwaite, T. A.; Chen, C.-h.; Poon, C.-D.; Terzis, A.;
Chen, A.; Hutchison, J. E.; Clark, M. R.; Wignall, G.; Londono, J. D.;
Superfine, R.; Falvo, M.; Johnson, C. S., Jr.; Samulski, E. T.; Murray, R.
W. J. Am. Chem. Soc. 1995, 117, 12537-12548.
(
54) Zamborini, F. P.; Leopold, M. C.; Hicks, J. F.; Kulesza, P. J.; Malik, M.
A.; Murray, R. W. J. Am. Chem. Soc. 2002, 124, 8958-8964.
(55) Zamborini, F. P.; Smart, L. E.; Leopold, M. C.; Murray, R. W. Anal. Chim.
Acta 2003, 496, 3-16.
(
56) Ibanez, F. J.; Growrishetty, U.; Crain, M. M.; Walsh, K. M.; Zamborini,
F. P. Anal. Chem. 2006, 78, 753-761.
J. AM. CHEM. SOC.
9
VOL. 130, NO. 2, 2008 625

A R T I C L E S
Iba n˜ ez and Zamborini
Figure 3. CA plots of films of (A) C6S Pd MPCs, (B) C8NH2 Pd MPCs, (C) C12NH2 Pd MPCs, and (D) TOABr Pd nanoparticles measured at -0.3 V
during repeated exposure to 100% H2 (b) and air (4) during film conditioning to reach stable currents. (Insets) Expanded plots of the regions indicated by
the dashed rectangles. The arrows after the * in (C) represent exposure to 9.6% H2.
6
C8NH2, C12NH2, and C16-18NH2, respectively, showing a
decrease in conductivity with increasing chainlength. The current
displayed by the film of C16-18NH2 Pd MPCs is again similar
to the background with no MPCs deposited. The current
displayed by the film of TOABr-coated Pd nanoparticles is not
10- A during the first 300 s of exposure (see inset), continued
to rise slightly when in 100% air, and then increased to 5.0 ×
-
4
10 A upon the second H2 exposure. This trend continued until
-
3
the film reached a final current of 1.2 × 10 A, which is an
increase of about 6 orders of magnitude as compared to the
factor of 1.4 increase for the film of C6S Pd MPCs. This shows
that the more weakly coordinated octylamines do not hinder
the Pd from reacting with H2. The irreversible increase in current
is likely due to restructuring of the Pd film during repeated
exposure and removal from 100% H2, which is well known for
-
8
-10
constant; it decreases from 3.5 × 10 to 2.0 × 10
A over
1
000 s, which is consistent with a diffusion process and ionic
conductivity. The final conductivity value is above the baseline,
which could be due to steady-state ionic conductivity or some
electron hopping through the film. In summary, the data in
Figure 2 show that films of alkylamine-coated Pd MPCs conduct
by an electron hopping process and those of TOABr-coated Pd
nanoparticles conduct mainly by ions.
Reactivity of Films of Alkylamine- and TOABr-Coated
Pd Nanoparticles to 100% H2. Figure 3 shows CA plots of
C6S, C8NH2, C12NH2, and TOABr Pd nanoparticles at a
voltage of -0.3 V where 100% H2 or 100% air (or N2 for C6S)
is flowing over the device at various times as indicated by H2
1
Pd materials.
Figure 3C shows that the current through a film of C12NH2
-
10
-10
Pd MPCs increased from about 1 × 10 to about 3.5 × 10
A and reversibly returned close to the baseline upon the first
400 s and second 300 s (see inset) exposure to H2. This is similar
to that for films of C6S Pd MPCs, but the behavior is reversible.
The * in the plot indicates that sensing was performed at this
time, but the baseline was unstable and constantly increasing
(vide infra). At a later time after the * (∼1150 s), the current
“
on” (b) and H2 “off” (4), respectively. Films of C16-18NH2
-5
Pd MPCs had very little conductivity and did not change in the
presence of 100% H2 for up to 2000 s and are not shown in
Figure 3. The current is stable for C6S, C8NH2, and C12NH2
Pd MPCs in 100% air initially as also shown in Figure 2. In
the presence of 100% H2, Figure 3A shows that the current
passing through a film of C6S Pd MPCs increased irreversibly
increased to about 2 × 10 A and then finally to a stable value
-4
of about 1.2 × 10 A (the arrows in the plot after the *
correspond to 9.6% H2 exposure). The films of C12NH2 Pd
MPCs are overall similar to those of C8NH2 Pd MPCs in that
they irreversibly increase in current to a value about 6 orders
of magnitude higher than the initial current, likely due to film
restructuring in both cases, but the kinetics are different for the
two films. Films of C8NH2 Pd MPCs restructure and stabilize
in a more continuous 500 s step in H2, whereas films of C12NH2
Pd MPCs initially show a somewhat stable, reversible response
to H2 that changes to a large, irreversible increase in current at
times greater than 700 s in H2. The longer time required to reach
the higher current is likely due to the longer alkyl chains slowing
down the reaction and film restructuring process. The final film
of C12NH2 Pd MPCs has a higher resistance compared to that
of C8NH2 Pd MPCs.
-
8
-8
from about 5.0 × 10 to 7.0 × 10 A over a 1000 s period
(
see inset for expanded plot). This reflects little reactivity with
H2, which we previously determined was likely due to poisoning
of the reaction between Pd and H2 by the strongly coordinated
thiolates surrounding the Pd nanoparticles.45 Ozone- and heat-
induced removal of C6S led to much greater reactivity with H2
and reversible sensing characteristics.45
In contrast to C6S Pd MPCs, the conductivity of films of
C8NH2 and C12NH2 Pd MPCs changes much more appreciably
in the presence of 100% H2 without any ozone or heat treatment.
Figure 3B shows that the current through the film of C8NH2
Figure 3D shows the current passing through a film of
TOABr-coated Pd nanoparticles upon exposure to 100% H2 and
-
9
Pd MPCs increased irreversibly from about 1 × 10 to 5 ×
626 J. AM. CHEM. SOC.
9
VOL. 130, NO. 2, 2008

Nanoparticles for Sensing and Catalysis Applications
A R T I C L E S
Figure 4 shows two different types of sensing behavior: those
that decrease in conductivity and those that increase in
conductivity in the presence of H2. The direction depends on
14,45
the baseline conductivity.
Films of C8NH2 Pd and C12NH2
Pd (after second conditioning) have baseline currents on the
-3
-4
order of 10 and 10 A and decrease in current in the presence
of H2, which is consistent with what has been observed
previously for well-connected, low-resistance Pd materials.1
This is due to the increased resistance of the PdHx compared to
,26
29
pure Pd. Films of C12NH2 Pd MPCs after the first conditioning
-
8
have a relatively smaller baseline current on the order of 10
A, and the current increases in the presence of H2 at concentra-
tions greater than 1.0%. The Pd nanoparticles in these films
are not well connected and the expansion in volume of the Pd
when forming PdHx leads to an overall decrease in resistance
due to closer spacing or formation of connections between
Figure 4. CA plots of films of (A) C8NH2 Pd MPCs, (B) C12NH2 Pd
MPCs (intermediate current), (C) C12NH2 Pd MPCs (high current), and
(
D) TOABr Pd nanoparticles exposed to H2 concentrations from 9.6 to
14
particles. We attribute the current decrease in the presence of
0
1
.08% as indicated in N2 carrier gas. The films were initially exposed to
00% N2, and the circles represent the point of exposure to the H2
H2 at concentrations of 0.50% and lower for these films to a
temporary increase in resistance upon small amounts of H
concentration indicated and open triangles represent the point of exposure
back to 100% N2.
30
diffusing into R-phase Pd at low H2 concentration. Eventually,
the resistance of the film should decrease once the expanded
R-â or â phase Pd forms. We believe that the slow kinetics of
this process at low H2 concentration leads to the reduction in
current, which would have eventually increased if exposed for
longer time. Even at higher concentrations (g1.0%), an initial
drop in current occurs first, but the faster phase change kinetics
eventually leads to increased current on the time scale of the
air. These films exhibited stable, reversible increases in current
in the presence of 100% H2 immediately and did not require
any conditioning (see inset). Table S.1 of the Supporting
Information summarizes the results of the conditioning during
initial exposure to 100% H2 obtained on three samples of each
type of film of Pd nanoparticles.
H2 Sensing with Films of Alkylamine- and TOABr-Coated
Nanoparticles. Figure 4 shows chronoamperometry (CA) plots
for selected films that were exposed to H2 concentrations from
30
exposure. The increase in current for films of TOABr-coated
Pd nanoparticles is not well understood. The films conduct
ionically, but there may be some electron hopping as well. The
expanded volume of PdHx could affect both types of conductiv-
ity. Increased thermal motion upon H2 adsorption into the films
could also increase the current similar to that observed for films
9
.6% down to 0.08% H2. The films are initially in 100% N2
and then exposed to various concentrations of H2 (indicated by
circles) and then back to 100% N2 (indicated by triangles). The
sensing experiments in Figure 4 occurred after conditioning the
films until they were stable as discussed in Figure 3, except for
those of TOABr-coated Pd nanoparticles, which did not require
conditioning. Figure 4A shows that the film of C8NH2 Pd MPCs
exhibited a very sharp and reversible decrease in current in the
presence of 9.6-1.0% H2. The decrease in current is less
pronounced below 1.0% H2 but is noticeable for concentrations
as low a 0.08% H2 for a 100 s exposure. Figure 4B and C shows
the chemiresistive response to H2 for a film of C12NH2 Pd
MPCs after the first and second conditioning, respectively, as
discussed in Figure 3. After the first conditioning (see the * in
57
of C6S-coated Au MPCs in the presence of CO2 gas. More
experiments are needed to better understand the mechanism.
We also observed an increase in conductivity for TOABr-coated
Au nanoparticles in the presence of H2, but the Pd metal plays
an important role since films of TOABr-coated Pd were much
more sensitive, stable, and reversible.
Reactivity of Alkylamine- and TOABr-Coated PdAg and
PdAu Nanoparticles to 100% H2. We next studied the H2
response of films of PdAg and PdAu alloy nanoparticles. These
alloys are significant because Ag is cheaper than Pd, which
could reduce the cost of sensors or catalysts based on these
materials. From a chemical standpoint, PdAg has been shown
to be more sensitive since Ag can adsorb more H into its lattice
and PdAu alloys have shown faster response times due to Au
hindering the R-â phase transition during H diffusion into
-
8
Figure 3C), the current through the film is on the order of 10
A and the current increases in the presence of H2 down to 1.0%,
but curiously decreases for lower H2 concentrations down to
0
.11%. The response for this film is not totally reversible as
indicated by the constantly shifting baseline current to higher
values. Figure 4C shows the current response of the same film
after it reached a higher, more stable baseline current on the
9-11
Pd.
CVs of films of the alloy nanoparticles were very similar
to those of the pure Pd and are not shown.
Figure 5A shows the change in current for a film of C8NH2
PdAg when first exposed to 100% H2. The current increased
-
4
order of 10 A. The response completely changed, where the
current decreased in the presence of H2 down to concentrations
of 0.11%, similar to films of C8NH2 Pd MPCs. Figure 4D shows
the change in current of a selected film of TOABr-coated Pd
nanoparticles in the presence of H2. The film exhibited a stable,
reversible increase in current for concentrations ranging from
-
10
-8
from about 1.0 × 10
to 1.2 × 10 A, which is ∼2 orders
5
8
of magnitude. Interestingly, this is significantly smaller than
the 6 order of magnitude increase observed for C8NH2 Pd films
in Figure 3. The presence of less than 10% Ag in the alloy led
to much higher resistance in the film following restructuring in
9
.6% down to 0.11% without any pretreatment or film condi-
100% H2. The reactivity and restructuring of these nanoparticle
tioning. There is a slight drift in the baseline over time due to
a decrease in the baseline ionic conductivity with time as shown
in Figure 2.
(
57) Choi, J.-P.; Coble, M. M.; Branham, M. R.; DeSimone, J. M.; Murray, R.
W. J. Phys. Chem. C 2007, 111, 3778-3785.
J. AM. CHEM. SOC.
9
VOL. 130, NO. 2, 2008 627

A R T I C L E S
Iba n˜ ez and Zamborini
Figure 5. CA plots of films of (A) C8NH2 PdAg and (B) C12NH2 PdAu
measured at -0.3 V during repeated exposure to 100% H2 (b) and air (4)
during film conditioning to reach stable currents. (Insets) Expanded plots
of the regions indicated by the dashed rectangles.
films are clearly very sensitive to metal composition. Similarly
prepared films of C12NH2 PdAg films exhibited very low
baseline currents and did not change significantly in the presence
of 100% H2 (data not shown). Figure 5B shows the change in
current for a film of C12NH2 PdAu MPCs in the presence of
Figure 6. CA plots of films of (A) C8NH2 PdAg MPCs, (B) C12NH2
PdAu MPCs, and (C) TOABr PdAg nanoparticles exposed to H2 concentra-
tions from 9.6 to 0.08% as indicated in N2 carrier gas. The films were
initially exposed to 100% N2, and the circles represent the point of exposure
to the H2 concentration indicated and open triangles represent the point of
exposure back to 100% N2.
1
00% H2. In contrast to films of C12NH2 PdAg, the C12NH2
-7
PdAu film exhibited a rapid jump in current to 10 A after
about 300 s total in H2 (first two exposures) and eventually
-
3
increased to about 1.0 × 10 A, similar to films of C12NH2
Pd MPCs. One major difference between the PdAu alloy and
pure Pd, though, is that the current constantly increased in air
and N2 upon only two exposures to 100% H2, whereas pure Pd
showed some initial reversibility to H2 during the first two
exposures. As with TOABr Pd, films of TOABr PdAg nano-
particles exhibited stable, reversible increases in current im-
mediately in the presence of H2. Table S.2 (Supporting
Information) summarizes the behavior of three different samples
of Pd alloy films that were conditioned with 100% H2 and air.
H2 Sensing with Films of Pd Alloy Nanoparticles. Figure
3
(
10- A) that decreased in the presence of H2. The addition of
less than 10% Ag altered the film restructuring process and H2
sensing mechanism. Figure 6B shows the change in current for
a film of C12NH2 PdAu MPCs during H2 exposure. The baseline
-
5
current is on the order of 10 A and the current decreased in
the presence of 9.6 to 0.08% H2, similar to films of C8NH2 Pd
and C12NH2 Pd after the second conditioning, but very different
from films of C12NH2 PdAg MPCs, which did not respond to
H2. Figure 6C shows the response for a film of TOABr PdAg
nanoparticles. The current increased in the presence of 9.6 to
0.11% H2 similar to films of TOABr Pd (Figure 4C). The
baseline currents and response direction are not as sensitive to
metal composition as the alkylamine-coated MPCs for the metal
ratios studied. The explanation for the different types of sensing
(increase or decrease in current) is the same as that described
for the films of pure Pd nanoparticles.
6
shows the chemiresistive response to H2 for selected alloy
films following conditioning to stable currents as discussed in
-
8
Figure 5. Having a low initial conductivity (∼10 A), the
current for films of C8NH2 PdAg MPCs increased in the
presence of H2 concentrations from 9.6 to 0.08% (Figure 6A).
This is different from films of C8NH2 Pd MPCs (Figure 4A),
which have a larger stable baseline current after conditioning
(
2 2
58) We believe the lower initial current for C8NH PdAg compared to C8NH
Surface FTIR Spectroscopy Characterization. We used
FTIR spectroscopy to probe the organic portion of the films
before and after exposure to H2 to gain insight into the film
restructuring process and sensing mechanism. Figure 7A-D
shows surface FTIR spectra for films of C8NH2 Pd, C8NH2
Pd is caused by a smaller amount of MPCs drop-cast deposited onto the
electrode surface in this case (see Experimental Section) and irregular film
2
deposition during drying. When preparing films of C8NH PdAg from a
more concentrated (70 mg/mL) solution, the film displayed much higher
-7
current (∼1.0 × 10 A) in line with the shorter alkyl chain and Ag in the
alloy. Note that the differences in film thickness and initial conductivity
did not change the conditioning and sensing behavior discussed next.
628 J. AM. CHEM. SOC.
9
VOL. 130, NO. 2, 2008

Nanoparticles for Sensing and Catalysis Applications
A R T I C L E S
Figure 7. Surface FTIR spectra of films of (A) C8NH2 Pd, (B) C8NH2 PdAg, (C) C12NH2 Pd, and (D) C12NH2 PdAu MPCs obtained before and after
exposure to 100% H2 until they reached stable currents. Films were exposed to 100% H2 for a total time of 500, 400, 800, and 300 s, for A-D, respectively.
The number displayed next to each spectrum is the current (in amps) passing through similarly prepared films deposited onto electrodes and exposed to H2
in parallel with the FTIR samples. The films of C12NH2 Pd MPCs shows two FTIR spectra after exposure to H2 that correspond to conditioning to intermediate
current and high current.
PdAg, C12NH2 Pd, and C12NH2 PdAu, respectively, before59
and after exposure to 100% H2 that were placed in the same
chamber as electrodes containing the same films for conductivity
measurements. Before exposure, all of the spectra show absor-
bance peaks similar to those previously reported for alkylamine
between 2000 and 2250 cm , whereas dinitriles adsorbed to
-1
-1
63,69
Cu surfaces showed stretches in the 2080-2100 cm range.
The CtN stretch for isocyanides (R-NtC) adsorbed to Pd
3
6
67,70,71
-1
nanoparticles or Pd films
is 2150-2170 cm , similar
to the peak we observed, but the spectra for nanoparticles also
displayed peaks at 2100-1700 and 1700-1500 cm , assigned
to a doubly bridged, and triply bridged isocyanide ligand on
Pd, respectively. It is difficult to envision how we could
produce an isocyanide from an amine in our case. We considered
the formation of R-CHdNH, but this would be expected to
have a lower CdN stretch around 1600-1700 cm . Our data
is most consistent with the formation of an alkyl nitrile, which
6
0
49
-1
monolayers on Pd and Au. For example, the spectrum for
the film of C8NH2 Pd MPCs before H2 displays peaks at 3280,
-
1
36
1
604, and 1000 cm , which corresponds to the N-H stretch,
4
9
N-H bend, and C-N stretch, respectively, as indicated. The
peaks at 3000-2850 cm are due to the asymmetric and
symmetric CH3 and CH2 stretches, and the peak at 1462 cm
-
1
-
1
-1 72
is due to a CH2 scissor mode. In addition, we observed a sharp
-
1
peak at 2167 cm , which is in the region of a CtN or CtC
stretching vibration. We believe this may be due to a CtN
stretch that arises from converting the R-CH2-NH2 to a R-Ct
N by oxidation of amines catalyzed by Pd during the synthesis
as reported previously for Cu(II)61 and Pd(II) complexes. The
CtN stretch for free alkyl nitriles (gas or liquid) is generally
has a lower stretching frequency relative to the free CtN due
to π back-bonding with the Pd metal surface.
do further work to determine this peak assignment conclusively.
63,69
We need to
Figure 7A shows the FTIR spectrum of a film of C8NH2 Pd
MPCs before and after exposure to 100% H2 for 500 s
62
-9
-3
corresponding to currents of 3.2 × 10 and 1.4 × 10 A,
respectively. There are four main changes in the spectrum
following exposure to H2 and the increase in current. First, the
intensity of the CH2 and CH3 stretches decreased significantly.
In particular, the peak height of the asymmetric CH2 stretch
decreased by 80%. Figure 3 earlier showed that the film
restructures and increases in conductivity upon the first exposure
to 100% H2. The IR data show that the alkylamines are removed
from the film during this process. The second change is the
-
1
around 2230-2270 cm and sensitive to the coordination
environment.6
3-66
Acetonitrile65,67 and benzonitrile64,65,68 ad-
sorbed to various metal surfaces exhibited CtN stretches
(
(
(
(
(
(
(
59) We note that all of the spectra of these films before any H
2
exposure show
very little or no changes after 4 days in ambient air.
60) Rao, C. R. K.; Lakshminarayanan, V.; Trivedi, D. C. Mater. Lett. 2006,
6
0, 3169.
61) Maeda, Y.; Nishimura, T.; Uemura, S. Bull. Chem. Soc. Jpn. 2003, 76,
2
399-2403.
62) Wang, J.-R.; Fu, Y.; Zhang, B.-B.; Cui, X.; Liu, L.; Guo, Q.-X. Tetrahedron
Lett. 2006, 47, 8293-8297.
-1
loss of the peak at 2167 cm . If this peak is CtN, then it is
63) Loo, B. H.; Lee, Y. G.; Frazier, D. O. J. Phys. Chem. 1985, 89, 4672-
4
676.
well known that nitriles can be reduced to amines in the presence
64) Mrozek, M. F.; Wasileski, S. A.; Weaver, M. J. J. Am. Chem. Soc. 2001,
23, 12817-12825.
65) Solomun, T.; Christmann, K.; Baumgartel, H. J. Phys. Chem. 1989, 93,
1
(69) Loo, B. H.; Lee, Y. G.; Frazier, D. O. Chem. Phys. Lett. 1985, 119, 312-
316.
(70) Murphy, K. L.; Tysoe, W. T.; Bennet, D. W. Langmuir 2004, 20, 1732-
7
199-7208.
(
66) Storhoff, B. N.; Lewis, H. C. J. Coord. Chem. ReV. 1977, 23, 1-29.
(
67) Murphy, K.; Azad, S.; Bennett, D. W.; Tysoe, W. T. Surf. Sci. 2000, 467,
1738.
1
-9.
(71) Stapleton, J. J.; Daniel, T. A.; Uppili, S.; Cabarcos, O. M.; Naciari, J.;
Shashidhar, R.; Allara, D. L. Langmuir 2005, 21, 11061-11107.
(72) Ortiz-Hernandez, I.; Williams, C. T. Langmuir 2007, 23, 3172-3178.
(68) Oranskaya, O. M.; Semenskaya, I. V.; Filimonov, V. N. React. Kinet. Catal.
Lett. 1976, 5, 135-139.
J. AM. CHEM. SOC.
9
VOL. 130, NO. 2, 2008 629

A R T I C L E S
Iba n˜ ez and Zamborini
of Pd and H2.73 The third and fourth change are a sharpening
similar to that of C8NH2 Pd films. The asymmetric CH2
stretched decreased by ∼62% and the initially broad peak for
the NH2 stretch split into three sharper peaks.
and slight shift to higher wavenumber for the N-H bend around
-
1
1
3
600 cm
and a splitting of the broad NH2 stretch at
-
1
280 cm into 2 (or 3) sharper peaks. These latter differences
Taken together with the current data, Figure 7 shows a strong
correlation between monolayer desorption and change in film
conductivity during the first exposure to 100% H2. Films of
C8NH2 Pd MPCs, C12NH2 PdAu MPCs, and C12NH2 Pd
indicate a change occurs in the coordination environment
between the R-NH2 and Pd upon exposure to H2. This might
be expected considering many of the alkylamines desorbed from
the surface. This could cause orientation changes, differences
in hydrogen bonding, or other interactions for the remaining
alkylamines. The peak at higher wavenumbers is consistent with
(second conditioning) increased in conductivity by 5-6 orders
of magnitude after a 500, 300, and >800 s exposure, respec-
tively, and showed a correspondingly large loss of the alky-
lamine monolayer in the IR spectra (80, 62, and 36%,
respectively). In addition, the nitriles were reduced to amines
(for pure Pd) and the coordination environment of the NH2
groups was altered. This shows that monolayer desorption and
coordination changes accompany film restructuring and con-
ductivity increases. Films of C8NH PdAg and C12NH Pd
7
4
non-hydrogen-bonded amines as observed in dilute solutions.
Figure 7B shows the FTIR spectrum of a film of C8NH2
PdAg MPCs before and after exposure to 100% H2 for 400 s
-
10
-9
corresponding to currents of 1.2 × 10
and 7.0 × 10 A,
respectively. The spectrum has some small differences compared
to the film of C8NH2 Pd MPCs before exposure to H2. First,
2
2
-
1
the peak attributed to nitriles at 2167 cm is much smaller in
intensity and there are two sharp peaks in the NH2 stretching
region as opposed to one broad peak. Adding less than 10%
Ag to the reaction clearly reduced the formation of nitriles and
changed the coordination environment of the amines. This
suggests that many of the Ag atoms reside at the surface of the
nanoparticles.7 After exposure to H2, the spectrum is drastically
different compared to pure Pd. Most notably, the intensity of
the CH2 and CH3 stretches barely decreased (only 4% for the
asymmetric CH2 stretch) and there is a new peak that appears
near 1700 cm . Similar to pure Pd, the nitrile peak disappeared
and a better defined third peak appeared at higher wavenumbers
in the NH2 stretching region. This peak is lowest in intensity
out of the three peaks for the film of PdAg MPCs, but is largest
in intensity for the Pd MPCs.
MPCs (first conditioning) showed a much smaller 1-2 order
of magnitude conductivity change after 400 and 800 s,
respectively, and a correspondingly much smaller loss of
alkylamines from the surface (4 and 12%, respectively). Small
changes in the spectrum include loss of the nitrile peak and
changes in the NH2 stretch. The IR and conductivity data show
that for pure Pd, the longer C12NH2 monolayer enhances
stability and reduces the reactivity to 100% H2. The presence
of less than 10% Ag enhanced stability of the monolayer and
reduced the amount of film restructuring for C8NH2-coated
MPCs compared to pure Pd whereas the presence of less than
5
-1
10% Au reduced monolayer stability and increased the restruc-
turing for C12NH2-coated MPCs compared to pure Pd. This
may be due to stronger bonding between alkylamines and PdAg
as compared to alkylamines and PdAu.
Figure 7C shows the FTIR spectrum of a film of C12NH2
Pd MPCs before H2, after the first conditioning (800 s in 100%
H2), and after the second conditioning until the film reached
Atomic Force and Optical Microscopy Characterization.
We used optical microscopy and AFM to directly image various
films before and after exposure to 100% H2 to determine if
morphological changes occur during the film conditioning/
restructuring process. Figure 8A shows an optical image and
Figure 8B shows the corresponding AFM image in the same
area as indicated by the dashed box of a film of C8NH2 Pd
MPCs deposited on an electrode device before exposure to H2.
Figure 8C and D shows the optical and corresponding AFM
image, respectively, of the same area after exposure to 100%
H2 for 500 s, compressed air for 200 s, and ambient air for 5
min during the AFM setup. The optical image shows a change
in the morphology with the appearance of several black features
in the image. The corresponding AFM image shows that these
are ∼100 nm tall raised features (or islands) that form on the
surface. The arrows in Figure 8D show the same area on the
surface as in Figure 8B. Morphological changes upon incorpora-
tion of H into Pd and subsequent removal are well known to
-
10
stability. The corresponding currents are 3.0 × 10 , 1.0 ×
-
9
-4
1
0 , and 6.2 × 10 A, respectively. The spectrum before H2
exposure is very similar to that of the film of C8NH2 Pd MPCs,
where the NH2 stretch is broad and there is a noticeable
absorbance band attributed to nitriles. After the first condition-
ing, there is a small 12% decrease in the asymmetric CH2 stretch
and after the second, there is a 36% loss of this peak. The NH2
peaks remained fairly broad with the presence of a small broad
peak appearing around 3313 cm , instead of splitting into three
sharp peaks as observed for films of C8NH2 Pd. Another
similarity to films of C8NH2 Pd MPCs is that the nitrile peak
disappeared after the first conditioning.
-1
Figure 7D shows the FTIR spectrum of a film of C12NH2
PdAu MPCs before and after conditioning with 100% H2 for
-
10
-4
3
00 s, corresponding to currents of 1.8 × 10 and 6.0 × 10
A, respectively. The spectrum before H2 is similar to the film
of C12NH2 Pd MPCs, except that there is no evident peak for
the nitrile at 2167 cm . As with Ag, this shows that the
presence of less than 10% Au affected the formation of nitriles.
There are also small differences in the NH2 stretching and
bending region. After exposure to H2, the spectrum is very
1
occur. The morphological changes observed by microscopy are
consistent with the 5-6 order of magnitude conductivity
increase and 80% loss of C8NH2 ligands from the MPCs
observed from CA plots and surface FTIR data, respectively.
In addition, powder X-ray diffraction (XRD) data on C8NH2
Pd MPCs show that the domain size of the particles increases
from 2.4 to 12.8 nm after 100% H2 exposure as determined by
the peak width of the Pd(111) peak and the Scherrer equation
(Figure S.1, Supporting Information). This is consistent with
the morphological changes observed with AFM and optical
images. Figure 8E and F show optical and AFM images,
-
1
(
73) Sachtler, W. M. H.; Huang, Y. Appl. Catal. A 1999, 182, 365-378.
74) Pavia, D. L.; Lampman, G. M.; Kriz, G. S. Introduction to Spectroscopy;
Harcout Brace College: New York, 1996.
(
(75) Hostetler, M. J.; Zhong, C.-J.; Yen, B. K. H.; J, A.; Gross, S. M.; Evans,
N. D.; Porter, M.; Murray, R. W. J. Am. Chem. Soc. 1998, 120, 9396-
9
397.
630 J. AM. CHEM. SOC.
9
VOL. 130, NO. 2, 2008

Nanoparticles for Sensing and Catalysis Applications
A R T I C L E S
where ib is the initial baseline current in 100% N2, ir is the
current in the presence of H2/N2 mixture, and ∆i ) (ir - ib). A
negative value is equal to a decrease in the current upon
exposure to H2 and vice-versa. Figure 9 shows the average
calibration curves plotting the % response (y-axis) versus the
H2 concentration (x-axis) for the six tested films. The points
and curves represent the average of three samples. Figure 9A
and C shows the response of films whose conductivity increased
in the presence of H2 from 0.0 to 0.50% and from 0.0 to 9.6%
H2, respectively. Figure 9B and D shows the response of films
whose conductivity decreased in the presence of H2 from 0.0
to 0.50% and from 0.0 to 9.6% H2, respectively. All the films
displayed fairly linear behavior below 0.50% H2 (not forced
through the origin). The behavior from 0.0 to 9.6% H2 depended
upon the stabilizer surrounding the particles. Films of alky-
lamine-coated MPCs exhibited nonlinear behavior above 0.50%
due to the R-â phase transition, which occurs anywhere from
7
1,2
0
.3 to 2.0% H2, and saturation of H in Pd at higher
concentrations. Films of TOABr-stabilized nanoparticles were
also nonlinear but did not reach saturation up to 9.6%. The
slopes in Figure 9A reflect the following sensitivity order for
films that increase in current: TOABr Pd (22.4) > TOABr PdAg
(
18.8) > C8NH2 PdAg (6.6). The slopes in Figure 9B reflect
Figure 8. Optical (A, C, E) and AFM (B, D, F) images of a film of C8NH2
Pd MPCs before (A, B) and after (C, D) conditioning by exposure to 100%
H2 for 500 s and air for 200 s. (E, F) shows the same film after an additional
the following sensitivity order for films that decrease in
current: C12NH2 Pd (at intermediate current) (-10.5) >
C12NH2 PdAu (-4.9) > C8NH2 Pd (-2.9) > C12NH2 Pd (at
high current) (-0.8). All of the films easily detect H2 below
the explosive limit. Films of C8NH2 Pd and PdAg MPCs
exhibited the lowest detection limit (0.08%), whereas films of
C12NH2 PdAu and C8NH2 PdAg displayed the fastest response
times of 10-15 and 10-20 s, respectively. Others have shown
that PdAu and PdAg alloy films show faster response times
1
h exposure to air. Circles labeled #1 represent areas were the film healed,
and circles labeled #2 represent areas where the films changed irreversibly
after exposure to 1 h in air.
respectively, of the same area of the same surface after another
1
h of exposure to ambient air. Several of the black features in
the optical image and bright features in the AFM image
disappeared, showing that there are slow morphological changes
that occur over longer periods of time after exposure to H2.
Many of the regions appear to heal themselves as indicated by
the circle marked #1, but several regions were also altered
irreversibly as indicated by the circle marked #2 (Figure 8C
and E).
9
-11
and greater sensitivity, respectively.
This is consistent with
the fast conditioning and response of C12NH2 PdAu MPC films.
However, it is important to note that the film thickness (or
nanoparticle loading) is not constant for these films, which could
also affect the response and recovery times due to different
diffusion lengths for H into the film. Films of PdAu alloys
To directly correlate morphology changes with changes in
current and the FTIR data, we drop-cast deposited films of
C8NH2 Pd, C12NH2 Pd, C8NH2 PdAg, and C12NH2 PdAu
MPCs onto electrodes and obtained optical and AFM images
along with current data before and after conditioning with 100%
H2. Figure S.2 of the Supporting Information shows the results.
In general, films of C8NH2 Pd and C12NH2 PdAu MPCs
showed the largest morphology changes, which is consistent
with the 5-6 order increase in current and large loss of
alkylamine monolayer. Films of C12NH2 Pd and C8NH2 PdAg
showed very small morphology changes, consistent with the
smaller loss of monolayer in the FTIR and smaller increase in
current for the C8NH2 PdAg MPC film. Films of C12NH2 Pd
exhibit a large 5-6 order of magnitude increase in current, but
the slower kinetics of this process may lead to smaller
morphological changes in the film. Powder XRD data obtained
on these films are also consistent with the conductivity, FTIR,
and microscopy data (Figure S.1, Supporting Information).
Sensor Comparisons. The analytical signal used for the
detection of H2 is % response as described by the following
equation:
1
1
described in the literature were less sensitive to H2; however,
our films of C12NH2 PdAu MPCs were more sensitive
compared to those of C12NH2 Pd MPCs at high current. We
were unable to directly compare films of C8NH2 Pd with C8NH2
PdAg since they have different sensing mechanisms. The
response of films of TOABr PdAg were similar to those of
TOABr Pd at 0.50% and below but 2-3 times higher at 1.0%
and above. Table 1 displays the response direction, limit of
detection, % response measured at 1% H2, and range of response
times for 3 different samples of each type of the 6 films studied.
Table S.3 of the Supporting Information shows all of the data
for each sample measured, including averages and standard
deviations.
Conclusions
We prepared and studied the H2 reactivity of drop-cast films
of chemically synthesized alkylamine-coated Pd, PdAg (10:1)
and PdAu (10:1) MPCs and films of tetraoctylammonium
bromide (TOABr)-stabilized Pd and PdAg (10:1) nanoparticles.
The as-prepared films are highly reactive to H2 and do not
require O3 or thermal treatment as previously observed for films
of C6S-coated Pd MPCs. The films can be placed into two
categories: those that decrease in current in the presence of H2
%
response ) (i - i )/i × 100% ) ∆i/i × 100%
r
b
b
b
J. AM. CHEM. SOC.
9
VOL. 130, NO. 2, 2008 631

A R T I C L E S
Iba n˜ ez and Zamborini
Figure 9. Calibration curves showing the percent response versus H2 concentration from 0.00 to 0.50% (A, B) and 0.0 to 9.6% (C, D) for each sensor
described in this study. The sensors are divided into those that increase in current in the presence of H2 with positive slopes (A, C) and those that decrease
in the presence of H2 with negative slopes (B, D). The points on the curves represent the average percent response from three sensor devices at each H2
concentration. The linear regression fits in (A, B) are not forced through the origin and the lines in (C, D) are present as a guide to the eye.
Table 1. Sensing Characteristics of the Six Tested Films
type of sensor response
direction
)
increase
response time
a
in current
limit of detection (% H
S#2
2
)
percent response at 1.0% H
S#2
2
2
at 1.0% H (s)
S#1
S#3
S#1
S#3
TOABr Pd
TOABr PdAg
C8NH2 PdAg
C12NH2 Pd (medium current)
0.21
0.11
0.11
0.11
0.21
0.11
0.08
0.11
0.21
0.11
0.11
0.11
13.0
32.5
6.9
9.2
28.9
9.6
13.9
27.1
10.5
67.7
40-55
60-70
10-20
50-90
69.2
44.5
type of sensor response
direction
)
decrease
response time
a
in current
limit of detection (% H
2
)
percent response at 1.0% H
S#2
2
2
at 1.0% H (s)
S#1
S#2
S#3
S#1
S#3
C8NH2 Pd
C12NH2 Pd (high current)
C12NH2 PdAu
0.08
0.11
0.08
0.08
0.11
0.08
0.08
0.11
0.08
-8.8
-1.8
-7.0
-10.0
-1.9
-11.6
-4.3
-3.2
-14.7
20-30
10-20
10-15
a
Note that the film thickness (or nanoparticle loading) is not constant for these films. This could have an affect on the response time due to different
lengths for H diffusion into the films.
and those that increase. Films of C8NH2 Pd, C12NH2 Pd (second
conditioning), and C12NH2 PdAu exhibit large 5-6 order of
magnitude irreversible increases in current upon the first
exposure to 100% H2, which is accompanied by a large loss of
the alkylamine monolayer and noticeable morphological changes
in the case of C8NH2 and C12NH2 PdAu films. After this
irreversible conditioning/restructuring, the films exhibit stable,
reversible decreases in current in the presence of H2 concentra-
tions down to 0.08%. Films of C12NH2 Pd (first conditioning)
and C8NH2 PdAg exhibit a much smaller 1-2 order of
magnitude irreversible increase in current upon the first exposure
to 100% H2, which is accompanied by very little loss of the
alkylamine monolayer and correspondingly small morphological
changes in the film. Afterward, the films exhibit reversible
increases in current in the presence of H2 concentrations down
to 0.08%, which is stable only for the case of PdAg MPCs.
Films comprised of TOABr-coated Pd and PdAg exhibit
reversible, stable increases in current in the presence of H2
concentrations down to 0.11% without any pretreatment or
conditioning. Alkylamine-coated Pd MPCs change in current
through higher resistance of PdHx or lower resistance during
volume increases of PdHx; however, the sensing mechanism of
ionically conductive TOABr-coated Pd nanoparticles is less
understood. All of the films are easy to synthesize on a large
scale, and the devices are simple to construct, leading to
responses to H well below the explosive limit. For sensing,
2
films of TOABr-coated Pd and PdAg alloy nanoparticles are
the most promising of all the films because of their high
sensitivity, wider detection range, and no need for pretreatment
or conditioning. Out of the films of alkylamine-coated MPCs,
those of C8NH PdAg MPCs are the best because they exhibit
2
the highest sensitivity, lowest limit of detection, are the most
stable, have fast response, and require very little conditioning
in 100% H2 to reach stable, sensing behavior. They would also
be slightly lower in cost due to the incorporation of Ag into
the material, which is cheaper than Pd. C12NH2 PdAu films
were fairly sensitive and demonstrated the fastest response with
H2 during the conditioning process and sensing. Catalysis
applications for these materials need further exploration.
Acknowledgment. We gratefully acknowledge the National
Science Foundation (CHE-0518561) and the Kentucky Science
632 J. AM. CHEM. SOC.
9
VOL. 130, NO. 2, 2008

Nanoparticles for Sensing and Catalysis Applications
A R T I C L E S
and Engineering Foundation for partial support of this research.
We also acknowledge Andres J. Puppato from the Department
of Chemistry at the University of Louisville for providing
valuable help in the use of NMR. We thank Kevin M. Walsh,
Mark M. Crain, and Ana Sanchez from the Department of
Electrical and Computer Engineering at the University of
Louisville for use of the clean room to microfabricate the
electrode devices.
nanoparticles, powder X-ray Diffraction data, optical and AFM
images of the alkylamine-coated Pd and Pd alloy films before
and after conditioning with H2, and table showing the sensor
response data for each sample used in this study (including
average response, standard deviation, relative standard deviation,
and percent success for detecting a specific H2 concentration
for each film). This information is available free of charge via
the Internet at http://pubs.acs.org
Supporting Information Available: Tables showing the
conditioning data for films of pure Pd nanoparticles and Pd alloy
JA076000H
J. AM. CHEM. SOC.
9
VOL. 130, NO. 2, 2008 633