Vapour phase transfer hydrogenation of α,β-unsaturated carbonyl compounds. Thermodynamic and experimental studies

Article abstract of DOI:10.1016/j.apcata.2015.11.046

This paper presents the first systematic thermodynamic study of the vapour phase transfer hydrogenation of α,β-unsaturated carbonyl compounds at temperatures: 423.15-723.15 K. Calculations were made for four compounds, namely: acrolein, α-methylacrolein, β-methylacrolein and methyl vinyl ketone. The Gibbs free energies and equilibrium mole fractions (EMFs) were calculated for transfer hydrogenation with ethanol and 2-propanol as hydrogen donors. It was noted that for transfer hydrogenation and hydrogenation with hydrogen the formation of the unsaturated alcohol (UOL) is the least thermodynamically favoured reaction and that saturated alcohol (SOL) and saturated aldehyde or ketone (SAL or SON) are the main products. A set of eight carbonyl compounds have been transfer hydrogenated with ethanol and 2-propanol in the presence of MgO as the catalyst. The main conclusions are that: (a) the reduction of a carbonyl group into a carbinol group occurs with a very high selectivity, (b) for almost all carbonyl compounds, except acrolein, the reactivity of 2-propanol highly exceeded that shown by ethanol and (c) the high chemoselectivity of transfer hydrogenation of acrolein with alcohols resulted from the kinetic control caused by the presence of magnesium oxide.

Full text of DOI:10.1016/j.apcata.2015.11.046

Applied Catalysis A: General 511 (2016) 131–140
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Vapour phase transfer hydrogenation of ␣,␤-unsaturated carbonyl
compounds. Thermodynamic and experimental studies
∗
Marek Gli n´ ski , Urszula Ulkowska
Warsaw University of Technology (Politechnika), Faculty of Chemistry, Chair of Chemical Technology, Noakowskiego 3, 00-664 Warsaw, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 7 August 2015
Received in revised form 5 November 2015
Accepted 29 November 2015
Available online 4 December 2015
This paper presents the first systematic thermodynamic study of the vapour phase transfer hydrogena-
tion of ␣,␤-unsaturated carbonyl compounds at temperatures: 423.15–723.15 K. Calculations were made
for four compounds, namely: acrolein, ␣-methylacrolein, ␤-methylacrolein and methyl vinyl ketone. The
Gibbs free energies and equilibrium mole fractions (EMFs) were calculated for transfer hydrogenation
with ethanol and 2-propanol as hydrogen donors. It was noted that for transfer hydrogenation and hydro-
genation with hydrogen the formation of the unsaturated alcohol (UOL) is the least thermodynamically
favoured reaction and that saturated alcohol (SOL) and saturated aldehyde or ketone (SAL or SON) are
the main products.
Keywords:
Thermodynamic analysis
␣
,␤-unsaturated carbonyl compounds
A set of eight carbonyl compounds have been transfer hydrogenated with ethanol and 2-propanol
in the presence of MgO as the catalyst. The main conclusions are that: (a) the reduction of a carbonyl
group into a carbinol group occurs with a very high selectivity, (b) for almost all carbonyl compounds,
except acrolein, the reactivity of 2-propanol highly exceeded that shown by ethanol and (c) the high
chemoselectivity of transfer hydrogenation of acrolein with alcohols resulted from the kinetic control
caused by the presence of magnesium oxide.
Hydrogenation
Transfer hydrogenation
Chemoselectivity
©
2015 Elsevier B.V. All rights reserved.
1
. Introduction
of hydrogenation of crotonaldehyde [1]. In the article, the authors
presented a figure which enables a rough estimation of the Gibbs
free energy change of the mentioned reactions at a given tempera-
ture (298, 400, 600 and 700 K). However, no method of calculation
was cited and no thermodynamic data were shown.
Chemoselective hydrogenation of ␣,␤-unsaturated carbonyl
compounds to the corresponding unsaturated alcohols (UOLs) is
a topic of many recent studies, due to the fact that these alcohols
are very important and versatile intermediates for pharmaceuti-
cal, polymer, fragrance and food industries. The reduction of these
unsaturated carbonyl compounds by gaseous dihydrogen over
transition or noble metal catalysts usually results in hydrogenation
of the C C bond. This happens, according to literature, because of
two main reasons, first, thermodynamics favors the hydrogenation
of the C C bond over the C O bond by c.a. 35 kJ mol [1] and sec-
ond, the C C bond shows higher reactivity than the C O bond [2].
Although in many articles from the field of heterogeneous cataly-
sis both statements are frequently repeated [3–6] it seems that the
former statement has its origin in only one article about selectivity
Recently, the latter case for the preference of hydrogenation of
the C C bond over the C O bond was cited in literature and a value
−
1
was given (127.0 kJ mol ) [7], however, no details were provided.
In our opinion, the value must be erroneous in the light of the data
presented above and by us in this work.
Catalytic transfer hydrogenation (CTH) is a reaction in which one
organic compound, called the hydrogen donor (D), gives a pair of
its hydrogen atoms to another compound (hydrogen acceptor—A)
in the presence of a catalyst (Scheme 1). It is a convenient way to
reduce many organic compounds for three reasons. First, gaseous
hydrogen is not used as the reductant; second, the reaction is per-
formed under normal pressure, which excludes using high pressure
equipment; and third, high selectivities towards the desired prod-
ucts are very often observed. On the other hand, the high selectivity
is connected with the restriction of the type of organic groups which
can be reduced by this method. Among functional groups prone to
the reduction by CTH, the reduction of the carbonyl group in vari-
ous saturated aldehydes and ketones by alcohols used as hydrogen
donors in the presence of heterogeneous catalysts is very well doc-
−
1
Abbreviations: A, hydrogen acceptor; CTH, catalytic transfer hydrogenation;
D, hydrogen donor; D/A, donor/acceptor molar ratio; EMF, equilibrium mole frac-
tion; SAL, saturated aldehyde; SOL, saturated alcohol; SON, saturated ketone; UAL,
unsaturated aldehyde; UOL, unsaturated alcohol; UON, unsaturated ketone.
∗ Corresponding author. Fax: +48 22 628 27 41.
E-mail address: marekg@ch.pw.edu.pl (M. Gli n´ ski).
http://dx.doi.org/10.1016/j.apcata.2015.11.046
0
926-860X/© 2015 Elsevier B.V. All rights reserved.

1
32
M. Gli n´ ski, U. Ulkowska / Applied Catalysis A: General 511 (2016) 131–140
Scheme 1. Catalytic transfer hydrogenation (CTH). Hydrogen acceptor—e.g., carbonyl compound, hydrogen donor—e.g., aliphatic alcohol.
umented [8–13]. Therefore, it has been anticipated that the ease of
the reduction of the carbonyl group in the CTH might be preserved
also for ␣,␤-unsaturated carbonyl compounds, which could result
in a high chemoselectivity towards unsaturated alcohols. Indeed,
CTH of ␣,␤-unsaturated carbonyl compounds very often leads to
the formation of unsaturated alcohols as the main products but
the reaction chemoselectivity is not always as high as expected
desired products under given reaction conditions, to understand
some peculiarities occurring in a set of reaction pathways of
CTH between ␣,␤-unsaturated carbonyl compounds and ethanol
or 2-propanol, and to find plausible generalizations. To authors’
knowledge such a description has not been published yet. The
present work fills this gap.
The second aim of this work was to look closer at the unpre-
dictably (by thermodynamics) low reactivity of 2-propanol shown
in the presence of magnesium oxide in the transfer hydrogenation
of acrolein than that of ethanol. Another aspect of the studies was
to compare the reactivities of both alcohols in transfer hydrogena-
tions of many other carbonyl compounds in aim to find plausible
regularities.
[
14–16].
Among hydrogen donors the secondary alcohols are mainly
used, with 2-propanol as the most frequent one, although pri-
mary alcohols: ethanol, 1-butanol or even methanol have also been
studied [17]. The explanation of the popularity of 2-propanol is
the result of its accessibility, a low price, low toxicity, as well as
relatively high volatility, which is a great advantage during the sep-
aration of products. What is also important, the alcohol belongs to
a group of secondary alcohols which seem to be better hydrogen
donors than primary ones.
2. Experimental
2.1. Thermodynamic calculations
Although in literature there are publications in which the higher
reactivity of ethanol than 2-propanol in the transfer hydrogena-
tion of acrolein in the presence of MgO has been documented,
the authors did not discuss this fact at all, probably due to a
modest knowledge about the thermodynamic description of such
reactions at that time [17,18]. In the light of the difficulties met
during interpretation of the results of CTH tests with the participa-
tion of various ␣,␤-unsaturated carbonyl compounds taken from
literature and ours, we find our own very preliminary thermody-
namic description of the transfer hydrogenation of acrolein with
various alcohols which have been published recently [19] insuffi-
cient. First, in the mentioned description each reaction has been
considered separately, which, of course, is only a very rough esti-
mation, and does not enable the calculation of the thermodynamic
compositions of the reactants under equilibrium. Second, for our
calculations of the transfer hydrogenation we have chosen only one
temperature (673 K), which is too high for most hydrogenations of
multiple carbon–carbon or carbon–oxygen bonds. Third, we have
made our calculations only for acrolein, so it is impossible to draw
general conclusions about the reduction of a whole group of ␣,␤-
unsaturated carbonyl compounds. Unfortunately, there have been
no previous reports of thermodynamics of transfer hydrogenation
of ␣,␤-unsaturated carbonyl compounds with alcohols as hydro-
gen donors, so the observed experimental phenomena could not
be explained on the basis of thermodynamic considerations.
In short, it can be seen that the thermodynamic background
for the hydrogenation/transfer hydrogenation of ␣,␤-unsaturated
carbonyl compounds cited in literature is scarce and uncertain.
We believe that for designing heterogeneous catalytic systems
which would chemoselectively hydrogenate the C O bond in
The thermodynamic calculations leading to the evaluation of
the Gibbs function (ꢀG), the equilibrium constant (K) and equilib-
0
2
rium mole fractions (EMFs) were based on data (enthalpies ꢀH 98,
entropies S 02 98, and molar heat capacities Cp) received from NIST
database [20], from the group contribution methods [21–23] and
from the experimental data found in literature [24]. The missing
data were calculated using known methods [21,22]. The details of
calculations together with basic thermodynamic data are given in
the supplementary materials.
Our calculations were made for the temperature range of
23.15–723.15 K, for normal pressure, which corresponds to typ-
ical conditions at which all vapour phase transfer hydrogenation
reactions were performed.
4
2
.2. Catalyst preparation and characterization
Magnesium oxide was prepared by thermal decomposition
of Mg(OH)2 whose preparation is described elsewhere [19]. The
sieved fraction of the hydroxide of 0.16–0.40 mm was calcined first
at 873 K for 1 h in a stream of dry air, then for 5 h in a stream of
dry deoxygenated nitrogen and stored under nitrogen. The spe-
2
−1
3
−1
cific area of MgO was 99.7 m g , the pore volume 0.529 cm g .
The detailed characterization of thus prepared MgO is given in a
previous paper for the same batch of catalyst [19].
2
.3. Reagents
Acetophenone (99%), benzaldehyde (>99%), cyclohexanone
>99%), hexanal (97%), 5-hexen-2-one (allylacetone) (99%) and
-methylbutan-2-one (methyl isopropyl ketone, 99%) were all
(
3
␣
,␤-unsaturated carbonyl compounds, a deeper understanding
of thermodynamics is needed. This is why the principal aim
of the paper was to perform a comprehensive thermodynamics
analysis of hydrogenation/transfer hydrogenation reactions, with
gaseous dihydrogen or alcohols as reductants, of four chosen ␣,␤-
unsaturated carbonyl compounds. The obtained thermodynamic
data would be very helpful to determine equilibrium yields of
Aldrich products. They were dried over anhydrous MgSO , distilled
4
under normal/reduced pressure in the atmosphere of dry nitrogen
and kept under nitrogen in Schlenk-type containers with the excep-
tion of both aldehydes for which the second distillation was made to
remove contamination with carboxylic acids formed during contact
withatmosphericoxygenand the productswere distilleddirectlyto

M. Gli n´ ski, U. Ulkowska / Applied Catalysis A: General 511 (2016) 131–140
133
Table 1
List of possible reactions (R, R’ and R” = H or Me).
Number
Reaction
1
2
3
4
5
6
7
7
8
8
9
9
1
1
E
P
E
P
E
P
0E
0P
Schlenk-type containers. Benzophenone (99%, Aldrich) was twice
crystallized from ethanol. Acrolein (90%, Aldrich) was purified as
described before [19] and stored under nitrogen in a Schlenk-type
container at 243 K in a freezer. Ethanol (p.a., anhydrous 99.8%), 2-
propanol (p.a.) both from POCh Gliwice, Poland, and 1-propanol
(99%, Aldrich) were distilled over metallic sodium and kept dry in
Schlenk-type containers under nitrogen.

1
34
M. Gli n´ ski, U. Ulkowska / Applied Catalysis A: General 511 (2016) 131–140
Table 2
2.4. Catalytic activity measurements
Calculated Gibbs free energies for hydrogenation of acrolein, ␣-methylacrolein, ␤-
methylacrolein and methyl vinyl ketone with dihydrogen in gas phase under normal
pressure at various temperatures.
The measurements were carried out in a fixed-bed tubular
quartz reactor equipped with a concentric tube where a thermo-
couple was placed, and heated by an electric furnace. A sample of
the catalyst (250 ± 5 mg) was loaded into the reactor in a stream
of dry nitrogen. The activity measurements were performed in the
temperature range 423–573 K. A solution of a carbonyl compound
in ethanol or 2-propanol at a given molar ratio was dosed using
a microdosing pump with a Liquid Hourly Space Velocity (LHSV):
Compound
T
[K]
ꢀGr(i) [kJ mol−1]
(1)
(3)
(3 + 4)
4
23.15
−5.76
−58.79
−52.23
−45.58
−38.86
−32.07
−25.23
−18.35
−51.10
−44.22
−37.30
−30.38
−23.48
−16.62
−9.82
−79.74
−67.05
−54.20
−41.21
−28.12
−14.93
−1.67
473.15
523.15
0.50
6.87
573.15
623.15
673.15
723.15
13.31
19.83
26.41
33.04
−7.83
−1.47
4.98
11.50
18.07
24.70
31.37
−5.27
0.99
−
1
3
−
1
3
h
into a stream of nitrogen (50 cm min ) which was passed
through the catalyst bed. Before typical measurements of activity,
the catalyst was maintained for 60 min at the first (lowest) reaction
temperature, in the stream of reactants, and the products formed
during that initial period were discarded.
423.15
−73.26
−60.24
−47.07
−33.77
−20.37
−6.89
473.15
523.15
573.15
623.15
673.15
723.15
423.15
473.15
2
.5. Analytical determinations
6.65
−47.81
−41.46
−35.03
−28.54
−21.99
−15.39
−8.75
−67.76
−55.28
−42.65
−29.89
−17.03
−4.08
The reaction products were analysed by GC using HRGC KONIK
(
0
Spain) equipped with a TRACER wax capillary (length 30 m,
.25 mm i.d.) and an FID. A selected n-alkane, depending on the
volatility of the reactants, from the C –C range was used as
523.15
573.15
7.36
13.81
20.33
26.92
33.56
-3.19
2.57
623.15
673.15
723.15
423.15
8
14
an internal standard. Compounds were identified by GC–MS (HP-
890N with a 5973N mass detector) and by comparison of the
retention time with that of the internal standard.
8.93
6
−75.28
−69.33
−63.30
−57.17
−50.97
−44.69
−38.33
−84.16
−72.30
−60.30
−48.18
−35.96
−23.65
−11.25
473.15
523.15
573.15
623.15
673.15
723.15
8.35
14.13
19.92
25.71
31.50
3
. Results and discussion
In the work of Campo et al. [6] concerning the catalytic hydro-
genation of ␣,␤-unsaturated carbonyl compounds it is mentioned
that thermodynamics favours the hydrogenation of the C C bond
•
•
•
only the C O bond (1);
only the C C bond (3);
both C C and C O bonds (3 + 4).
−
1
over the C O bond by approximately 35 kJ mol
[1]. The same
value has been cited in other works [3–5]. However, our calcula-
tions made at the early stage of this work for the hydrogenation of
␤
-methylacrolein (crotonaldehyde), as it was published in the orig-
The values of the Gibbs free energy changes for all elementary
−
1
inal work of Vannice and Sen [1], give values 42.5 and 42.3 kJ mol
reactions: the formation of unsaturated alcohol (UOL) (1), saturated
aldehyde/ketone (SAL/SON) (3) and saturated alcohol (SOL) (3 + 4),
carried out at seven different temperatures, from 423.15 to 723.15 K
with a 50 deg step, are summarized in Table 2. It can be seen that for
all three reactions, the Gibbs free energy values increase with tem-
perature, which means that the reactions are becoming less and less
thermodynamically favourable. Reaction (1) was found thermo-
dynamically favourable at temperatures no higher than 473.15 K,
except in the case of ␣-methylacrolein, for which a negative value of
the Gibbs free energy change was also noted below 532.15 K. Reac-
tion (3) is favoured in the whole range of temperatures. Reaction
for 423 and 723 K, respectively. Due to this discrepancy between
the value found in literature and the values calculated by us and
also the lack of data for hydrogenation of other compounds of this
type, we decided to extend our studies to other ␣,␤-unsaturated
carbonyl compounds.
Four ␣,␤-unsaturated carbonyl compounds were selected for
thermodynamic calculations: acrolein – the simplest unsaturated
aldehyde, its ␣- and ␤-methyl derivatives, and methyl vinyl ketone
–
the simplest unsaturated ketone (Scheme 2).
The list of equations which describe the possible reactions which
can occur when an ␣,␤-unsaturated carbonyl compound reacts
with gaseous dihydrogen or a hydrogen donor is given in Table 1.
Calculations for three groups of reactions were performed:
(
3 + 4) is in general thermodynamically favourable. It is only not
favoured in the case of ␣-methylacrolein and ␤-methylacrolein at
23.15 K.
7
Based on the results presented in Table 2, the exact temper-
•
•
•
hydrogenation of ␣,␤-unsaturated carbonyl compounds (Eqs.
–4);
dehydrogenation of alcohols: ethanol and 2-propanol (Eqs. 5 and
);
transfer hydrogenation of ␣,␤-unsaturated carbonyl compounds
with ethanol (Eqs. 7E, 8E, 9E and 10E) and with 2-propanol (Eqs.
atures at which ꢀGr(1) > 0, which means that the formation of
unsaturated alcohols (UOLs) is not thermodynamically favoured,
were obtained. They are: 469, 485, 465 and 451 K for acrolein,
-methylacrolein, ␤-methylacrolein and methyl vinyl ketone,
respectively.
1
6
␣
It has been stated in literature that thermodynamics favours
the hydrogenation of the C C bond over that of the C O bond by
7
P, 8P, 9P and 10P).
−
1
c.a. 35 kJ mol
[1]. According to our calculations, at 423.15 and
3.1. Thermodynamic analysis of the vapour-phase hydrogenation
723.15 K the appropriate values for acrolein, ␣-methylacrolein, ␤-
reaction of acrolein, ˛-methyl- and ˇ-methylacrolein, and methyl
vinyl ketone with dihydrogen
methylacrolein and methyl vinyl ketone are equal 53.0 and 51.4,
43.3 and 41.2, 42.5 and 42.3, and 72.1 and 69.8 kJ mol , respec-
tively.
−1
For hydrogenation of each ␣,␤-unsaturated carbonyl compound
three single reaction pathways were analysed, namely hydrogena-
tion of:
The results of the above mentioned calculations are in line with
those we obtained using the method of group contributions devel-
oped by Van Krevelen and Chermin [25]. For acrolein it gave us the

M. Gli n´ ski, U. Ulkowska / Applied Catalysis A: General 511 (2016) 131–140
135
Scheme 2. ␣,␤-Unsaturated carbonyl compounds–substrates in hydrogenation and transfer hydrogenation reactions.
Scheme 3. Catalytic transfer hydrogenation of acrolein with ethanol.
Table 3
Saturated alcohols (SOL-s) were dominating hydrogenation
products in a lower range of temperatures, at 423.15 K their EMFs
values exceeded 0.99 with the exception of methyl ethyl carbinol
for which this value reached c.a. 0.91. An increase of the EMFs values
for saturated aldehydes or ketones (SAL-s/SON-s) at the expense of
SOL-s has been noted with temperature. They become the main
products above 573 K or in the case of ethyl methyl ketone, already
at 523 K.
Summing up these results, it can be seen that the main prod-
ucts of the reactions are SAL/SON and SOL. Based on the analysis
of the obtained values for the four studied ␣,␤-unsaturated car-
bonyl compounds we have shown that the structure of a given
compound greatly influences the degree to which the hydrogena-
tion of the C C bond is preferred over that of the C O bond. It has
been concluded that temperature strongly influences the SAL/SOL
ratio in the hydrogenation of studied ␣,␤-unsaturated carbonyl
compounds.
Calculated EMFs for hydrogenation of acrolein, ␣-methylacrolein, ␤-methylacrolein
and methyl vinyl ketone with dihydrogen, in gas phase under normal pressure at
various temperatures. Molar ratio H2/UAL(UON) = 6.
Compound
T
[K]
EMFs
UAL/UON
UOL
SAL/SON
SOL
−4
−4
−4
−4
−4
−4
−
−
4
4
5
5
6
6
7
4
4
5
5
6
6
7
4
4
5
5
6
6
7
4
4
5
5
6
6
7
23.15
73.15
23.15
73.15
23.15
73.15
23.15
23.15
73.15
23.15
73.15
23.15
73.15
23.15
23.15
73.15
23.15
73.15
23.15
73.15
23.15
23.15
73.15
23.15
73.15
23.15
73.15
23.15
<10
<10
<10
<10
<10
<10
<10
<10
0.0032
0.0280
0.1462
0.4279
0.7210
0.8733
0.9019
0.0023
0.0208
0.1164
0.3760
0.6821
0.8275
0.7770
0.0043
0.0358
0.1771
0.4788
0.7491
0.8422
0.7566
0.0906
0.3665
0.7073
0.8883
0.9561
0.9806
0.9889
0.9968
0.9720
0.8538
0.5719
0.2772
0.1150
0.0468
0.9977
0.9792
0.8836
0.6231
0.3087
0.1209
0.0418
0.9957
0.9641
0.8228
0.5197
0.2377
0.0928
0.0333
0.9094
0.6335
0.2927
0.1117
0.0439
0.0190
0.0091
4
0.0002
0.0018
0.0116
0.0511
4
0.0001
0.0002
−4
−4
−4
−4
<10
<10
<10
<10
−4
−4
<10
<10
0.0008
0.0089
0.0510
0.1803
0.0001
0.0002
0.0005
0.0008
−
4
−4
<10
<10
<10
−4
−4
−
3.2. Thermodynamic analysis of the vapour-phase
dehydrogenation of ethanol and 2-propanol
<10
4
0.0001
0.0015
0.0130
0.0646
0.2095
<10
0.0001
0.0002
0.0004
0.0007
The dehydrogenation propensity of ethanol and 2-propanol
is the key to understanding their thermodynamic capability for
hydrogenation of organic acceptors (Table 4). It can be seen that the
dehydrogenation of both alcohols is not favoured at low tempera-
tures. The ꢀG value decreases with temperature and, as calculated,
becomes negative above 605.6 K for ethanol and 502.3 K for 2-
propanol. By comparison of these two values it can be concluded
that 2-propanol is more prone to dehydrogenation than ethanol, so
the former should be a better hydrogen donor. Indeed, the review
of publications concerning CTH has shown that 2-propanol is the
most frequently used hydrogen donor. Its popularity as the donor,
besides its low toxicity, and a low price, has a strong thermody-
namic background (Table 4).
−4
−4
−4
−4
−4
<10
<10
<10
<10
<10
−4
<10
−4
<10
−4
<10
−
−
−
4
4
4
0.0001
0.0004
0.0020
<10
<10
<10
EMF—equilibrium mole fraction; UAL—unsaturated aldehyde; UON—unsaturated
ketone; UOL—unsaturated alcohol; SAL—saturated aldehyde; SON—saturated
ketone; SOL—saturated alcohol.
following values: 57.2 and 57.9 kJ mol 1, at 423.15 and 723.15 K,
respectively.
−
3
.3. Thermodynamic analysis of the vapour-phase transfer
hydrogenation of acrolein, ˛- and ˇ-methylacrolein, and methyl
vinyl ketone with ethanol or 2-propanol as hydrogen donors
Calculations of equilibrium compositions of the reactants under
study in the temperature range of 423.15–723.15 K with a molar
ratio of hydrogen to the ␣,␤-unsaturated carbonyl compound equal
to 6 have been performed (Table 3). The equilibrium mole frac-
tions (EMFs) of the substrates in the postreaction mixture are
close to zero at low temperatures and increase insignificantly with
temperature. The EMFs for all UOLs are very low regardless of
the temperature and do not exceed 0.0008, so these products are
not favoured by thermodynamics. The EMFs for UAL-s (or UON-
s) clearly indicate that for acrolein and methyl vinyl ketone the
conversion is almost quantitative. In contrast, for the other com-
pounds, namely ␣-methylacrolein and ␤-methylacrolein, the EMFs
are larger and reach nearly 0.2 at the highest studied temperature.
In contrast to the hydrogenation with dihydrogen, in the ther-
modynamic analysis of the transfer hydrogenation reaction with
alcohols the Gibbs free energy of formation of a chosen hydro-
gen donor, as well as its dehydrogenation product (an aldehyde
or ketone) must be taken into account (Scheme 3).
The values of the Gibbs free energy change for all elementary
reactions are collected in Tables 5 and 6.
The analysis of the results of the calculations of the Gibbs free
energies of transfer hydrogenation with ethanol as the hydrogen
donor (Table 5) indicates that:

1
36
M. Gli n´ ski, U. Ulkowska / Applied Catalysis A: General 511 (2016) 131–140
Table 4
Table 6
Calculated Gibbs free energies (Eqs. 5 and 6) and EMFs for dehydrogenation of
ethanol and 2-propanol in gas phase under normal pressure at various temperatures.
Calculated Gibbs free energies for transfer hydrogenation of acrolein, ␣-
methylacrolein, ␤-methylacrolein and methyl vinyl ketone with 2-propanol in gas
phase under normal pressure at various temperatures.
−
Alcohol
T
[K] ꢀGr[kJ mol 1
]
EMFs
Compound
T
[K]
ꢀGr(i) [kJ mol−1]
Alcohol Acetaldehydea H2
(
7P)
(9P)
(9P + 10P)
EtOH
423.15 22.32
0.9196 0.0402
0.7772 0.1114
0.5426 0.2287
0.2918 0.3541
0.1258 0.4371
0.0504 0.4748
0.0210 0.4895
0.5880 0.2060
0.2958 0.3521
0.1124 0.4438
0.0402 0.4799
0.0156 0.4922
0.0068 0.4966
0.0032 0.4984
0.0402
4
5
5
6
6
7
73.15 16.28
23.15 10.18
0.1114
0.2287
0.3541
0.4371
0.4748
0.4895
0.2060
0.3521
0.4438
0.4799
0.4922
0.4966
0.4984
423.15
473.15
523.15
573.15
623.15
673.15
723.15
423.15
473.15
523.15
573.15
623.15
673.15
3.49
3.92
4.43
4.99
5.61
6.27
6.99
1.41
1.95
2.54
3.17
3.85
4.56
5.31
3.98
4.41
4.92
5.48
6.11
6.78
7.51
6.06
5.99
5.90
5.81
5.70
5.58
5.45
−49.55
−48.81
−48.02
−47.19
−46.30
−45.37
−44.40
−41.85
−40.80
−39.75
−38.71
−37.70
−36.75
−35.88
−38.57
−38.04
−37.48
−36.86
−36.21
−35.52
−34.80
−66.03
−65.91
−65.74
−65.50
−65.20
−64.83
−64.40
−61.25
−60.21
−59.08
−57.86
−56.57
−55.20
−53.78
−54.77
−53.40
−51.95
−50.42
−48.82
−47.17
−45.46
−49.27
−48.45
−47.53
−46.54
−45.48
−44.35
−43.18
−65.67
−65.46
−65.18
−64.83
−64.41
−63.92
−63.36
73.15
23.15
73.15
4.02
-2.17
-8.40
23.15 -14.65
2
-PrOH
423.15
9.25
3.42
-2.44
-8.33
4
5
5
6
6
7
73.15
23.15
73.15
23.15 -14.23
73.15 -20.14
23.15 -26.05
723.15
423.15
473.15
523.15
573.15
EMF—equilibrium mole fraction.
a
Acetone for 2-PrOH.
Table 5
623.15
673.15
723.15
423.15
Calculated Gibbs free energies for transfer hydrogenation of acrolein, ␣-
methylacrolein, ␤-methylacrolein and methyl vinyl ketone with ethanol in gas
phase under normal pressure at various temperatures.
4
5
5
6
73.15
23.15
73.15
23.15
Compound
T
[K]
ꢀGr(i) [kJ mol 1
−
]
(
7E)
(9E)
(9E + 10E)
423.15
473.15
523.15
573.15
623.15
673.15
723.15
423.15
473.15
523.15
573.15
623.15
673.15
723.15
423.15
473.15
523.15
573.15
623.15
673.15
723.15
423.15
473.15
523.15
573.15
623.15
673.15
723.15
16.56
16.78
17.04
17.34
17.66
18.02
18.40
14.48
14.81
15.15
15.52
15.90
16.30
16.72
17.05
17.27
17.53
17.83
18.16
18.53
18.91
19.13
18.85
18.52
18.16
17.75
17.32
16.86
−36.48
−35.95
−35.41
−34.84
−34.24
−33.63
−32.99
−28.79
−27.94
−27.13
−26.36
−25.65
−25.01
−24.47
−25.50
−25.19
−24.86
−24.51
−24.16
−23.78
−23.40
−52.96
−53.05
−53.12
−53.15
−53.14
−53.09
−52.98
−35.11
−34.50
−33.85
−33.17
−32.46
−31.72
−30.96
−28.63
−27.68
−26.71
−25.72
−24.71
−23.68
−22.64
−23.13
−22.73
−22.30
−21.84
921.36
−20.87
−20.36
−39.53
−39.74
−39.94
−40. 13
−40.30
−40.44
−40.54
673.15
723.15
•
•
•
irrespective of the type of an ␣,␤-unsaturated carbonyl com-
pound the formation of UOL is not thermodynamically favoured
in the whole range of temperatures (423.15–723.15 K);
for all ␣,␤-unsaturated aldehydes SOL is the most thermodynam-
ically favoured product, for methyl vinyl ketone SON and SOL are
practically equal thermodynamically favoured;
the values of the Gibbs free energies of transfer hydrogenation
with 2-propanol are always lower in comparison to the appropri-
ate values for ethanol. The gap equals 13.07 and 11.41 kJ mol
for 423.15 and 723.15 K, respectively.
−
1
Calculations of EMFs for transfer hydrogenation of studied
␣,␤-unsaturated carbonyl compounds with ethanol (Table 7) or
2-propanol (Table 8) revealed that at the temperature range of
23.15–723.15 K a quantitative conversion of every carbonyl com-
pound can be attained. This is accompanied by the presence of
SAL/SON and SOL as the only products. For ethanol as the hydrogen
donor SOL is the main product with the exception of methyl vinyl
ketone, for which SON dominates. For 2-propanol SOL is in every
case the main product, its EMF varies between 0.68 and 0.99. At
low temperatures (423.15–473.15 K) for ␣,␤-unsaturated aldehy-
des SOL is the only product; its EMF is in the range of 0.97–0.99.
4
•
•
•
irrespective of the type of an ␣,␤-unsaturated carbonyl com-
pound the formation of UOL is not thermodynamically favoured
in the whole range of temperatures (423.15–723.15 K);
the formation of SAL/SON and SOL are practically equally thermo-
dynamically favoured, with the exception of methyl vinyl ketone
for which SON predominates over SOL;
3
.4. Reactivity of ethanol and 2-propanol as hydrogen donors in
the vapour-phase CTH of various carbonyl compounds in the
presence of MgO
The reactivity of ethanol and 2-propanol as hydrogen donors
to eight carbonyl compounds (Scheme 4) has been studied in
the presence of MgO as the catalyst. Three of these compounds,
namely: hexanal, methyl isopropyl ketone and cyclohexanone, pos-
sess only a carbonyl group prone to hydrogenation. The next three,
namely benzaldehyde, acetophenone and benzophenone, also have
a phenyl substituent, which could be hydrogenated to a cyclohexyl
group, although such a reaction does not occur under the stud-
very weak dependence of all the Gibbs free energies of transfer
hydrogenation with ethanol on temperature has been observed.
Based on the data presented in Table 6, the following conclusions
have been put forward for 2-propanol as the hydrogen donor:

M. Gli n´ ski, U. Ulkowska / Applied Catalysis A: General 511 (2016) 131–140
137
Scheme 4. Carbonyl compounds used as hydrogen acceptors.
Table 7
Table 8
Calculated EMFs for transfer hydrogenation of acrolein, ␣-methylacrolein, ␤-
methylacrolein and methyl vinyl ketone with ethanol in gas phase under normal
pressure at various temperatures. Molar ratio EtOH/UAL(UON) = 6.
Calculated EMFs for transfer hydrogenation of acrolein, ␣-methylacrolein, ␤-
methylacrolein and methyl vinyl ketone with 2-propanol in gas phase under normal
pressure at various temperatures. Molar ratio 2-PrOH/UAL(UON) = 6.
Compound
T [K]
EMFs
Compound
T
[K]
EMFs
UAL/UON
UOL
SAL/SON
SOL
UAL/UON
UOL
SAL/SON
SOL
−
−
−
4
4
4
−4
−4
−4
−4
−4
4
4
5
5
6
6
7
4
4
5
5
6
6
7
4
4
5
5
6
6
7
4
4
5
5
6
6
7
23.15
<10
<10
<10
<10
<10
<10
<10
<10
<10
0.3572
0.3538
0.3515
0.3500
0.3489
0.3480
0.3472
0.2935
0.2972
0.3024
0.3092
0.3176
0.3277
0.3396
0.4132
0.4036
0.3962
0.3905
0.3857
0.3816
0.3780
0.9100
0.8723
0.8332
0.7944
0.7570
0.7217
0.6887
0.6428
0.6462
0.6484
0.6499
0.6509
0.6517
0.6521
0.7065
0.7027
0.6973
0.6902
0.6814
0.6706
0.6578
0.5867
0.5962
0.6033
0.6087
0.6129
0.6162
0.6188
0.0900
0.1277
0.1668
0.2056
0.2430
0.2782
0.3112
423.15
473.15
523.15
573.15
623.15
673.15
723.15
423.15
473.15
523.15
573.15
623.15
673.15
723.15
423.15
473.15
523.15
573.15
623.15
673.15
723.15
423.15
473.15
523.15
573.15
623.15
673.15
723.15
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
0.0174
0.0263
0.0368
0.0488
0.0617
0.0754
0.0894
0.0125
0.0196
0.0287
0.0399
0.0532
0.0688
0.0868
0.0229
0.0335
0.0457
0.0591
0.0733
0.0880
0.1030
0.3041
0.3063
0.3086
0.3108
0.3128
0.3147
0.3164
0.9826
0.9737
0.9632
0.9512
0.9383
0.9246
0.9105
0.9875
0.9804
0.9713
0.9601
0.9468
0.9312
0.9130
0.9771
0.9665
0.9543
0.9409
0.9266
0.9118
0.8968
0.6959
0.6937
0.6914
0.6892
0.6872
0.6853
0.6836
−4
−4
−4
−4
−4
−4
−4
−4
−4
−4
−4
−4
−4
73.15
23.15
73.15
23.15
73.15
23.15
23.15
73.15
23.15
73.15
23.15
73.15
23.15
23.15
73.15
23.15
73.15
23.15
73.15
23.15
23.15
73.15
23.15
73.15
23.15
73.15
23.15
<10
−4
<10
−
−
−
4
4
4
−4
0.0001
0.0002
0.0003
0.0005
<10
−4
<10
−4
<10
−
4
0.0001
<10
<10
<10
<10
<10
<10
<10
−4
−4
−4
−4
−4
−4
<10
<10
−
−
−
4
4
4
0.0001
0.0002
0.0005
0.0009
0.0014
0.0022
0.0001
0.0002
0.0005
0.0008
0.0013
0.0020
0.0028
<10
<10
<10
−
−
4
0.0001
0.0002
0.0004
4
0.0001
0.0001
−
−
−
4
4
4
−4
−4
<10
<10
<10
<10
<10
−4
−4
<10
<10
−4
−4
<10
<10
−
−
4
−4
−4
0.0001
0.0001
0.0002
0.0003
<10
<10
<10
4
<10
−
4
0.0001
0.0001
<10
0.0001
−4
−4
−4
−4
−4
−4
−4
−4
−4
−4
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
−4
−4
−4
<10
<10
<10
−4
−4
−4
<10
<10
<10
−4
−4
−4
<10
<10
<10
−4
−4
−4
<10
<10
<10
−4
−4
−4
<10
<10
<10
−4
−4
−4
<10
<10
<10
EMF—equilibrium mole fraction; UAL—unsaturated aldehyde; UON—unsaturated
ketone; UOL—unsaturated alcohol; SAL—saturated aldehyde; SON—saturated
ketone; SOL—saturated alcohol.
EMF—equilibrium mole fraction; UAL—unsaturated aldehyde; UON—unsaturated
ketone; UOL—unsaturated alcohol; SAL—saturated aldehyde; SON—saturated
ketone; SOL—saturated alcohol.
ied conditions. The last two, acrolein and 5-hexen-2-one, are ␣,␤-
and ␥,␦-unsaturated carbonyl compounds, respectively. It has been
found that the transfer hydrogenation of a carbonyl group in stud-
ied compounds proceeds very selectively with the formation of
alcohols (Figs. 1–5).
It has been found that the transfer hydrogenation of a carbonyl
group in studied compounds proceeds very selectively with the
formation of alcohols. An exception has only been observed for
hexanal, for which aldol condensation occurs as a side reaction,
diminishing the yield of 1-hexanol. In contrast, a very high chemos-
electivity towards unsaturated alcohols has been noted for acrolein
and 5-hexenone, despite the fact that in their case the reduction of
the carbonyl group is not favoured thermodynamically unlike the
hydrogenation of the C C bond.
It has been shown that in the presence of MgO as the catalyst
2-propanol is more reactive than ethanol in the transfer hydrogena-

1
38
M. Gli n´ ski, U. Ulkowska / Applied Catalysis A: General 511 (2016) 131–140
Fig. 1. Catalytic transfer hydrogenation of hexanal and benzaldehyde with ethanol
−1
−1
GHSV (N2) = 4000 h .
(
E) or 2-propanol (P); T = 473 K, D/A = 3, LHSV = 3 h
,
Fig. 3. Catalytic transfer hydrogenation of acetophenone and benzophenone with
ethanol (E) or 2-propanol (P); T = 573 K, D/A = 3 for acetophenone, D/A = 8 for ben-
zophenone, LHSV = 3 h , GHSV (N2) = 4000 h . Black—conversion, white—yield of
carbinol.
Black—conversion, white—yield of carbinol.
−
1
−1
Fig. 4. Catalytic transfer hydrogenation of 5-hexen-2-one with ethanol (E) or
−
1
−1
GHSV (N2) = 4000 h .
2-propanol (P); D/A = 3, LHSV = 3 h
,
Black—conversion,
white—yield of 5-hexen-2-ol.
Fig. 2. Catalytic transfer hydrogenation of methyl isopropyl ketone and cyclo-
hexanone with ethanol (E) or 2-propanol (P); T = 473 K, D/A = 3 for methyl
isopropyl ketone, D/A = 4 for cyclohexanone, LHSV = 3 h , GHSV (N2) = 4000 h
Black—conversion, white—yield of carbinol.
only a half of those attained for ethanol. The reason for this unpre-
dictable behaviour is yet unknown. The question arises whether or
not 2-propanol shows low reactivity in the transfer hydrogenation
of all ␣,␤-unsaturated carbonyl compounds in the presence of MgO.
−
1
−1
.
tion of various types of carbonyl compounds including aliphatic and
aromatic aldehydes, aliphatic ketones, cyclic ketones, aralkyl and
aryl ketones as well as 5-hexen-2-one. This corresponds well with
the results of our calculations, which show that 2-propanol is ther-
modynamically a better hydrogen donor than ethanol. However,
3.5. Vapour-phase transfer hydrogenation of acrolein with
ethanol in the presence of MgO. The origin of reaction
chemoselectivity
2
-propanol shows a markedly low reactivity during transfer hydro-
For this part of work ethanol has been chosen as the hydro-
gen donor due to its higher reactivity than 2-propanol, to study
the origin of reaction chemoselectivity in vapour-phase transfer
genation of acrolein. It has been observed that at 473 and 523 K the
yields of allyl alcohol with 2-propanol as the hydrogen donor are

M. Gli n´ ski, U. Ulkowska / Applied Catalysis A: General 511 (2016) 131–140
139
Fig. 7. Catalytic transfer hydrogenation of allyl alcohol (UOL) with ethanol; D/A = 6,
−
1
−1
LHSV = 3 h , GHSV (N2) = 4000 h . Black—conversion, grey—yield of 1-propanol
SOL).
(
Fig. 5. Catalytic transfer hydrogenation of acrolein with ethanol (E) or 2-propanol
Table 9
−
1
−
1
(
P); D/A = 6, LHSV = 3 h , GHSV (N2) = 4000 h . Black—conversion, white—yield of
Transformations of propionaldehyde, allyl alcohol and 1-propanol in the presence
of MgO as the catalyst. LHSV = 3 h ; GHSV (N2) = 4000 h .
allyl alcohol.
−1
−1
Compound
T [K]
Conv. [%]
Yield [%]
UOL
–
UAL
10a
SAL
–
423
13
473
523
423
473
523
423
473
523
4
3
0
tr
2
0
–
–
–
–
–
–
–
–
3a
1
–
–
–
–
–
–
a
–
0
0
–
–
–
b
b
b
tr
tr
tr
tr
c
c
UOL—unsaturated alcohol, UAL—unsaturated aldehyde, SAL—saturated aldehyde.
a
2
0
0
-methyl-2-pentenal.
.1%.
.3%.
b
c
Additionally, catalytic tests in the presence of MgO were per-
Fig. 6. Catalytic transfer hydrogenation of propionaldehyde (SAL) with ethanol;
D/A = 6, LHSV = 3 h , GHSV (N2) = 4000 h . Black—conversion, grey—yield of 1-
propanol (SOL).
formed with propionaldehyde, allyl alcohol and 1-propanol in the
absence of ethanol. The results are collected in Table 9. The aim
of these tests was to check if any reaction of the possible prod-
ucts derived from transfer hydrogenation of acrolein could react
further in the presence of the catalyst, hence obscuring the results
obtained earlier. We have found that in the range of temperatures of
−1
−1
hydrogenation of acrolein with alcohols in the presence of MgO.
We have examined the transformations of the products of acrolein
transfer hydrogenation, namely allyl alcohol (UOL) and propi-
onaldehyde (SAL) in the presence of ethanol as the hydrogen donor
4
23–523 K, magnesium oxide is not active in either the isomeriza-
tion of allyl alcohol into propionaldehyde or in the dehydrogenation
of 1-propanol into propionaldehyde. Under the same conditions
propionaldehyde undergoes aldol condensation with the formation
of 2-methyl-2-pentenal (Scheme 5). However, in this case, the con-
version of propionaldehyde decreased strongly with temperature
due to deactivation of the catalyst (Table 9; UAL yield).
(Figs. 6 and 7).
The comparison of conversion in catalytic tests performed with
propionaldehyde (Fig. 6) and allyl alcohol (Fig. 7) as the hydrogen
acceptors from ethanol reveal that in the presence of MgO, only the
former is reduced to 1-propanol with high yields.
The above mentioned facts lead to the conclusion that the very
high chemoselectivity towards allyl alcohol in the transfer hydro-
genation of acrolein is caused by a kinetic control induced by
magnesium oxide as the catalyst. It seems obvious that the struc-
ture of the hydrogen acceptor plays a crucial role in the observed
behaviour of the catalyst; otherwise, acrolein would be (transfer)
hydrogenated preferentially into a SAL/SOL mixture.
It was found that magnesium oxide shows a residual activity
in allyl alcohol transfer hydrogenation with ethanol, a 3% yield of
the product has been only noted at the highest studied tempera-
ture (523 K). We have found that in the range of temperatures of
4
23–523 K, the catalyst is not active in the transfer hydrogenation
of acrolein with ethanol into propionaldehyde.
Therefore, the lack of 1-propanol and the product of aldol con-
densation of propionaldehyde (viz. 2-methyl-2-pentenal), which
always accompanies the transformations of the aldehyde in minute
amounts (Scheme 5), indicate that the formation of propionalde-
hyde in transfer hydrogenation of acrolein with ethanol is strongly
inhibited.
4. Conclusions
The main achievement of this work is a comprehensive ther-
modynamic description of vapour phase transfer hydrogenation

1
40
M. Gli n´ ski, U. Ulkowska / Applied Catalysis A: General 511 (2016) 131–140
Scheme 5. Aldol condensation of propionaldehyde.
of four representatives of ␣,␤-unsaturated carbonyl compounds:
acrolein, ␣-methylacrolein, ␤-methylacrolein and methyl vinyl
ketone at 423.15–723.15 K with alcohols (EtOH or 2-PrOH) as
hydrogen donors and comparison of the obtained data with those
calculated for hydrogenation of the same set of compounds with
dihydrogen. The calculations were performed in accordance to
well-known methods and the Gibbs free energies and equilib-
rium mole fractions (EMFs) were obtained. The results obtained
from both modes of hydrogenation reveal that the EMFs for the
UOLs are very low (<0.0009) at all temperatures for all four stud-
ied compounds. When an ␣,␤-unsaturated carbonyl compound is
hydrogenated with dihydrogen at 423.15 K, the EMF for SAL/SON is
in the range of 0.002–0.091 (the exact value depends on the struc-
ture). The values obtained for transfer hydrogenation differ from
those from hydrogenation with dihydrogen in that even at the low-
est temperature the SAL/SOL ratio is much greater, i.e. 0.36:0.64
for transfer hydrogenation of acrolein with ethanol and is strongly
influenced by the structure of the hydrogen acceptor as well as
hydrogen donor. Based on the analysis of the calculated values for
these four ␣,␤-unsaturated carbonyl compounds we have shown
that the structure of a given compound greatly influences the extent
to which the hydrogenation of the C C bond is preferred over the
hydrogenation of the C O bond.
chemoselectivity of the transfer hydrogenation of acrolein with
alcohols is caused by a kinetic control induced by magnesium oxide
as the catalyst.
Acknowledgements
The authors gratefully acknowledged the financial support for
this work from the National Science Center (NCN–Narodowe Cen-
trum Nauki) Poland (Grant No. 0275/B/H03/2011/40).
We would like to thank prof. Tadeusz Hofman from the
Department of Physical Chemistry Faculty of Chemistry Warsaw
University of Technology (Politechnika) for valuable discussions
and for encouragement to us to overcome the difficulties found
during calculations.
Appendix A. Supplementary data
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.apcata.2015.11.
046.
References
Our measurements of the reactivity of ethanol and 2-propanol
in the transfer hydrogenation of chosen carbonyl compounds:
hexanal, methyl isopropyl ketone, cyclohexanone, benzaldehyde,
acetophenone, benzophenone, 5-hexen-2-one and acrolein in the
presence of magnesium oxide as the catalyst revealed that:
[1] M.A. Vannice, B. Sen, J. Catal. 115 (1989) 65–78.
[
[
[
2] V. Ponec, Appl. Catal. A: Gen. 149 (1997) 27–48.
3] P. Claus, H. Hofmeister, Ch. Mohr, Gold Bull. 37 (2004) 181–186.
4] P. Claus, Appl. Catal. A: Gen. 291 (2005) 222–229.
[5] F. Braun, J.I. Di Cosimo, Catal. Today 116 (2006) 206–215.
[
[
6] B. Campo, C. Petit, M.A. Volpe, J. Catal. 254 (2008) 71–78.
7] E. Gebauer-Henke, J. Grams, E. Szubiakiewicz, J. Farbotko, R. Touroude, J.
Rynkowski, J. Catal. 250 (2007) 195–209.
•
•
a very high selectivity of the reduction of a carbonyl group into
carbinol group for all studied compounds except hexanal has
been observed;
for all carbonyl compounds except acrolein the reactivity of 2-
propanol highly exceeded that shown by ethanol.
[8] R.A.W. Johnstone, A.H. Wilby, I.D. Entwistle, Chem. Rev. 85 (1985) 129–170.
9] G.K. Chuah, S. Jaenicke, Y.Z. Zhu, S.H. Liu, Curr. Org. Chem. 10 (2006)
639–1654.
10] J.R. Ruiz, C. Jiménez-Sanchidrián, Curr. Org. Chem. 11 (2007) 1113–1125.
[11] M. Gli n´ ski, Appl. Catal. A: Gen. 349 (2008) 133–139.
[
1
[
[
[
[
12] M. Gli n´ ski, React. Kinet. Catal. Lett. 97 (2009) 275–279.
13] M. Gli n´ ski, React. Kinet. Mech. Catal. 99 (2010) 93–98.
14] M.A. Aramendía, V. Borau, C. Jiménez, J.M. Marinas, A. Porras, F. Urbano, Catal.
Lett. 50 (1998) 173–177.
It has been proved that in the presence of magnesium oxide the
[
[
15] T.M. Jyothi, T. Raja, K. Sreekumar, M.B. Talawar, B.S. Rao, J. Mol. Catal. A:
Chem. 157 (2000) 193–198.
16] J.J. Ramos, V.K. Díez, C.A. Ferretti, P.A. Torresi, C.R. Apesteguía, J.I. Di Cosimo,
Catal. Today 172 (2011) 41–47.
reactivity of ethanol in the transfer hydrogenation of acrolein is
higher than that with 2-propanol, despite the fact that from the
thermodynamic standpoint, secondary alcohols are better hydro-
gen donors than primary ones. The chemoselectivity of vapour
phase transfer hydrogenation of acrolein to allyl alcohol and its
yield are the highest noted in literature. Although thermodynam-
ics favours other products, the presence of magnesium oxide as
the catalyst controls the reaction kinetically to give such a high
chemoselectivity.
It has been shown that in the transfer hydrogenation of pro-
pionaldehyde and allyl alcohol with ethanol in the presence of
magnesium oxide only the former is reduced to 1-propanol with
high yields. The formation of propionaldehyde in the transfer
hydrogenation of acrolein with ethanol is strongly inhibited, which
is recognized by the lack of 1-propanol in the products. It has also
been proved that the catalyst is not active in either the isomeriza-
tion of allyl alcohol into propionaldehyde or in the dehydrogenation
of 1-propanol into propionaldehyde. In the light of the thermo-
dynamic calculations, it can be assumed that the observed high
[
[
17] S.A. Ballard, H.D. Finch, D.E. Winkler, Adv. Catal. 9 (1957) 754–763.
18] T. Kobyli n´ ski, S. Malinowski, W. Zatorski, Rocz Chem. 42 (1968) 1899–1904,
CA: 70: 105586u.
[19] M. Gli n´ ski, U. Ulkowska, Catal. Lett. 141 (2011) 293–299.
[
20] S.E. Stein, R.L. Brown, Structures and Properties Group Additivity Model in
NIST Chemistry WebBook, NIST Standard Reference Database Number 69, in:
P.J. Linstrom, W.G. Mallard (Eds.), National Institute of Standards and
Technology, Gaithersburg, MD, 2013, p. 20899, http://webbook.nist.gov,
(
retrieved July 9, 2013).
[
[
21] D.N. Rihani, L.K. Doraiswamy, Ind. Eng. Chem. Fund. 7 (1968) 375–380.
22] N.P. Chopey, Handbook of Chemical Engineering Calculations, third ed.,
McGraw-Hill, New York, 2004.
[
[
23] M.W. Chase, J. Phys. Chem. Ref. Data, Monograph 9 (1998) 1.
24] C.L. Yaws, Yaws’ Handbook of Thermodynamic Properties for Hydrocarbons
and Chemicals, Knovel, 2009 (electronic ISBN: 978-1-60119-797-9), http://
app.knovel.com/web/toc.v/cid:kpYHTPHC09/viewerType:toc/root slug:yaws-
handbook-thermodynamic/url slug:yaws-handbook-thermodynamic/?,
(
retrieved November 28, 2011).
[
25] J.P. de Wilde, Enthalpy and Entropy of organic compounds from group
contributions, Delft Univ. of Technology, Memorandum LR M-635, Report
PML.