Aqueous phase reforming of industrially relevant sugar alcohols with different chiralities

Article abstract of DOI:10.1021/cs501894e

The influence of substrate chirality was studied in the aqueous phase reforming (APR) over Pt/Al₂O₃ in a continuous fixed-bed reactor at 225 °C. Two epimeric sugar alcohols, namely sorbitol and galactitol, were used for performance comparison. For the very first time galactitol was used in the APR process. The reliability of the liquid-phase product analysis was considerably improved due to application of a spiking technique for qualitative analysis and subsequent peak fitting for quantification. A detailed analysis of the gas and liquid phases showed almost the same behavior of both polyols, apart from the formation of several intermediate oxygenates. This implies that industrially relevant feedstocks with different chiralities can be efficiently processed in APR, since the chirality of the initial substance does not affect the reaction rate and selectivity to the final products. However, the observed difference in the liquid phase composition should be considered, because the catalyst should be active in the conversion of both polyols and oxygenates. Finally, possible reaction pathways were proposed and discussed from a thermodynamic point of view.

Full text of DOI:10.1021/cs501894e

Subscriber access provided by UNIV OF NEBRASKA - LINCOLN
Article
Aqueous phase reforming of industrially
relevant sugar alcohols with different chirality
Lidia Godina, Alexey Kirilin, Anton Tokarev, and Dmitry Yu. Murzin
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/cs501894e • Publication Date (Web): 02 Apr 2015
Downloaded from http://pubs.acs.org on April 5, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Catalysis is published by the American Chemical Society. 1155 Sixteenth
Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 49
ACS Catalysis
1
2
3
4
5
6
7
8
9
Aqueous phase reforming of industrially relevant
sugar alcohols with different chirality
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Lidia I. Godina1, Alexey V. Kirilin1,2, Anton V. Tokarev1, Dmitry Yu. Murzin1*
1Laboratory of Industrial Chemistry and Reaction Engineering, Process Chemistry Centre,
Åbo Akademi University, FI-20500 Turku, Finland
2current address: Dow Benelux B.V., P.O. Box 48, building 443, 4530 AA, Terneuzen,
Netherlands
*dmurzin@abo.fi
Abstract
The influence of substrate chirality was studied in the aqueous phase reforming (APR) over
Pt/Al2O3 in a continuous fixed-bed reactor at 225°C. Two epimeric sugar alcohols namely
sorbitol and galactitol were used for performance comparison. For the very first time galactitol
was used in the APR process. The reliability of the liquid-phase products analysis was
considerably improved due to application of a spiking technique for qualitative analysis and
subsequent peak fitting for quantification. A detailed analysis of the gas and liquid phases
showed almost the same behavior of both polyols, apart from formation of several intermediate
oxygenates. This implies that industrially relevant feedstock with different chirality can be
efficiently processed in APR, since chirality of the initial substance does not affect the reaction
1
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 49
1
2
3
4
5
6
7
8
9
rate and selectivity to the final products. However, the observed difference in the liquid phase
composition should be considered, because the catalyst should be active in conversion of both
polyols and oxygenates. Finally, possible reaction pathways were proposed and discussed from
the thermodynamic point of view.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Keywords: aqueous phase reforming, sorbitol, galactitol, Pt/Al2O3, thermodynamics.
2
ACS Paragon Plus Environment

Page 3 of 49
ACS Catalysis
1
2
3
1. Introduction
4
5
6
7
8
9
Over the last decade the volatility of crude oil prices became one of the main driving forces for
fuel production from renewable resources, especially from biomass. The abundance of renewable
carbohydrates determines the choice of aqueous phase reforming (APR) as a promising
technology, which possess certain advantages. First, the net effect of this approach to green-
house emissions is zero since all CO2 produced was captured from the atmosphere during the
process of carbohydrate formation. A second benefit is that comparing with one of the most
abundant processes, such as steam reforming, APR allows to reduce energy consumptions
needed for hydrogen production1,2 Even if the yields of hydrogen in steam reforming of sorbitol
are practically the same as in the APR (50-60%), the operation temperature is substantially lower
in the APR (225°C and 525-625°C)3. The third advantage is that water, which is used as a
solvent, acts also as a reactant, thus being converted into additional hydrogen via the water-gas
shift reaction (WGS).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The method of APR allows one to use a wide range of substrates with methanol, ethylene
glycol, glycerol, xylitol and sorbitol being mostly studied1,4–6. Sorbitol is currently produced in
high amounts from glucose7–13; on top of that a large number of recent papers is focused on a
direct cellulose conversion to sorbitol14–23. Galactitol could be produced from Siberian larch
(Larix sibirica) wood. This tree species area covers significant parts of Russia and Canada. The
concentration of arabinogalactans is up to 20 wt. % in dry wood. The main component of these
polysaccharides is galactose (an average ratio of units of galactose : arabinose : glucuronic acid
is about 5.6:1:0.08)24. The other sources for galactitol could be galactoglucomannans, which are
found in large quantities in softwood25 and can be recovered in mechanical pulping process from
the waste waters24,26,27
.
The influence of substrate chirality on the APR parameters such as selectivity and product
distribution has not been properly addressed. Deutsch et al.28 carefully examined an aqueous-
3
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 49
1
2
3
4
5
6
7
8
9
phase hydrogenolysis (APH) of polyols ranged from C3 to C6 except for talitol and iditol over a
Ru/C catalyst. The obtained results are essential for the present study, since conditions of APH
are very similar to APR. In this process reaction rates showed a dependence on chirality, but not
affected by the carbon chain length of a reactant. Molecules with erythro-sequences of hydroxyl
groups, where in the Fischer projection two pairs of the same substituents are located on the
same side of two adjacent chiral centers, showed the highest rates in hydrogenolysis due to steric
or adsorption phenomena involving the Ru surface. Meanwhile, Liu2 reported same hydrogen
production rates for mannitol and sorbitol in the APR. According to the data available in the
literature, the influence of chirality on APR of polyols is unclear.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Thus, there is an apparent need to investigate the influence of chirality on APR of sugar
alcohols. A comparative study of sorbitol and galactitol was performed in the present work to
unravel a potential impact of chirality in this section. Additionally, a thorough analysis of the
reaction products was performed in order to establish the foundations for the very complex
reaction network. Thermodynamic calculations were performed for various plausible reaction
pathways in order to exclude unrealistic thermodynamically forbidden reactions.
2. Experimental
2.1. Catalyst
A commercially available catalyst was provided by Sigma–Aldrich. The catalyst has a loading
of 1 wt.% of Pt deposited on γ-Al2O3. The catalyst was compressed into pellets and then crushed
and sieved to 125–250 µm fraction prior to the experiments.
4
ACS Paragon Plus Environment

Page 5 of 49
ACS Catalysis
1
2
3
4
5
2.2. Catalyst characterization
2.2.1 CO chemisorption
6
7
8
9
The platinum dispersion was evaluated from CO titration by means of CO pulse chemisorption
apparatus (Micromeritics, AutoChem 2900). The following program was used for a catalyst
pretreatment prior analysis: heating from 25 to 50°C at 10°C/min in He, dwell for 30 min, gas-
switch to H2, heating up to 250°C with the rate 5°C/min, dwell for 2 h, followed by flushing for
60 min in He at 250°C. In the first step the sample of 0.10 g was pre-reduced in a U-shape quartz
tube, while in the second the surface hydrogen was removed. After this procedure the catalyst
was cooled to ambient temperature and titrated by CO pulses (10 vol.% CO in He). A Pt/CO
stoichiometry was assumed to be 1:1.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.2.5. Temperature-programmed reduction
The sample of 0.12 g of catalyst was reduced in a U-shape quartz tube during the TPR by
means of the AutoChem 2900 instrument with the following program: heating from room
temperature to 700°C with the rate 10°C/min. The reduction was performed by 5% H2 in Ar; a
thermal conductivity detector (TCD) was used to measure the hydrogen uptake.
2.3 APR experiments
The experimental setup for APR is shown in Figure 1. A stainless-steel reactor was located in a
furnace; the catalyst bed was placed in the middle of the reactor between two layers of quartz
sand. The polyol solution was fed with an HPLC pump. The reactor was connected to nitrogen
(1% He) and hydrogen gas lines. Gas-liquid separation took place in a T-shape connector with 4
mm inner diameter above a bottle for liquid waste products. There is no need in additional
cooling or any extra gas-liquid separation equipment, since constant nitrogen co-feeding during
5
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 49
1
2
3
4
5
6
7
the reaction (flow is 25 ml/min), forms a trickling regime. Liquid samples were periodically
taken via a sampling loop for an offline HPLC analysis, while the gas samples were analyzed
online by means of Micro-GC.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Experimental set-up for aqueous phase reforming.
For all experiments 1.0 g of the fractionized 1% Pt/Al2O3 catalyst was used. The experimental
studies were done at 225°C and 29.7 bar. For APR experiments 3.6 wt.% water solutions of a
substrate were used. The flow rates ranged from 0.1 to 1.0 ml/min; which corresponds to the
weigh- hour space velocities of 0.22-2.16 h-1 calculated as mass of the substrate per mass of the
catalyst per hour (gsubst x gcat−1 x h−1).
In total three experiments were done. In the first and the second experiments sorbitol and
galactitiol 3.6 wt. % solutions were fed within the flow rates range from 0.1 to 1.0 ml/min, as
mentioned above. The experimental sequence of the third experiment is illustrated in the Figure
2. Sorbitol solution (3.6 wt. %) was fed, followed by galactitol solution (3.6 wt.%), then a
mixture of polyols (1:1, 3.6 wt.% in total) was fed followed by final feeding with sorbitol
solution.
6
ACS Paragon Plus Environment

Page 7 of 49
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. TOS scheme for the sequential experiment, where sorbitol (sorb), galactitol (gal) and
mixture of both polyols (s+g) was fed.
The catalyst was reduced in situ prior to all experiments in a hydrogen flow (40 ml/min). The
following program was used: heating in a hydrogen flow from room temperature to 250°C with
the rate 5°C/min, dwelling for 2 hours, cooling to the operating temperature. Residual hydrogen
was removed by flushing with N2 (25 ml/min) for 20 min.
Gas products were quantitatively analyzed online by a micro-GC (Agilent Micro-GC 3000A)
equipped with four columns: Plot U, OV-1, Alumina and Molsieve.
The analysis of liquid products was performed by means of HPLC (Agilent 1100) equipped
with an Aminex HPX-87H column, which was quantitatively calibrated for the anticipated
reaction products (external standard method). The analysis was performed at 45°C under
isocratic conditions and flow rate 0.6 ml/min. The mobile phase consisted of 5 mM H2SO4 water
solution. The products were analyzed by using a refractive index detector (RI). Peak
identification was based on retention times. Eighty-nine substances were used in a spiking
technique for additional improvement of the confidence of identification. A full list of substances
is presented in the Supporting Information, Table S1. The selection of the substances was based
on the data reported in references29–36. Peak fitting was applied to increase the accuracy of
quantification, because the poor resolution of most the peaks, corresponding to polyols and
sugars. Fitting procedure was performed in the data-analyzing software OriginPro 9.037.
7
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 49
1
2
3
4
5
A total organic carbon analysis (TOC) was applied to monitor the carbon balance in liquid
samples. The observed values were close to 95%.
6
7
8
9
The following equations were used to calculate the results of experiments.
Conversion of the substrates:
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(Cfeedstock
)
Conversion (%) 
100
(Cinput
)
, (1)
where ν(Cfeedstock) – molar flow of carbon in the unreacted substrate [mol/min], ν(Cinput) –
molar flow of carbon in the incoming flow of substrate [mol/min].
The results of APR of sorbitol and galactitol were compared based on selectivity to the
reaction products. For alkanes, CO2, CO and products in the liquid phase selectivity was
calculated in the following way:
Selectivity to
product X
n(CX )
(%) =
×100
n(Cgas )
, (2)
where ν(Cx)– molar flow of carbon, contained in a product [mol/min].
Moles of carbon in a product is determined as:
(Cx)  Cx vn , (3)
where Cx - molar flow of product [mol/l*min]; v – volumetric flow [l/min]; n – number of
carbon atoms in a product.
An additional parameter “reforming ratio” RR (H2/CO2) was introduced to calculate the
selectivity to hydrogen, since hydrogen is produced both from polyols and via the water-gas shift
(WGS) reaction:
H2O
C6 H14O6 6CO  7H2 6CO2 13H2
Selectivity to hydrogen is defined in the following way:
n(H2 )
1
Selectivity to H2 (%) =
×
×100
n(Cgas ) RR
, (4)
8
ACS Paragon Plus Environment

Page 9 of 49
ACS Catalysis
1
2
3
where ν(H2)– moles of H2 formed, and RR=13/6 for sorbitol and galactitol.
4
5
6
7
8
3. Results and discussion
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3.1. Catalyst characterization
The commercial catalyst 1% Pt/Al2O3 was completely reduced prior to APR experiments as
evidenced by a corresponding TPR profile in the Figure 3. The reduction peak of the surface
PtO2 species, weakly interacting with the support, starts at 39°C. Reduction of Pt oxides
interacted with Al2O3 more strongly occurs at 188°C. A high temperature peak observed at
38
372°C can be assigned to the formation of PtO2-Al2O3 complex . The reduction of the support
occurred at the temperatures above 580°C due to interactions between Pt and Al2O3 and
formation of Pt-Al alloy39.
Surface properties of the catalyst were elucidated by means of pulse CO chemisorption. The
Pt/Al2O3 catalyst has metal dispersion 52% with metallic surface area 129 m2/g metal and an
average particle diameter 2.2 nm.
Figure 3. TPR profile of Pt/Al2O3 catalyst.
9
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 49
1
2
3
3.2. Test of polyols
4
5
6
3.2.1. Conversion of substrate and carbon distribution between phases.
7
8
9
Previously similar types of catalysts were used for APR of various polyols29,40–50, despite a
concern that such support as Al2O3 can be transformed into boehmite under experimental
conditions. Copeland et al demonstrated that alumina support is reasonably stable in the presence
of such a polyol as glycerol, even being tested in overheated water41. This observation
contradicts with XRD data obtained by Doukkali et al51. Severe transformations of Pt/Al2O3
were confirmed for the same reaction due to incorporation of water molecules in the catalyst
matrix51. Similar conditions were applied in the current work. However, support alteration could
be a slow and time-dependent process. Duration of experiments probably plays a crucial role.
According to Fig. 4 and 5, no significant deactivation was observed during 135 hours of time on
stream for the whole experimental program. The conversion level, as well as selectivity to
hydrogen, decreased slightly during the first 20 hours due to the initial deactivation.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fig. 6 displays conversion dependence on weight-hour space velocity (WSHV) for all three
experiments with sorbitol, galactitol and the mixture thereof. Conversion levels varied from 15 to
100%. A small discrepancy between the conversion level of sorbitol and galactitol is clearly
visible in Fig. 4 and 6. This difference can be attributed to probable differences in the catalyst
wetting, rather than the properties of the substances. This hypothesis was supported by the
results of the third sequential experiment (see insertion in Fig. 6), which gave exactly the same
levels of conversion for both polyols and their 1:1 mixture.
10
ACS Paragon Plus Environment

Page 11 of 49
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Performance of 1% Pt/Al2O3 catalyst as a function of time on stream (TOS) at 225°C,
29.7 bar and flow rate 0.2 ml/min: conversion of the substrates. ”Sorb” – the experiment with
sorbitol only, “gal” – the experiment with galactitol, ”s+g” – sequential experiment with both
polyols (see Section 2.3).
Figure 5. Performance of 1% Pt/Al2O3 catalyst as a function of time on stream (TOS) at 225°C
and flow rate 0.2 ml/min: selectivity to H2. ”Sorb” – the experiment with sorbitol only, “gal” –
the experiment with galactitol, ”s+g” – sequential experiment with both polyols (see Section
2.3).
11
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 49
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. Conversion of the substrate as a function of WHSV for sorbitol and galactitol at 225°C
and 29.7 bar. ”Sorb” – the experiment with sorbitol only, “gal” – the experiment with galactitol,
”sorb+gal” – sequential experiment with both polyols (see Section 2.3).
The distribution of carbon during APR experiments was studied by means of TOC along with
the analysis of the gas phase composition; the results for sorbitol are shown in Figure 7a. The
mass balance based on the carbon content observed is shown as an upper segment of the diagram
being close to 95% in both experiments. The distribution of carbon between the gas and the
liquid products was found to be very similar for both sugar alcohols. Remarkably almost 42%
carbon (fed to the reactor in the form of polyol) is still present in the liquid phase in the form of
oxygenated products, whereas conversion of the initial substrate already reached 100%.
12
ACS Paragon Plus Environment

Page 13 of 49
ACS Catalysis
1
2
3
4
5
6
7
Figure 7. Carbon distribution between the liquid and gas phases and total carbon balance versus
conversion in APR of sorbitol at 225°C and 29.7 bar: a) the whole conversion region, b) carbon
distribution at 62 % conversion.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3.2.2 Selectivity to gas-phase products
The selectivity to the gas phase products was calculated according to equations (2) and (4).
The results obtained are provided in Fig. 8 and 9. It was found that the selectivity to some
gaseous products (H2, CO and CO2) were similar for both polyols in the whole range of substrate
conversion (20-100%). Selectivity to hydrogen decreased significantly with an increase in
conversion due to consumption of hydrogen in hydrogenation reactions, in contrast to a rapid
growth of the total selectivity to alkanes. The last parameter is much lower compared to
hydrogen selectivity (see Fig. 7), as it was already shown by Kirilin et al.29. The selectivity to
CO was close to zero indicating that almost all carbon monoxide was converted to CO2 via the
water-gas shift reaction (WGS) under experimental conditions, as demonstrated by Haruta and
co-workers52. It is noteworthy, that selectivity to alkanes appeared to be slightly different in the
sequential experiment, which probably happened due to minor changes in the equipment with
time. Sensitivity to alkanes changed between the first two experiments (sorbitol, galactitol), and
the third one (sequential experiment). However, all data points from the sequential experiment
lay in a row, indicating the same selectivity in the case of both polyols and their mixture.
13
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 49
1
2
3
Figure 8. Selectivity to H2 and CO2 versus carbon conversion in APR of sorbitol and galactitol
4
5
at 225°C and 29.7 bar.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 9. Selectivity to CO and alkanes versus carbon conversion in APR of sorbitol and
galactitol at 225°C and 29.7 bar.
The detailed comparison of alkanes is provided in Fig. 10 and 11. Selectivity to alkanes for
both substrates was identical showing no effect of chirality on this parameter (see sequential
experiment). The discrepancy between different experiments is explained above. For all
hydrocarbons, selectivity increases with conversion. It is noteworthy that some amounts of
branched pentane and iso-hexane were detected in the gas phase, indicating some skeletal
isomerisation process during APR. However, iso-butane or unsaturated compounds were not
observed. Traces of heptane were also found, although it contains more carbon atoms than the
starting C6-polyol. However, for concentrations of heptane detected some variation in
experimental data was observed (Fig. 11), which might be caused by limited volatility of
hydrocarbons heavier than pentane at room temperature and atmospheric pressure.
14
ACS Paragon Plus Environment

Page 15 of 49
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 10. Selectivity to linear alkanes versus carbon conversion in APR of sorbitol and
galactitol at 225°C and 29.7 bar.
Figure 11. Selectivity to branched alkanes and heptane versus carbon conversion in APR of
sorbitol and galactitol at 225°C and 29.7 bar.
15
ACS Paragon Plus Environment

ACS Catalysis
Page 16 of 49
1
2
3
3.2.3 Selectivity to liquid-phase products
4
5
6
7
8
9
In depth studies of liquid-phase composition during polyol conversion were performed mainly
for aqueous-phase dehydration/hydrogenation (APD/H) on Pt/Al2O3-SiO230,31, Pt/TiO2+ZrO2-
WOx32,33, Ru/C and Re-Ru/C34. Oxygenates were analyzed in the APR of sorbitol on Ni and
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
35
29
bimetallic Ni-Pd catalysts supported on Al2O3, ZrO2 and CeO2 , as well as Pt/Al2O3 .
Identification of oxygenates was mostly based on by direct HPLC analysis or in combination
with GC-MS. In the work of Kirilin et al. analysis of intermediates in the APR of sorbitol on
Pt/Al2O3 involved also solid-phase microextraction36. More than 250 of intermediates with 50
major compounds were detected.
In the present study the composition of the liquid phase was thoroughly elucidated by means of
HPLC, which allowed identifying approximately 62% of carbon stored in all intermediates. The
distribution of identified and unknown carbon is showed at 68% conversion of sorbitol in Fig.
7b. Considering that all gas carbon was identified, one can conclude that approximately 80% of
carbon is counted among the known substances.
Selectivity to identified liquid products was calculated according to equations (2) and (3). The
results obtained in the current study are combined with the results from the previous studies on
APDH of sorbitol over different catalysts30–32, and are shown in Table 1. However, the
concentration of some substances was lower than the detection limit, therefore those species
were not inserted in Table 1 (the full list of compounds is provided in Table S1). It should be
also noted that the denominator in equation (2) contains the amount of carbon in the gas phase ),
which means that selectivity could exceed 100%, if carbon remains mostly in the liquid phase.
The composition of the liquid phase in the current study can be easily obtained from Table 1,
since the selectivity values are shown for very close levels of conversion: 62 and 65% for
sorbitol and galactitol correspondingly. The product distribution is very similar for the samples
obtained during APR of both polyols; in both cases similar sugars, polyols, alcohols, acids,
16
ACS Paragon Plus Environment

Page 17 of 49
ACS Catalysis
1
2
3
4
5
6
7
8
9
carbonyl compounds and furans were found. One notable exception is the presence of distinct
primary intermediates, such as glucose and galactose. No galactose was found in the sample
from sorbitol and only traces of glucose appeared in the sample from galactitol. However, this
method does not give a good separation of sugars and sugar alcohols, therefore only substances
with high concentration can be taken into account. Interestingly that even if the selectivity to
arabitol was also significantly higher in the case of galactitol, xylitol was also found among the
products, indicating that diastereomerisation reaction can take place. It is worth noting that such
substances as tartaric and fumaric acids (2,3-dihydroxybutendioic and 2-butenedioic acids
correspondingly) were found as the products of sorbitol conversion, but not in the case of
galactitol.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Comparison with APDH data, when additional hydrogen is supplied, revealed higher
selectivity to alcohols with a longer chain (hexanol, pentanol) and lower selectivity to methanol
and ethanol. It could be assumed that additional hydrogen would increase the total amount of
reactions of hydrogenolysis C-O bond, preventing C-C bond cleavage. All catalysts used for
APDH were prepared using supports with acidic properties, which influenced the reaction
pathways and resulted in higher concentrations of cyclic products compared to APR. The polyol
transformation started from formation of cyclic 1,4-sorbitan and isosorbide, which were found to
be the main intermediates at early stages of the APDH reaction. It leads to significantly higher
selectivity to heterocyclic compounds, such as furanic or pyranic compounds. At the same time,
isosorbide was not found during APR of sorbitol or galactitol, although the conversion level was
close to 60%.
Identification of isosorbide could be obstructed due to presence of glycerol among the reaction
products, which should be taken into account. Retention times of isosorbide and glycerol are
very close, which can be easily seen in Fig. 12, presenting a part of HPLC chromatogram with
19 curves. Every curve represents the same liquid sample obtained during APR of galactitol,
17
ACS Paragon Plus Environment

ACS Catalysis
Page 18 of 49
1
2
3
4
5
6
7
8
9
which was mixed with a solution of some substances, listed in a Table 1. The purple curve
represents the sample with glycerol addition, while the green one corresponds to isosorbide
addition, the others do not contain any of these substances. Therefore the peak with the retention
time of about 13.51 min can be assigned with a high level of certainty to glycerol.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure. 12. Glycerol and isosorbide peaks in the galactitol sample.
Comparison with APDH data revealed similar values of selectivity to some products.
Nevertheless, additional experiments are needed to evaluate the influence of additional hydrogen
on the liquid-phase products.
Table 1. Selectivity to substances found in the liquid phase during the APR of sorbitol and
galactitol and APDH of sorbitol.
Substances\Experiments
Galactitola
Sorbitolb
Ref. 47c
Ref31;d
Ref 45e
Alcohols
Selectivity [mol C %]
2-methylcyclopentanol
-
-
Tracesf
Traces
Traces
Traces
0
-
-
-
2-methylpentanol
butane-1,2-diol
butane-1,2,4-triol
butane-2,3-diol
butanol-1
-
-
-
0.47
0.64
-
-
-
0g
0
-
0
0
-
-
-
0.70
0
Traces
Traces
-
-
0.50
-
butanol-2
-
18
ACS Paragon Plus Environment

Page 19 of 49
ACS Catalysis
1
2
3
4
5
6
7
8
9
ethanol
2.02
0
3.02
1.46
Traces
-
0.60
2.10
-
ethylene glycol
glycerol
0
1.75
0.09
0
-
1.36
0.10
0
-
4.00
1.40
0
hexane-1,2-diol
hexane-1,2,6-triol
hexane-2,5-diol
hexanol-1
Traces
Traces
-
-
-
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0.18
0
0.19
0
-
-
7.02
2.40
1.10
-
hexanol-2
0
0
Traces
Traces
Traces
Traces
Traces
9.50
-
hexanol-3
0
0
-
-
-
isopropanol
methanol
0
0
-
6.96
0
5.20
0
0.30
0
0.40
7.80
0.30
-
pentane-1,2-diol
pentanol-1
0.06
0.04
1.39
1.45
0.06
0.02
1.35
1.74
3.70
-
pentanol-2
Traces
Traces
Traces
propane-1,2-diol
propanol-1
-
3.60
1.10
1.10
Acids
acetic acid
0.54
0
0.82
0
0
-
-
-
-
-
-
-
-
-
-
butyric acid
fumaric acid
hexanoic acid
lactic acid
0
2.80
0
0.69
0.04
0.10
0
-
-
0
-
0.07
0.15
0
Traces
-
pentanoic acid
propanoic acid
pyruvic acid
tartaric acid
Traces
-
0
0
Traces
-
1.71
0
0.08
0.03
-
-
-
-
Adehydes/Ketones
3-hydroxy-2-butanone
3,4-hexanedione
acetone
1.49
-
1.59
-
-
Traces
Traces
-
-
-
-
-
0.53
0.25
0.20
19
ACS Paragon Plus Environment

ACS Catalysis
Page 20 of 49
1
2
3
4
5
6
7
8
9
butanone
-
0
-
-
0
-
Traces
10.53
10.53
Traces
Traces
4.39
2.00
0.20
hexanone-3
hexanone-2
pentane-2,3-dione
pentanone
6.30
1.10
-
-
0
-
0
-
-
2.00
-
-
0.20
-
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
pentanone-3
-
-
Furans/Pyranes
2-acetylfuran
0.02
0.02
-
Traces
Traces
Traces
Traces
-
5.90
5.00
-
-
0.50
0.30
-
2-methyltetrahydrofuran
2-methyltetrahydropyran
2,4-dimethyltetrahyrdofuran
2,5-dimethyltetrahydrofuran
5-hydroxymethyl-2-furaldehyde
tetrahydrofuran
0
0
-
-
-
-
0.06
0.06
-
0.04
0.07
-
5.30
-
0.60
-
Traces
1.00
43.00
1.20
13.70
0
tetrahydrofurfuryl alcohol
tetrahydropyran
0.45
0
0.17
0
-
-
2.90
0.30
6.20
tetrahydropyran-2-methanol
0
0
Traces
Sugars, sugar alcohols
1,4-sorbitan
-
0.56
0.003
0.10
0
-
0.15
0.11
0
185.96
78.50
27.00
D-(+)-arabitol
D-(+)-glucose
D-galactose
D-sorbitol
-
-
-
-
-
-
-
-
-
14.06
0
74.56
-
-
galactitol
15.79
0
-
-
-
isosorbide
0
54.39
20.50
27.00
meso-erythritol
xylitol
1.12
0.99
0.95
1.05
-
-
-
-
-
-
a – APR of galactitol over 1% Pt/Al2O3 at 225°C, 29.7 bar and 65% conversion;
b – APR of sorbitol over 1% Pt/Al2O3 at 225°C, 29.7 bar and 62% conversion;
c – APDH of sorbitol over 2.2% Pt/ZrO2+TiO2-WOx (20:80) at 220°C, 22 bar and 91.5%
conversion32;
d – APDH of sorbitol over 4% Pt/zirconia phosphate at 245°C, 26.1 bar and 58% conversion31;
e – APDH of sorbitol over 4% Pt/Al2O3-SiO2 at 225°C, 29.3 bar and 74.6% conversion30;
20
ACS Paragon Plus Environment

Page 21 of 49
ACS Catalysis
1
2
3
4
5
6
7
f – traces - if the compound is detected, but its amount is lower than 0.005 mol / mol sorbitol
converted;
g
– zero entries indicate that a substance concentration is below sensitivity of RI detector (the
detection limit strongly depends on the type of a substance and can be in a range 8-800 mmol/L,
assuming signal-to-noise ratio S/N=3).
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3.2.4 Influence of stereoselectivity and possible reaction pathways
A certain number of sugar alcohols can be produced from plant biomass. They can differ in the
chain length and in chirality. An increase in the carbon chain length implies an increase in the
number of possible stereoisomers: two for tetraols, three for pentaols and six for hexitols. These
stereoisomers differ in their internal bonds, which sometimes has a profound influence on their
properties. Solubility of galactitol in water can serve as a good example being approximately
twenty times lower than solubility of sorbitol at temperatures between 15 and 100°C (see Fig.
13)53.The differences in properties of these stereoisomers can be attributed to conformational
differences in chirality of these molecules. Sorbitol has a strong 1,3-nonbonded interaction
between C2 and C4 hydroxyl groups being in zig-zag conformation, which destabilize it by 6.28
kJ/mol, and four 1,3-syn-axial interactions (two between C3 oxygen and C1 and C5 hydrogens,
and one between C5 oxygen and C6 hydrogen), which additionally destabilize this conformation
by 7.53 kJ/mol54. As a result, D-sorbitol exists in water solution and in the crystalline state
predominantly in sickle (bent) conformation, while D-galactitol keeps zig-zag conformation.
Even if these minor differences could be anticipated to have influence much APR process
performed at 225°C, they made a significant difference for hydrogenolysis studied at 205-
240°C28. Assuming high similarities between APDH and APR processes, it can be concluded
that Pt and Ru catalysts behave differently.
21
ACS Paragon Plus Environment

ACS Catalysis
Page 22 of 49
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 13. Solubility of alditols according to Kroschwitz and Howe-Grant equation53.
As it was shown in Figs. 8-11 and Table 1, the chirality of substrates does not have a
considerable influence on the selectivity to the final products or to the intermediates to a large
extent, which is an advantage for an industrial process, where the feedstock can be a mixture of
isomers. However, there is a notable difference between sorbitol and galactitol behavior at early
stages of APR reaction, when dehydrogenation of the substrate leads to formation of sugars -
galactose from galactitol and glucose from sorbitol, respectively.
The scheme of the main reaction pathways, which is based on the products analysis, is
displayed in Fig. 14. The substrate is C6-polyol, which can be transformed into C5-polyol via
consecutive dehydrogenation/decarbonylation steps 1,2 occurring on the metal sites30,47,55
.
Subsequent repetitions of these steps result in a complete conversion of the substrate to CO and
hydrogen. On Pt/Al2O3 almost all CO reacts with water to form CO2 and additional H2 via the
WGS reaction. Intermediate aldols can desorb from the surface. Aldehydes C4-C6 can be
transformed into stable cyclic products (substituted tetrahydrofurans and tetrahydropyrans) via
dehydration 3 occurred on acidic sites. Hydrogenolysis of primary hydroxyl-groups 4 gives
molecules with a non-reactive methyl-group at the end, and less reactive secondary hydroxyl-
groups. These secondary alcohols are either transformed into ketones via dehydrogenation 1 or
dehydration 3, or undergo complete reduction to form alkanes 5 depending on the conditions.
Since Pt favors dehydrogenation and C-C bond breaking, the main alkanes produced will be
22
ACS Paragon Plus Environment

Page 23 of 49
ACS Catalysis
1
2
3
alkanes with low molar mass – ethane and propane. Desorbed aldehydes are easily converted
4
5
into gemi-diols, which can be dehydrogenated on Pt with formation of acids (see Fig. 15)50,56–59
.
6
7
8
9
Aldol condensation is also preferred to Cannizzaro disproportionation, when the reacting
aldehyde contains alpha-hydrogen atom60.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 14.The main reaction pathways in APR of sorbitol and galactitol.
23
ACS Paragon Plus Environment

ACS Catalysis
Page 24 of 49
1
2
3
4
5
6
7
Figure 15. Formation of acids from aldehydes via gemi-diol dehydrogenation.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
All possible optically active hexoses and pentaols formed starting from D-galactitol and D-
sorbitol are shown in Figure 16, assuming that only steps 1 and 2 take place. Neither glucose, nor
xylitol can be produced from galactitol via this path, however, they were found in small amounts
among the liquid-phase products. Transition metals like Au, Pd or Ru are known to catalyze the
racemization of alcohols being used as a heterogeneous catalyst in dynamic kinetic resolution in
combination with enzymes61–63. However, Pt itself was not so widely applied probably due to
interactions with enzymes64,65. Thus, one can tentatively assume that diastereomerization also
takes place during APR (see Fig. 17).
24
ACS Paragon Plus Environment

Page 25 of 49
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 16. Galactitol and sorbitol dehydration/decarbonylation in APR.
Figure
17.
Diastereomerization
of
carbohydrates
on
Pt
surface
via
dehydrogenation/hydrogenation steps.
At 225°C and 30 bar the ionization constant of water66 is Kw = 5.597×10-12 corresponding to
pH = 5.628, which can promote reactions requiring protonation steps. Typically heterogeneous
catalytic isomerization of alkanes requires strong acidity, thus demanding another explanation of
25
ACS Paragon Plus Environment

ACS Catalysis
Page 26 of 49
1
2
3
4
5
branched alkane formation. Pinacol-pinacolone rearrangement of desorpbed aldehydes can be a
plausible explanation for branching (see Fig. 18).
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 18. Pinacol-pinacolone rearrangement with subsequent reduction to branched alkane.
Despite the fact that retro-aldol reaction assumed in many papers to be one of the possible
reaction paths, no clear evidences were presented in the literature and in the current study for
presence of this reaction during APR process on Al2O3- or carbon-based catalysts. This reaction
requires some specific conditions, as a special type of support or addition of a basic promoter.
Base-catalyzed retro-aldol reaction allowed production of significant amounts of glycerol from
sorbitol and both glycerol and ethylene glycol from xylitol during hydrogenolysis on Pt/NaY
zeolite67 or in presence of Ca(OH)268–70. It should be noted that two molecules of glycerol can be
formed from sorbitol only through transformations to ketohexose, but not aldohexose. However,
transformation of aldehyde form to keto form should occur rather easy. Therefore, the exact
mechanism of dehydrogenation is still unclear. The distribution of gas and liquid products in our
experiments does not confirm retro-aldol reaction to play a significant role for APR
transformations. However, this reaction is thermodynamically possible as described below.
Secondary alcohols with non-substituted neighboring atoms do not undergo C-C bond
cleavage, which was nicely shown by Vilcoqc et al.32
26
ACS Paragon Plus Environment

Page 27 of 49
ACS Catalysis
1
2
3
4
5
6
7
8
9
As mentioned above, one of the unexpected products was n-heptane, which contains an extra
carbon atoms then the substrate. Formation of hexane was detected by Jiang et al71 during APR
of xylitol over Pt/HZSM-5 and Ni/HZSM-5 catalysts. This fact was explained by ketones
formation from xylitol with subsequent aldol-condensation and hydrodeoxygenation; other
explanation was based on Fischer-Tropsch synthesis. It should be noted that only branched
alkanes could be produced after aldol-condensation of ketones, while the authors did not mention
it they observed linear or branched hexane. No Fischer-Tropsch synthesis occurs during APR,
which was proven by additional tests of the complementary water-gas shift reaction, where the
relationship between the APR and the WGS was thoroughly studied72–74. Thus, formation of
compounds with longer carbon chain could be a result of a sequence of condensation reactions.
A possible pathway is illustrated in Fig. 19, showing a reversible aldol and subsequent croton
condensations. This process is reversible and under these experimental conditions a retro-aldol
reaction can occur too. Branched alkanes were not previously detected previously for the APR
process over Pt/Al2O3, however, an addition of a solid acid as SiO2-Al2O3 promoted formation of
about 5% of branched alkanes75 to achieve a total selectivity up to 5%. Substantial quantities of
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
isoparafins were observed for APR of sorbitol over Ni/HZSM-5 catalyst76,77
.
Figure 19. Extension of a carbon chain via aldol and croton condensation and retro-aldol
reaction.
27
ACS Paragon Plus Environment

ACS Catalysis
Page 28 of 49
1
2
3
3.3. Thermodynamic aspects
4
5
6
7
8
9
Changes in standard Gibbs energy in the conversion of glycerol to different intermediates with
the same chain length and in the conversion of acetone to liquid fuels were nicely demonstrated
by Simonetti and Dumesic78. Aqueous and vapor phase reforming of ethylene glycol is discussed
by Kandoi et al79. Thermodynamic analysis and Gibbs energy minimization was performed for
APR of methanol, acetic acid and ethylene glycol80, glycerol81, polyols82 and C6-compounds83.
These calculations were performed to reveal thermodynamic feasibility of a number of reactions,
which could be a part of a complex APR network.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
A reaction is considered to be thermodynamically favorable if the free Gibbs energy of the
reaction is below zero ΔG°(r)<0. Direct calculation of all possible ΔG°(r) for each possible
reaction is challenging due to a huge number of possible reactants. Meanwhile some values will
be helpful to elucidate the overall reaction network. In most cases several examples are presented
for the same reaction types (Table 2).
Gas-phase thermodynamic properties of some reactants and products were calculated using
HSC Chemistry 6.084, which has an extensive database, but not covering many reactants
involved in APR. Properties of those compounds were estimated via molecular modeling
software HyperChem 8.085, using a semi-empirical quantum-mechanical method PM3. Free
Gibbs energies of formation ΔG°(f, 498°K) of some compounds are provided in Supported
Information. In some cases the values calculated in HCS 7 are provided for a liquid (l) or
aqueous (aq) phases instead of a gas phase (g). Gibbs formation energies, which were estimated
in the HyperChem software, should be taken into account as approximate values for such
substances as: glyceraldehyde, aldotetrose, aldopentose, aldohexose, ketohexoses and polyols
like tetraol, pentitol and hexitol. The reason is that even if geometry optimization was performed,
the values could differ for different optical isomers.
28
ACS Paragon Plus Environment

Page 29 of 49
ACS Catalysis
1
2
3
4
5
6
7
In most cases Gibbs energies calculated using HCS 7 differ from the HyperChem values,
except for hydrogenolysis and dehydration reactions. Despite these differences the values, which
were estimated in HyperChem, follow the same trends.
8
9
Gibbs energies for the most meaningful reactions are shown in Table 2, and discussed below.
Methanation of CO is thermodynamically favorable (-96.66 kJ/mol). However, it was
demonstrated, that methanation does not occur under experimental conditions of APR on Pt, Re,
Rh, PtRe and RhRe catalysts supported on Al2O3 or carbon74,86,87. Thus, methane formation in
the current case mostly happens via C-C bond breaking in acetaldehyde.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Presence of thermodynamically favorable (-20.5 kJ/mol) water-gas shift reaction was also
proven under APR conditions74,86,87
.
The values of Gibbs energy were calculated for the reduction of carbonyl group. A reversed
reaction of dehydrogenation is shown as paths 1 in Fig. 14. Dehydrogenation is proposed to be
the first step of the APR mechanism. However, reduction was found to be even slightly favorable
for sugars according to HCS 7 calculations (-15.86 kJ/mol for reduction of glucose to sorbitol).
For shorter molecules, as aldopentoses and glyceraldehyde, the values should be very similar in
accordance with HyperChem values. Reduction of ethanedial to 1,2-propanediol is also
thermodynamically favorable (-71.63 kJ/mol). Hydrogenation is slightly favorable for
nonsubstituted molecules like hexanal and propionaldehyde (close to -10 kJ/mol). The case with
ketones is the opposite: reduction of acetone and hexanones is unfavorable (3-4 kJ/mol). At the
elevated temperature, like 200-250°C, the reaction should be predominantly under
thermodynamic control. Therefore, reduction of aldehydes should prevail over reduction of
ketones. This leads to the idea that the reverse reaction of dehydrogenation of secondary
hydroxyl groups should dominate over dehydrogenation of primary ones, unlike it is proposed in
the part 3.2.4. However, the Gibbs energy values are close to zero, thus the equilibrium constant
29
ACS Paragon Plus Environment