Exploring the myth of nascent hydrogen and its implications for biomass conversions

Article abstract of DOI:10.1002/asia.201200557

Iron (and to a lesser extent manganese) in the wall of a 316 stainless steel (SS) reactor is responsible for the hydrogenation of cyclohexanone to cyclohexanol when using an aqueous formic acid solution under high temperature and pressure water (HTPW) conditions. However, not only dilute formic acid but also aqueous solutions of several other organic and mineral acids in the presence of iron are active in this reaction covering a range of aldehydes and ketones, even under ambient conditions. The stoichiometry, kinetics, and the possible mechanisms of both dihydrogen production as well as of the hydrogenation of the model compound cyclohexanone were examined. The reduction is essentially stoichiometric with respect to metallic iron, and the conversions are highly dependent on the speed of stirring as well as temperature and reactant concentrations. Importantly, it is established unequivocally that water participates in dihydrogen gas formation (hydrogen atoms originate from both the acid and water molecules) and facilitates substrate reduction. Copyright

Full text of DOI:10.1002/asia.201200557

DOI: 10.1002/asia.201200557
Exploring the Myth of Nascent Hydrogen and its Implications for Biomass
Conversions
[a]
Viktꢀria Fꢁbos, Alexander K. L. Yuen, Anthony F. Masters, and Thomas Maschmeyer*
Dedicated to Prof. Sir John Meurig Thomas, FRS on the occasion of his 80th birthday
Abstract: Iron (and to a lesser extent
manganese) in the wall of a 316 stain-
less steel (SS) reactor is responsible for
the hydrogenation of cyclohexanone to
cyclohexanol when using an aqueous
formic acid solution under high tem-
perature and pressure water (HTPW)
conditions. However, not only dilute
formic acid but also aqueous solutions
of several other organic and mineral
acids in the presence of iron are active
in this reaction covering a range of al-
dehydes and ketones, even under ambi-
ent conditions. The stoichiometry, ki-
netics, and the possible mechanisms of
both dihydrogen production as well as
of the hydrogenation of the model
compound cyclohexanone were exam-
ined. The reduction is essentially stoi-
chiometric with respect to metallic
iron, and the conversions are highly de-
pendent on the speed of stirring as well
as temperature and reactant concentra-
tions. Importantly, it is established un-
equivocally that water participates in
dihydrogen gas formation (hydrogen
atoms originate from both the acid and
water molecules) and facilitates sub-
strate reduction.
Keywords: acid · hydrogenation ·
iron · nascent hydrogen · water
[
2]
Introduction
drogen through the solution in the absence of zinc. Conse-
quently, it has been asserted that this so-called “nascent hy-
drogen” formed in situ is of different reactivity to gaseous,
molecular hydrogen. Although several theories have been
The concept of “nascent hydrogen” has long been invoked
to explain some mysterious form of hydrogen exhibiting en-
hanced reducing power. The field of hydrothermal biomass
conversions has not been entirely immune from this percep-
[3–11]
proposed
to explain the peculiar activity of this in situ-
generated hydrogen, the exact details are still a matter of
conjecture.
[1]
tion, even though a decade ago Laborda et al. had already
asked of nascent hydrogen in their title “when is it going to
be over?” They pointed out that “nascent hydrogen is…a
fictitious concept”. However, their findings seem not to
have penetrated the community fully, with discussions of
nascent hydrogen currently still appearing in prestigious
journals, as some ephemeral, ill-defined material, imbued
with special powers. This treatment may be fine empirically,
but it does not help in the rational development of our disci-
pline.
Dilute, aqueous acid solutions are well-known to produce
dihydrogen in the presence of particular metals (e.g., iron or
zinc). Indeed, in reporting the discovery of hydrogen, Cav-
endish wrote “I know of only three metallic substances,
namely zinc, iron and tin, that generate inflammable air by
solution in acids; and those only by solution in the diluted
[sulphuric acid]…”. It is surprising that only a few papers
have investigated the utility of such systems for the reduc-
tion of organic substrates, using the “nascent hydrogen”
[2b]
[12–17]
“
Nascent hydrogen” can be produced on metal surfaces
formed in situ.
Furthermore, there appear to be no re-
by the reduction of protons, and it is well known that the ox-
idation of a reactive metal such as zinc in acidic media re-
ports of the hydrogenation of ketones using dilute acid solu-
tions and metallic iron. This paper concerns the reactivity of
a model ketone, cyclohexanone, with iron in a range of acids
(Scheme 1), so as to determine the mechanism and, there-
fore, the stoichiometry of the hydrogenation to probe the
“nascent hydrogen” concept.
[2]
sults in the production of gaseous dihydrogen. Under
acidic conditions zinc is also able to act as a chemical reduc-
[2]
ing agent. For example, if ferric chloride is added to a mix-
ture of zinc and sulfuric acid, the ferric salt is quickly re-
duced, but no such reduction occurs by bubbling gaseous hy-
[
a] Dr. V. Fꢀbos, Dr. A. K. L. Yuen, Prof. Dr. A. F. Masters,
Prof. Dr. T. Maschmeyer
School of Chemistry
The University of Sydney
NSW 2006, Australia
Fax : (+61)293513329
E-mail: th.maschmeyer@chem.usyd.edu.au
Homepage: www.chem.usyd.edu.au/research/maschmeyer.html
Scheme 1. Formation of cyclohexanol (2) and cyclohexene (3) from cyclo-
hexanone (1).
Chem. Asian J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
&
&
&
These are not the final page numbers! ÞÞ

FULL PAPER
It should be noted that the present study is the companion
In considering the mechanism of the hydrogenation, there
are two in-principle possibilities:
1) A special (possibly catalytic) site is created on the
metal surface during the treatment of the iron with the acid
solution, and this site plays a crucial role in substrate reduc-
[18]
piece to its predecessor,
that under the conditions used for hydrothermal upgrading
HTU) of lignocelluosic biomass, the walls of typical stain-
in which it was demonstrated
(
less steel reactors are able to effect ketone reductions in the
presence of formic acid, itself a product typically formed
under the HTU conditions.
tion, using the liberated H formed from the decomposition
2
of the acid. If this were the case, it would be reasonable to
expect that additional H gas should increase the conversion
2
of cyclohexanone.
Results and Discussion
2) The alternative to the above scenario is that surface-
bound species, possibly surface hydrides, resulting from the
decomposition of the acid and surface metal oxidation, are
responsible for the reduction, in which case no promoting
influence of additional hydrogen should be observed.
To examine these alternatives, the reductions were per-
formed under inert and reactive atmospheres (these gases
were added in addition to any
Sulfuric, hydrochloric, phosphoric, acetic, lactic, and pro-
pionic acids were tested as reagents for cyclohexanone re-
duction in the presence of iron. The reactions were per-
formed under both HTPW and ambient conditions. Table 1
shows the conversion of cyclohexanone using different aque-
H generated in situ). As no no-
2
ticeable differences in conver-
[
a]
Table 1. The conversion of cyclohexanone in aqueous acid solutions.
Proton/hydrogen source Molar equiv.
Entry T [8C] p [bar] Name pK i Iron H-source
sions under N , air, or H2 at-
2
Conversion [%] Selectivity [%]
mospheres
were
observed
a
1
1
2
3
(
within the experimental error,
[
[
[
[
b]
b]
b]
b]
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
250
250
250
250
100
100
100
100
100
100
100
100
100
100
100
RT
RT
60
60
60
60
1
1
1
1
1
1
1
1
1
1
1
1
1
H
H
2
3
SO
PO
4
À3/2.0
2.2/7.2/12.3
3.9
4.7
3.8
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0.2
0.6
0.2
0.5
0.5
1.5
0.5
1.0
0.5
0.5
0.5
0.5
1.2
0.9
0.5
0.5
0.5
0.1 20.4
0.1 11.6
0.1 4.8
0.1 7.2
0.4 16.1
0.4 18.4
0.4 25.0
0.4 30.9
0.4 12.7
0.4 3.3
0.4 11.9
0.4 5.9
0.4 15.1
0.4 0.0
0.4 0.0
0.4 18.2
0.4 13.6
100
17.2
100
99.1
100
100
100
100
100
100
100
100
100
0
0
82.8
0
0.9
0
0
0
0
0
0
0
0
0
0
0
0
0
Æ5%), it would appear that the
second alternative is the correct
one, that is, hydrogen or a relat-
ed species produced in situ on
the metal surface is responsible
for the reduction of cyclohexa-
none (Table 2), the likely candi-
date being a transient iron-hy-
dride species.
To explore whether soluble
iron complexes (which might
indicate a homogeneous cata-
lytic system) were formed
during the reaction under either
HTPW or ambient conditions,
liquid phase in-situ infrared
4
Lactic acid
CH COOH
HCOOH
HCOOH
H
H
3
3.8
2
SO
SO
4
À3/2.0
À3/1.99
À4
2
4
HCl
PO
0
1
2
3
4
5
6
7
H
3
4
2.2/7.2/12.3
4.8
CH
Lactic acid
Lactic acid
Propionic acid 4.9
H
H
3
COOH
3.9
3.9
2
–
0
100
100
2
SO
4
À3/2.0
HCOOH
3.8
[
t
1
a] Conditions: Cyclohexanone (1, 330 mg, 3.4 mmol) in 5 mL of acid solution in a round-bottomed flask,
reaction =1 h; no formation of cyclohexane was detected in entries 1–17. [b] Cyclohexanone (1, 120 mg,
.2 mmol) in 2 mL of acid solution in a sapphire NMR tube, treaction =4 h.
ous solutions of these acids. In general, the stronger the acid
the lower the pK value), the higher the conversion ach-
Table 2. The conversion of cyclohexanone under different atmospheres.
(
a
Entry T [8C] Atm.
Acid
Amount of Fe [mg] Conversion of
ieved after one hour. The highest conversions were obtained
using sulfuric, formic, or hydrochloric acids. Furthermore, as
expected for any given acid, the higher its concentration, the
higher the yield of cyclohexanol that can be observed. The
lack of any further reaction of cyclohexanol is consistent
with the fact that this has only been observed at higher tem-
peratures. The dehydration of cyclohexanol occurs to
a small extent at 1408C; however, when the temperature is
increased to, for example, 3758C, cyclohexene is the major
~
1 bar
1 [%]
1
2
3
50
50
50
air
N
H2
HCOOH 500
HCOOH 500
HCOOH 500
15.7
16.1
14.5
2
Conditions: Cyclohexanone (1, 330 mg, 3.4 mmol) in 5 mL of 1.4m
HCOOH solution; t=2 h.
spectroscopic (IR) measurements were recorded during the
reaction. However, during these reductions neither iron-car-
bonyl nor iron-hydride signals were detected (irrespective of
whether sulfuric or formic acid was used). Nonetheless, as
was shown previously, the formation of catalytically inac-
tive iron salts (formate or sulfate) occurred (see the Sup-
porting Information for time-resolved IR spectra).
[19]
reaction product.
Use of either propionic acid or gaseous H as the reduc-
2
tant under ambient conditions did not result in any conver-
sion, which is consistent with the results that were observed
under HTPW conditions in the presence of both SSS and Fe
[18]
[18]
metal.
&
&
&2
&
www.chemasianj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 0000, 00, 0 – 0
ÝÝ These are not the final page numbers!

Exploring the Myth of Nascent Hydrogen
When analyzing the gaseous product(s) formed Table 3. The effect of iron to acid molar ratio variation on the conversion of cyclohex-
anone in 0.9m H
2
SO
4
solution at 1008C after 1 hour.
(
both in the presence and absence of cyclohexa-
Entry Acid
Amount of Fe [mg] Molar ratio of iron to H
2
SO
4
Conversion of
[%]
none), only hydrogen could be detected. In fact, it
is instructive to measure the instantaneous concen-
1
1
2
3
H
H
H
2
2
2
SO
SO
SO
4
4
4
250
1
2
6
16.1
21.9
28.7
tration of H gas in the exit stream as the rate of
2
500
hydrogen production decreases over time as per
Figure 1. Furthermore, addition of cyclohexanone
1500
added to the acid solution, ensuring the total consumption
of the iron powder over several hours. Using a burette
system, developed for the precise measurement of gas con-
sumption or generation (shown schematically in Figure 6),
the exact amount of hydrogen gas generated during the re-
action could be determined. A comparison of the moles of
the hydrogen produced to the moles of iron added and com-
pletely consumed during the reaction showed that iron
reacts with one equivalent of a monoprotic acid (error of
Æ15%), thus indicating that iron reacts close to stoichio-
metrically and ultimately forming iron salts such as formate
or sulfate species.
To investigate whether water was necessary for the reac-
tion, experiments were carried out using a solution of
2
Figure 1. Instantaneous concentration of H gas in the exit stream from
the reaction of 0.08m HCOOH solution with 1.3 mg iron at RT in the
presence and absence of 1 g cyclohexanone (1).
9
8.15 wt% formic acid in different dry solvents such as etha-
nol, methanol, 2-propanol, and dimethyl sulfoxide, affording
reaction media with an overall maximum water content of
0.15 wt%. Notably, neither gas production nor conversion of
cyclohexanone could be detected by gas chromatography
under these relatively dry conditions. Hence, water plays
a crucial role in both, the formation of hydrogen gas and, in-
dependently, the reduction of the substrate. The effect upon
varying the quantity of water on the reaction was also
tested. As Figure 2 shows, a small amount of water (5 wt%)
in ethanol is enough to induce some reaction of cyclohexa-
none, but the extent of hydrogenation increases with in-
creasing water content. This observation is consistent with
the results in the accompanying report, obtained in stainless
significantly reduces the H evolution rate, thereby clearly
demonstrating a competitive reaction between the reduction
2
of cyclohexanone and H production.
2
To gain a better understanding of this hydrogenation pro-
cess and its mechanism, the stoichiometry of the reaction
was probed and the roles of iron, water, and the composi-
tion of the hydrogen gas generated were investigated. For
this purpose, the determination of the exact amount of hy-
drogen formed during the reaction was required, and thus
precise hydrogen evolution measurements were performed.
Firstly, the amount of iron used was varied and its effect
on the reaction studied. As expected, a larger amount of
iron (and therefore reactive surface) results in a higher con-
version of cyclohexanone. However, if the proportionality of
this increase is considered (raising the mass of the iron by
[18]
steel reactors, in which the addition of water to the reac-
tion increased the conversion of cyclohexanone to cyclohex-
600% results in a further 12.6% conversion), it is clear that
the effect is not very pronounced (Table 3). Although the
larger amount of iron should result in higher cyclohexanone
conversions, it is also responsible for a much more rapid
acid consumption, releasing unproductive hydrogen gas and
thereby exhausting the acid supply needed for the produc-
tion of the reductively active metal hydride (or a closely-re-
lated reductant). It is apparent from these observations that
the rate of H production is greater than the rate of cyclo-
2
hexanone reduction and that it is likely a common inter-
mediate is involved in both processes. Significantly, when
the iron-to-acid molar ratio was 1, no iron powder remained
after reaction for 1 hour, whereas when that ratio exceeded
1, iron powder remained.
To determine the exact stoichiometry of the iron to the
Figure 2. The conversion of cyclohexanone (1) into cyclohexanol (2) in
hydrogen produced in situ, a very small amount of iron was
ethanol/water at RT.
Chem. Asian J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
3
&
&
&
These are not the final page numbers! ÞÞ

Thomas Maschmeyer et al.
FULL PAPER
anol. Indeed, such higher reactivity in dilute acids was al-
ready observed by Cavendish (see above).
As water therefore appears essential for the reaction to
occur, an obvious question is whether the hydrogen formed
in situ is derived solely from the acid (as is mirrored by the
reaction equations commonly used) or whether water is also
To rule out any peculiarities in the set-up of our system
that would give rise to these observations, the temperature
dependence of the hydrogen evolution was investigated so
that the activation energies could be determined. Thus, the
treatment of sulfuric acid with iron powder was monitored
at three different temperatures. As the plots of the volume
of produced hydrogen as a function of time in Figure 4
[2b]
a proton source in the H generation. To explore this ques-
2
tion, the hydrogen ion concentration was measured by titra-
tion before and after the reaction of the iron with the acid
solution. Given that a relatively high acid concentration was
used, the change of the hydrogen ion concentration should
result only from the consumption of acid. Hence, the
amount of hydrogen produced was compared to the change
in hydrogen ion concentration. In these experiments, it was
found that one mole of monoprotic acid generates one mole
of dihydrogen. This unequivocally shows that, in the absence
of additional substrate, half the atoms of the hydrogen gas
produced derive from the acid and half from the water.
To investigate the quantity of hydrogen gas produced
when reduction of a substrate also occurs, gas evolution was
monitored in the presence and absence of cyclohexanone.
The difference in the moles of hydrogen evolved was com-
pared to the yield of cyclohexanol formed (calculated from
GC results). The effect of cyclohexanone on the net volume
of hydrogen that evolved from the reaction of a fixed
amount of iron and sulfuric acid is shown in Figure 3. As-
Figure 4. Time course of the hydrogen produced in situ from a mixture of
1
5 mL of 0.8m
1.8 mmol) iron powder at temperatures of 258C (&), 358C (*), and
58C (~).
2 4 2 4
H SO (12 mmol of H SO ) solution and 109 mg
(
4
show, temperature has a significant influence on the rate of
H evolution, as only a 108C increase in the temperature re-
2
sults in five-fold faster evolution of hydrogen gas. The initial
reaction rates were determined from the slope of the linear
fit of the first 10 data points in Figure 4 at 25, 35, and 458C,
and are summarized in Table 4. With constant initial condi-
Table 4. The calculated initial reaction rates (vinit) at different tempera-
tures in the reaction of 15 mL of 0.8m H
2
SO
4
solution and 109 mg iron
powder.
À1
T [8C]
v
init [mmol(H
2
)min
]
À3
À5
25
35
45
5.2ꢂ10 Æ4.8ꢂ10
À2
À4
À4
1.8ꢂ10 Æ1.60ꢂ10
Figure 3. Hydrogen gas produced from 15 mL of 1m H
2
SO
) and 108 mg (1.8 mmol) of iron powder in the absence
*) and in the presence of (*) 1 g (10 mmol) cyclohexanone (1) at
58C. The conversion of cyclohexanone was 2.0%.
4
solution
À2
4.1ꢂ10 Æ2.72ꢂ10
(
(
2 4
20 mmol H SO
2
tions, the temperature dependence of the initial rates of the
suming ideal gas behavior, the reduction in the number of
moles of H2 exiting the reaction vessel, calculated from
hydrogen production gives an activation energy of 81.9Æ
À1
8.5 kJmol , a value that is within the same range as the
À1
DV(H ), was found to be 0.194 mmol. As the conversion of
67.9 kJmol reported for an Armco iron electrode in 0.5m
2
[20]
cyclohexanone under the above conditions was measured to
be 2%, this corresponds to the reduction of 0.209 mmol of
cyclohexanone. Taken together, these data indicate that
H SO solution, albeit under very different conditions.
2 4
In order to probe any effect of the stirring speed, the hy-
drogenation was carried out at two different rates of agita-
tion. It was found that the stirring speed has a significant
effect on the conversion of cyclohexanone (see Table 5). In
fact, tripling the stirring rate halved the conversion in the
case of sulfuric acid. This result suggests that the reduction
of cyclohexanone is slower than the hydrogen evolution,
1
mole of active hydrogen, generated in situ, has the poten-
tial to reduce 1 mole of cyclohexanone (experimental error
under the above conditions was measured to be Æ7%). This
result confirms that substrate hydrogenation competes with
the formation of dihydrogen (cf. Figure 1.).
&
&
&4
&
www.chemasianj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 0000, 00, 0 – 0
ÝÝ These are not the final page numbers!

Exploring the Myth of Nascent Hydrogen
Table 5. The conversion of cyclohexanone (1) at different rates of stir-
ring.
1008C, it increases constantly with time at room tempera-
ture. This is in good agreement with the finding that increas-
ing the temperature significantly increases the rate of the
hydrogen evolution (see Figure 4). Additionally, it also re-
flects the depletion of the acid after 1 hour at 1008C. Conse-
quently, at the end of 5 hours of reaction time, a much
higher conversion can be achieved at room temperature (ap-
proximately double) than at 1008C.
Entry
Acid
Stirring rate
Conversion of 1 [%]
1
2
H
H
2
SO
SO
4
4
750 rpm
250 rpm
5.4
10.3
2
2 4
Conditions: 5 mL 0.9m aqueous H SO solution, 0.33 g (3.4 mmol) cyclo-
hexanone, and 0.5 g (9 mmol) iron powder at RT for 1 h.
The rate of the hydrogen evolution can also be controlled
by continual addition of small aliquots of the acid solution
to the iron powder (or vice versa). As Table 6 shows, when
that is, slowing the generation of hydrogen gas formed in
situ improves the yield of cyclohexanol.
Combining these two observations, cyclohexanone conver-
sion was plotted as a function of time for a slowly stirred
Table 6. The conversion of cyclohexanone (1, 0.33 g, 3.4 mmol) in 1.8m
(
250 rpm) reaction at room temperature and at 1008C (see
aqueous H
2
SO
4
solution and 0.5 g (9 mmol) iron at 1008C for 3 h.
Figure 5). While the conversion plateaus after 1 hour at
Entry
Acid
Reagent added dropwise
Conversion of 1 [%]
1
2
H
H
2
2
SO
SO
4
4
acid
Fe
51.6
21.7
acid is added dropwise to the reaction mixture containing
the iron powder, the conversion is much higher than if iron
is added in small portions to the acid. Therefore, when the
iron-to-acid ratio is kept high (by the presence of a small
amount of acid) and slow hydrogen evolution is maintained,
an improved reaction of cyclohexanone results.
Finally, to determine whether a dilute acid solution in the
presence of iron can be used for the hydrogenation of sub-
strates other than cyclohexanone, different aldehydes, ke-
tones, and unsaturated compounds were tested. Table 7
shows that the carbonyl groups of benzaldehyde, acetophe-
none, hexanone, and levulinic acid are susceptible to hydro-
genation using aqueous formic or sulfuric acid in the pres-
ence of iron. As no reaction occurred using cyclohexene or
toluene under these conditions,
Figure 5. The conversion of cyclohexanone (0.33 g, 3.4 mmol) in 5 mL of
0
.9m aqueous H
2 4 2 4
SO solution (4.3 mmol H SO ) with 0.25 g of iron
(
4.2 mmol) at RT (*) and at 1008C (&).
it is clear that the present re-
duction process is selective for
aldehyde and ketone groups.
Furthermore, the selectivities of
the reactions to the correspond-
ing alcohols were 100%, as de-
termined by GC-FID and GC-
MS (see the Experimental Sec-
tion). Hence, the foregoing ob-
servations do not require the
invocation of “nascent hydro-
gen” and are consistent with es-
tablished mechanistic steps ac-
cepted for carbonyl reductions.
Generally, for catalytic hy-
drogenations, hydride transfer
from the H-donor (MH ) to the
2
substrate can take place accord-
ing
a
to
two
mechanisms:
Figure 6. Schematic diagram of the gas burette equipment for the determination of the volume of the pro-
duced/consumed gas (Th: thermostat, MSB: magnetic stir bar, MS: magnetic stirrer, ST: stopper, R: reactor
RC: reflux condenser, M: manometer containing octane, GB: gas burette, MH: mercury holder, T: tap, for 3-
way tap 2: 1. state, open to vacuum; 2. state, open to gas burette; 3. state, open to reaction vessel)
metal-templated, concerted
process (direct H-transfer) or
a metal-hydride-mediated mul-
Chem. Asian J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
5
&
&
&
These are not the final page numbers! ÞÞ

Thomas Maschmeyer et al.
FULL PAPER
Table 7. The reduction of different substrates.
Entry
Substrate
Acid
SO
HCOOH
SO
HCOOH
SO
HCOOH
SO
HCOOH
SO
HCOOH
SO
HCOOH
Conversion [%]
1a
1b
2a
2b
3a
3b
4a
4b
5a
5b
6a
6b
cyclohexene
cyclohexene
toluene
H
2
4
0
0
0
0
60.9
64.4
11.7
6.8
3.0
5.1
38.0
12.0
H
2
4
toluene
benzaldehyde
benzaldehyde
acetophenone
acetophenone
hexanone
hexanone
levulinic acid
levulinic acid
H
2
4
Scheme 2. Proposed general mechanism for the stoichiometric ionic hy-
[
22]
drogenation of ketones.
H
2
4
H
2
4
hydrogen generation is very fast. Therefore, it is conceivable
that the hydrogenation of cyclohexanone in the presence of
iron in an aqueous acid solution proceeds via an ionic hy-
drogenation.
H
2
4
Conditions: 0.5 g iron (9 mmol), 3.4 mmol substrate, 5 mL of 1.4m
HCOOH or H SO solution (7 mmol acid), T=508C, t=2 h.
2
4
The oxidation of metallic iron is apparently coupled to
the reduction of protons to form transient surface-bound
[21]
tistep process (hydridic route). The direct H-transfer pro-
ceeds via the formation of a complex in which both the
donor and the substrate are bound to the metal, and subse-
quent insertion of the ketone into the MÀH bond is re-
quired. If it is assumed that this is the case in the current
system, then both the acid and water molecules, as well as
the cyclohexanone, would need to bind to the metal surface
at the site(s) of hydrogenation, thereby competing with the
formation of dihydrogen.
Scheme 3. Proposed mechanism for the formation of transient iron-hy-
dride species capable of ketone reduction.
By contrast, the hydridic route involves the formation of
an intermediate metal hydride, followed by hydride transfer
to the substrate. Thus, the H-donor and the substrate inter-
act separately with the metal at different stages of the reac-
tion. Transition metals generally accomplish the reduction
+
iron hydrides [FeÀH] (Scheme 3). This reactive species
either participates in the substrate reduction or reacts with
another proton in the system to generate dihydrogen, which
is no longer active for substrate reduction and escapes from
the liquid phase as gas, thus leaving iron(II)-ions. However,
the participation of water needs to be included as it was
shown that the hydrogen evolved originates from both the
acid and the water. Therefore, the iron hydride species, [FeÀ
[21]
via the hydridic route. However, although the scope and
utility of ionic hydrogenations is yet to be fully defined,
recent evidence increasingly suggests that such mechanisms
[22]
are becoming more widely recognized. For example, the
catalytic ionic hydrogenation of ketones occurs by proton
transfer to a ketone from a cationic metal dihydride, fol-
+
H] , might react with a water molecule producing an iron
hydride hydroxide complex [Fe(H)(OH)], which might also
be formed directly from the interaction of iron with a water
molecule. This species then either reduces the substrate or
decomposes during dihydrogen generation, the iron remain-
ing as iron(II) oxide. Regardless of which process occurs
(possibly both do), the formation of a metal hydride species
is required to act as a hydrogen donor, finally generating an
[22]
lowed by hydride transfer from a neutral metal hydride.
The use of this process for stoichiometric hydrogenation
of a variety of C=C and C=O bonds has also been reported.
[23]
Kursanov and co-workers developed ionic hydrogenations
using stoichiometric quantities of CF CO H as the acid cou-
3
2
pled with HSiEt3 as the hydride-transfer reagent. Several
studies have demonstrated that stoichiometric hydrogena-
tion of C=O bonds is possible by metal hydrides upon reac-
II
iron salt, in which iron has been oxidized to Fe . Hence
“nascent hydrogen” is not hydrogen at all, but most likely
a metal-hydride or hydride–hydroxide complex generated
from both the acid and water molecules.
[24,25]
tion with acids.
Furthermore, previous studies of the ki-
netics of the stoichiometric ionic hydrogenation of alde-
hydes and ketones revealed that the rate of hydrogenation
decreased as the reaction proceeded, due to the progressive
It should be noted that although, in principle, it is possible
for iron(0) to participate in a redox process to generate an
[
26]
2+
consumption of the acid.
Bullock et al.
The mechanism proposed by
intermediate [Fe(H) ] complex, no evidence for the for-
2
[22]
IV
is consistent with all the observations in-
mation of Fe materials was found when examining the
volving a pre-equilibrium proton transfer to the substrate,
followed by rate-determining hydride transfer, as shown in
Scheme 2.
These observations are in good agreement with the pres-
ent results, which show that the reaction slows upon the acid
consumption or exhaustion at higher temperature, when the
mixture. Furthermore, their role in the reaction is inconsis-
tent with the stoichiometric investigations that indicated
that 1 mole of iron produces 1 mole of hydrogen. Therefore
1 mole of iron salt is expected.
Two possible paths are conceivable for the reduction of
cyclohexanone (Scheme 4). Protonation of either cyclohexa-
&
&
&6
&
www.chemasianj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 0000, 00, 0 – 0
ÝÝ These are not the final page numbers!

Exploring the Myth of Nascent Hydrogen
posed as the common intermediate for both substrate reduc-
tion and dihydrogen formation, which conclusively obviates
any need for the notion of “nascent hydrogen”.
Finally, this chemistry needs to be considered when de-
signing HTU cascades and the iron in the reactor walls
needs to be protected in some way. Alternatively, the reac-
tivity of iron might be employed usefully throughout the
cascading system. Studies in this regard are the subject of
further investigations.
Experimental Section
Scheme 4. Possible mechanisms for cyclohexanone reduction at an iron
surface based on analogous, homogeneous reaction steps.
Cyclohexanone, cyclohexanol, n-decane (Sigma–Aldrich), and formic
(
Riedel-de Haen), acetic, sulfuric (Merck), hydrochloric (Ajax Finechem
Pty Ltd), and lactic acids (all 99% purity), as well as iron powder
Merck) were used without further purification. Quantitative analysis of
the organic compounds was performed using a Shimadzu GC-17A Gas
Chromatograph (GC) fitted with Restek column (RTX5, 30 mꢂ
.25 mmꢂ0.25 mm) and equipped with a flame ionization detector (FID)
(
none (Scheme 4a) or iron (Scheme 4b) may occur as a first
step. When cyclohexanone is protonated, a hydride transfer
is required for the formation of cyclohexanol. Both iron hy-
a
0
using n-decane as an internal standard. All mass balances (relative to cy-
clohexanone) were quantitative within Æ10% or better (most were quan-
titative within Æ5% or better). Infrared spectra were recorded with a Re-
actIR 1000 Reaction Analysis System fitted with a high-pressure silicon
+
dride [FeÀH] and [Fe(H)(OH)] may act as a hydride-
+
donor. In addition, hydride transfer from [FeÀH] can
occur, followed by the protonation of the intermediate iron
complex, thus generating cyclohexanol. As stoichiometric
ionic hydrogenation with a metal hydride as both a proton
(
SiComp) probe head. In-situ high-pressure IR experiments were carried
out in a custom-manufactured 300 mL stainless steel reactor equipped
with a high-pressure silicon (SiComp) probe head and high-pressure gas
system. Water contents were determined by Karl-Fischer titration.
[22]
and hydride donor has been reported,
it is conceivable
that the [Fe(H)(OH)] complex could reduce cyclohexanone.
The reduction might proceed via a concerted hydrogen-
Hydrogenation of Cyclohexanone under HTPW Conditions using
Different Acids in a High-Pressure Sapphire NMR Tube
[27]
transfer mechanism as it has been suggested in the case of
A
dilute aqueous acid solution (0.5–2m, 2 mL) of cyclohexanone
[28]
the homogeneous Shvo catalyst.
(
120 mg) and n-decane (72 mg) as an internal standard were measured
Hence, the existence of “nascent hydrogen” can be ex-
plained by the formation of a reactive, surface-bound iron-
hydride complex. Given that the term “nascent hydrogen”
can have a connotation of some form of dihydrogen, its use
is misleading. The reduction of the substrate molecule com-
petes with the evolution of dihydrogen gas, which itself is
not active for substrate reduction in the presence of metallic
iron or iron salts.
into a 7 mL sapphire high-pressure NMR tube with a titanium top. Iron
powder (250–550 mg) was added to the reaction mixture. The tube was
closed quickly and heated to 2508C in a sand bath. After a reaction time
of 4 h, the tube was cooled to ambient temperature. Significant pressure
build-up was observed upon opening. The reaction mixture was extracted
with dichloromethane (3ꢂ5 mL), separated, and the organic phase was
dried over sodium sulfate. After filtration, samples were taken and ana-
lyzed by GC. Quantification was achieved by comparing integration
values to those derived from standard curves (5 different concentrations)
for the starting material and products, using authentic samples. In each
GC run, n-decane was used as an internal standard. The injector and de-
tector ports were maintained at 2508C. A heating profile of 408C for
Conclusions
5 min (isothermal), followed by an increase to 1308C at a rate of
À1
1
08Cmin was employed. Retention times were: cyclohexene 5.7; cyclo-
hexanol 11.1; cyclohexanol 11.4; and n-decane 14.5 min.
Aqueous solutions of several mineral and organic acids are
suitable for the hydrogenation of cyclohexanone in the pres-
ence of iron powder under both HTPW and ambient condi-
tions. Investigations of the stoichiometry of the reaction
show that one mole of iron reacts with one mole of monop-
rotic acid to either produce dihydrogen or to reduce one
mole of cyclohexanone. Additionally, water is necessary for
the reduction. The hydrogen gas produced originates from
both the acid and water molecules. Reducing the rate of hy-
drogen gas generation, by decreasing the speed of stirring
and/or the reaction temperature, results in a large improve-
ment in the conversion of cyclohexanone. Sulfuric and
formic acids are suitable for the hydrogenation of different
substrates, with the reduction being selective for aldehydes
and ketones. An ionic hydrogenation mechanism is present-
ed and importantly, a reactive iron-hydride species is pro-
Hydrogenation of Cyclohexanone under Ambient Conditions using
Different Acid Solutions
Dilute aqueous acid solutions (0.9–2.7m, 5 mL) of cyclohexanone
(330 mg) and n-decane (180 mg) as an internal standard were measured
into a 25 mL round-bottomed flask fitted with a reflux condenser. The re-
action mixture was stirred and heated to 25–1008C. After rapid addition
of 250–500 mg of iron at the reaction temperature, the stirring was con-
tinued at this temperature for 1–5 h. The reaction mixtures were separat-
ed by extraction with dichloromethane (2ꢂ15 mL) and the extracts dried
over sodium sulfate. After filtration, the samples were analyzed by GC.
To select different gaseous atmospheres under which to perform the reac-
tion, the headspace in the reaction apparatus was purged with the rele-
vant gas; to this end, the condenser tap was closed and a balloon filled
with either air, nitrogen, or hydrogen was attached to the flask.
When aiming for a slow evolution of hydrogen, the reduction was carried
out by the addition of either the acid or the metallic iron in 6 small por-
tions, at half-hourly intervals, to the reaction mixture. When sulfuric acid
Chem. Asian J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
7
&
&
&
These are not the final page numbers! ÞÞ

Thomas Maschmeyer et al.
FULL PAPER
(
860 mg, 98% w/w) was added in portions, the reaction mixture con-
tained cyclohexanone (330 mg), iron powder (500 mg), and n-decane
180 mg) as an internal standard. In the case of iron (500 mg) added in
when the rotating spoon was turned to pour the iron powder into the
acid solution.
(
For monitoring the volume of the produced hydrogen gas, the mercury
level in the burette and the manometer was continuously adjusted by
moving the mercury holder. The volume of the gas was read and record-
ed at certain intervals until completion of the reaction. The maximum de-
tectable gas volume with our burette system was 25 mL. For determina-
tion of the exact amount of hydrogen after the total consumption of iron
powder, the reaction mixture was vigorously stirred for an additional 1 h
after the reaction to expel all the remaining dissolved hydrogen in the so-
lution to the gas burette. For substrate hydrogenation under these condi-
tions, cyclohexanone (1 g) was also added to the acid solution before
closing the system.
portions, the reaction mixture contained cyclohexanone (330 mg), sulfuric
acid (860 mg, 98% w/w), and n-decane (180 mg) as an internal standard.
To test different substrates for the reduction, 3.4 mmol of substrate (cy-
clohexene, toluene, benzaldehyde, acetophenone, hexanone, or levulinic
acid) were used in place of cyclohexanone.
Hydrogenation of Cyclohexanone under Ambient Conditions using
Different Solvents
To investigate the reduction of cyclohexanone in non-aqueous media, ex-
periments were carried out in different solvents. Previously dried ethanol,
methanol, 2-propanol, or dimethyl sulfoxide (5 mL) was measured into
a 25 mL round-bottomed flask fitted with a reflux condenser under a ni-
trogen atmosphere. Cyclohexanone (330 mg), n-decane (180 mg) as an in-
ternal standard, and concentrated formic acid (410 mg, water content
+
Determination of the H Ion Concentration in the Acid Solution Before
and After the Reaction
+
The H concentration was determined by a simple titration method.
1
8
.85%) was added. The reaction mixture was stirred and heated to 70–
08C. After rapid addition of 500 mg of iron, the stirring was continued
Formic acid and sulfuric acid were both tested in the hydrogen evolution
measurement. The titration was carried out by the general acid-base neu-
tralization methods using sodium hydroxide (0.2m) as a base. In the case
of sulfuric acid, methyl orange was applied as an indicator, while phe-
nolphthalein was used in the case of formic acid.
at the reaction temperature for 2 h. The reaction mixtures were worked-
up by the same method described above.
To investigate the effect of water concentration, different water/ethanol
mixtures (5 mL, 0–100% w/w water content) were measured into
a 25 mL round-bottomed flask. Formic acid (410 mg, water content
1
.85%), cyclohexanone (330 mg), and n-decane (180 mg) as an internal
standard were added before the addition of the iron powder.
Investigating the Effect of Cyclohexanone on Hydrogen Gas Evolution
Acknowledgements
The reaction vessel used was a gas-tight, double-walled round-bottom,
cylindrical quartz reactor fitted with a Dresher-type gas bubbler via
a gas-tight B24 Quickfit joint. Coolant water maintained at 258C was
pumped through the outer compartment with a Julabo Thermostat/Circu-
lator. Formic acid solution (15 mL of 0.088m) was poured into the reac-
tion vessel, equipped with an additional stir bar. Argon was continuously
bubbled through the magnetically stirred solution at a controlled flow
The authors thank the Australian Research Council for funding
DP0987166) and (T.M.) for a Future Fellowship (FT0990485), We are
also grateful to Istvan T. Horvꢀth for the loan of a high-pressure sapphire
reactor and Lꢀszlꢄ T. Mika for providing an infra red spectroscopy.
(
À1
rate of 27.4 mLmin . This flow of argon carried any hydrogen formed in
[
[
1] F. Laborda, E. Bolea, M. T. Baranguan, J. R. Castillo, Spectrochim.
Acta Part B 2002, 57, 797–802.
2] a) T. E. Thorpe, J. Chem. Soc. Trans. 1882, 41, 287–296; b) H. Cav-
endish, Philos. Trans. R. Soc. London 1766, 56, 141–184.
3] C. A. Wurtz, The Atomic Theory, Appleton, New York, 1881.
4] J. Osann, Prakt. Chem. 1853, 58, 385–391.
5] F. Paneth, E. Klever, K. Peters, Z. Elektrochem. Angew. Phys.
Chem. 1927, 33, 102–107.
6] D. Tommasi, Ber. Dtsch. Chem. Ges. 1878, 11, 345.
7] D. Tommasi, Ber. Dtsch. Chem. Ges. 1878, 12, 1701.
8] D. Tommasi, Chem. Chem. News J. Industrial Science 1880, 40, 171.
9] W. A. Noyes, J. Am. Chem. Soc. 1901, 23, 460–463.
10] J. H. Reedy, E. D. Biggers, J. Chem. Educ. 1942, 403–406.
11] J. Meija, A. D’Ulivo, Anal. Bioanal. Chem. 2008, 392, 771–772.
12] G. F. Pavelko, Dopov. Nats. Akad. Nauk Ukr. 2005, 9, 157–158.
13] A. Gaiffe, R. Pallaud, Compt. Rend. 1962, 254, 3099–3100.
14] M. F. Abidova, V. K. Pitsaris, Dokl. Akad. Nauk UzSSR 1968, 25,
the headspace of the reactor out through the bubbler, gas-tight brass
tubing, and a moisture trap containing activated 3 ꢃ molecular sieve pel-
lets (~50 g, Fluka) into an online gas chromatograph (Shimadzu, GC-
1
7A) for the quantification of the hydrogen gas. The GC was custom-
[
[
[
modified for gas injection and analyses as follows. Attached to the GC
was a 6-port sampling and switching valve with a 5 mL sample loop,
through which the headspace gas from the reactor flowed continuously.
To analyze the headspace gas, the valve was switched such that the gas
aliquot in the sample loop was carried with argon into the GC, which
was equipped with a stainless steel-packed column (Alltech, 1/8 in. diam-
eter, 6 ft, containing 5 ꢃ 60/80 mesh molecular sieve), maintained at
[
[
[
[
4
08C and a thermal conductivity detector (TCD) maintained at 2008C.
[
[
[
[
[
Calibration of the GC was carried out using accurately prepared mixtures
of hydrogen in argon (BOC).
Monitoring the Exact Amount of Hydrogen Gas Produced During the
Treatment of Aqueous Acid Solution with Iron
3
6–37.
Using the equipment shown schematically in Figure 6, the exact amount
of gas generated or consumed during the reaction was determined. A
double-walled quartz reactor was filled with aqueous acid solution (0.8–
[15] D. Kondrat’ev, Khim. Zhizn 1981, 9, 72–73.
[16] V. V. Stetsik, Khim. Zhizn 1981, 9, 73–74.
[17] F. Mondragon, H. Itoh, K. Ouchi, Proc. Intersoc. Energy Convers.
Eng. Conf. Proc. Int. Conf. Coal Sci. 1983, 18–21.
2
m, 15 mL), and iron powder (30–110 mg) was measured into the rotata-
ble spoon (Figure 1) on the side of the reactor. The reactor was fitted
with a reflux condenser, purged with hydrogen by closing the taps of the
burette T2 and manometer T3, where the mercury level was the highest,
to avoid the presence of air. After the whole system (the reactor, burette,
and manometer) was filled with hydrogen by opening T2 (3. state) and
T3, the mercury and octane levels were adjusted to the same height.
Both the reactor and the burette were thermostated separately. The bu-
rette temperature was maintained at 258C in all cases, while the reactor
was thermostated to 25–458C. After the temperature remained constant
for 15 min, the octane levels in the two stems of the manometer (M)
were balanced by moving the mercury holder (MH). The reaction started
[18] V. Fꢀbos, A. F. Masters, T. Maschmeyer, Chem. Asian J. 2012, DOI:
10.1002/asia.2012000556.
[19] R. C. Crittendon, E. J. Parsons, Organometallics 1994, 13, 2587–
2591.
[20] Lj. M. Vra cˇ ar, D. M. Dra zˇ i c´ , Corros. Sci. 2002, 44, 1669–1680.
[21] S. Gladiali, E. Alberico, Chem. Soc. Rev. 2006, 35, 226–236.
[22] R. M. Bullock, Chem. Eur. J. 2004, 10, 2366–2374.
[23] D. N. Kursanov, Z. N. Parnes, N. M. Loim, Synthesis 1974, 633–651.
[24] P. L. Gaus, S. C. Kao, K. Youngdahl, M. Y. Darensbourg, J. Am.
Chem. Soc. 1985, 107, 2428–2434.
[25] D. H. Gibson, Y. S. El-Omrani, Organometallics 1985, 4, 1473–1475.
&
&
&8
&
www.chemasianj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 0000, 00, 0 – 0
ÝÝ These are not the final page numbers!

Exploring the Myth of Nascent Hydrogen
[
[
26] J.-S. Song, D. J. Szalda, R. M. Bullock, C. J. C. Lawrie, M. A.
Rodkin, J. R. Norton, Angew. Chem. 1992, 104, 1280–1282; Angew.
Chem. Int. Ed. Engl. 1992, 31, 1233–1235.
27] C. P. Casey, S. W. Singer, D. R. Powell, R. K. Hayashi, M. J. Kavana,
J. Am. Chem. Soc. 2001, 123, 1090–1100.
[28] Y. Shvo, D. Czarkie, Y. Rahamim, J. Am. Chem. Soc. 1986, 108,
7400–7402.
Received: June 22, 2012
Published online: && &&, 0000
Chem. Asian J. 2012, 00, 0 – 0
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
9
&
&
&
These are not the final page numbers! ÞÞ

FULL PAPER
Biomass Conversion
Ironing out the problem of “nascent
hydrogen”: Aqueous acid solutions are
suitable for hydrogenation of ketones
and aldehydes in the presence of iron
powder under both HTPW and ambi-
ent conditions. Investigations on the
stoichiometry of the reactions show
that one mole of iron reacts with one
mole of monoprotic acid to reduce one
mole of substrate. Additionally, water
is necessary for the reduction. The
hydrogen gas produced originates from
both the acid and the water molecules.
The nature of “nascent hydrogen” is
revealed and an ionic hydrogenation
mechanism proposed.
Viktꢀria Fꢁbos, Alexander K. L. Yuen,
Anthony F. Masters,
Thomas Maschmeyer*
&&&& — &&&&
Exploring the Myth of Nascent
Hydrogen and its Implications for Bio-
mass Conversions
&
&
&10
&
www.chemasianj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 0000, 00, 0 – 0
ÝÝ These are not the final page numbers!