ORIGINAL PAPER

Catalytic Generation of Hydrogen from Formic acid and itsDerivatives: Useful Hydrogen Storage Materials

Bjorn Loges • Albert Boddien • Felix Gartner •

Henrik Junge • Matthias Beller

Published online: 11 May 2010

� Springer Science+Business Media, LLC 2010

Abstract In this account the concept of using formic acid

as a hydrogen storage material is presented. Catalytic

reduction of carbon dioxide and heterogeneously catalyzed

decomposition of formic acid to hydrogen and carbon

dioxide are briefly discussed. In the main part the historic

development and recent examples of homogeneously cat-

alyzed hydrogen generation from formic acid are covered

in detail.

Keywords Formic acid � Hydrogen � Hydrogen storage �Energy � Catalysis � Ruthenium � Carbon dioxide

1 Introduction

A sufficient and benign supply of energy is one of the most

important challenges for the future of human society [1–3].

Among the various known energy carriers, hydrogen could

play an important role. One of the major obstacles to use

hydrogen for energy applications however, is the efficient

storage and handling of hydrogen. Compared to traditional

fuels, hydrogen has a very high gravimetric energy density,

but its volumetric energy density at atmospheric conditions

is too low. To obtain a balanced gravimetric and volu-

metric hydrogen storage density, pure hydrogen can be

stored in compressed gaseous or liquid form [4, 5]. In

addition, it can be adsorbed on porous materials, e.g.

zeolites, metal organic frameworks, or polymers of intrin-

sic microporosity [6–9]. Moreover, inorganic and organic

hydrogen storage materials are known in which chemical

bonds are formed and broken upon storage and release of

hydrogen. Here, mainly hydrides have been explored,

including covalent hydrides such as boranes, salts such as

magnesium hydride, metallic hydrides of the MHx type,

and complex hydrides such as Li(AlH4) [4, 7, 10–17]. In

recent years also few organic materials (‘‘organic

hydrides’’) have emerged as materials for covalent hydro-

gen storage. Due to their easier handling and inherent

energy efficiency especially liquid compounds like meth-

anol and formic acid appear to be practical [18, 19].

In this review the use of formic acid for hydrogen

storage will be discussed. Carbon dioxide hydrogenation to

formic acid, and formic acid decomposition in the presence

of heterogeneous catalysts will be covered briefly, and by

no way comprehensively. Recent advances in homoge-

neous hydrogen generation from (aqueous) formic acid/

metal formate solutions and from formic acid amine mix-

tures will be described in detail.

2 Formic Acid For Hydrogen Storage: The Concept

Based on formic acid and carbon dioxide a sustainable

cycle for energy storage can be conceived (Fig. 1). For

energy storage, carbon dioxide is converted to formic acid

or a formate derivative either electrochemically [20, 21] or

by catalytic hydrogenation [22–24]. The resulting material

is a liquid at ambient conditions, either pure formic acid, an

adduct containing formic acid, or an inorganic formate in

aqueous solution, and can thus be handled, stored, and

transported easily. On the other side of the cycle energy is

released either in a direct formic acid fuel cell, or by

selective on-demand decomposition into carbon dioxide

and hydrogen, which can be used directly in an appropriate

hydrogen oxygen fuel cell [25–30]. If pure hydrogen is

B. Loges � A. Boddien � F. Gartner � H. Junge � M. Beller (&)

Leibniz-Institut fur Katalyse e.V. an der Universitat Rostock,

Albert-Einstein-Str. 29a, 18059 Rostock, Germany

e-mail: matthias.beller@catalysis.de

123

Top Catal (2010) 53:902–914

DOI 10.1007/s11244-010-9522-8

required the gases may also be separated using membrane

techniques [31].

The hydrogen content of pure formic acid is 43 g/kg or

52 g/L, its release from formic acid is thermodynamically

downhill by DG� = –32.8 kJ/mol at room temperature [24]

(Scheme 1).

As early as 1978, the electrochemical reduction of carbon

dioxide to formic acid and its subsequent decomposition on

Pd/C for energy storage have been proposed by Williams

et al. [32]. Later on, a system for solar energy conversion by

reduction of aqueous carbonate was proposed by Halmann

in 1983 [33]. In 1986, a similar concept was described by

Wiener et al. They proposed a Pd/C catalyst to decompose

aqueous formate solutions to obtain hydrogen [34–36].

However, both approaches have not led to any application.

Two decades later, the research on the use of carbon dioxide

for energy storage has been resumed recently [19, 37–39].

Compared to methanol, a competing hydrogen carrier,

the hydrogen density of formic acid is considerably lower

(Table 1). However, to the best of our knowledge, there is

currently no procedure to obtain hydrogen from methanol

at ambient temperature. Additionally, regarding potential

hazards of formic acid, it may generally be considered less

hazardous than methanol. Methanol is highly flammable

and exhibits a metabolic toxicity which affects the central

nervous system and may lead to blindness [40]. Formic

acid is primarily a strong acid with an immediate corrosive

effect causing severe burns (see Table 1) [41]. An addi-

tional hazard of pure formic acid is its decomposition to

gases. On the other hand, dilute formic acid is approved as

a food additive [42].

Obviously, for the actual energy storage cycle, both the

formation and the decomposition of formic acid have to be

improved. However, most research groups focused only on

one of these two steps. In the next two sections, the reduc-

tion of carbon dioxide and heterogeneous decomposition of

formic acid are discussed briefly. Then, the first approaches

and current developments in the homogeneously catalyzed

hydrogen generation from formic acid are shown.

3 The Reduction of Carbon Dioxide to Formic Acid

In the first step of the proposed cycle, formic acid is formed

by reduction from carbon dioxide. One option is the

CatalyticRelease

CatalyticStorage

H2

from Renewable Resources

H2

Usage

CO2

HCO H (derivative)2

Fig. 1 A catalytic cycle for hydrogen storage in formic acid

Scheme 1 The hydrogenation

of carbon dioxide to formic acid

and/or derivatives [24]

Table 1 Comparison of properties: methanol versus formic acid

Methanol Formic acid

Molecular mass 32.042 g/mol 46.026 g/mol

Gravimetric hydrogen density 125 g/kg 43 g/kg

Volumetric hydrogen density 99 g/L 52 g/L

Hazard codes T, F C

Risk statements (R-sentence) 11–23/24/25–39/23/24/25 10–35

Boiling point 65 �C 101 �C

Vapor pressure (20 �C) 130.3 hPa 42.0 hPa

Explosion limits (lower - upper) 6 - 36 vol% 18 - 57 vol%

Flash point 11 �C 48 �C

Workplace exposure limit 200 ppm 5 ppm

LD50 (oral, rat) 5,628 mg/kg 1,100 mg/kg

The data has been derived from material safety data sheets available at commercial suppliers. Material safety data sheets are available from all

suppliers of chemical products upon request or via a website after login, e.g., www.chemdat.info or www.sigmaaldrich.com

Top Catal (2010) 53:902–914 903

123

electrochemical reduction of carbon dioxide in water,

which is covered in Ref. [20, 21]. For a sustainable storage

cycle, this approach depends on the availability of ‘‘car-

bon dioxide-free’’ electricity. Another option is the syn-

thesis of formic acid by hydrogenation of carbon dioxide,

reviewed in [22–24]. However, a major issue of this

reaction are the unfavorable thermodynamics: Formic acid

is formed under reaction conditions, i.e., elevated pres-

sure, but generally decomposes as soon as the pressure is

relieved.

For sustainable hydrogen storage, the hydrogenation of

carbon dioxide depends on the availability of ‘‘carbon

dioxide-free’’ hydrogen and energy. As a guideline, the

additional energy (DG�) required for this reaction is equiv-

alent to approximately 0.14 mol of hydrogen per mole of

formic acid, if the lower heating value of hydrogen in the

fuel cell reaction is considered (242 kJ/mol) [43].

Recently, the use of task specific ionic liquids in order to

facilitate the isolation of formic acid was reported [44, 45].

As a current state of the art catalyst system an iridium PNP-

pincer catalyst has been developed, which reached activi-

ties up to a turnover frequency (TOF) of 150,000 per hour

and an overall turnover number (TON) of 3,500,000,

respectively [46]. It has also been proposed that both

electrochemical reduction and hydrogenation of CO2 could

rely on sunlight as a sustainable source of energy: Elec-

tricity may be obtained via photovoltaics, and hydrogen

may then obtained by electrolysis [21, 32–38]. Alterna-

tively, hydrogen may be obtained by direct photochemical

water splitting in the future [3].

4 Hydrogen Generation from Formic Acid and Its

Derivatives

The second step of the cycle for hydrogen storage is the

selective hydrogen generation from formic acid. This step

also liberates carbon dioxide that may be recycled. It is

important to note that the decomposition of formic acid

may occur via two different pathways (Scheme 2):

dehydrogenation/decarboxylation (A) and dehydration/

decarbonylation (B) [47]. Both reactions are thermody-

namically downhill at standard conditions. As carbon

monoxide is a catalyst poison for fuel cell catalysts, only

the dehydrogenation/decarboxylation (A) pathway is of

interest for hydrogen generation, while the carbon mon-

oxide concentration due to decomposition via the other

pathway must be minimized to the ppm level! Only

reaction systems that selectively catalyze pathway A will

be considered here.

4.1 Decomposition of Formic Acid With

Heterogeneous Catalysts

The heterogeneously catalyzed decomposition of formic

acid to hydrogen and carbon dioxide has been reported

first by Sabatier in 1912 [48]. Since then, this reaction has

been employed in heterogeneous catalysis to study

adsorption and desorption processes during the decom-

position of formic acid vapors as a model substrate on

surfaces [49–53]. An early example was reported by Ri-

enacker et al., who performed this reaction with formic

acid vapors on surfaces of copper–gold and silver–gold

alloys [54]. Later, Rienacker et al. employed the decom-

position of formic acid to measure the activity of many

types of heterogeneous catalysts, mainly metals and alloys

of transition metals, such as iron, nickel, copper, palla-

dium, silver, platinum, and gold [55–57]. Another known

heterogeneously catalyzed decomposition of formic acid is

the photocatalytic conversion on titanium dioxide or other

nanoparticles [58–60], which occurs during photocatalytic

wastewater treatment [61]. As an application of catalytic

formic acid decomposition, Hyde and Poliakoff et al. have

proposed formic acid or formates as hydrogen sources for

hydrogenation reactions ‘‘without gases’’. Here, formic

acid is decomposed in a pressure reactor on a platinum

catalyst at 450 �C, and the resulting supercritical fluid is

fed to a second reactor, where the substrate and further

solvent is added. The product can be collected down-

stream after decompression of the supercritical fluid from

the reactor. This principle has been successfully applied

for hydrogenation of alkenes, alkynes, and carbonyl

compounds [62–64].

With respect to hydrogen generation, Williams et al.

used Pd/C (1 wt% Pd) to obtain around 55 mL of hydrogen

from 4 mol/L aqueous formic acid within 10 min. The

carbon dioxide obtained was trapped in a column filled

with potassium hydroxide pellets [32]. In the approach of

Wiener et al. in the mid-1980s, 900 mL hydrogen was

evolved from a 4 mol/L aqueous sodium formate solution

within 20 min at 70 �C in the presence of a Pd/C catalyst

(10 wt% Pd) [34]. They used a barium hydroxide solution

as carbon dioxide trap. Recently, Xing et al. reinvestigated

the decomposition of formic acid/sodium formate solutions

with noble metal catalysts supported on charcoal. They

obtained around 1,250 mL of gas from a solution of 1 eq.

Scheme 2 Formic acid

decomposition pathways, and

their thermodynamic properties

([114], see also [24])

904 Top Catal (2010) 53:902–914

123

sodium formate in 3 eq. formic acid with a Pd-Au/C cat-

alyst that had been co-deposited with CeO2 at 92 �C, which

corresponds to a TOF of 227 h-1 [65]. Another example is

the report of Iglesia et al., who selectively decomposed

formic acid on gold or platinum on alumina at 80 �C at

rates of more than 1,000 mol H2 per gram gold per hour,

which corresponds to a TOF of about 5 h-1. According to

the authors, it depends on the size of Au domains whether

formic acid dehydrogenation or dehydration, or even water

gas shift reaction (WGS) occurs [66]. The electrochemical

decomposition of formic acid, which is of interest for

wastewater treatment, has also been reported. Carbon

dioxide and hydrogen are generated separately on different

electrodes, and hydrogen can be obtained in high purity

(99.999%) [67].

4.2 Hydrogen Generation from Formic Acid and

Alkaline Formates

In homogeneous catalysis, an early report of a catalytic

decomposition of formic acid to carbon dioxide and

hydrogen dates back to the 1960s, when Coffey showed

formic acid decomposition with several platinum, ruthe-

nium and iridium phosphine complexes [68]. Coffey’s

most active catalyst was [IrH3(PPh3)3] with an initial 8890

turnovers per hour in a refluxing solution of formic acid in

acetic acid (Table 2, entry 1). Though metal carbonyls

were formed, no free carbon monoxide could be detected.

Four years later, Forster and Beck used rhodium and irid-

ium iodocarbonyl compounds in the presence of hydroiodic

acid, achieving a TOF of 4.4 h-1 in 70% aqueous formic

acid at 100 �C (Table 2, entry 2) [69]. Studying homoge-

neously catalyzed WGS, it was found that Ru3(CO)12

catalyzes the decomposition of formate in basic media at

75 �C (Table 2, entry 3) [70]. Later, Otsuka et al. studied

the selective dehydrogenation of formic acid in order to

characterize intermediates for the (homogeneously) cata-

lyzed water gas shift reaction. Their platinum(0) complex

[Pt(2-Pr3P)3] catalyzed the decomposition of formic acid in

acetone/water at 20 �C at a rate of 25 turnovers within the

first 15 min (Table 2, entry 4). A mechanism for the role of

this complex in the water gas shift reaction was proposed,

involving the decomposition of formic acid via the unstable

platinum formate [PtH(HCO2)(2-Pr3P)2] [71]. Strauss,

Whitmire and Shriver showed that the cyclometalated

rhodium triphenylphosphine complex 1 (Scheme 3) is a

precursor for [Rh(HCO2)(PPh3)3], which could be isolated.

The decomposition of formic acid was carried out in a

Table 2 Hydrogen generation from 5 mL TEAF with several homogeneous catalyst precursors and heterogeneous catalysts

Catalyst Substrate T/ �C TON TOF/h-1 Conv. Selectivity Remarks References

1 [IrH3(PPh3)3] HCO2H in acetic

acid (0.75 mol/L)

118 [11,000 8890 n.a. No free CO

detected

[68]

2 RhI/NaI HCO2H in

water 70%

100 n.a. 4.4 n.a. Significant

amount of CO

[69]

3 [Ru3(CO)12] HCO2H in

ethoxyethanol/water

75 50 600 50% n.a. Performed in a

pressure vessel

[70]

4 [Pt(2-Pr3P)2] HCO2H in

acetone/water 70%

20 25 100 Full n.a. [71]

5 1 HCO2H in

toluene 70%

20 [3.5 0.06 n.a. n.a. [72]

6 2 HCO2H/HCO2Na in

water 70%

20 ‘‘Several

hundred’’

3.3 n.a. n.a. [73]

7 RhCl3�3H2O/

NaNO2

HCO2H in water 90 n.a. 126 12.5% n.a. Titration of aq. cat.

soln. with HCO2H

[74]

8 3 HCO2H in

acetone (0.35 mol/L)

R.T. 2,000 (?) *500 Full No free CO,

H2O detected

Performed in NMR

spectrometer

[76]

9 [cp*Mo(PMe3)3H] HCO2H in

pentane (13 mmol/L)

80 n.a. n.a. n.a. n.a. [77]

10 4 HCO2H in THF-d8

(63 mmol/L)

80 10 \2 Full n.a. Performed in NMR

spectrometer

[78]

11 6 HCO2H/HCO2Na

in water

25 80 30 n.a. No free CO

detected

[82]

12 7 HCO2H in water

(2 mol/L)

90 \10,000 14,000 Near

full

No free CO

detected

Also performed in

pressure vessels

[84]

13 [Ru(H2O)6](tos)2/

TPPTS

HCO2H/HCO2Na 9:1

in water (4 mol/L)

120 [40,000 460 n.a. No CO

detected

Continuous setup [87]

The values have been calculated from the data obtained from the respective references

Top Catal (2010) 53:902–914 905

123

solution of formic acid in toluene at 20 �C, although the

turnover frequency of 0.06 h-1 for this catalyst is rather

low (Table 2, entry 5) [72]. In 1982 Trogler and co-

workers reported a binuclear platinum triethylphosphine

catalyst 2 (Scheme 3) that decomposed an aqueous solu-

tion of formic acid in the presence of sodium formate with

a TOF of 3.3 h-1 at 20 �C and at a constant rate of several

100 turnovers (Table 2, entry 6) [73].

Rhodium, iridium, ruthenium, and palladium chloride in

the presence of sodium nitrite, but without any other

ligand, were found to catalyze formic acid decomposition

at 90 �C in aqueous solution by King and Bhattacharyya

(max. TOF 126 h-1) in 1995 (Table 2, entry 7) [74].

Another binuclear complex for formic acid decomposi-

tion was presented by Puddephat et al. in 1998. This binu-

clear ruthenium phosphine complex [Ru2(l-CO)(CO)4

(l-dppm)2] (3) was the most active complex for this reaction

at that time, achieving a TOF of 500 h-1 after quantitative

formic acid decomposition (15 min; Table 2, entry 8). The

reaction was performed in NMR tubes at room temperature

in a solution of formic acid and acetone-d6. The authors

identified various intermediate hydrides and formate com-

plexes by NMR spectroscopy, and isolated [Ru2H(l-H)

(l-CO)(CO)2(l-dppm)2]. In the presence of triethylamine

they successfully performed the reverse reaction, achieving

a maximum HCO2H:NEt3 ratio of 1.2:1 [75, 76].

The formic acid decomposition with an early transition

metal complex, [cp*Mo(PMe3)3H], was reported by Parkin

et al. in 2002 (Table 2, entry 9) [77]. Another approach has

been followed by Lau et al. in 2003. To explore the acid-

ities of Ru–H and Mo-H or W–H hydrides, they

synthesized heterobinuclear bisphosphine complexes 4

containing ruthenium and molybdenum or tungsten for the

interconversion of carbon dioxide/hydrogen and formic

acid (Scheme 4). In NMR studies, they identified a com-

plex bearing a bridging hydride, which was not stable as a

formate, but could be isolated as a tetrafluoroborate.

However, the activity of these complexes does not exceed a

TOF of 30 h-1 for carbon dioxide hydrogenation and 2 h-1

for formic acid decomposition (Table 2, entry 10) [78].

Ogo, Fukuzumi et al. observed stoichiometric hydrogen

evolution treating [Rh(III)(1,4,7-triazacyclononane)

(HCO2)2](OTf) 5 with sodium formate, and they obtained a

dihydride carbonyl complex [79]. Using 1H NMR and mass

spectroscopy together with deuteration experiments and

other spectroscopic studies, they concluded that hydrogen

is generated from the diformate complex 5 by protonation.

They also demonstrated that hydrogen is in fact evolved

from the formate ligands, and that a rhodium carbonyl

complex is formed. Based on their former investigations on

carbon dioxide hydrogenation [80, 81], Fukuzumi et al.

used [Rh(cp*)(bipy)(H2O)](SO4) (6) and similar com-

plexes for the hydrogen generation from aqueous formic

acid with a maximum of 80 turnovers within 5 h at

pH = 3.8 (Table 2, entry 11). Combining spectroscopic

studies and DFT calculations, they demonstrated that for-

mic acid decomposition occurs via a formate complex and

a hydride complex [82]. More recently, he demonstrated

that heteronuclear iridium–ruthenium complexes are highly

active catalysts for hydrogen generation in an aqueous

solution under ambient conditions giving a TOF of about

426 h-1 [83]. Focusing on increased hydrogen output,

Himeda reported the iridium catalyzed decomposition of

formic acid/sodium formate in aqueous solution at 90 �C

[84]. With the iridium complex 7 similar to the system used

earlier in the group of Fukuzumi, he achieved the highest

activity for non-phosphine-containing catalysts below

100 �C with an initial TOF of 14,000 h-1 at 90 �C

(Table 2, entry 12; Scheme 4).

In 2008, Laurenczy et al. and our group independently

revisited the concept of formic acid as a hydrogen storage

material. Based on their expertise in the hydrogenation of

Scheme 3 Selected catalyst precursors for the decomposition of

formic acid 1979–1998

Ru M

R

Ph2P PPh2OC

CO

Lau et al., 2003

M = Mo, WR = H, Me

RhO

HN

NH

NHH

O

O

HO

OTf-

Ogo, Fukuzumi et al., 2005

Ru OH2

N

N

SO42-

Ogo et al., 2002

2+

+

4 5 6

Ir OH2

N

N

Himeda, 2009

2+

7

OH

SO42-

HO

Scheme 4 Selected catalyst precursors for the decomposition of formic acid 2003–2009

906 Top Catal (2010) 53:902–914

123

carbonate, Laurenczy et al. used a water-soluble ruthenium

tppts (tris-m-sulfonated triphenylphosphine trisodium salt,

or 3,30,300-phosphinidynetris(benzenesulfonic acid) triso-

dium salt) catalyst, which released hydrogen from aqueous

solutions of formic acid/sodium formate (9:1) at tempera-

tures of 70–120 �C. At 25 �C, hydrogen generation is also

observed, but conversion is slow. Notably, hydrogen can be

obtained at pressures between 1 and 220 bar, and no loss of

catalytic activity is observed up to 750 bar [85–87]. Under

this rather high pressure, carbon dioxide hydrogenation

takes places with the remaining product gases when the

reaction turns slightly basic after full conversion of formic

acid due to the presence of formate. Remarkably, no carbon

monoxide was detected in the gas phase by high pressure

infrared spectroscopy. In a continuous setup, formic acid is

fed to the autoclave and product gases are released from it.

A TOF of 460 h-1 was obtained, and the catalyst was

stable for more than 90 h, reaching more than 40,000

cycles (Table 2, entry 13). High pressure NMR provided

information for proposing a mechanism with two compet-

ing pathways of formic acid decomposition, one involving

a monohydride, the other a dihydride catalyst species

(Scheme 5).

Starting from the bisphosphine tetraaqua ruthenium

complex 8, formate replaces a water ligand to form com-

plex 8-I. A second water molecule is lost when b-elimi-

nation of a hydrogen atom from formate occurs, forming

the carbon dioxide hydride complex 8-II. Carbon dioxide is

replaced by water to give the monohydride complex 8-III.

Concluding the monohydride mechanism, formic acid

replaces a water ligand and protonates the hydride to form

dihydrogen in 8-IV, which is subsequently displaced by

water to reform formate 8-I. Monohydride 8-III is also the

starting point for the dihydride mechanism. Instead of

formic acid, a formate ion coordinates to give the hydride

formate species 9-I. b-elimination of hydrogen forms the

carbon dioxide complex 9-II, which eliminates carbon

dioxide to form 9-III. From this complex, hydrogen is

eliminated upon addition of a molecule of formic acid. The

authors suggested that the reaction mainly occurs via the

dihydride mechanism. Further investigations on the heter-

ogenization of these ruthenium phosphine catalysts were

successful, but catalyst activity is still very low [88].

4.3 Hydrogen Generation From Formic Acid Amine

Mixtures

Our own approach on formic acid decomposition for

hydrogen generation is based on the formic acid amine

adducts as substrate/solvent, which are also known as tri-

alkylammonium formates. Such compounds are also well

known as hydrogen donors in transfer hydrogenation

reactions [89–93], in particular triethylammonium formate

(TEAF, HCO2H/NEt3 5:2) [94]. In the past there have been

some observations on hydrogen generation during these

transfer hydrogenations, especially with ruthenium and

rhodium catalysts [89, 95, 96]. However, the potential for

hydrogen generation at ambient conditions has never been

explored before our work.

At the start of our work the selective decomposition of

formic acid triethylamine adducts was investigated apply-

ing several homogeneous and heterogeneous catalysts

(Table 3) [97]. With most heterogeneous catalysts

(Table 3, entries 1–7) no or only little hydrogen was

evolved at ambient temperature. In TEAF only Pd/C,

which has been suggested for aqueous formate decompo-

sition for energy storage [32, 34], was active for about

20 min, but then deactivated, too. In order to establish a

first system for hydrogen generation, we also investigated

several soluble transition metal compounds (Table 3,

entries 8–17). Among these, only two ruthenium(II) arene

complexes (Table 3, entries 14–17) were active for more

than 2 h, being [RuCl2(p-cymene)]2 (10) the most active.

Slow formation of hydrogen is also observed with rho-

dium(III) chloride, whereas ruthenium(III) chloride, cop-

per(I) iodide, iron(II) chloride, and iron(III) chloride as

Scheme 5 Mechanism of

formic acid decomposition

catalyzed by the Ru/tppts

system (P = tppts; charges are

omitted for clarity) [87]

Top Catal (2010) 53:902–914 907

123

well as [cpFe(CO)2I] are not active under these conditions.

Interestingly, with rhodium(III) chloride, hydrogen pro-

duction increased four times when the experiment was

continued 6 h [97].

The robustness of the TEAF/10 system for hydrogen

generation was investigated adding water and other sol-

vents including some ionic liquids to the substrate

(Table 4, entries 1–9) [98, 99]. The addition of water,

ethanol or N,N0-dimethylformamide (DMF) has no or only

little effect on the system’s performance. Dimethylsulfox-

ide (DMSO) and tetrahydrofuran (THF) decrease the

hydrogen output. Among the ionic liquids, 1-methyl-3-

octylimidazolium tetrafluoroborate did not have a negative

effect. We also found that the addition of a small amount of

alkaline or earth alkaline bromides or iodides is beneficial

and increased catalyst activities by a factor of up to five

(Table 4, entries 10–12). More recently, Shi and co-

workers have successfully investigated the formic acid/10

system using task specific ionic liquids and sodium formate

as a base [100].

Having established the [RuCl2(p-cymene)]2 (10) as a

reliable and robust system for the catalytic decomposition

Table 3 Hydrogen generation from 5 mL TEAF with several homogeneous catalyst precursors and heterogeneous catalysts

Catalyst T/ �C nmetal/lmol 2 h 3 h

Vgas/mL TON Vgas/mL TON

1 CuO on alumina 13 wt% 80 59.5 1.4 \1 1.7 \1

2 Fe2O3 on silica 5 wt% 40 19.1 0.4 No H2 0.9 No H2

3 Nano Fe2O3 40 59.5 0.6 No H2 – –

4 Nickel on silica 60 wt% 80 595 6.9 \1 7.5 \1

5 Pd/C 10 wt% 40 19.1 19 20 20 22

6 Ru on Fe3O4 5 wt% 40 59.5 12.4 4.2 13.0 4.5

7 RuOx on Fe2O3 5 wt% 40 59.5 2.9 1.0 3.2 1.1

8 CuI 40 59.5 4.3 1.5 6.1 2.1

9 FeCl2 120 120 2.9 \1 4.8 \1

10 FeCl3 120 120 3.9 \1 6.4 \1

11 [cpFe(CO)2I] 40 59.5 3.6 No H2 4.1 No H2

12 RhCl3�xH2O 40 19.1 1.9 2.0 4.3 4.5

13 RuCl3�xH2O 40 19.1 0.6 No H2 – –

14 [RuCl2(benzene)]2 40 59.5 30 10 46 16

15 [RuCl2(p-cymene)]2 40 19.1 12 13 17 19

16 [RuCl2(p-cymene)]2 40 59.5 41 14 61 21

17 [RuCl2(p-cymene)]2 26.5 1,191 215 3.7 345 5.9

Table 4 Hydrogen generation from 5 mL TEAF/59.5 lmol 10 with additives at 40 �C

Catalyst Additive Amount 2 h 3 h References

Vgas/mL TON Vgas/mL TON

1 [RuCl2(p-cymene)]2 – – 41 14 61 21 [97]

2 [RuCl2(p-cymene)]2 Water 5 mL 41 14 62 21 [98]

3 [RuCl2(p-cymene)]2 Ethanol 5 mL 44 15 67 23 [98]

4 [RuCl2(p-cymene)]2 DMF 1 mL 47 16 69 24 [98]

5 RuCl2(p-cymene)]2 DMSO 1 mL 29 10 35 12

6 RuCl2(p-cymene)]2 THF 1 mL 35 12 52 18

7 RuCl2(p-cymene)]2 1-methyl-3-octylimidazolium tetrafluoroborate 1 mL 43 15 66 22

8 RuCl2(p-cymene)]2 1-butyl-3-methylimidazolium tetrafluoroborate 1 mL 31 11 48 16

9 RuCl2(p-cymene)]2 1,3-dimethylimidazolium dimethylphosphate 1 mL 3.5 1.2 4.4 1.5

10 RuCl2(p-cymene)]2 Potassium bromide 595 lmol 101 34 150 51 [99]

11 RuCl2(p-cymene)]2 Magnesium bromide 595 lmol 78 27 120 41 [99]

12 RuCl2(p-cymene)]2 Potassium iodide 595 lmol 279 96 338 116 [99]

908 Top Catal (2010) 53:902–914

123

of TEAF, we investigated the effect of the amine and its

concentration in more detail [97–99].

In a first approach, only tertiary amines and ammonia

were studied, later on other nitrogen-containing com-

pounds such as imines, amides and heterocycles com-

pounds were also added to our study. As shown in Fig. 2, a

relation between the added amine and the activity of the

resulting hydrogen generation system was established. In

general, more basic compounds generate systems with

higher activity. The most active system was obtained

applying a 5:3 mixture of formic acid and DBN (1,5-

diazabicyclo(4.3.0)non-5-ene).

The concentration of the amine also strongly affects the

rate of hydrogen generation. We investigated this effect for

triethylamine and N,N-dimethylbutylamine. In control

experiments, no catalytic hydrogen generation is observed

in the absence of amine. A higher concentration of amine is

beneficial for hydrogen production. For NEt3/HCO2H, an

increase of catalyst activity (TON after 3 h) from 1.2 for a

1:10 mixture to 76 for a 3:4 mixture is observed. For

BuNMe2/HCO2H the TON after 3 h increased from 2.3 for

1:10 mixture to 41 for a 4:5 mixture [98]. For DBN/

HCO2H, 58 turnovers were achieved in 3 h for a 2:5

mixture, and 90 turnovers for a 3:5 mixture [99].

Noteworthy, using the ruthenium phosphine complex

[RuCl2(PPh3)3] we found that hydrogen generation

increased by more than an order of magnitude. A similar

activity is obtained with an in situ catalyst prepared from

ruthenium trichloride hydrate and triphenylphosphine in

DMF. We explored this phenomenon investigating a number

Fig. 2 The influence of different nitrogen containing compounds on

hydrogen evolution from 2:5 mixtures with formic acid using

[RuCl2(p-cymene)]2 (10) at 40 �C [99]

Table 5 Hydrogen generation from 5 mL TEAF with ruthenium/triphenylphosphine catalysts at 40 �C [98]

Ruthenium precursor nRu/lmol Phosphine ligand Ru:P 2 h 3 h References

Vgas/mL TON Vgas/mL TON

1 – – PPh3 – 0.0 – 0.0 – [98]

2 RuCl2(PPh3)3 5.95 – – 260 891 261 893 [97]

3 RuCl3�xH2O 5.95 PPh3 1:3 202 691 204 700 [98]

4 RuBr3�xH2O 5.30 PPh3 1:3.4 230 882 232 891 [98]

5 RuBr3�xH2O 17.1 PPh3 1:3.4 1,154 1,375 1,238 1,475 [98]

6 [RuCl2(p-cymene)]2 6.05 PPh3 1:3 22 73 24 81 [98]

7 [RuCl2(p-cymene)]2 19.0 PPh3 1:3 47 50 52 56 [98]

8 [RuI2(p-cymene)]2 5.97 PPh3 1:3 13 44 17 57 [98]

9 [RuCl2(benzene)]2 5.95 – – 2.5 8.7 3.8 13 [98]

10 [RuCl2(benzene)]2 5.95 PPh3 1:1 1.6 5.5 2.0 6.9 [98]

11 [RuCl2(benzene)]2 5.95 PPh3 1:3 106 361 124 425 [98]

12 [RuCl2(benzene)]2 5.95 PPh3 1:6 116 397 131 450 [98]

13 [RuCl2(benzene)]2 5.95 PPh3 1:20 133 454 147 505 [98]

14 [RuCl2(benzene)]2 19.4 PPh3 1:3 324 340 419 440 [98]

15 [RuCl2(benzene)]2 29.8 PPh3 1:3 450 308 671 459 [98]

16 [RuCl2(C10H16)]2a 5.99 PPh3 1:3 47 159 51 173 [98]

17 Ru(methylallyl)2 COD 5.96 PPh3 1:3 20 68 23 79 [98]

18 Ru(acac)3 5.93 PPh3 1:3 1.1 3.9 1.2 4.0 [98]

19 RuCl2(bipy)2 5.96 PPh3 1:3 0.2 0.85 0.3 1.0 [98]

20 Shvo0s cat.b 5.90 PPh3 1:3 0.0 – 0.0 – [98]

a) [RuCl2(C10H16)]2 = bischloro(l-chloro)bis[(1,2,3,6,7,8,h)-2,7-dimethyl-2,6-octadien-1,8-diyl]diruthenium (IV); b) pretreatment at 120 �C,

reaction at 50 �C

Top Catal (2010) 53:902–914 909

123

of ruthenium precursors and phosphine ligands for the for-

mation of several in situ catalysts. First, ruthenium precur-

sors were investigated with triphenylphosphine in DMF

(Table 5). While ruthenium precursors such as ruthe-

nium(III) chloride and bromide showed a high initial activ-

ity, they are deactivated after 20 min. Significant activity

after the 20 initial minutes was best achieved with ruthenium

(II) g6-arene complex [RuCl2(benzene)]2 (14). With this

precursor, the highest catalyst activities are observed at

lower ruthenium concentrations (Table 5, entries 12, 14, 15).

Varying the ruthenium to phosphine ratio (Table 5, entries

9–13), we observed that three PPh3 ligands are needed to

obtain an improved activity. Adding 6 or 20 equivalents of

PPh3 also increases catalyst activity, but less dramatically.

Next, different phosphine ligands were studied using

precursor 14 and a Ru:P ratio of 1:6 (Table 6). The best

performance among monodentate phosphines is observed

for 14/triphenylphosphine. Among bidentate phosphines,

14/1,2-bis(diphenylphosphino)ethane (dppe) catalyst per-

forms best, but with a long induction period. Only little

hydrogen generation is observed in the presence of alkyl

phosphines.

To prove the concept of using the hydrogen from formic

acid for electricity generation, we coupled one of our

systems to a fuel cell. The gas evolved from a mixture of

5HCO2H/4HexNMe2 in the presence of [RuCl2(benzene)]2

(14)/6 PPh3 at room temperature, contained only hydrogen,

carbon dioxide, and traces of argon. However, to prevent

possible poisoning of the fuel cell by traces of starting

material that might have been below the limit of detection

of our GC, we filtered the gas through a short column of

activated charcoal. After an initial phase of higher activity,

constantly 26 mW (at 370 V) are obtained during 42 h

[98]. After full conversion of formic acid, the system can

be re-activated simply by adding formic acid, which is an

indicator for the robustness of the catalyst.

Using ruthenium in combination with the dppe ligand,

the system was further developed towards practical appli-

cations [101]. Applying 20 mL of 5HCO2H/4HexNMe2 as

substrate a TON of 5,716 within 3 h at 40 �C was reached,

which constituted the highest activity for our approach to

hydrogen generation so far. After gas evolution had ceased,

the catalyst was reactivated 10 times by addition of formic

acid. In the first run, a prolonged induction period was

Table 6 Hydrogen generation from 5 mL TEAF with [RuCl2(benzene)]2 (14)/ligand catalysts at 40 �C [98]

Ruthenium

precursor

nRu/

lmol

Ligandb Ru:P 2 h 3 h References

Vgas/

mL

TON Vgas/

mL

TON

1 [RuCl2(benzene)]2 19.1 PCy3 1:6 1.7 1.8 2.0 2.1 [98]

2 [RuCl2(benzene)]2 19.1 BuPAd2 1:6 1.2 1.3 1.5 1.6 [98]

3 [RuCl2(benzene)]2 19.1 P(o-tolyl)3 1:6 11 12 16 17 [98]

4 [RuCl2(benzene)]2 19.1 P(furyl)3 1:6 164 175 198 211 [98]

5 [RuCl2(benzene)]2 19.1

NPPh

Ph

1:6 4.5 4.8 6.1 6.5 [98]

6 [RuCl2(benzene)]2 19.1 dppm 1:6 2,799 2,624 2,985 2,798

7 [RuCl2(benzene)]2 19.1 dppe 1:6 238 254 1,289 1,376 [98]

8 [RuCl2(benzene)]2 19.1 dppp 1:6 694 740 799 852 [98]

9 [RuCl2(benzene)]2 19.1 dppb 1:6 23 25 27 29 [98]

10 [RuCl2(benzene)]2 19.1 dppf 1:6 73 78 87 93 [98]

11 [RuCl2(benzene)]2 19.1 1,2-bisdiphenylphosphinobenzenea 1:6 5.5 5.9 12 13 [98]

12 [RuCl2(benzene)]2 19.1 Xantphos 1:6 22 24 35 37

13 [RuCl2(benzene)]2 19.1 (S)-1-[(RP)-2-(diphenylphosphino)ferrocenyl]-ethyldi(3,5-

xylyl)phosphine

1:6 41 44 62 66

14 [RuCl2(benzene)]2 19.1

NPCy

Cy

P Cy

Cy

1:6 32 34 49 52

a Pre-treatment 3 min at 40 �C in an ultrasonic bathb dppm bis(diphenylphosphino)methane, dppe 1,2-bis(diphenylphosphino)ethane, dppp 1,3-bis(diphenylphosphino)propane, dppb 1,4-

bis(diphenylphosphino)butane, dppf 1,10-bis(diphenylphosphino)ferrocene

910 Top Catal (2010) 53:902–914

123

observed. The activity did not decrease significantly within

the next 10 runs, and an overall TON [ 60,000 was

achieved (Fig. 4a). Based on these results, a continuous

reactor was set up with 9.55 lmol [RuCl2(benzene)]2 14/6

dppe 17.5 mL HexNMe2 (Fig. 3).

To start the reaction, 4.75 mL formic acid were added.

Then addition of formic acid was continued at a rate of

0.74 mL/h. The gas flow was monitored by a flow meter

and the hydrogen content was measured with a hydrogen

gas sensor, sustained by a GC of the gas collected every

24 h. The system worked for more than 11 days using

commercial 99% formic acid from BASF SE as received.

Gas output and hydrogen concentration were virtually

constant during this time, and no signs of catalyst deacti-

vation were observed (Fig. 4b). Overall, 260,000 turnovers

were achieved, corresponding to TOF [ 900 h-1. This

concept was proven to work in a small prototype model car

driven by a hydrogen/air fuel cell, which has been coupled

to an onboard hydrogen generation system using formic

acid and a similar catalyst.

Recently, the group of Wills studied hydrogen genera-

tion from TEAF with a tethered half-sandwich complexes

of rhodium 11 and ruthenium 12 at room temperature,

obtaining a TOF of around 490 h-1 [102]. In the same

publication, they also presented studies with different

transition metal complexes at 120 �C, among them

[RuCl2(dmso)6]. This is an active catalyst (TOF =

18,000 h-1), which is also stable and recyclable (TON =

25,000, after four cycles). However, addition of one

equivalent of phosphine led to deactivation, and the for-

mate-bridged binuclear complex [Ru2(HCO2)2(CO)4

(PPh3)2] 13 is observed (Scheme 6).

Another interesting aspect of the catalytic decomposi-

tion of formic acid is its photochemical acceleration. In the

early 1990s, there have been two independent reports of

light-accelerated reactions of formic acid and transition

metal complexes in solution. Onishi reported that

Fig. 3 Continuous setup for

hydrogen generation from

formic acid [101]

h

/ L

time / min

Fig. 4 a Recycling of a [RuCl2(benzene)]2/dppe catalyst for hydrogen generation from 5HCO2H/4HexNMe2. b Gas output and hydrogen

concentration from continuous formic acid decomposition with [RuCl2(benzene)]2/dppe in HexNMe2 [101]

Scheme 6 Complexes by Wills et al. for the dehydrogenation of

formic acid

Top Catal (2010) 53:902–914 911

123

irradiation with a 400 W Hg vapor lamp with Pyrex filter

accelerates the reaction of formic acid with [HCo(Ph-

P(Et)2)4], performed in THF at 30 �C, from 0.09 turnovers

to 1.6 turnovers within 3 h, and 3.1 in 6 h [103]. They also

observed a shift of the hydride signal in 1H NMR from a

well defined quintet at -14.76 ppm of the original complex

to a broad singlet -12.47 ppm upon the addition of formic

acid, which did not change neither after ageing the com-

plex for 6 h, nor after subsequent irradiation. At the same

time, King et al. showed that hydrogen is evolved from

aqueous formate solutions with chromium hexacarbonyl

under irradiation with a TON of 18 after 1 h in a setup

similar to Onishi’s [104]. They proposed that one carbon

monoxide ligand dissociates from the chromium center

upon photolysis, which is replaced by a weakly coordinated

solvent molecule (methanol). When formate is added,

hydrogen is evolved. Interestingly this reaction is inhibited

by the addition of pyridine.

While investigating the decomposition of TEAF with

various ruthenium phosphine catalysts, we observed that

this process is also accelerated by illumination with sun-

light [105]. Though the photochemistry of ruthenium

phosphine hydride and carbonyl complexes [106, 107] and

ruthenium g6-arene complexes [108–111] is well-known,

this behavior has rarely been explored for catalysis [112,

113]. When our previous ruthenium phosphine containing

catalyst systems were irradiated with visible light, gas

evolution increased up to 11 times (Fig. 5). We have

studied several phosphine ligands in combination with

[RuCl2(benzene)]2 (14), but the effect is only observed for

aryl phosphines. RuCl3•xH2O and [Ru(cod)(methylallyl)2]

are suitable ruthenium sources, too. The best catalyst pro-

ductivity is observed with a 14/dppe catalyst, where gas

evolution increased from 407 to 2,804 turnovers, which is

an almost 7-fold increase. The highest activity with a

monodentate ligand is observed with PPh3 as ligand.

Additionally, the reaction was performed at different

temperatures. In all cases experiments under irradiation

performed significantly better than dark reactions, so a

temperature effect can be ruled out. Notably, our system is

still active at low temperatures such as 0 �C.

Analyzing the rate of gas evolution, it was found that

irradiation generates faster an active species, and then

accelerates the actual reaction. Based on these observa-

tions, and on further NMR-spectroscopic investigations, we

proposed the mechanism shown in Scheme 7 for this light

accelerated dehydrogenation of formic acid (Scheme 7).

Starting from the ruthenium precursor 14, a ruthenium

hydride phosphine complex 15-I is formed, while the g6-

benzene ligand is cleaved under irradiation. Formic acid

adds to this complex, possibly via intermediates, and a

dihydrogen formate ruthenium complex 15-II is formed.

Dihydrogen loss from 15-II is as well accelerated by light

as it is the subsequent b-elimination of carbon dioxide from

the formate 15-III to reform the ruthenium hydride phos-

phine complex 15-I. Additionally, irradiation prevents the

catalyst from being deactivated. This allows one trigger the

hydrogen generation from formic acid simply by switching

on and off the light source.

5 Conclusion

Although the decomposition of formic acid, especially with

heterogeneous catalyst, has been often studied as a model

reaction for catalyst characterization, only recently this

reaction has received significant attraction for hydrogen

generation. In general, formic acid allows for a simple and

benign storage of hydrogen. On one hand, the industrial

production of formic acid via direct catalytic hydrogenation

of carbon dioxide seems not too distant. On the other hand,

already today several methods for hydrogen generation

Ru

deactivated species

HCO2H

HCO2H

[RuCl2(benzene)]2

PAryl3LnRu + ...

CO2

H2

O

O

H

(PAryl3)mLn-1Ru

O

O

H

Ln

RuLn

H

L

L

L = H-, Cl-, benzenem = 1-3n = 1-3

H H(PAryl3)m

h·

(PAryl3)m

15-I

15-II15-III

14

h·

h·

h·

Scheme 7 Mechanism for the light-accelerated decomposition of

formic acid [105]

Fig. 5 Acceleration of ruthenium-phosphine catalyzed hydrogen

generation from 5 mL TEAF with light (19.1 lmol Ru, 40 �C). aNo hydrogen detected by GC. b No dark experiment, experiment

performed under lab conditions (environmental light)

912 Top Catal (2010) 53:902–914

123

from formic acid are available awaiting application.

Depending on the developed catalysts hydrogen can be

generated either at higher temperatures ([100 �C) and high

pressure as well as at ambient conditions and low temper-

atures. It has been shown by us that systems invented in the

lab can be actually used in ‘‘real world’’ electric applica-

tions, and that scale-up is viable. Yet a less expensive cat-

alyst system is desirable, and investigations are ongoing.

Hydrogen cleaning is easy, but for a higher performance

and prolonged lifetime of the fuel cell, hydrogen and carbon

dioxide should be separated. This would also lead to a better

control of the fuel cell performance, and a higher energy

density of the entire system.

In the future formic acid obtained from sustainable

resources might be also of considerable importance as

hydrogen storage material. Due to its intrinsic properties,

we believe that formic acid will be especially valuable for

niche energy applications, such as small portable devices.

Acknowledgements This work has been supported by the State of

Mecklenburg-Vorpommern, the BMBF, and the DFG (Leibniz-prize

and research training group 1213).

References

1. Bockris JOM (1972) Science 176:1323

2. Crabtree GW, Dresselhaus MS, Buchanan MV (2004) Phys

Today 57:39

3. Lewis NS, Nocera DG (2006) Proc Natl Acad Sci USA

103:15729

4. Eberle U, Felderhoff M, Schuth F (2009) Angew Chem

121:6732

5. Schlapbach L, Zuttel A (2001) Nature 414:353

6. Thomas KM (2007) Catal Today 120:389

7. van den Berg AWC, Arean CO (2008) Chem Commun 668

8. Murray LJ, Dinca M, Long JR (2009) Chem Soc Rev 38:1294

9. Morris RE, Wheatley PS (2008) Angew Chem 120:5044

10. Schuth F, Bogdanovic B, Felderhoff M (2004) Chem Commun

2249

11. Biniwale RB, Rayalu S, Devotta S, Ichikawa M (2008) Int J

Hydrogen Energy 33:360

12. Crabtree RH (2008) Energy Environ Sci 1:134

13. Hu YH, Ruckenstein E (2008) Ind Eng Chem Res 47:48

14. Graetz J (2009) Chem Soc Rev 38:73

15. Hamilton CW, Baker RT, Staubitz A, Manners I (2009) Chem

Soc Rev 38:279

16. Marrero-Alfonso EY, Beaird AM, Davis TA, Matthews MA

(2009) Ind Eng Chem Res 48:3703

17. Collins DJ, Zhou HC (2007) J Mater Chem 17:3154

18. Makowski P, Thomas A, Kuhn P, Goettmann F (2009) Energy

Environ Sci 2:480

19. Johnson TC, Morris DJ, Wills M (2010) Chem Soc Rev 39:81

20. Benson EE, Kubiak CP, Sathrum AJ, Smieja JM (2009) Chem

Soc Rev 38:89

21. Arakawa H, Aresta M, Armor J, Barteau M, Beckman E, Bell A,

Bercaw J, Creutz C, Dixon DA, Dixon D, Domen K, DuBois D,

Eckert J, Fujita E, Gibson D, Goddard W, Goodman D, Keller J,

Kubas G, Kung H, Lyons J, Manzer L, Marks T, Morokuma K,

Nicholas K, Periana R, Que L, Rostrup-Nielson J, Sachtler W,

Schmidt L, Sen A, Somorjai G, Stair P, Stults B, Tumas W

(2001) Chem Rev 101:953

22. Leitner W (1995) Angew Chem 107:2391

23. Jessop PG, Joo F, Tai CC (2004) Coord Chem Rev 248:2425

24. Jessop PG (2008). In: de Vries JG, Elsevier CJ (eds) The

handbook of homogeneous hydrogenation, vol 2. Wiley-VCH,

Weinheim, p 489

25. de Bruijn FA, Papageorgopoulos DC, Sitters EF, Janssen GJM

(2002) J Power Sources 110:117

26. Ahluwalia RK, Wang X (2008) J Power Sources 180:122

27. Russell AE, Ball SC, Maniguet S, Thompsett D (2007) J Power

Sources 171:72

28. Giddey S, Ciacchi F, Badwal S (2005) Ionics 11:1

29. Tingelof T, Hedstrom L, Holmstrom N, Alvfors P, Lindbergh G

(2008) Int J Hydrogen Energy 33:2064

30. Larminie J, Dicks A (2003) Fuel cell systems explained. Wiley,

Chichester

31. Bernardo P, Drioli E, Golemme G (2009) Ind Eng Chem Res

48:4638

32. Williams R, Crandall RS, Bloom A (1978) Appl Phys Lett

33:381

33. Halmann M, Ulman M, Aurian-Blajeni B (1983) Sol Energy

31:429

34. Wiener H, Sasson Y, Blum J (1986) J Mol Catal 35:277

35. Zaidman B, Wiener H, Sasson Y (1986) Int J Hydrogen Energy

11:341

36. Wiener H, Zaidman B, Sasson Y (1989) Sol Energy 43:291

37. Enthaler S (2008) ChemSusChem 1:801

38. Joo F (2008) ChemSusChem 1:805

39. Fukuzumi S (2008) Eur J Inorg Chem 1351

40. Methanol. In Rompp Online (Thieme Chemistry, Stuttgart 2010)

41. Formic acid. In Rompp Online (Thieme Chemistry, Stuttgart

2010)

42. Diluted formic acid is on the US Food and Drug Administration

list of food additives. (US Code of Federal Regulations: 21 CFR

186.1316, 21 CFR 172.515)

43. Joseck F, Placet M, vol 2009. Hydrogen Analysis Resource

Center, U.S. Department of Energy

44. Zhang Z, Xie Y, Li W, Hu S, Song J, Jiang T, Han B (2008)

Angew Chem 120:1143

45. Zhang Z, Hu S, Song J, Li W, Yang G, Han B (2009)

ChemSusChem 2:234

46. Tanaka R, Yamashita M, Nozaki K (2009) J Am Chem Soc

131:14168

47. Akiya N, Savage PE (1998) AIChE J 44:405

48. Sabatier P, Mailhe A (1912) Compt Rend 152:1212

49. Trillo JM, Munuera G, Criado JM (1972) Catal Rev 7:51

50. Columbia MR, Thiel PA (1994) J Electroanal Chem 369:1

51. Tamaru K (1997) Appl Catal A 151:167

52. Jarvi TD, Stuve EM (1998) Electrocatalysis 75

53. Hinshelwood CN, Hartley H (1923) J Chem Soc Trans 123:1333

54. Rienacker G (1936) Z Anorg Allg Chem 227:353

55. Rienacker G, Hansen N (1956) Z Anorg Allg Chem 285:28356. Rienacker G, Volter J (1959) Z Anorg Allg Chem 302:299

57. Rienacker G, Mueller H (1968) Z Anorg Allg Chem 357:255

58. Willner I, Goren Z (1986) J Chem Soc Chem Commun 172

59. Kaneco S, Kurimoto H, Shimizu Y, Ohta K, Mizuno T (1999)

Energy 24:21

60. Sun B, Smirniotis PG (2003) Catal Today 88:49

61. Hoffmann MR, Martin ST, Choi W, Bahnemann DW (1995)

Chem Rev 95:69

62. Garcia-Verdugo E, Liu Z, Ramirez E, Garcia-Serna J, Fraga-

Dubreuil J, Hyde JR, Hamley PA, Poliakoff M (2006) Green

Chem 8:359

63. Hyde JR, Poliakoff M (2004) Chem Commun 1482

Top Catal (2010) 53:902–914 913

123

64. Hyde JR, Walsh B, Singh J, Poliakoff M (2005) Green Chem

7:357

65. Zhou X, Huang Y, Xing W, Liu C, Liao J, Lu T (2008) Chem

Commun 3540

66. Ojeda M, Iglesia E (2009) Angew Chem 121:4894

67. Kilic EO, Koparal AS, Ogutveren UB (2009) Fuel Process

Technol 90:158

68. Coffey RS (1967) Chem Commun 18:923a

69. Forster D, Beck GR (1971) Chem Commun 994:1072

70. Laine RM, Rinker RG, Ford PC (1977) J Am Chem Soc 99:252

71. Yoshida T, Ueda Y, Otsuka S (1978) J Am Chem Soc 100:3941

72. Strauss SH, Whitmire KH, Shriver DF (1979) J Organomet

Chem 174:C59

73. Paonessa RS, Trogler WC (1982) J Am Chem Soc 104:3529

74. King RB, Bhattacharyya NK (1995) Inorg Chim Acta 237:65

75. Gao Y, Kuncheria J, Yap GPA, Puddephatt RJ (1998) Chem

Commun 2365

76. Gao Y, Kuncheria JK, Jenkins HA, Puddephatt RJ, Yap GPA

(2000) J Chem Soc Dalton Trans 3212

77. Shin JH, Churchill DG, Parkin G (2002) J Organomet Chem

642:9

78. Man ML, Zhou Z, Ng SM, Lau CP (2003) Dalton Trans 3727

79. Ogo S, Nishida H, Hayashi H, Murata Y, Fukuzumi S (2005)

Organometallics 24:4816

80. Hayashi H, Ogo S, Fukuzumi S (2004) Chem Commun 2714

81. Ogo S, Kabe R, Hayashi H, Harada R, Fukuzumi S (2006)

Dalton Trans 4657

82. Fukuzumi S, Kobayashi T, Suenobu T (2008) ChemSusChem

1:827

83. Fukuzumi S, Kobayashi T, Suenobu T (2010) J Am Chem Soc

132:1496

84. Himeda Y (2009) Green Chem 11:2018

85. Fellay C, Dyson PJ, Laurenczy G (2008) Angew Chem

120:4030

86. Laurenczy G, Fellay C, Dyson P (2008) WO2008/047312

87. Fellay C, Yan N, Dyson PJ, Laurenczy G (2009) Chem Eur J

15:3752

88. Gan W, Dyson PJ, Laurenczy G (2009) React Kinet Catal Lett

98:205

89. Fujii A, Hashiguchi S, Uematsu N, Ikariya T, Noyori R (1996) J

Am Chem Soc 118:2521

90. Noyori R, Hashiguchi S (1997) Acc Chem Res 30:97

91. Gladiali S, Alberico E (2004) In: Beller M, Bolm C (eds)

Transition metals for organic synthesis, vol 2. Wiley-VCH,

Weinheim, p 145

92. Gladiali S, Alberico E (2006) Chem Soc Rev 35:226

93. Matharu DS, Morris DJ, Clarkson GJ, Wills M (2006) Chem

Commun 3232

94. Wagner K (1970) Angew Chem 82:73

95. Himeda Y, Onozawa-Komatsuzaki N, Sugihara H, Arakawa H,

Kasuga K (2003) J Mol Catal A 195:95

96. Ogo S, Abura T, Watanabe Y (2002) Organometallics 21:2964

97. Loges B, Boddien A, Junge H, Beller M (2008) Angew Chem

120:4026

98. Boddien A, Loges B, Junge H, Beller M (2008) ChemSusChem

1:751

99. Junge H, Boddien A, Capitta F, Loges B, Noyes JR, Gladiali S,

Beller M (2009) Tetrahedron Lett 50:1603

100. Li X, Ma X, Shi F, Deng Y (2010) ChemSusChem 3:71

101. Boddien A, Loges B, Junge H, Gartner F, Noyes JR, Beller M

(2009) Adv Synth Catal 351:2517

102. Morris DJ, Clarkson GJ, Wills M (2009) Organometallics

28:4133

103. Onishi M (1993) J Mol Catal 80:145

104. Linn DE, King RB, King AD (1993) J Mol Catal 80:151

105. Loges B, Boddien A, Junge H, Noyes JR, Baumann W, Beller M

(2009) Chem Commun 4185

106. Roundhill DM (1994) Photochemistry and photophysics of

metal complexes. Plenum Press, New York

107. Geoffroy GL, Bradley MG (1977) Inorg Chem 16:744

108. Zelonka RA, Baird MC (1972) Can J Chem 50:3063

109. Bennett MA, Smith AK (1974) J Chem Soc Dalton Trans 233

110. Hayashida T, Nagashima H (2002) Organometallics 21:3884

111. Lackner W, Standfest-Hauser CM, Mereiter K, Schmid R, Kir-

chner K (2004) Inorg Chim Acta 357:2721

112. Morton D, Cole-Hamilton DJ (1988) J Chem Soc 1154

113. Morton D, Cole-Hamilton DJ, Utuk ID, Paneque-Sosa M, Lopez

Poveda M (1989) J Chem Soc Dalton Trans 489

114. NIST Chemistry Webbook. http://webbook.nist.gov (accessed

29.01.2010)

914 Top Catal (2010) 53:902–914

123