Selective metal-catalyzed transfer of H2 and CO from polyols to alkenes

Article abstract of DOI:10.1002/cssc.201200843

Transmission of alcohols achieved: A method for the direct transfer of the CHOH function from simple polyols to alkenes has been developed. In a dual-reactor system, successive iridium-catalyzed dehydrogenations and decarbonylations of polyols such as glycerol and sorbitol generates a low pressure of syngas, which is directly used in exsitu alkene hydroformylation. Copyright

Full text of DOI:10.1002/cssc.201200843

CHEMSUSCHEM
COMMUNICATIONS
DOI: 10.1002/cssc.201200843
Selective Metal-Catalyzed Transfer of H and CO from
2
Polyols to Alkenes
J. Johan Verendel, Michael Nordlund, and Pher G. Andersson*
[a]
[a]
[a, b, c]
To reduce our dependency on fossil fuels, we have to develop
practical methods for converting renewable feedstocks into
suitable starting materials for organic synthesis. The polyols
that make up cellulosic biomass have a high O/C ratio and are
thus very hydrophilic, contrary to the petroleum products
We envisioned that this process could also be applied to
polyols, producing highly pure syngas. C –C polyols are fre-
2
6
[8]
quently appointed as future platform chemicals because they
[9]
are inexpensive and readily available from polysaccharides.
As these chemicals contain C and O in a 1:1 ratio, the reaction
should theoretically degrade them completely to CO and H2,
[
1]
around which most chemical industries are built. Hence, the
transformation of the secondary alcohols (ÀCHOH functionali-
ties) found in sugars to more lipophilic and synthetically useful
compounds is a major challenge. Among the most highly de-
veloped methods for transforming biomass is gasification at
high temperature (typically 700–9008C) to yield a gas-mixture
with the only by-product being an extra equivalent of H gas
2
per molecule. Furthermore, if the H and CO produced were
2
used directly in the hydroformylation of alkenes, the result
would be an efficient transfer of biomass-derived ÀCHOH func-
tionalities into high-value chemicals. Hydroformylation is one
of the most commercially important homogeneous metal-cata-
[2]
rich in hydrogen and carbon monoxide, that is, syngas.
[10]
Syngas is a versatile reagent that can be used in many reac-
tions including alcohol or Fischer–Tropsch synthesis to pro-
lyzed reactions, and its realization without high-pressure cyl-
inders of flammable and toxic gases would be an additional
benefit of the degradation of polyols to syngas.
[
3]
duce fuels and chemicals. Recently, aqueous-phase reforming
has enabled syngas production from aqueous solutions of
[
4]
polyols and sugars. Although this has allowed H –CO mix-
2
tures to be produced more selectively and under milder condi-
[
5]
tions, it still requires high temperatures and pressures.
Recently, Olsen and Madsen reported the selective removal
of mono-alcohols by using iridium catalysts, liberating hydro-
[
6]
gen and carbon monoxide gas (Scheme 1) and Ho et al. dem-
onstrated the same overall metal-catalyzed reaction using pho-
[
7]
tochemical activation.
Scheme 2. Apparatus used to transfer ÀCHOH from polyols through iridium-
catalyzed dehydrogenation–decarbonylation to ex situ alkenes through rho-
dium-catalyzed hydroformylation.
Scheme 1. Iridium-catalyzed tandem dehydrogenation–decarbonylation of
an alcohol liberates H and CO in a 1:1 ratio.
2
Here, we demonstrate the hydroformylation of alkenes using
a low-pressure mixture of H and CO that is produced ex situ
2
[
a] Dr. J. J. Verendel, M. Nordlund, Prof. P. G. Andersson
Department of Chemistry—BMC
Uppsala University
Husargatan 3, Box 576, 751 23, Uppsala (Sweden)
Fax: (+46)018-471-3818
E-mail: pher.andersson@biorg.uu.se
by the selective iridium-catalyzed dehydrogenation-decarbony-
lation of polyols. A simple dual-reactor setup (Scheme 2) was
used to convert olefins to aldehydes in high yields using inex-
pensive C –C polyols as syngas sources.
3
6
For the hydroformylation reaction, we chose to use Wilkin-
son’s hydroformylation catalyst, Rh(H)(CO)(PPh ) , one of the
[b] Prof. P. G. Andersson
3
3
Department of Organic Chemistry
Stockholm University
Arrhenius Laboratory, SE-106 91, Stockholm (Sweden)
E-mail: phera@organ.su.se
few hydroformylation catalysts that have proven efficient even
[11]
under very mild reaction conditions. As hydroformylation is
sensitive to the configuration of the steric bulk around the
double bond and also sometimes accompanied by undesired
[
c] Prof. P. G. Andersson
School of Chemistry
University of KwaZulu-Natal
University Road, Westville, Durban 4041 (South Africa)
[12]
alkene isomerization, we used styrene as the model alkene
while developing this system.
We began by using the primary mono-alcohol 2-(naphtha-
len-2-yl)ethanol, 1 (Figure 1), as the syngas source and reaction
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cssc.201200843.
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 426 – 429 426

CHEMSUSCHEM
COMMUNICATIONS
www.chemsuschem.org
Figure 1. Substrates for dehydrogenative decarbonylation in reactor A (see
Scheme 2).
[
a]
Table 1. Various H
2
/CO sources for the hydroformylation of styrene.
Scheme 3. Iridium-catalyzed dehydrogenation of diol 2 to produce an a-hy-
droxyaldehyde that tautomerizes to yield 5.
Entry
Reactor A Reactor B
[
b]
[d]
substrate
solvent
T
cat. loading
yield [mol%]
RCHO/EtPh
[
c]
[
8C]
[mol%]
1
2
3
4
5
6
7
8
1
2
2
2
3
3
4
4
4
Mes
Mes
Mes
DGDE
Mes
DGDE
DGDE
DGDE
DGDE
185
185
185
210
185
210
210
210
210
5
5
98
85
58
84
26
64
53
46
66
0
11
16
13
10
23
21
20
28
Following the successful transfer of H2 and CO from
a mono-alcohol to styrene, we tested the structurally similar
2.5
2.5
5
2.5
2.5
1
1,2-diol 2 as syngas source (Table 1, entry 2). The catalyst load-
ing and alcohol/alkene ratio, calculated based on the number
of ÀCHOH groups, were maintained at 5 mol% and 1.5:1, re-
spectively; thus, these quantities were 10 mol% and 0.75:1
when calculated based on the number of mols of 2. The
alkene was again fully consumed over 44 h, but the yield of al-
dehydes dropped to 85%; the hydrogenated product ethyl-
benzene was also obtained in 11% yield. This suggested that
H2 accumulated in the system, and the likely cause of the
build-up was revealed upon examining the reaction mixture.
The relatively inert keto alcohol 5 was detected as a major
component of the mixture (Scheme 3). Hence, the a-hydroxyal-
dehyde intermediate formed from the dehydrogenation of 2
was trapped as its keto alcohol tautomer 5, resulting in an ini-
[
e]
9
1
[a] t=44 h. Reaction conditions in reactor A (cf. Scheme 2): 1.5 mol CHOH
units per mol alkene in reactor B; [alcohol]=0.39m; (S)-BINAP was used
as ligand. Reaction conditions in reactor B: styrene (900 mmol, 0.9m, in
C H ), RT. See the Supporting Information for details. [b] Mes=mesitylene,
6 6
DGDE=diethyleneglycol diethyl ether. [c] Catalyst loading with respect to
the number of CHOH functions. [d] RCHO=Aldehyde. Yields were deter-
mined by performing H NMR spectroscopy. See the Supporting Informa-
1
tion for details. [e] 2.0 equiv of CHOH units with respect to styrene.
conditions similar to those previously reported for dehydro-
tial build-up of H that caused some of the styrene to be hy-
2
[
6]
genative decarbonylation (Table 1, entry 1).
[Ir(cod)Cl]2
drogenated. Dehydrogenation of a secondary alcohol, 1-
phenyl-2-propanol, under the same conditions was very slow,
supporting the idea that 5 was formed by tautomerization
rather than from the direct dehydrogenation of the secondary
CHOH functionality of 2. Furthermore, the doubly dehydrogen-
ated 6 was not detected during the dehydrogenative decar-
bonylation of 2. When 2 was subjected to dehydrogenation–
decarbonylation using only 2.5 mol% catalyst with respect to
the number of ÀCHOH groups (i.e., 5 mol%), styrene conver-
sion was incomplete, but a 58% yield of aldehydes was still
achieved (Table 1, entry 3).
(2.5 mol%; COD=1,5-cyclooctadiene) and (S)-BINAP [2,2’-bis(di-
phenylphosphino)-1,1’-binaphthyl; 5.0 mol%] were used as
precatalysts. To ensure the complete conversion of styrene, 1.5
equivalents of the alcohol (with respect to the alkene) were
used. The system was sealed under Ar and reactor A was
heated to 1858C for 44 h. The pressure in the system was
monitored, and the partial pressure of produced gases was ap-
1
proximately constant at around 130 mbar. H NMR analysis of
the hydroformylation solution showed complete conversion of
styrene and 98% yield of aldehydes (linear/branched=3:2).
This preference for the linear product is contrary to what is ob-
When we tested tetraol 3 as source of H and CO, the yield
2
[
11b]
served under a full atmosphere of syngas,
where the
of aldehydes from hydroformylation was low (Table 1, entry 5).
As 3 was poorly soluble in mesitylene, we turned to diethylene
glycol diethyl ether (DGDE) as a solvent. Olsen and Madsen
found that dehydrogenative decarbonylation is slower in
glymes, likely due to competing coordination of the ether
branched aldehyde is the major product. Although regioselec-
tive hydroformylation of styrenes has been problematic in the
past, catalyst development has allowed the selective synthesis
of both branched and linear products under a variety of reac-
[
10a]
[6]
tion conditions.
Analysis of the residues in the syngas-pro-
groups to the catalyst. However, the boiling point of DGDE is
ducing reactor A revealed that 93% of the alcohol had been
consumed, giving b-methylnaphthalene in 85% yield. Small
amounts of the corresponding aldehyde were also detected.
higher than that of mesitylene, and this solvent thus permitted
a higher working temperature that we hoped would compen-
sate for its slowing of the dehydrogenative decarbonylation.
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 426 – 429 427

CHEMSUSCHEM
COMMUNICATIONS
www.chemsuschem.org
Indeed, diol 2 and tetraol 3 yielded 84 and 64% aldehydes, re-
spectively, using 2.5 mol% catalyst with respect to the number
of CHOH groups (entries 4 and 6).
Table 2. Screening of polyol substrates in dehydrogenative decarbonyla-
tion.
[a]
Entry
Reactor A
substrate
Reactor B
yield [mol%]
RCHO/EtPh
The naturally occurring hexitol sorbitol was tested as a sub-
strate under the same reaction conditions (1.5 equivalents of
CHOH equivalents with respect to the alkene; sorbitol/sty-
rene=1:4) and resulted in a 53% yield of aldehydes (Table 1,
entry 7). When the catalyst loading was lowered even further,
to 1 mol% per CHOH unit, the yield dropped to 46% (entry 8).
Increasing the amount of sorbitol to 33 mol% with respect to
alkene (2.0 equiv CHOH units) resulted in almost full conver-
sion of styrene and 66% yield of aldehydes (Table 1, entry 9).
Although a moderate yield of aldehydes was obtained, 25%
of the alkene was hydrogenated to ethylbenzene. Hence, for
[
d]
cat. loading
7
[
b]
[c]
[
mol%]
[equiv]
1
2
3
4
5
6
7
4
4
4
1.0
1.0
3.0
1.0
1.0
1.0
1.0
–
66
55
51
60
54
48
83
28
9
0
32
34
46
3
1.0
1.5
–
–
–
d-mannitol
xylitol
meso-erythritol
glycerol
–
[
a] t=44 h. Reaction conditions in reactor A (see Scheme 2): 2.0 mol
CHOH units per mol alkene in reactor B, (S)-BINAP was used as ligand for
iridium. See the Supporting Information for details. [b] Catalyst loading
with respect to number of CHOH units. [c] Equivalents of additive with re-
spect to sorbitol. [d] RCHO=Aldehyde. Yields were determined by using
longer polyols the effect of H build-up was more pronounced,
2
possibly because several dehydrogenations–isomerizations can
occur. The metal-catalyzed isomerization of a-hydroxyalde-
hydes to hydroxyketones has been reported and is frequently
1
H NMR spectroscopy. See the Supporting Information for details.
[
13]
used for aldose–ketose transformations.
In an attempt to
counter this process, methyl benzoyl formate, 7, was added to
The complete dehydrogenative decarbonylation should theo-
reactor A. a-Keto-esters such as 7 are good H acceptors and
have been used as substrates in transfer hydrogenations.
retically produce seven equivalents of H , six equivalents of
2
2
[14]
CO, and three long-chain hydrocarbons (Scheme 5). Hence,
We hoped that the hydrogenation of 7 would store a fraction
of the H , thus counteracting its accumulation (Scheme 4).
2
2
Scheme 4. Principle of H storage by iridium-catalyzed transfer hydroge-
nation.
Scheme 5. Production of syngas and hydrocarbons from triglycerides.
Indeed, a significantly smaller amount of ethylbenzene was de-
tected after the reaction, but the conversion of styrene drop-
ped significantly, possibly because a lesser fraction of the iridi-
um catalyst was available for dehydrogenative decarbonylation
a 1:3 mixture of glycerol and 1-hexadecanol was subjected to
dehydrogenative decarbonylation in diethyleneglycol diethyl
ether at 2108C with ex situ hydroformylation of styrene.
As predicted, the mixture hydroformylated four equivalents of
styrene with our apparatus. After 66 h, a 1:1 isomeric mixture
of aldehydes was obtained in 98% yield from reactor B and
pentadecane was isolated in 89% yield from reactor A.
(Table 2, entries 2 and 3).
Other polyol substrates could also be used. Unsurprisingly,
d-mannitol produced similar results as sorbitol (Table 2,
entry 4). The pentitol xylitol produced a slightly lower yield of
aldehydes, and more ethylbenzene, compared to sorbitol
(
entry 5) and the tetraol erythritol gave rise to even more eth-
In summary, we have developed a catalytic method for the
direct transfer of CO and H2 from polyols to alkenes using
a dual-reactor system. The homogeneous iridium-catalyzed de-
hydrogenative decarbonylation of alcohols at 185–2108C pro-
ylbenzene (entry 6). Glycerol is another naturally occurring
polyol that, as a by-product of biodiesel production, finds only
limited use in the production of bulk chemicals exists in
[
15]
a large surplus. Here, glycerol degradation for the ex situ hy-
droformylation of styrene produced an 83% yield of aldehydes
and only 3% ethylbenzene (entry 7). A mixture of glycerol and
long-chain aliphatic alcohols can be obtained from triglycer-
duces a low (<0.5 bar) pressure of H and CO, which is suffi-
2
cient to feed an ex situ hydroformylation reaction. Catalyst
loadings of 1 mol% with respect to the number of CHOH func-
tions available were effective in the tandem dehydrogenation–
decarbonylation reaction. Although the degree of CHOH trans-
fer from polyols ranged between 25 and 75 mol%, further de-
velopment could increase the efficiency of the system, which is
[
16]
ides by carbonyl reduction, and we saw an opportunity to
directly use this mixture as a source of syngas for hydroformy-
lation and simultaneously as a source of hydrocarbon fuels.
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 426 – 429 428

CHEMSUSCHEM
COMMUNICATIONS
www.chemsuschem.org
clearly highly selective. Studies are in progress to further devel-
[3] a) O. Levenspiel, Ind. Eng. Chem. Res. 2005, 44, 5073–5078; b) Y.-W. Li, J.
Xu, Y. Yang in Biomass to Biofuels (Eds.: A. A. Vertꢁs, N. Qureshi, H. P. Bla-
schek, H. Yukawa), Blackwell Publishing, Oxford, 2010, pp. 123–139;
c) P. J. A. Tijm, F. J. Waller, D. M. Brown, Appl. Catal. A 2001, 221, 275–
op the concept of direct H /CO-transfer from biomass-derived
2
polyols to synthetic intermediates.
2
82.
4] a) J. C. Serrano-Ruiz, R. Luque, A. Sepulveda-Escribano, Chem. Soc. Rev.
011, 40, 5266–5281; b) G. W. Huber, J. A. Dumesic, Catal. Today 2006,
11, 119–132.
5] a) R. D. Cortright, R. R. Davda, J. A. Dumesic, Nature 2002, 418, 964–967;
b) G. W. Huber, J. W. Shabaker, J. A. Dumesic, Science 2003, 300, 2075–
[
Experimental Section
2
1
General procedure for dehydrogenative decarbonylation with
ex situ hydroformylation: [Ir(cod)Cl]2 (6.05 mg, 9 mmol), (S)-BINAP
[
(
11.21 mg, 18 mmol), and polyol (300 mmol) were weighed into re-
2077; c) R. R. Soares, D. A. Simonetti, J. A. Dumesic, Angew. Chem. 2006,
actor A and Rh(H)(CO)(PPh ) (8.30 mg, 9 mmol) was weighed into
3
3
118, 4086–4089; Angew. Chem. Int. Ed. 2006, 45, 3982–3985.
reactor B. To reactor A was attached a cold-finger reflux condenser
and to reactor B a 1–2000 mbar pressure reader. The system was
purged with argon and benzene (1.0 mL) and styrene (103 mL,
[6] E. P. K. Olsen, R. Madsen, Chem. Eur. J. 2012, 18, 16023–16029.
[7] H.-A. Ho, K. Manna, A. D. Sadow, Angew. Chem. 2012, 124, 8735–8738;
Angew. Chem. Int. Ed. 2012, 51, 8607–8610.
[
8] a) J. J. Bozell, G. R. Petersen, Green Chem. 2010, 12, 539–554; b) T.
Werpy, G. Petersen, Vol. I: Results of Screening for Potential Candidates
from Sugars and Synthesis Gas, U. S. Department of Energy, Pacific
Northwest National Laboratory, 2004.
9
(
00 mmol) were added to reactor B by using a cannula. Solvent
3.5 mL) was added to reactor A, and the system pressure was
equalized with the surroundings. The solutions were stirred for
0 min at room temperature before reactor A was immersed in
1
[
9] J. J. Verendel, T. L. Church, P. G. Andersson, Synthesis-Stuttgart 2011,
a thermostatted silicone oil bath. Upon complete reaction, the
system was cooled to room temperature and vented. The hydro-
1649–1677.
[
10] a) B. Breit in Topics in Current Chemistry, Vol. 279 (Ed.: M. Krische),
Springer, Berlin, 2007, pp. 139–172; b) M. Beller, K. Kumar, in Transition
Metals for Organic Synthesis (Eds.: M. Beller, C. Bolm), Wiley-VCH, Wein-
heim, 2008, pp. 28–55.
1
formylation reaction was analyzed by using H NMR spectroscopy
with 1,3,5-trimethoxybenzene as internal standard. See the Sup-
porting Information for detailed experimental procedures.
[
11] a) D. Evans, J. A. Osborn, G. Wilkinson, J. Chem. Soc. A 1968, 3133–
3142; b) C. K. Brown, G. Wilkinson, J. Chem. Soc. A 1970, 2753–2764.
[
[
12] B. Breit, W. Seiche, Synthesis 2001, 0001–0036.
13] a) R. W. Nagorski, J. P. Richard, J. Am. Chem. Soc. 2001, 123, 794–802;
b) H. Zhao, J. E. Holladay, H. Brown, Z. C. Zhang, Science 2007, 316,
Acknowledgements
We are grateful to Prof. R. Madsen, Dr. T. L. Church, and Mr.
E. P. K. Olsen for useful discussions. Energimyndigheten (the Swed-
ish Energy Agency), Nordic Energy Research (N-INNER II), the
Swedish Research Council (VR), the Knut and Alice Wallenberg
Foundation, and VR/SIDA supported this work.
1
597–1600; c) V. Choudhary, A. B. Pinar, S. I. Sandler, D. G. Vlachos, R. F.
Lobo, ACS Catal. 2011, 1, 1724–1728; d) Y. Romꢂn-Leshkov, M. Moliner,
J. A. Labinger, M. E. Davis, Angew. Chem. 2010, 122, 9138–9141; Angew.
Chem. Int. Ed. 2010, 49, 8954–8957.
ˇ
[
[
14] a) D. Sterk, M. S. Stephan, B. Mohar, Tetrahedron Lett. 2004, 45, 535–
37; b) J. W. Yang, B. List, Org. Lett. 2006, 8, 5653–5655; c) K. M. Stew-
5
ard, E. C. Gentry, J. S. Johnson, J. Am. Chem. Soc. 2012, 134, 7329–7332.
15] a) A. A. Koutinas, C. Du, R. H. Wang, C. Webb in Introduction to Chemi-
cals from Biomass (Eds.: J. H. Clark, F. E. I. Deswarte), Wiley, Chichester,
2008, pp. 77–101; b) B. Sels, E. D’Hondt, P. Jacobs in Catalysis for Re-
newables (Eds.: G. Centi, R. A. van Santen), Wiley-VCH, Weinheim, 2007,
pp. 223–255.
[16] a) T. Turek, D. L. Trimm, N. W. Cant, Catal. Rev. 1994, 36, 645–683; b) K.
Noweck, W. Grafahrend in Ullmann’s Encyclopedia of Industrial Chemistry,
Wiley-VCH, Weinheim, 2006.
Keywords: alcohols
·
decarbonylation
·
homogeneous
catalysis · hydroformylation · transition metals
[
[
1] J.-P. Lange in Catalysis for Renewables (Eds.: G. Centi, R. A. van Santen),
Wiley-VCH, Weinheim, 2007, pp. 21–51.
2] a) M. Arshadi, A. Sellstedt in Introduction to Chemicals from Biomass
(
Eds.: J. H. Clark, F. E. I. Deswarte), Wiley, Chichester, 2008, pp. 143–178;
b) S. R. A. Kersten, W. P. M. van Swaaij, L. Lefferts, K. Seshan in Catalysis
for Renewables (Eds.: G. Centi, R. A. van Santen), Wiley-VCH, Weinheim,
2
007, pp. 119–145; c) P. Mondal, G. S. Dang, M. O. Garg, Fuel Process.
Received: November 5, 2012
Technol. 2011, 92, 1395–1410.
Published online on January 9, 2013
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 426 – 429 429