Amine grafted silica supported CrAuPd alloy nanoparticles: Superb heterogeneous catalysts for the room temperature dehydrogenation of formic acid

Article abstract of DOI:10.1039/c5cc02371h

Herein we show that a previously unappreciated combination of CrAuPd alloy nanoparticles and amine-grafted silica support facilitates the liberation of CO-free H₂ from dehydrogenation of formic acid with record activity in the absence of any additives at room temperature. Furthermore, their excellent catalytic stability makes them isolable and reusable heterogeneous catalysts in the formic acid dehydrogenation.

Full text of DOI:10.1039/c5cc02371h

ChemComm
View Article Online
View Journal
COMMUNICATION
Amine grafted silica supported CrAuPd alloy
nanoparticles: superb heterogeneous catalysts
for the room temperature dehydrogenation of
formic acid†
Cite this:DOI: 10.1039/c5cc02371h
Received 21st March 2015,
Accepted 3rd June 2015
Mehmet Yurderi,a Ahmet Bulut,a Nurdan Caner,a Metin Celebi,a Murat Kayab and
DOI: 10.1039/c5cc02371h
Mehmet Zahmakiran*a
www.rsc.org/chemcomm
Herein we show that a previously unappreciated combination of these,6 the difficulties met during their isolation-recovery
of CrAuPd alloy nanoparticles and amine-grafted silica support steps hinder their practical application for on-board systems.
facilitates the liberation of CO-free H2 from dehydrogenation of With this concern, the current research has been focused on the
formic acid with record activity in the absence of any additives at development of practical heterogeneous catalysts7,8 that exhibit
room temperature. Furthermore, their excellent catalytic stability significant activity under mild conditions with painless synthesis
makes them isolable and reusable heterogeneous catalysts in the and recovery routes. In spite of the tremendous labour, the majority
formic acid dehydrogenation.
of the reported heterogeneous catalysts for FA dehydrogenation
needs extra additives (e.g. HCOONa, LiBF4 etc.,) and elevated
Hydrogen (H2) is considered to be a promising energy carrier temperatures.7 To date, only a few heterogeneous catalysts8 have
for our future society as it is light-weight and has a high energy been found to provide notable activities in the additive-free FA
density (142 MJ kgÀ1), almost three times higher than that of dehydrogenation at low temperatures. In this regard, the develop-
natural gas (55 MJ kgÀ1).1 In addition to being light weight and ment of highly active, selective and reusable solid catalysts that
possessing high energy density, hydrogen is also an environ- operate at low temperatures for FA dehydrogenation in the
mentally friendly energy vector to end users when combined absence of additives is of great importance. Herein, we present
with proton exchange membrane fuel cells (PEMFC), since only a facile synthesis of CrAuPd alloy nanoparticles (NPs) supported
water and small amounts of heat are the by-products when it is on 3-aminopropyltriethoxysilane functionalized silica, hereafter
utilized in PEMFC.2 However, controlled storage and release of referred to as CrAuPd/N-SiO2, and their excellent catalysis in the
hydrogen are still technological barriers in the fuel cell based additive-free dehydrogenation of FA at room temperature.
hydrogen economy.1,2 In this context, formic acid (FA, HCOOH),
CrAuPd/N-SiO2 can reproducibly be prepared through a simple
which is one of the major stable and non-toxic liquid products impregnation route followed by subsequent sodium borohydride
formed in biomass processing, has attracted recent attention as a (NaBH4) reduction in water all at room temperature.9 After
suitable hydrogen carrier for fuel cells designed for portable use.3 In centrifugation, copious washing with water, CrAuPd/N-SiO2 was
the presence of metal catalysts, FA can catalytically be decomposed isolated as a gray powder and characterized by multi-pronged
via dehydrogenation (HCOOH - H2 + CO2) and dehydration techniques. The molar composition of the as-prepared catalyst
(HCOOH - H2O + CO) pathways.3 The selective dehydrogenation was found to be Cr0.15Au0.40Pd0.60 (0.21% wt Cr, 1.26% wt Au
of FA is indispensable for the production of ultrapure H2, since toxic and 1.65% wt Pd loadings correspond to 4.0 mmol Cr, 6.6 mmol
carbon monoxide (CO) produced by the dehydration pathway Au and 16.0 mmol Pd, 0.98 mmol NH2/g SiO2) by inductively
significantly reduces the activity of the Pt catalyst in PEMFC.4 coupled plasma-optical emission spectroscopy (ICP-OES) and the
Recently, serious efforts have been made in the development of ninhydrin method.10 The conventional transmission electron micro-
homogeneous catalysts for the selective dehydrogenation of FA.5 scopy (CTEM), high resolution-TEM (HRTEM), scanning TEM-
Even though, the notable activities have been reported by some energy dispersive X-ray spectroscopy (STEM-EDX) and high angle
annular dark field-scanning transmission electron microscopy
a Department of Chemistry, Science Faculty, Yuzuncu Yıl University, 65080, Van,
Turkey. E-mail: zmehmet@yyu.edu.tr; Web: www.nanomatcat.com
b Department of Chemical Engineering and Applied Chemistry, Atilim University,
06836, Ankara, Turkey
(HAADF-STEM) investigations were performed to examine the
size, morphology and the composition of the CrAuPd/N-SiO2
catalyst. CTEM images of CrAuPd/N-SiO2 given in Fig. 1(a) and
(b) reveal the presence of CrAuPd NPs. The mean particle size for
the images given in Fig. 1(a) and (b) was found to be ca. 2.6 nm
by the particle size analysis of 4100 non-touching particles
† Electronic supplementary information (ESI) available: Detailed synthesis protocols,
experimental procedures, activity results, TEM, FTIR, XPS, and GC characterization.
See DOI: 10.1039/c5cc02371h
This journal is ©The Royal Society of Chemistry 2015
Chem. Commun.

View Article Online
Communication
ChemComm
Fig. 1 (a) CTEM image and the corresponding size histogram, (b) CTEM and TEM-EDX spectrum, (c) HRTEM image of CrAuPd/N-SiO2, (d) P-XRD
patterns of CrAuPd/N-SiO2, Cr/N-SiO2, Au/N-SiO2, Pd/N-SiO2, (e) DR-UV-vis spectra of CrAuPd/N-SiO2 and Au/N-SiO2, (f) HAADF-STEM image and line
scan analysis spectrum of CrAuPd/N-SiO2.
(Fig. 1(a), inset). STEM-EDX analysis of a large number of different Pd (5.67 eV), Au (5.54 eV) and Cr (4.5 eV). This electron transfer in
domains on the CrAuPd/N-SiO2 surface revealed the presence of Cr, CrAuPd/N-SiO2 has the potential deign itself with the high activity in
Au, and Pd in the analyzed regions (Fig. 1(b), inset). The HRTEM FA dehydrogenation (vide infra) as the increase in the electron
image of CrAuPd/N-SiO2 given in Fig. 1(c) displays the highly density of catalytically active Pd and Au centers facilitates metal-
crystalline nature of the NPs on N-SiO2.
formate formation in the FA decomposition, which enhances the
The crystalline fringe distance was measured to be 0.21 nm, rate of the catalytic dehydrogenation of FA.12
which is different from the (111) spacing of the face-centered cubic
The catalytic activities of Cr0.15Au0.40Pd0.60/N-SiO2 together
(fcc) Au (0.235 nm),8e Pd (0.223 nm)8e and the (110) spacing of with its monometallic, bimetallic and trimetallic counterparts
Cr (0.263 nm).11 In addition to this, the powder X-ray diffraction in different molar compositions were investigated in the dehy-
(P-XRD) pattern of the as-synthesized catalyst (Fig. 1(d)) exhibits a drogenation of aqueous FA solution (0.20 M in 10.0 mL H2O) at
diffraction peak in the position between Pd(111) and Au(111), whose 298 K and their results are given in Fig. 2(a)–(c). Evidently, the
position also differs from Cr(110).11 The diffuse reflectance UV-vis Cr0.15Au0.40Pd0.60/N-SiO2 catalyst provides a better activity than
(DR-UV-Vis) spectrum taken for solid powders of CrAuPd/N-SiO2 mono and bimetallic catalysts prepared by the same method.
(Fig. 1(e)) shows almost no surface plasmon resonance band for Au From Fig. 2(a) and (b), it is clear that Pd is the crucial active
NPs, whereas AuN-SiO2 has surface plasmon resonance bands near metal in all catalysts; without Pd addition Cr1.0/N-SiO2, Au1.0
/
440 nm. This surface plasmon resonance quenching caused by N-SiO2, and Cr0.49Au0.51/N-SiO2 catalysts show almost no activity. On
alloying was also observed in PdAu8e and PdAg8f alloy NPs. Further- the other hand, the initial activity of monometallic Pd (Pd1.0/N-SiO2)
more, the formation of an alloy structure was also confirmed by cannot resume as the active sites are easily deactivated by poisonous
HAADF-STEM-line analysis (Fig. 1(f)). When the distribution of Cr, CO intermediate, which is one of the important obstacles for
Au and Pd in the randomly chosen particle was assessed by using the application of monometallic catalysts in FA dehydrogena-
the line scanning analysis in the STEM-EDX mode, we can clearly tion.3 Although, bimetallic Cr0.48Pd0.52/N-SiO2 and Au0.45Pd0.55
see the overlapping of Cr, Au and Pd signals that can be assignable N-SiO2 catalysts show better activities than Pd1.0/N-SiO2, their
to the formation of alloy structure in CrAuPd NPs. Additionally, X-ray performances are still far inferior to the trimetallic Cr0.15Au0.40Pd0.60
/
/
photoelectron spectroscopy (XPS) results reveal that the binding N-SiO2. The morphological investigation by CTEM analyses (Fig. S2,
energies for Pd 3d and Au 4f in CrAuPd/N-SiO2 are shifted to the ESI†) reveals that mono and bimetallic catalysts have similar
lower values with respect to those in Pd/N-SiO2 and Au/N-SiO2, shapes and sizes with respect to Cr0.15Au0.40Pd0.60/N-SiO2. The
respectively; whereas the Cr 2p binding energy is shifted to a higher enhanced catalytic activity of Cr0.15Au0.40Pd0.60/N-SiO2 may be
value compared with that of Cr/N-SiO2 (Fig. S1, ESI†). These shifts attributed to its special composition and surface electronic state
are indicative of transfer of some electrons from Cr to Pd and Au to in the formed alloy structure,13 which was further supported by
equalize the Fermi level owing to the difference of work functions of the result of a control experiment in which the physical mixture
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2015

View Article Online
ChemComm
Communication
Fig. 2 The volume of gas (CO2 + H2) (mL) versus time (min) graphs for the catalytic dehydrogenation of aqueous FA solution (0.20 M in 10.0 mL H2O)
catalyzed by (a) monometallic, (b) bimetallic, (c) trimetallic catalysts (in all [catalyst] = 2.67 mM), (d) Cr0.15Au0.25Pd0.60/N-SiO2 catalyst at different [CrAuPd]
concentrations at 298 K, (e) Cr0.15Au0.25Pd0.60/N-SiO2 catalyst at different temperatures, and (f) reusability and recyclability performances of the
Cr0.15Au0.25Pd0.60/N-SiO2 catalyst in the FA dehydrogenation (0.20 M in 10.0 mL H2O) at 298 K.
of Cr0.17/N-SiO2, Au0.23/N-SiO2 and Pd0.60/N-SiO2 exhibits lower indicates that Cr0.15Au0.40Pd0.60/N-SiO2 catalyzed additive-free FA
activity than the Cr0.15Au0.40Pd0.60/N-SiO2 catalyst in FA dehydro- dehydrogenation is close to first-order with respect to the catalyst
genation under identical conditions (Fig. S3, ESI†). The generated concentration within the investigated concentration window.
gas obtained during Cr0.15Au0.40Pd0.60/N-SiO2 catalyzed FA dehy-
In addition to the catalyst concentration, we also investigated
drogenation was analyzed by gas chromatography (GC), infrared the effect of temperature on the rate of FA dehydrogenation by
spectroscopy (FTIR) and NaOH trap experiment (Fig. S3–S6, ESI†). performing the catalytic reaction at different temperatures
Their results revealed that the generated gas is a mixture of (Fig. 2(f)). It is apparent that the Cr0.15Au0.40Pd0.60/N-SiO2 catalyst
H2 and CO2 with a H2 : CO2 molar ratio of 1.0 : 1.0 where CO works more effectively at elevated temperatures (4298 K), and
was below the detection limit (i.e. o10 ppm). In other words, the initial TOF values are 1280, 2030, 3360 and 4700 mol H2 mol
these experiments point to the important fact that CO-free H2 catalystÀ1 hÀ1 at 308, 318, 328 and 338 K, respectively. The
generation can be achieved in the absence of additives even reaction rates for each catalyst concentration were calculated
at room temperature from an aqueous FA solution for fuel from the linear portion of each plot.
cell applications4 by utilizing a Cr0.15Au0.40Pd0.60/N-SiO2 catalyst.
The values of observed rate constants kobs determined from
Fig. 2(d) shows the plot of the volume of generated gas (CO2 + H2) the linear portions of the volume of generated gas (CO2 + H2)
versus time for Cr0.15Au0.40Pd0.60/N-SiO2 catalyzed dehydrogena- versus reaction time plots at five different temperatures are
tion of aqueous FA solution (0.20 M in 10.0 mL H2O) at different used to obtain Arrhenius and Eyring plots (Fig. S8 and S9, ESI†)
catalyst concentrations ([CrAuPd]). We found that only 2.0 mol% to calculate activation parameters. Using these plots, apparent
of Cr0.15Au0.40Pd0.60/N-SiO2 can catalyze dehydrogenation of FA activation energy (Ea), apparent activation enthalpy (DHaa) and
at complete conversion with an initial TOF value of 730 mol apparent activation entropy (DSaa) values were calculated to be
H2 mol catalystÀ1 hÀ1. To the best of our knowledge, this is the 49.8 kJ molÀ1, 47.4 kJ molÀ1 and À65.1 J molÀ1 KÀ1, respectively.
highest TOF value ever reported for FA dehydrogenation at room Assuming that the apparent activation parameters calculated
temperature using a heterogeneous catalyst without utilizing any from the macroscopic kinetic data given above are relevant to the
additives (Table S1, ESI†) and is even comparable to most homo- most critical activation step in the FA dehydrogenation mecha-
geneous catalysts.6 More importantly; the dehydrogenation of FA nism, one can argue that the small positive value of apparent
can be completed within 10 min at 499% conversion measured activation energy and the large negative value of apparent
at 298 K. The reaction rates at each catalyst concentration were activation entropy imply the presence of an associative mecha-
calculated from the linear portion of each plot given in Fig. 2(d). The nism in the transition state.14 Then, to understand the effect of
logarithmic plot of the hydrogen generation rate versus catalyst the surface grafted amine group on the catalytic reactivity of
concentration (Fig. S7, ESI†) gives a line with a slope of 1.13 which CrAuPd NPs, we conducted a control experiment, in which the
This journal is ©The Royal Society of Chemistry 2015
Chem. Commun.

View Article Online
Communication
ChemComm
activity of CrAuPd NPs supported on amine-free SiO2 was inves- elemental analyses of recovered Cr0.15Au0.40Pd0.60/N-SiO2 catalyst
tigated in the FA dehydrogenation under identical conditions. We samples and reaction solutions indicated that metal and –NH2
found that an amine-free SiO2 supported CrAuPd catalyst contents of the catalysts remained intact after recycle and reusability
(Cr0.13Au0.27Pd0.60/SiO2) provides lower catalytic performance experiments. Additionally, removing the Cr0.15Au0.40Pd0.60/N-SiO2
(TOF = 120 mol H2 mol catalystÀ1 hÀ1 and 42% conversion) than catalyst from the reaction solution can completely stop the
that of Cr0.15Au0.40Pd0.60/N-SiO2 (Fig. S10, ESI†). The lower reac- dehydrogenation of FA. These results are indicative of the high
tivity of Cr0.15Au0.40Pd0.60/N-SiO2 can be explained by the absence stability of CrAuPd NPs against leaching throughout the cata-
of –NH2 functionalities on the support material, which may have a lytic runs. In summary, amine grafted silica supported CrAuPd
direct impact on the FA adsorption/storage process as well as the NPs have been reproducibly synthesized and preliminary char-
nucleation and growth of the CrAuPd NPs on the support surface. acterized by the combination of multi-pronged techniques. The
The CTEM image of the Cr0.13Au0.27Pd0.60/SiO2 catalyst (Fig. S11, resulting Cr0.15Au0.40Pd0.60/N-SiO2 catalyst revealed a record
ESI†) reveals the existence of highly clumped particles with activity (730 mol H2 mol catalystÀ1 hÀ1) and excellent conversion
respect to Cr0.15Au0.40Pd0.60/N-SiO2, which shows the stabilizing (499%) converging to that of the existing state of the art homo-
effect of surface grafted amine groups.15 Additionally, Yamashita genous catalytic systems available for the room temperature
et al.16 reported that Pd or Ag@Pd NPs supported on a resin dehydrogenation of FA in the absence of any additives. Under-
bearing –N(CH3)2 acted as more efficient organic supports than standing of the existing synergistic effects in the Cr0.15Au0.40Pd0.60
/
those of bearing –SO3H, –COOH and –OH in the catalytic decom- N-SiO2 catalyst remains an active area of research in our group.
position of FA. According to the results of their detailed mecha- More detailed experimental studies on the mechanism of
nistic studies, it was found that O–H bond cleavage is facilitated Cr0.15Au0.40Pd0.60/N-SiO2 catalyzed dehydrogenation of FA are still
with the assistance of the –N(CH3)2 group and leads to the underway. This uniquely active, selective and stable catalytic
formation of metal-formate species along with a –[N(CH3)2H]+ material has strong potential to be exploited in practical/techno-
group, which, then, undergo further dehydrogenation to produce logical applications, where FA is utilized as a viable hydrogen
H2 and CO2. In the light of these results, it is reasonable to carrier in mobile fuel cell applications.
understand that the surface grafted amine existing in our support
acts as a proton scavenger and provides a basic environment
around CrAuPd NPs, which benefits the O–H bond dissociation
Notes and references
1 L. Schlapbach and A. Zu¨ttel, Nature, 2001, 414, 353.
that is subsequently associated with the C–H bond cleavage from
the metal-formate intermediate to release H2.
2 J. A. Turner, Science, 1999, 285, 687.
3 M. Yadav and Q. Xu, Energy Environ. Sci., 2012, 5, 9698.
4 K. V. Kordesch and G. R. Simader, Chem. Rev., 1995, 95, 191.
5 T. C. Johnson, D. J. Morris and M. Wills, Chem. Soc. Rev., 2010, 39,
81–88.
6 A. Boddien, D. Mellmann, F. Gaertner, R. Jackstell, H. Junge,
P. J. Dyson, G. Laurenczy, R. Ludwig and M. Beller, Science, 2011,
333, 1733.
The catalytic stability of the Cr0.15Au0.40Pd0.60/N-SiO2 catalyst
was investigated by performing recycling and reusability experi-
ments (Fig. 2(f)). When all of the FA was converted to CO2 and
H2 in a particular cycle, more FA was added into the solution
and the reaction was continued up to five consecutive catalytic
cycles. It was found that the Cr0.15Au0.40Pd0.60/N-SiO2 catalyst
shows high stability and retains 70% of its initial activity and
provides 95% of conversion without CO generation after the
5th consecutive cycle. Isolability and reusability characteristics
of Cr0.15Au0.40Pd0.60/N-SiO2 were also tested in the FA dehydro-
genation under identical conditions. After the complete dehydro-
genation of FA, the catalyst was isolated as a dark gray powder and
bottled under a nitrogen atmosphere. Then, it was re-dispersed in
the aqueous FA solution. This re-dispersed catalyst preserved 50%
of its initial activity with 80% conversion of FA to CO2 and H2 even
after the 5th catalytic reuse. CTEM analyses of recovered samples
7 A. Bulut, M. Yurderi, Y. Karatas, M. Zahmakiran, H. Kivrak,
M. Gulcan and M. Kaya, Appl. Catal., B, 2014, 160, 514.
8 (a) K. Tedsree, T. Li, S. Jones, C. W. A. Chan, K. M. K. Yu,
P. A. J. Bagot, E. A. Marquis, G. D. W. Smith and S. C. E. Tsang,
Nat. Nanotechnol., 2011, 6, 302; (b) Z. L. Wang, J. M. Yan, H. L. Wang,
Y. Ping and Q. Jiang, J. Mater. Chem. A, 2013, 1, 12721; (c) Z. L. Wang,
J. M. Yan, Y. Ping, H. L. Wang, W. T. Zheng and Q. Jiang, Angew.
Chem., Int. Ed., 2013, 52, 4406; (d) Z. L. Wang, H. L. Wang, J. M. Yan,
Y. Ping, S. II. O, S. J. Li and Q. Jiang, Chem. Commun., 2014, 50, 2732;
¨
(e) O. Metin, X. Sun and S. Sun, Nanoscale, 2013, 5, 910; ( f ) S. Zhang,
¨
O. Metin, D. Su and S. Sun, Angew. Chem., Int. Ed., 2013, 52, 3681;
(g) A. Bulut, M. Yurderi, Y. Karatas, M. Zahmakiran, H. Kivrak,
M. Gulcan and M. Kaya, Appl. Catal., B, 2015, 164, 324.
9 See the ESI† for details.
10 I. Taylor and A. G. Howard, Anal. Chim. Acta, 1993, 271, 77.
11 J. He, Y. Zhang and E. Y.-X. Chen, ChemSusChem, 2013, 6, 61.
after 5th consecutive catalytic run of the recycling and reusability 12 J. S. Yoo, F. A. Pedersen, J. K. Nørskov and F. Studt, ACS Catal., 2014,
4, 1226; J. A. Herron, J. Scaranto, P. Ferrin, S. Li and M. Mavrikakis,
ACS Catal., 2014, 4, 4434.
13 R. Ferrando, J. Jellinek and R. L. Johnston, Chem. Rev., 2008, 108, 845.
experiments (Fig. S12 and S13, ESI†) show a slight increase in the
average particle size of Cr0.15Au0.40Pd0.60/N-SiO2 (2.9 and 4.2 nm,
respectively) consistent with the observed decrease in the activity 14 K. A. Connors, Theory of Chemical Kinetics, VCH Publishers,
New York, 1990.
at the end of these experiments. The XPS spectrum of the reused
sample shows that there is no significant change in the chemical
¨
15 M. Zahmakiran, M. Tristany, K. Philippot, K. Fajerwerg, S. Ozkar
and B. Chaudret, Chem. Commun., 2010, 46, 2938.
states of Pd, Au and Cr (Fig. S14, ESI†). Moreover, ICP-OES and 16 K. Mori, M. Dojo and H. Yamashita, ACS Catal., 2013, 3, 1114.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2015