Rhodium catalyzed hydroformylation with formaldehyde and an external H 2-source

Article abstract of DOI:10.1016/j.tetlet.2013.02.049

The efficiency of the syngas-free rhodium catalyzed hydroformylation of olefins with formaldehyde can be significantly improved by the addition of hydrogen gas or formic acid.

Full text of DOI:10.1016/j.tetlet.2013.02.049

Tetrahedron Letters 54 (2013) 2209–2211
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Rhodium catalyzed hydroformylation with formaldehyde and an external
H2-source
Marcus Uhlemann a, Stephan Doerfelt b, Armin Börner a,c,
⇑
a Leibniz-Institut für Katalyse an der Universität Rostock e.V., Albert-Einstein Straße 29a, 18059 Rostock, Germany
b Miltitz Aromatics GmbH, PF 1355, 06756 Bitterfeld-Wolfen, Germany
c Institut für Chemie der Universität, Rostock A.-Einstein-Str. 31, 18059 Rostock, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
The efficiency of the syngas-free rhodium catalyzed hydroformylation of olefins with formaldehyde can
be significantly improved by the addition of hydrogen gas or formic acid.
Ó 2013 Elsevier Ltd. All rights reserved.
Received 3 January 2013
Revised 13 February 2013
Accepted 15 February 2013
Available online 24 February 2013
Keywords:
Hydroformylation
Olefin
Rhodium
Formaldehyde
Hydrogen
Introduction
complexes based on tripodal triphosphines have been suggested.6
Rosales et al. gave evidence that in strong contrast to the hydrofor-
The hydroformylation of olefins, that means the addition of CO
and H2 (syngas) to olefins, is one of the largest homogeneously cat-
alyzed reaction in industries.1 In this respect exclusively catalysts
on the basis of cobalt and rhodium are used. They are frequently
modified by trivalent phosphorus compounds as ligands. Annually
more than 10 million tons of aldehydes are produced by this meth-
odology.2 It is interesting to note, that with the exception of some
products in pharmaceutical, agrochemical, and fragrance industry,3
the reaction is mainly employed for the large-scale production of
bulk chemicals. Besides the high price of the catalyst (rhodium)
the work with highly toxic and gaseous CO is another argument
against the use of this reaction in academic labs and for the man-
ufacturing of fine chemicals. Meanwhile also the increasing price
for CO may decide upon the economy of small scale hydroformyla-
tions. Non-toxic and cheap substitutes for syngas would overcome
these problems.4 One of the most promising alternatives is
represented by paraformaldehyde or an aqueous solution of form-
aldehyde (formalin) in combination with a phosphorus modified
Rh-catalyst.5 The catalyst must be active in the decarbonylation
of formaldehyde as well as in the hydroformylation of the olefin.
Up to now most efforts have been dedicated to the optimization
of the catalysts. Thus, for the decarbonylation of aldehydes Rh-
mylation with syngas a double excess of bidentate phosphorus li-
gands in comparison to rhodium supports the decomposition of
paraformaldehyde as well as the subsequent hydroformylation
reaction.7 Recently, Morimoto and co-workers showed that a com-
bination of two catalysts, Rh-BINAP and rhodium complexes bear-
ing Xantphos-type ligands, can efficiently mediate the reaction
with formaldehyde in toluene.8 They assumed that the former cat-
alyst decomposes formaldehyde to give syngas and the second
immediately utilizes it for the hydroformylation of the olefin. Un-
der these conditions several aldehydes could be produced in good
yields and excellent n-regioselectivity within 20 h. A similar cata-
lytic mechanism as suggested for the hydroformylation with syn-
gas was suggested. In addition, Taddei and co-workers noted that
the reaction can be accelerated by the effect of microwaves.9
Bearing the known dependency of some Rh-catalyzed hydro-
formylations with syngas on the hydrogen partial pressure in
mind,10 we investigated the reaction with formaldehyde in the
presence of hydrogen gas (Scheme 1). We had hoped that the com-
peting hydrogenation of the starting olefin can be suppressed.
Results and discussion
Reactions with formaldehyde (37 % aqueous solution) were
conducted at 10 bar H2 in a 25 mL reactor. In first trials the precat-
alyst was prepared by reaction of [{Rh(cod)}2(l-Cl)2] with BINAP.
⇑
Corresponding author. Tel.: +49 381 1281 202; fax: +49 381 1281 51202.
E-mail address: armin.boerner@catalysis.de (A. Börner).
0040-4039/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tetlet.2013.02.049

2210
M. Uhlemann et al. / Tetrahedron Letters 54 (2013) 2209–2211
without H2 significantly improved conversion, yield of desired
CHO
aldehyde, and n-regioselectivity could be achieved. That means
turnover numbers (TON) as well as turnover frequencies (TOF)
were ca. two- or threefold higher than without hydrogen. Also in
comparison to the trial where syngas (ratio CO/H2 = 1:1, 10 bar) in-
stead of formaldehyde was used (Table 2, run 7), a dramatic
improvement of reactivity and selectivity was noted. With dppf
as the ligand an extremely high n-regioselectivity of 25 was ob-
tained (run 9). In order to elucidate, whether also other hydrogen
sources can be used, formic acid was added to the reaction (Table 2,
runs 10–12). Also under these conditions with all three catalysts
the reaction with formaldehyde is clearly supported. But in com-
parison to the reaction with hydrogen gas a higher isomerization
tendency of the starting olefin was found.
R
l
HCHO, Rh-cat.
+ H2
isomerized
olefin
+
+
R
OHC
R
R
b
Scheme 1. Hydroformylation with formaldehyde in the presence of hydrogen gas.
Table 1
Rh-catalyzed hydroformylation of 1-octene with formaldehyde/H2 with BINAP as
liganda
To investigate the scope of the new method in the subsequent
trials also other olefins were subjected to these reaction conditions
(Table 3).
Run
Solvent
pH2
Conv.b
(%)
Yieldb
(%)
l/bb
Isom.b,c
(%)
Octaneb
(%)
(bar)
By using a Rh(dppp) catalyst without exception with all sub-
strates improved reaction rates were achieved in comparison to
the reaction without hydrogen gas. This holds especially for the
reaction with styrenes (runs 1–10). Remarkably, a decrease of the
l/b-ratio yielded under the new conditions. With allylbenzene a
strong isomerization activity was noted, also in the absence of
hydrogen gas and the yield of the desired aldehyde was low (Ta-
ble 3, runs 15 and 16). N-Vinylphthalimide and butylacrylate suf-
fered partial hydrogenation (runs 18 and 20).
1
2
3
4
Toluene
Toluene
THF
—
10
—
34
79
27
59
30
69
23
56
1.8
3.9
1.9
2.4
4
5
4
2
—
5
—
1
THF
10
a
Conditions: [1-octene] = 1 M; 1.2 equiv formaldehyde, [{Rh(cod)}2(l-Cl)2]/BIN-
AP/1-octene = 1:4:1000, 90 °C, solvent, 6 h.
b
Determined by GC.
c
Contained mainly isomerized olefins and traces of other internal aldehydes.
Finally, the reactions were scaled-up in a 1 L batch reactor under
a reduced H2-pressure of 5 bar. Under this modified technical set-
up the rate accelerating effect of hydrogen on the hydroformylation
remained clearly visible, but now isomerization of the starting ole-
fin became a side reaction to some extent (e.g., with 1-octene: 17%).
1-Octene was used as the substrate at 90 °C in toluene. To our de-
light, we found that results obtained under these conditions were
superior to those obtained without hydrogen gas (Table 1).
Thus, within 6 h reaction time, conversion of olefin and yield of
the desired aldehyde could be almost doubled in comparison to the
parent set-up (Table 1, runs 1 and 2). Interestingly also a consider-
able increase in n-regioselectivity was found. The degree of
isomerization of the starting olefin was almost the same under
both conditions. Only a small amount of octane as a product of
the hydrogenation was produced under hydrogen. Similar results
were obtained in THF as the solvent. Since THF has some disadvan-
tages in industrial scale, all subsequent trials were performed in
toluene.
In the next attempts we investigated the effect of phosphorus
ligands. The reaction time was shortened to 2 h and the tempera-
ture was increased to 120 °C. Table 2 clearly shows that also dppp
[1,3-bis(diphenylphosphino)propane] and dppf [1,10-bis(diphenyl-
phosphino)ferrocene] can be advantageously used as ligands in-
stead of BINAP. In all instances in comparison to the reaction
Summary and conclusion
In conclusion, a new protocol for the syngas-free hydroformyl-
ation has been discovered, which should be particularly useful for
the production of aldehydes on a small scale. Pivotal aspect is that
for the acceleration of the reaction and for the improvement of the
regioselectivity additional hydrogen sources like gaseous hydrogen
or formic acid are used.11 We assume that hydrogen supports
the hydrogenolysis of the Rh-acyl complex in a late stage of the
catalytic cycle.12
Experimental section
Typical procedure for the hydroformylation.
Table 2
Rh-catalyzed hydroformylation with 1-octene with formaldehyde/H2 using different P-ligandsa
Run
Ligand
pH2 (bar)
Conv.b (%)
Yieldb,c (%)
l/bb
1.9
Isom.b,d (%)
Octane (%)
1
2
4
5
7
8
9
10
11
12
BINAP
BINAP
dppp
dppp
dppp
dppf
—
10
—
30
81
38
84
26
46
93
68
97
75
28
72
32
68
26
25
55
59
68
40
2
8
5
11
—
20
22
8
—
1
1
5
—
1
16
1
5
2.3
1.3
1.3
1.0
6.5
25.0
2.1
1.4
6.9
10
e
—
—
dppf
10
f
BINAP
dppp
dppf
—
—
—
f
24
33
f
2
a
b
c
d
e
f
Conditions: [1-octene] = 0.5 M, 1.2 equiv formaldehyde, [Rh(cod)2(
Determined by GC.
Yield of aldehydes.
Contained mainly isomerized olefins and traces of other internal aldehydes.
2 h, without formaldehyde, but with syngas (CO/H2 = 1/1, 10 bar).
Without hydrogen gas, but with formic acid (1.0 equiv).
l-Cl)2]/ligand/1-octene = 1:4:1000, 120 °C, toluene.

M. Uhlemann et al. / Tetrahedron Letters 54 (2013) 2209–2211
2211
Table 3
Rh(dppp)-catalyzed hydroformylation of different olefins with formaldehyde/H2
a
Run
Olefin
pH2 (bar)
Conv.b (%)
Yieldb,c (%)
l/bb
Isom.b,d (%)
Hydr. (%)
1
2
4
5
7
Styrene
Styrene
4-Cl-styrene
4-Cl-styrene
4-MeO-styrene
4-MeO-styrene
4-tBu-styrene
4-tBu-styrene
Cyclooctene
—
10
—
10
—
10
—
10
—
10
—
10
—
10
—
23
91
14
81
13
77
11
82
<1
7
61
88
90
99
11
46
2
23
83
14
76
13
74
11
78
<1
7
61
85
1
2
6
1.1
0.6
0.8
0.5
0.6
0.4
0.7
0.5
—
—
—
—
—
—
—
—
—
—
—
—
—
89
96
—
—
—
12f
—
8
—
5
—
3
—
4
—
—
—
3
-
1
8
9
10
11
12
13
14
15
16
17
18
19
20
Cyclooctene
—
3,3-Dimethyl-1-butene
3,3-Dimethyl-1-buten
Allylbenzene
11
22
2.3
2.1
0.7
0.1
—
Allylbenzene
N-Vinylphthalimidee
N-Vinylphthalimidee
Butylacrylate
Butylacrylate
5
18
2
10
—
10
28
—
17
71
3.7
38
a
b
c
d
e
f
Conditions: [olefin] = 0.5 M, 1.2 equiv formaldehyde, [{Rh(cod)}2(
Determined by GC.
Yield of aldehyde.
Contained mainly isomerized olefins or other internal aldehydes (see also footnote f).
N-Vinylphthalimid = 0.05 M.
Concerns alcohols formed.
l-Cl)2]/dppp/olefin = 1:4:1000, 120 °C, toluene, 20h reaction time.
2. Process Economics Program Report 21E OXO ALCOHOLS, Syed Naqvi, SRI
Consulting, Menlo Park, California 94025, USA, 2010.
[{Rh(cod)}2(l-Cl)2] (5 lmol), phosphorus ligand (20 lmol), and
solvent (6.3 mL) were added to a pressure reactor (25 mL). After
addition of formaldehyde (37 % in H2O, 6 mmol) and substrate
(5 mmol) the reaction mixture was purged with nitrogen for sev-
eral times and heated with stirring (1000 rpm) to the indicated
reaction temperature. Once the desired temperature was achieved,
H2 was introduced and the reaction mixture was stirred for the
indicated reaction time. After cooling to room temperature the
gas was released and the reaction mixture was again purged with
nitrogen. Dodecane was added as an internal standard and brine in
order to separate the phases. The organic phase was finally ana-
lyzed by GC.
3. Whiteker, G. T.; Cobley, C. Top. Organomet. Chem. 2012, 42, 35.
4. Morimoto, T.; Kakiuchi, K. Angew. Chem., Int. Ed. 2004, 43, 5580.
5. (a) Ahn, H. S.; Han, S. H.; Uhm, S. J.; Seok, W. K.; Lee, H. N.; Korneeva, G. A. J. Mol.
Catal. A: Chem. 1999, 144, 295; (b) Okano, T.; Kobayashi, T.; Konishi, H.; Kiji, J.
Tetrahedron Lett. 1982, 23, 4967.
6. (a) Beck, C. M.; Rathmill, S. E.; Park, Y. J.; Chen, J.; Crabtree, R. H.; Liable-Sands,
L. M.; Rheingold, A. L. Organometallics 1999, 18, 5311; (b) Kreis, M.; Palmelund,
A.; Bunch, L.; Madsen, R. Adv. Synth. Catal. 2006, 348, 2148.
7. (a) Rosales, M.; Arrieta, F.; Baricelli, P.; Gonzalez, A.; Gonzalez, B.; Guerrero, Y.;
Moratinos, C.; Pacheco, I.; Perez, H.; Urdaneta, J. Catal. Lett. 2008, 126, 367; (b)
Rosales, M.; Gonzáles, A.; Gonzáles, B.; Moratinos, C.; Pérez, H.; Urdaneta, J.;
Sánchez-Delgado, R. A. J. Organomet. Chem. 2005, 690, 3095.
8. Makado, G.; Morimoto, T.; Sugimoto, Y.; Tsutsumi, K.; Kagawa, N.; Kakiuchi, K.
Adv. Synth. Catal. 2010, 352, 299.
9. Airiau, E.; Gerard, N.; Mann, A.; Salvadori, J.; Taddei, M. Synlett 2011, 199.
10. (a) Kubis, Ch.; Selent, D.; Sawall, M.; Ludwig, R.; Neymeyr, K.; Baumann, W.;
Franke, R.; Börner, A. Chem. Eur. J. 2012, 18, 8780; (b) Kubis, C.; Ludwig, R.;
Sawall, M.; Neymeyr, K.; Börner, A.; Wiese, K.-D.; Hess, D.; Franke, R.; Selent, D.
ChemCatChem 2010, 2, 287.
Acknowledgments
We acknowledge financial support by AiF Projekt GmbH und
stimulating discussions with Mr. J. Jimenez Pinto (Bitterfeld-
Wolfen), Dr. J. Norinder, and Dr. D. Selent (Rostock).
11. When we tested the dual system consisting of two different catalysts described
in Ref.
8 under our conditions, no rate accelerating effect on the
hydroformylation was observed.
12. Preliminary spectroscopic investigations did not give clear experimental
support for this hypothesis, in particular with respect to observations
described in Ref. 10. Probably the hydroformylation with formaldehyde
References and notes
proceeds via
a different mechanism in comparison to the reaction with
1. (a)Rhodium Catalyzed Hydroformylation; van Leeuwen, P. W. N. M., Claver, C.,
Eds.; Kluver Academic Publishers: Dordrecht, Netherlands, 2000; (b) Franke, R.;
Selent, D.; Börner, A. Chem. Rev. 2012, 112, 5675.
syngas. Therefore more detailed kinetic investigations are scheduled.