A general and convenient method for the rhodium-catalyzed decarbonylation of aldehydes

Article abstract of DOI:10.1002/adsc.200600228

A practical protocol for the decarbonylation of a wide range of aldehydes has been developed by using commercially available RhCl₃· SH₂O and dppp in a diglyme solution. This method gives rise to decarbonylated products in good to high yield and is particularly useful because of its experimental simplicity, high generality and excellent level of functional group tolerance. The reaction has been applied in a tandem Oppenauer oxidation-decarbonylation sequence, which removes a hydroxymethyl group in one operation.

Full text of DOI:10.1002/adsc.200600228

FULL PAPERS
DOI: 10.1002/adsc.200600228
A General and Convenient Method for the Rhodium-Catalyzed
Decarbonylation of Aldehydes
Michael Kreis,a Anders Palmelund,a Lennart Bunch,a and Robert Madsena,*
a
Center for Sustainable and Green Chemistry, Department of Chemistry, Building201,
Technical University of Denmark, 2800 Lyngby, Denmark
Fax : (+45)-4593-3968; e-mail: rm@kemi.dtu.dk
Received: May 17, 2006; Accepted: August 16, 2006
Supportinginformation for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: A practical protocol for the decarbonyla- tional group tolerance. The reaction has been ap-
tion of a wide range of aldehydes has been devel- plied in a tandem Oppenauer oxidation-decarbonyla-
oped by usingcommercially available RhCl ·3H2O tion sequence, which removes a hydroxymethyl
3
and dppp in a diglyme solution. This method gives group in one operation.
rise to decarbonylated products in good to high yield
and is particularly useful because of its experimental Keywords: aldehydes; decarbonylation; Oppenauer
simplicity, high generality and excellent level of func- oxidation; rhodium; tandem reaction
Introduction
thesis projects.[7] Based on these observations we de-
cided to search for a more practical catalytic method.
In 1965 Tsuji and Ohno described the decarbonyla- Herein, we describe the development of a more con-
tion of aldehydes with stoichiometric amounts of Wil- venient procedure for the decarbonylation of alde-
kinsonꢀs catalyst [Rh
(PPh3)3Cl].[1] Three years later hydes by the use of commercially available reagents.
G
the same authors reported the first example of a rho-
dium-catalyzed decarbonylation with 0.5% of
Rh(CO)
A
The high temperature did cause some side reactions
and in 1978 Doughty and Pignolet described a milder The initial studies were carried out with Rh(dppp)2Cl
N
procedure with Rh
phosphino)propane].[3] This catalyst was shown to per- ligand exchange from Rh
form the decarbonylation of heptanal and benzalde- in one step from RhCl3·3H2O.[8] Unfortunately, Rh-
hyde at temperatures between 115 and 1788C.[3] How-
(dppp)2Cl is poorly soluble in organic solvents like di-
ever, Rh(dppp)2Cl is not commercially available, oxane, toluene and xylene. Addition of water im-
A
N
AHCTREUNG
AHCTREUNG
ACHTREUNG
which may explain why relatively little work has been proved the solubility and it was found that a 10:1 mix-
published on the application of this method. There ture of dioxane and water dissolved the catalyst com-
are only a few examples in the literature where alde- pletely. However, decarbonylation of naphthaldehyde
hydes have been decarbonylated with Rh(dppp)2Cl at reflux in this mixture was very sluggish and re-
N
and in some of these cases the catalyst was generated quired several days for complete conversion. Further-
in situ from dppp and another Rh(I) complex more, the reaction was difficult to reproduce under
{Rh(CO)
there are also examples where the catalytic decarbon- formed in a closed vessel at 2008C in a microwave
ylation with Rh(dppp)2Cl has failed apparently due to oven. To our delight this experiment gave full conver-
decomposition of the catalyst due to its high sensitivi- sion of naphthaldehyde in 30 min with 5% of
ty towards air.[5] More recently, Rh(CO)
(triphos)SbF6 Rh(dppp)2Cl. The same result was obtained with 5%
was shown to decarbonylate aldehydes,[6] but this cat- of Rh
(COE)2Cl in the presence of 10% of dppp while
alyst is not easy to prepare. As a result, stoichiometric Rh(COE)2Cl or Rh(PPh3)3Cl in the absence of other
amounts of Wilkinsonꢀs catalyst are still beingused phosphine ligands only gave 10% conversion after
for decarbonylation of aldehydes in many total syn- 30 min. Interestingly, the active catalyst could also be
A
G
ACHTREUNG
A
ACHTREUNG
AHCTREUNG
A
ACHTREUNG
2148
ꢁ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 2148 – 2154

Rhodium-Catalyzed Decarbonylation of Aldehydes
FULL PAPERS
Table 1. Rhodium-catalyzed decarbonylation of 2-naphthal-
dehyde usingdifferent solvents.
generated in situ from commercially available
RhCl3·3H2O. Treatment of naphthaldehyde with 5%
of RhCl3·3H2O and 10% of dppp in the microwave
oven at 2008C gave complete conversion into naph-
thalene in 30 min. Duringthe experiment the pressure
in the vial increased to about 20 bars since carbon
monoxide is produced in the course of the reaction.
This makes it difficult to use a closed system for the
decarbonylation and may also lead to deactivation of
the rhodium catalyst. Therefore, different high boiling
solvents were screened to find a suitable alternative
to the dioxane/water mixture (Table 1).
The fastest reactions were observed by usingdi-
glyme or NMP with full conversion after 3 h (entries 1
and 2) whereas DMSO inhibited the reaction
(entry 3). In mesitylene, heptanol, diisobutyl ketone
and ethyl hexanoate (entries 4–7) the reaction also
went to completion, but reaction times of 10–20 h
were required. Accordingly, we selected diglyme as
the solvent of choice, since it allows a lower and more
constant reaction temperature than NMP.
[a]
Entry Solvent
Boiling
point
Time to reach full con-
version[b]
1
Diglyme
NMP
DMSO
Mesitylene
Heptanol
Diisobutyl
ketone
1628C
2028C
1898C
1648C
1768C
1688C
3 h
3 h
-
20 h
10 h
10 h
2[c]
3
[d]
4
5
6
7
Ethyl hexa-
noate
1668C
10 h
[a]
Reaction conditions: Naphthaldehyde (1.0 equiv.),
RhCl3·3H2O (5 mol%), dppp (10 mol%) in the corre-
spondingsolvent at reflux under argon.
By GC-MS.
Heated to 1908C.
Stalled after 24 h at 3% conversion.
[b]
[c]
[d]
With this procedure in hand a number of experi-
ments were then conducted in order to identify the
optimal phosphine ligand (Figure 1). We applied
mono-, bi- and tridentate phosphine ligands as well as
one aminophosphine ligand in the decarbonylation versions below 20%. Similarly, the tridentate ligands
with RhCl3·3H2O. 1 and 2 gave almost no conversion when the active
The initial results with monodentate ligands [PPh3, catalyst was generated in situ. Thus, only bidentate li-
PCy3, P(o-furyl)3 and PCy2(2-biphenyl)] all gave con- gands were further studied (Table 2).
ACHTREUNG
Figure 1. Phosphine ligands tested.
Adv. Synth. Catal. 2006, 348, 2148 – 2154
ꢁ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
2149

Michael Kreis et al.
FULL PAPERS
Table 2. Rhodium-catalyzed decarbonylation of 2-naphthal-
dehyde usingdifferent ligands.
determine the turnover frequency and compared the
results to the preformed catalyst usingphenylacetal-
dehyde as the startingmaterial. Applyingstandard
conditions with 0.05 mol% of rhodium catalyst gave
no conversion with the preformed catalyst, whereas
full conversion was achieved with the in situ formed
dppp catalyst. Even degassing of the reaction mixture
for 1 h by sonication under an argon atmosphere was
not sufficient to get any conversion with 0.05 mol%
[a]
Entry Ligand
Time Conversion[b] Bite angle (stan-
dard deviation)[c]
1
2
3
4
5
6
7
8
1
2
3 h
3 h
3 h
3 h
6 h
3 h
3 h
6 h
6 h
3 h
3 h
6 h
<5%
<5%
<5%
quant.
quant.
<5%
<5%
61%
dppe
dppp
dppb
3
858 (38)
918 (28)
988 (58)
of Rh(dppp)2Cl. Only a thorough degassing of the sol-
A
vent by refluxing overnight under an argon atmos-
phere and subsequent addition of the catalyst and the
startingmaterial lead to full conversion of the starting
4
838 (38)
104–1078
Xantphos
Davephos
dppf
BINAP
Walphos II
9
20%
material with 0.05 mol% of Rh(dppp)2Cl. For the
T
10
11
12
13
quant.
quant.
75%
968 (28)
928 (38)
purposes of comparison, this procedure was also ap-
plied for the in situ formed catalysts with dppp, dppf
Taniaphos I 3 h
quant.
and BINAP. The preformed catalyst Rh(dppp)2Cl and
T
[a]
the in situ formed catalysts with dppp and BINAP
showed full conversion towards the desired product
within 24 h. Fastest reactingwas the preformed
system with a turnover frequency (TOF) of 575 hÀ1
compared to the in situ formed systems with TOFs of
450 hÀ1 for BINAP and 390 hÀ1 for dppp. On the
other hand, the preformed catalyst with dppf initially
showed the fastest reaction of all the in situ formed
catalyst with a TOF of 560 hÀ1 but after 1 h the reac-
Reaction conditions: Naphthaldehyde (1.0 equiv.),
RhCl3·3H2O (5 mol%), ligand (10 mol%) in diglyme at
reflux under argon.
[b]
[c]
By GC-MS.
Bite angle according to the literature.[9]
When using the homologous ligands dppe, dppp, tion mixture started to turn dark and the formation of
and dppb, a maximum in reactivity was observed for toluene ceased while at the same time the formation
dppp. Only traces of product were observed with 3 of (Z)-2,4-diphenylbut-2-enal, an aldol condensation
and 4, which have a more rigid backbone and a small- product, started. Work-up after 24 h showed full con-
er bite angle. The more flexible ligand Xantphos[9] sumption of the startingmaterial and a mixture of tol-
with a large bite angle of 104–1078 gave an improved uene and (Z)-2,4-diphenylbut-2-enal. Hence for the
reactivity (61% yield after 6 h) but was still not com- preformed system as well as for the in situ formed sys-
parable to dppp whereas the mixed P,N ligand Dave- tems with dppp and BINAP a TON of>2000 could
phos[10] resulted in only minor amounts of the desired be achieved. While the preformed catalyst showed an
product. When dppf or BINAP were used, the decar- approximately 1.5 times faster reaction than the in
bonylation went to completion in 3 h. Surprisingly, situ formed catalyst, the in situ formed catalytic
the decarbonylation of naphthaldehyde with Sol- system showed a much higher tolerance towards air
phos,[11] a structural analogue of BINAP resulted in than the preformed catalyst. Since dppp reacts more
only traces of naphthalene. The chiral ferrocene li- or less as fast as BINAP and is much cheaper we se-
gands Walphos II[11] and Taniaphos I[11] showed a high lected this ligand for general use.
reactivity with the latter havinga similar reactivity as
To examine the reducingaegnt with the
in situ
dppp. The best ligands for the reaction, dppp, dppf, formed catalyst we carried out 31P NMR experiments
BINAP and Taniaphos I, all have an average bite in deuterated diglyme. Thus, RhCl3·3H2O and 2
angle between 91 and 968[9] and have a rather flexible equivalents of dppp were heated to reflux in deuterat-
backbone. Bidentate ligands with smaller bite angles ed diglyme under argon in the presence and in the ab-
like dppe, 3, and 4 gave almost no conversion while sence of 1 equivalent of benzaldehyde. NMR samples
Xantphos and dppb with a larger bite angle reacted were taken out and measured at 608C. Due to the
slowly. Hence, the most reactive ligands were dppp, much lower sensitivity of the rhodium-coordinated
dppf, BINAP and Taniaphos I. However, in the reac- phosphines and a partial precipitation of the rhodi-
tion with dppf the mixture quickly turned black which um-complex by coolingfrom 162 8C to 608C we were
indicates that either the ligand or the active rhodium only able to observe the non-coordinated phosphines.
catalyst is not sufficiently stable. With dppp, BINAP A spectrum of Rh
ACHTREUNG
and Taniaphos I the reaction mixture maintained a by usingD 2O as
a
clear yellow color with no sign of metal precipitation. RhCl3·3H2O with dppp and aldehyde showed after
To better quantify the catalytic activity of the dif- 1 min four singlet signals at À16.2 ppm (identified as
ferent ligands, we performed kinetic measurements to dppp), 29.5 ppm (identified as dppp dioxide) and
2150
www.asc.wiley-vch.de
ꢁ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 2148 – 2154

Rhodium-Catalyzed Decarbonylation of Aldehydes
FULL PAPERS
À16.4 ppm and 28.6 ppm (most likely the dppp mon- a,a,a-trifluorotoluene (entry 3, 78%), chlorobenzene
oxide, which showed À17.5 ppm and 28.7 ppm in pure (entry 4, 83%), toluene (entry 7, 76%), anisole
diglyme). After 10 min the signal of the free dppp (entry 8, 74%), and ethylbenzene (entry 13, 70%)
and the dppp monoxide disappeared leavingonly
were caused by work-up problems due to the volatili-
dppp dioxide in solution which shows that RhCl3 is ty of the products. In some cases, the products were
reduced by dppp which in turn is oxidized to non-co- isolated by direct distillation from the reaction mix-
ordinatingdppp dioxide. The spectra in the absence
of benzaldehyde showed the same result confirming glyme solution as shown by GC analysis. p-Nitroben-
that the aldehyde is not involved in the reduction. zaldehyde partially decomposed duringthe reaction
ture, but some product always remained in the di-
To determine the scope and limitations of the de- and the work-up (entry 6) and only gave 12% isolated
carbonylation procedure several different aldehydes yield together with some reduced by-products. Exten-
were subjected to the optimized conditions (Table 3). sion of the procedure towards enolizable aliphatic al-
Aldehydes containingether, ester, amino, cyano,
amide and imide groups as well as chlorides and fluo- (entry 10) and p-xylene (entry 11) in high yields (90%
rides are converted to the correspondingdecarbony- and 81%, respectively) with no sign of a competing
dehydes proceeded smoothly and gave nonane
lated products. Both electron-poor (entries 2–5) and aldol reaction. The decarbonylation of an a,b-unsatu-
electron-rich (entries 7–9) aromatic compounds react- rated aldehyde (entry 12) could also be achieved in
ed in good to high yield. The slightly lower yields of
Table 3. Decarbonylation of aldehydes.
Table 3. (Continued)
Entry
Aldehyde
Method[a]
A
Yield[b]
87%
12
70%[c]
94%
Entry
1
Aldehyde
Method[a]
A
Yield[b]
84%
13
14
A
A
2
A
87%
3
4
5
B
B
A
78%[c]
83%[c]
93%
15
16
C
87%
C
67%
6
7
8
A
B
12%
17
A
traces
76%[c]
74%[c]
[a]
Method A: A solution of the aldehyde (10.0 mmol),
RhCl3·3H2O (0.04–1.0 mmol) and dppp (0.08–2.0 mmol)
in diglyme was heated to reflux under argon. The prod-
uct was isolated by dilutingthe mixture with pentane and
washing5 times with water to remove diglyme. Purified
by flash chromatography. Method B: As in A, but the
product was isolated by continuous distillation from the
reaction mixture. Method C: As in A, but diglyme was
removed by kugelrohr distillation followed by purifica-
tion with flash chromatography.
A
9
A
A
A
97%
10
11
90%
[b]
[c]
Isolated yield.
81%[c]
Decreased yield due to the volatility of the product.
Adv. Synth. Catal. 2006, 348, 2148 – 2154
ꢁ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
2151

Michael Kreis et al.
FULL PAPERS
Table 4. Tandem Oppenauer-type oxidation-decarbonylation reac-
high yield (87%). More sterically demanding alde-
hydes like the a-branched hydratropaldehyde
(entry 13) and the di-ortho-substituted 2,4,6-trime-
thoxybenzaldehyde (entry 14) were decarbonylated in
good to excellent yields (70% and 94%). (Phthalimi-
dyl)acetaldehyde (entry 15) was also a good substrate
and was decarbonylated in a high yield (87%) with-
out attack on the imide group. Even the decarbonyla-
tion of the more complicated aldehyde in entry 16
with a double bond as well as a nitro and an amido
group succeeded in a good yield (67%) despite the
problematic nitro group. On the other hand, the a,a-
dibranched aldehyde in entry 17 was a poor substrate
for the reaction yieldingonly traces of the desired
product.
tion.[a]
Entry Alcohol
1[b]
Solvent
Catalyst
Yield[c]
54%
Al
A
Mesitylene
Bu)3
Al
Bu)3
N
2[b]
3[b]
4
Mesitylene
Mesitylene
56%
26%
To further expand the synthetic utility of the reac-
tion we decided to investigate the decarbonylation of
aldehyde 6 (Scheme 1).
Al
Bu)3
A
Benzophenone
Benzophenone
73%
K2CO3
AHCTREUNG
5
63%[d]
K2CO3
[a]
Reaction conditions: Alcohol (1.0 equiv.), RhCl3·3H2O
(4 mol%), dppp (8 mol%), K2CO3 (20 mol%), [Cp*IrCl2]2
(2 mol%) in benzophenone at 1708C under argon.
[b]
Reaction conditions: Alcohol (1.0 equiv.), RhCl3·3H2O
(5 mol%), dppp (10 mol%), Al(O-t-Bu)3 (30 mol%), benzo-
N
phenone (3.0 equiv) in mesitylene at reflux under argon.
Isolated yield.
25% of tetradecene was also formed.
[c]
[d]
aldehyde, which is then decarbonylated in the same
reaction mixture (Table 4).
Initially, 3,4-dimethoxybenzyl alcohol (Table 4,
entry 1) was reacted under the classical Oppenauer
conditions with aluminum tris(tert-butoxide) and ben-
zophenone in
a mesitylene solution containing
RhCl3·3H2O and dppp, and the mixture was heated
until the decarbonylation had gone to completion.
However, this only resulted in 54% yield of the de-
carbonylated product and the major problem seemed
Scheme 1. Consecutive Diels–Alder-decarbonylation reac-
tion.
This compound was recently prepared in our labo- to be the oxidation to the aldehyde. Similar unsatis-
ratory by a Diels–Alder reaction between diene 5 and factory results were obtained with 2-(4-methoxyphe-
acrolein.[12] Treatment of 6 with RhCl3·3H2O and nyl)ethanol and 2-(phthalimidyl)ethanol, which gave
dppp gave clean decarbonylation into decalin deriva- only poor to moderate yields (entries 2 and 3).
tive 7 without loss of the protectinggroups or rear-
Changing the solvent to NMP or diglyme resulted in
rangement of the double bond. The overall result of negligible product formation. Therefore, we decided
this two-step sequence is the addition of ethylene to to change the conditions for the hydrogen transfer ox-
diene 5, a transformation which is not feasible by a idation which can also be achieved with various tran-
one-step cycloaddition reaction due to the large sition metal catalysts, for example, the iridium com-
HOMO-LUMO gap.
plex [Cp*IrCl2]2 in the presence of potassium carbon-
We also performed the decarbonylation in combi- ate.[14] In fact, treatment of 2-(4-methylphenyl)ethanol
nation with the Oppenauer oxidation.[13] Hereby, a with [Cp*IrCl2]2, RhCl3·3H2O and dppp in benzophe-
primary alcohol is converted into the corresponding none gave 73% of p-xylene and the tandem reaction
2152
www.asc.wiley-vch.de
ꢁ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 2148 – 2154

Rhodium-Catalyzed Decarbonylation of Aldehydes
FULL PAPERS
went to completion without the addition of additional General Procedure for the Oppenauer-Type
catalyst duringthe course of the reaction (entry 4).
The tandem process also worked for a long-chain ali-
phatic alcohol although an olefin was formed as a by-
product in this case (entry 5).
Oxidation Decarbonylation Reaction
To
a
25-mL flask were added the primary alcohol
mol%), dppp
(2.0 mmol), RhCl3·3H2O (0.08 mmol,
4
(0.16 mmol, 8 mol%), [Cp*IrCl2]2 (0.04 mmol, 2 mol%),
K2CO3 (0.4 mmol, 20 mol%) and benzophenone (10 g). The
flask was equipped with a Liebigcondenser and then evacu-
ated and subsequently flushed with argon. This procedure
was repeated three times. The flask was put into a pre-
heated oil bath and the reaction mixture was heated to
1708C for 1–4 days. After coolingto approximately 50–60 8C
the solution was put on a silica gel column and purified by
flash chromatography.
Conclusions
In summary, we have established a versatile and easy
to handle procedure for the rhodium-catalyzed decar-
bonylation of aldehydes by usingcommercially avail-
able RhCl3·3H2O and dppp. The reaction tolerates a
wide range of functional groups and can be applied to
both aromatic and aliphatic aldehydes. This procedure
was also successfully used in a Diels–Alder decarbon-
ylation sequence, which introduces acrolein as an eth-
ylene synthon for the Diels–Alder reaction. Further-
more, we were able to employ the methodology in a
tandem Oppenauer-type oxidation-decarbonylation
reaction, which makes it possible to remove a hydroxy-
methyl group in one step. Currently, the mechanism
of the decarbonylation reaction is beingstudied in
further detail.
N-(E)-(4-Nitrostyryl)acetamide (Table 3, entry 16)
Synthesized accordingto the general procedure on a 0.16-
mmol scale affordinga yellow solid; yield: 24 mg
(0.11 mmol, 67%); Rf =0.1 (dichloromethane); 1H NMR
[300 MHz, (CD3)2CO]: d=9.68 (d, J=10.5 Hz, 1H, NH),
8.14 (d, J=8.6 Hz, 2H, H-5), 7.76 (dd, J=14.7, 10.5 Hz, 1H,
H-1), 7.60 (d, J=8.6 Hz, 2H, H-4), 6.27 (d, J=14.7 Hz, 1H,
H-2), 2.05 (s, 3H, CH3); 13C NMR [75 MHz, (CD3)2CO]: d=
169.4 (CO), 146.5 (C-6), 129.8 (C-3), 127.4, 125.8 (C-4, C-5),
111.6, 110.6 (C-1, C-2), 23.9 (CH3); MALDI-HR-MS: m/z=
229.0589, calcd. for C10H10N2O3Na: 229.0589.
(1R,2R,3S,10S)-1,2-(Isopropylidenedioxy)-3-[(tert-
butyldimethylsilyl)oxy]bicyclo[4.4.0]dec-5-ene (7)
Experimental Section
A
Synthesized accordingto the egneral procedure on a 33
General Remarks
mmol scale affordinga lihgt yellow solid; yield: 7 mg(21
1
mmol, 64%); Rf =0.47 (pentane/ethyl acetate, 4:1); H NMR
All chemicals were purchased from commercial sources and
used without purification. All reactions were carried out
under an inert atmosphere. Flash column chromatography
was performed with silica gel 60 (particle size 0.040–
0.063 mm). 1H NMR and 13C NMR spectra were recorded
on a Varian Mercury 300 spectrometer. Chemical shifts are
reported in parts per million (ppm) with the solvent chloro-
form as the internal standard (1H NMR CHCl3 dH =
7.26 ppm and 13C NMR CDCl3 dC =77.0 ppm). IR spectra
were recorded on a Perkin–Elmer 1600 Series FT-IR using
KBr plates. EI-MS were recorded on a Shimadzu GCMS-
QP5000 with a direct inlet. ESI-HR-MS were recorded on a
Ionspec 4.7 Telsla Ultima FT-MS.
(300 MHz, CDCl3): d=5.60–5.54 (m, 1H, H-6), 4.16 (dd, J=
5.1, 3.9 Hz, 1H, H-2), 3.83 (ddd, J=10.5, 4.7, 3.9 Hz, 1H, H-
3), 3.73 (dd, J=9.0, 5.1 Hz, 1H, H-1), 2.50–2.40 (m, 1H, H-
10), 2.30–2.20 (m, 1H, H-4 or H-7), 2.14 (dd, J=12.9,
4.7 Hz, 1H, H-4), 2.00–1.90 (m, 2H, H-4 or H-7), 1.90–1.80
(m, 1H, H-8 or H-9), 1.60–1.40 (m, 3H, H-8 or H-9), 1.54 (s,
3H, CH3C), 1.36 (s, 3H, CH3C), 0.91 (s, 9H, (CH3)3C), 0.10
(s, 3H, CH3Si), 0.10 (s, 3H, CH3Si); 13C NMR (50 MHz,
CDCl3): d=133.2 (C-5), 124.7 (C-6), 109.1 [Me2C(OR)OR],
74.8, 73.7, 71.2 (C-1, C-2, C-3), 38.1, 37.6 (C-4, C-10), 30.8
(C-7), 25.7 [(CH3)3C], 25.5, 20.4 (C-8, C-9), 18.0 [(CH3)3C],
À4.7 (CH3Si), À5.0 (CH3Si).
General Procedure for the Decarbonylation of
Aldehydes
Acknowledgements
We thank the Danish Technical Science Research Council,
the Lundbeck Foundation, the Holm Foundation and the
Leo Foundation for financial support. Center for Sustainable
and Green Chemistry is sponsored by the Danish National
Research Foundation. We thank Dr. Martin Kesselgruber
from Solvias for donation of the chiral ferrocene ligands.
To a 50-mL flask were added the aldehyde (10.0 mmol),
RhCl3·3H2O (0.3 mmol,
3 mol%), dppp (0.6 mmol, 6
mol%), and diglyme (25 mL). The flask was equipped with
a Liebigcondenser and then evacuated and subsequently
flushed with argon. This procedure was repeated three
times. The flask was put into a pre-heated oil bath and the
reaction mixture was heated to reflux (1628C) and kept at
reflux for 16 h. After cooling, the mixture was diluted with
pentane (30 mL) and washed with water (5”20 mL) and References
then dried (Na2SO4). The solvent was carefully evaporated
under reduced pressure and the decarbonylated products
were purified by column chromatography on silica gel.
[1] J. Tsuji, K. Ohno, Tetrahedron Lett. 1965, 3969–3971.
[2] K. Ohno, J. Tsuji, J. Am. Chem. Soc. 1968, 90, 99–107.
Adv. Synth. Catal. 2006, 348, 2148 – 2154
ꢁ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
2153

Michael Kreis et al.
FULL PAPERS
[3] D. H. Doughty, L. H. Pignolet, J. Am. Chem. Soc. 1978,
100, 7083–7085.
Tetrahedron 1995, 51, 11693–11702; f) G. H. Hakimela-
hi, S.-C. Tsay, J. R. Hwu, Helv. Chim. Acta 1995, 78,
411–420.
[4] a) G. S. Weatherhead, G. A. Cortez, R. R. Schrock,
A. H. Hoveyda, Proc. Natl. Acad. Sci. USA 2004, 101,
5805–5809; b) T. Shibata, N. Toshida, K. Takagi, J.
Org. Chem. 2002, 67, 7446–7450; c) R. K. Boeck-
man, Jr., J. Zhang, M. R. Reeder, Org. Lett. 2002, 4,
3891–3894; d) T. Morimoto, K. Fuji, K. Tsutsumi, K.
Kakiuchi, J. Am. Chem. Soc. 2002, 124, 3806–3807;
e) M. D. Meyer, L. I. Kruse, J. Org. Chem. 1984, 49,
3195–3199.
[5] a) T. Hansson, B. Wickberg, J. Org. Chem. 1992, 57,
5370–5376; b) R. McCague, C. J. Moody, C. W. Rees, J.
Chem. Soc., Perkin Trans. 1 1984, 165–174.
[6] C. M. Beck, S. E. Rathmill, Y. J. Park, J. Chen, R. H.
Crabtree, L. M. Liable-Sands, A. L. Rheingold, Orga-
nometallics 1999, 18, 5311–5317.
[7] See, for example: a) T. Kato, M. Hoshikawa, Y. Yagu-
chi, K. Izumi, Y. Uotsu, K. Sakai, Tetrahedron 2002, 58,
9213–9222; b) C.-m. Zeng, M. Han, D. F. Covey, J.
Org. Chem. 2000, 65, 2264–2266; c) M. Ikeda, Y. Kugo,
Y. Kondo, T. Yamazaki, T. Sato, J. Chem. Soc., Perkin
Trans. 1 1997, 3339–3344; d) F. E. Ziegler, M. Belema,
J. Org. Chem. 1997, 62, 1083–1094; e) M. Tanaka, T.
Ohshima, H. Mitsuhashi, M. Maruno, T. Wakamatsu,
[8] a) A. van der Ent, A. L. Onderdelinden, Inorg. Synth.
1990, 28, 90–92; b) B. R. James, D. Mahajan, Can. J.
Chem. 1979, 57, 180–187.
[9] Average ligand bite angles compiled from crystal struc-
tures: P. Dierkes, P. W. N. M. van Leeuwen, J. Chem.
Soc., Dalton Trans. 1999, 1519–1529.
[10] D. W. Old, J. P. Wolfe, S. L. Buchwald, J. Am. Chem.
Soc. 1998, 120, 9722–9723.
[11] For more details and other applications of the planar
and central chiral ferrocene ligands from Solvias AG
see, for example: a) F. Spindler, C. Malan, M. Lotz, M.
Kesselgruber, U. Pittelkow, A. Rivas-Nass, O. Briel, H.-
U. Blaser, Tetrahedron: Asymmetry 2004, 15, 2299–
2306; b) T. Sturm, W. Weissensteiner, F. Spindler, Adv.
Synth. Catal. 2003, 345, 160–164.
[12] C. S. Poulsen, R. Madsen, J. Org. Chem. 2002, 67,
4441–4449.
[13] C. F. de Graauw, J. A. Peters, H. van Bekkum, J. Husk-
ens, Synthesis 1994, 1007–1017.
[14] K.-i. Fujita, S. Furukawa, R. Yamaguchi, J. Organomet.
Chem. 2002, 649, 289–292.
2154
www.asc.wiley-vch.de
ꢁ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 2148 – 2154