Unusual deactivation in the asymmetric hydrogenation of itaconic acid

Article abstract of DOI:10.1016/j.jorganchem.2010.12.020

The formation of unusual Rh(III) substrate complexes from [Rh(DIPAMP)(MeOH)₂]BF₄ and itaconic acid has been detected which leads to the deactivation of the catalyst. The influence of different parameters on the formation of such complexes, namely substrate concentration, reaction time, temperature, acidic and basic additives, was investigated with different NMR methods. Two different Rh(III) substrate complexes are formed whose ratio is strongly dependent on substrate concentration and reaction time. The pH value of the solution shows a strong influence on the chemical shifts of the ³¹P NMR signals of such complexes. A catalyst-mediated esterification of itaconic acid in methanol was detected. Extended investigations provide detailed ¹H, ¹³C and ³¹P NMR data for the Rh(III) complexes and information about their stability in solution.

Full text of DOI:10.1016/j.jorganchem.2010.12.020

Journal of Organometallic Chemistry 696 (2011) 1760e1767
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Unusual deactivation in the asymmetric hydrogenation of itaconic acidq
*
T. Schmidt, W. Baumann, H.-J. Drexler, D. Heller
LIKAT, A.-Einstein-Str. 29a, 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 formation of unusual Rh(III) substrate complexes from [Rh(DIPAMP)(MeOH)2]BF4 and itaconic acid
has been detected which leads to the deactivation of the catalyst. The influence of different parameters
on the formation of such complexes, namely substrate concentration, reaction time, temperature, acidic
and basic additives, was investigated with different NMR methods. Two different Rh(III) substrate
complexes are formed whose ratio is strongly dependent on substrate concentration and reaction time.
The pH value of the solution shows a strong influence on the chemical shifts of the 31P NMR signals of
such complexes. A catalyst-mediated esterification of itaconic acid in methanol was detected. Extended
investigations provide detailed 1H , 13C and 31P NMR data for the Rh(III) complexes and information
about their stability in solution.
Received 11 October 2010
Received in revised form
13 December 2010
Accepted 14 December 2010
Keywords:
Homogeneous catalysis
Enantioselective hydrogenation
Itaconic acid
Catalyst deactivation
Rhodium alkyl complex
Ó 2011 Elsevier B.V. All rights reserved.
1. Introduction
The objective of this paper is to investigate the effect of different
parameters, namely substrate concentration, reaction time and pH
Enantioselective hydrogenation is an effective way to transform
itaconic acid and its derivatives into optically active methyl succi-
nates [1], which are also pharmaceutically relevant compounds [2].
However, when hydrogenation is carried out with the Rh-DIPAMP
system we [3] and others [4] observed deactivation of the catalyst.
A similar effect was reported when using a rhodium catalyst
modified with the Me-DuPhos ligand, thus suggesting that such
deactivation might be a common feature of Rh-diphosphane cata-
lysts for this type of substrate [3]. So far the reason for deactivation
of the catalyst during hydrogenation of itaconic acid was unknown.
Prochiral olefins like dehydroamino acids and dimethyl itaconate
coordinate to the Rh-catalyst via the CeC double bond and the
oxygen atom of the carbonyl group available in such compounds
[5,6] (Scheme 1 (1)). The same type of complexes is probably
formed between itaconic acid and Rh-diphosphane catalysts.
However, from a solution of [Rh(DIPAMP)(MeOH)2]BF4 and itaconic
acid we recently isolated an unusual Rh(III) alkyl complex (Scheme
1 (2)) (CCDC-690298) which is formed in a parallel running side
reaction [3]. Due to the high stability of the terdentate coordination
mode of the substrate to rhodium [7], the metal center is blocked
and thus cannot participate in hydrogenation which explains the
unusual behavior of the hydrogenation of itaconic acid promoted
by the Rh-DIPAMP system [8].
on the formation of complexes between catalyst and itaconic acid.
This investigation has been supported by an extended NMR study.
2. Experimental
2.1. General procedures
All experiments were carried out according to Schlenk tech-
niques [9] under argon. CH3OH was dried with Mg and distilled
under argon, CD3OD was dried with molecular sieves, distilled
under argon and then subjected to three cycles of freeze/thaw to
remove any residual oxygen.
2.2. Materials
[Rh(DIPAMP)(MeOH)2]BF4 was generated from [Rh(DIPAMP)
(NBD)]BF4 (UMICORE) via hydrogenation [10,11]. Itaconic acid was
recrystallized under argon prior to use. NEt3 was dried with NaOH
overnight and freshly distilled from Na prior to use. HBF4*2Et2O
(Fluka) was stored under argon.
2.3. Complex formation
Unless otherwise stated, catalyst-substrate complexes were formed
by either mixing itaconic acid and the aryl-bridged dimer [Rh(DIPAMP)
(BF4)]2 [11] in CH3OH (CD3OD in case of NMRexperiments) or byadding
a solution of [Rh(DIPAMP)(MeOH)2]BF4 to itaconic acid, to yield clear
orange solutions. At room temperature the orange colorof the solutions
q
Part II, for part I see Reference [3].
* Corresponding author. Tel.: þ49 381 1281 183.
E-mail address: detlef.heller@catalysis.de (D. Heller).
0022-328X/$ e see front matter Ó 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2010.12.020

T. Schmidt et al. / Journal of Organometallic Chemistry 696 (2011) 1760e1767
1761
The solution behavior of the Rh-DIPAMP system in the presence
of itaconic acid as observed by 31P NMR spectroscopy is different
from the one reported in the presence of either (Z)-(N)-methyl
acetamido cinnamate or dimethyl itaconate: these substrates coor-
dinate to [Rh(DIPAMP)(MeOH)2]BF4 to form two diastereomeric
substrate-adducts whose relative amounts and 31P NMR chemical
shifts are independent of the substrate concentration [6,10].
+
R2
O
R1
R3
O
P
P
P
P
R
R
h
h
X
*
*
O
O
R4
O
,
=
X
N
C
3.2. Time dependence of the 31P NMR spectra
(1)
( )
2
Scheme 1. Cationic Rh(I)- complex (1) and neutral Rh(III) complex (2).
As pointed out, a color change after preparation of the catalyst-
substrate complexes provides the hint that the solution composi-
tion changes over time. In order to investigate the time dependence
of the 31P NMR spectra, we recorded a series of NMR spectra of
solutions of [Rh(DIPAMP)(MeOH)2]BF4 and itaconic acid in CD3OD
at two different catalyst/substrate ratios 1:5 (Fig. 1 Supplementary
information.) and 1:40 (Fig. 2) over a time interval of several days.
The latter solution shows a dramatic change of the ratio between
the two catalyst-substrate complexes. Thirty minutes after sample
preparation, the relative amounts of the two catalyst-substrate
complexes (both type 2) are 90% and 10% respectively (Fig. 2a).
Already after three days (Fig. 2c), the amount of the former "minor"
substrate complex is higher (60%) than that of the initial "major"
diastereomer (40%). Furthermore, new signals appear which, to the
best of our knowledge do not correspond to complexes of itaconic
acid with [Rh(DIPAMP)(MeOH)2]BF4. After six days (Fig. 2d), the
relative amount of the initial "minor" complex has raised to 75%.
The concentration of the unknown species increases over time.
always fades away within a few minutes, indicating the conversion
from complex (1) to the substrate complex (2) (Scheme 1). All NMR
spectra recorded at room temperature therefore contain almost
exclusively catalyst-substrate complexes of type 2 (Scheme 1) [12].
2.4. Isolation of [Rh(DIPAMP)(methyl succinate)](2)
Single crystals of [Rh(DIPAMP)(methyl succinate)] (2) could be
isolated from a solution of 0.01 mmol Rh(DIPAMP)(MeOH)2]BF4 and
0.05 mmol itaconic acid in 1 mL MeOH either by slow diffusion of
diethyl ether (5 mL) or by addition of 0.05 mmol NEt3.
2.5. NMR experiments
Samples containing rhodium complexes were prepared under
argon in deoxygenated and dry CD3OD using Young NMR tubes
(Rotec Spintec GmbH). The catalyst concentration was 0.04 mmol/
mL unless otherwise stated. NMR experiments were carried out
either on a Bruker ARX-300 or on a Bruker ARX-400 spectrometer.
3.3. Origin of the new-appearing signals
As shown in Fig. 2 and Fig. 1 Supplementary information, new
signals emerge within a time range of several days in the 31P NMR
spectra of a solution of [Rh(DIPAMP)(MeOH)2]BF4 and itaconic acid
in CD3OD, independent on the substrate concentration. In the 1H
NMR spectra (ratio 1:5) the changes in the signals of the vinylidenic
CH2 protons over time are more informative (Fig. 3). In the spec-
trum recorded immediately after sample preparation (spectrum a,
Fig. 3) mainly the signal set of uncoordinated itaconic acid CH2 is
visible, already after five days (spectrum b, Fig. 3) the concentration
of a newly-formed species slightly exceeds that of itaconic acid. In
this species, the vinylidenic CH2 protons must have a chemical
environment similar to that experienced in itaconic acid as the
differences in the chemical shifts of the peaks are low. After 13 days
(Fig. 3c) the concentration of the new species is threefold that of
itaconic acid. Additionally, an H/D exchange of the vinylidenic CH2
protons is observed, leading to additional peaks.
3. Results and discussion
3.1. Influence of the substrate concentration
As previously highlighted [3], the hydrogenation kinetics of
itaconic acid catalyzed by [Rh(DIPAMP)(MeOH)2]BF4 is dependent
on substrate concentration: reaction rate decreases with increasing
substrate concentration under otherwise identical conditions. Fig. 1
shows the 31P NMR spectra of a solution of [Rh(DIPAMP)(MeOH)2]
BF4 and itaconic acid at different Rh/substrate ratios, recorded
immediately after sample preparation.
EvenataRh/substrateratioof2:1(Fig.1a)abouthalfofthecatalyst
(48%) is complexed by the substrate (red, orange underlined signals),
forming complexes of type 2 (Scheme 1). The other signals belong to
the h
6-bridged dimer [Rh(DIPAMP)BF4]2 (blue underlined) and to the
Due to the similarity of the signals of the newly-formed species
with those of itaconic acid we hypothesized that one of the possible
monomethyl itaconates had been formed (Scheme 2). To confirm
our assumption we let [Rh(DIPAMP)(MeOH)2]BF4 and itaconic acid
equilibrate in undeuterated methanol for 5 days and compared the
1H NMR spectrum with that of a solution of itaconic acid with no
added catalyst which had been left standing for 5 days as well
(Fig. 4).
solvent complex [Rh(DIPAMP)(MeOH)2]BF4 (brown underlined): the
equilibrium between these species is concentration-dependent [11].
At variance with what observed with dimethyl itaconate, at a ratio of
[Rh(DIPAMP)(MeOH)2]BF4 and itaconic acid of 1:1 (Fig. 1b) neither
the solvent complex nor aryl-bridged dimers are detectable a proof
that the Rh-DIPAMP/substrate complexof itaconic acid is more stable
than the one formed by dimethyl itaconate [6]. At room temperature,
the Rh/substrate complex of type 1 (Scheme 1) was not observed,
independent of the relative amount of itaconic acid present in solu-
tion. Two different, probably diastereomeric, substrate complexes of
type 2 (Scheme 1) can be detected in the NMR spectra. Surprisingly,
their relative amounts depend on the substrate concentration: at
a catalyst-substrate ratio of 1:40 (Fig. 1d) [13] the share of the minor
signals is 10% and therefore more than twofold the share at ratio 1:1
(4%). Furthermore, a drift of the chemical shift of the major signals
with variation of substrate concentration is visible and very prom-
inent when changing from a catalyst-substrate ratio of 2:1 to 1:1.
No new signals were observed in the absence of the catalyst. The
identity of the new species was confirmed by comparison with the
H
H
H
H
O
O
e
[
]
R (DIPAMP)(M H) BF4
h
O
2
H
O
H
O
H
O
O
O
O
Scheme 2. [Rh(DIPAMP)(MeOH)2]BF4 catalyzed esterification of itaconic acid.

1762
T. Schmidt et al. / Journal of Organometallic Chemistry 696 (2011) 1760e1767
Fig. 1. Room temperature 31P NMR spectra of solutions of [Rh(DIPAMP)(MeOH)2]BF4 and itaconic acid in CD3OD at different Rh/substrate ratios, recorded immediately after sample
preparation. Red, orange: catalyst-substrate complexes of type 2 (Scheme 1), brown: [Rh(DIPAMP)(MeOH)2]BF4, blue: dimeric
6-complexes of type [Rh(DIPAMP)BF4]2 [11].
h
1H NMR spectrum of an authentic sample of
b
-methyl itaconate
observed in spectra c and d (Fig. 2, framed signals) is the catalyst-
substrate complex with -methyl itaconate.
(reference spectrum in Fig. 2 Supplementary information). Metal
promoted esterification of itaconic acid with a titanium catalyst [14]
and of other organic carboxylic compounds [15] has been previously
reported in the literature.
b
3.4. Low-temperature NMR investigations of the formation of
catalyst-substrate complexes
A comparison of the 31P NMR spectrum of a solution of [Rh
(DIPAMP)(MeOH)2]BF4 and
b-methyl itaconate in CD3OD (Fig. 3
In an attempt to intercept the formation of intermediates we [3]
Supplementary information) proves that the newly-formed species
and Brown et al. [16] tried to freeze out the irreversible formation of
a
b
c
d
Fig. 2. Time-dependent 31P NMR spectra of a solution of [Rh(DIPAMP)(MeOH)2]BF4 and itaconic acid (Rh:substrate ¼ 1:40). The additional appearing signals are framed.

T. Schmidt et al. / Journal of Organometallic Chemistry 696 (2011) 1760e1767
1763
a
b
c
Fig. 3. Section of the time-dependent 1H NMR spectra of a solution of [Rh(Dipamp)(MeOH)2]BF4 and itaconic acid (1:5) in CD3OD (the signals correspond to the vinylidenic CH2
protons of the uncoordinated substrate).
the colorless Rh(III) alkyl complex ((2) in Scheme 1), whose X-ray
structure we had previously reported [3].
substrate coordination. In accordance with results by Brown et al.
[16], our findings confirm the proposed formation of a bis-chelate
complex of itaconic acid with the Rh-DIPAMP solvent complex,
followed by its irreversible transformation into a Rh(III) alkyl
complex [3].
A solution of [Rh(DIPAMP)(MeOH)2]BF4 in CD3OD was cooled to
ꢀ60 ꢁC and transferred to a precooled Schlenk tube containing
a solution of itaconic acid in the same solvent (Rh/substrate ratio
1:5). After stirring for 1 h, the orange solution was transferred to
a Young NMR tube and a 31P NMR spectrum was immediately
recorded at ꢀ60 ꢁC (Fig. 5, spectrum a).
3.5. Influence of acidic and basic additives and of the BFꢀ4 anion on
the 31P NMR signals
Beside the signals of the aryl-bridged Rh-DIPAMP dimers [11]
characteristic signals of a catalyst-substrate complex were found,
whereas the solvent complex could not be detected [17]. The J(31P,
31P) coupling of the observed substrate complex is 41 Hz and thus
in the range of the J(31P, 31P) coupling observed in the Rh-DIPAMP
substrate complex with dimethyl itaconate [6]. Additionally,
From preliminary investigations (see Supp. info of Reference [3])
we noticed that the chemical shifts of the 31P NMR signals of
a solution of [Rh(DIPAMP)(MeOH)2]BF4 and itaconic acid in CD3OD
differ from those observed in the 31P NMR spectrum of a solution
obtained dissolving single crystals of the complex [Rh(DIPAMP)
(methyl succinate)] in CD3OD. In the former sample, formation of
the complex [Rh(DIPAMP)(methyl succinate)] releases HBF4 in
solution which is absent in the latter sample.
a doublet (purple) is detected (
whose origin is unknown.
d
¼ 73.1 ppm, J(31P, 103Rh) ¼ 208 Hz)
Warming of the sample to room temperature leads to fading of
the orange color: the 31P NMR spectrum of the colorless solution is
the same as spectrum c reported in Fig. 1. If the solution is cooled
back to ꢀ60 ꢁC, the 31P NMR spectrum (spectrum b, Fig. 5) does not
reproduce spectrum a in Fig. 5. Spectrum b from Fig. 5 is the low-
temperature spectrum of the colorless Rh(III) alkyl species (Fig. 4
Supplementary information). The value of the J(31P, 31P) coupling
20.7 Hz and the color change give evidence of a drastic change in
Fig. 6 shows the 31P NMR spectra of complexes [Rh(DIPAMP)
(methyl succinate)] measured at different pH value or in the pres-
ence of variable amounts of BF4ꢀ. The complexes are either formed
"in situ" by reacting [Rh(DIPAMP)(MeOH)2]BF4 and itaconic acid in
the ratio of 1:1 (d) or by dissolving single crystals of [Rh(DIPAMP)
(methyl succinate)] in CD3OD (a). NEt3 was chosen as basic additive,
a solution of HBF4 in diethyl ether as the acidic one.
a
b
Fig. 4. 1H NMR spectra of the solutions of itaconic acid (with no added [Rh(DIPAMP)(MeOH)2]BF4, spectrum a; in the presence of [Rh(DIPAMP)(MeOH)2]BF4), spectrum b)
equilibrated over 5 days in CH3OH, after solvent evaporation and dissolution in CD3OD.

1764
T. Schmidt et al. / Journal of Organometallic Chemistry 696 (2011) 1760e1767
a
b
Fig. 5. 31P NMR spectra of [Rh(DIPAMP)(MeOH)2]BF4 after addition of five equivalents of itaconic acid at ꢀ60 ꢁC (a): measured after 1 h equilibration at ꢀ60 ꢁC (b) measured after
2 h equilibration at 25 ꢁC, followed by cooling to ꢀ60 ꢁC. Green: catalyst-substrate complex type 1, orange, red: catalyst-substrate complex type 2, blue: [Rh(DIPAMP)(BF4)]2, purple:
unknown species.
a
b
c
d
e
f
g
Fig. 6. Influence of basic and acidic additives and/or BFꢀ4 anion concentration on the 31P NMR spectra: a), b), e), f): single crystals of [Rh(DIPAMP)(methyl succinate)] dissolved in
CD3OD; c), d) g): in situ-preparation [Rh(DIPAMP)(MeOH)2]BF4 and itaconic acid (1:5) in CD3OD.

T. Schmidt et al. / Journal of Organometallic Chemistry 696 (2011) 1760e1767
1765
a
b
Fig. 7. 1H NMR spectrum sections (0e4.7 ppm) of a) [Rh(DIPAMP)(MeOH)2]BF4 and itaconic acid (1:5), b) dissolved single crystals of [Rh(DIPAMP)(methyl succinate)] in CD3OD.
Examples are reported in the literature showing the influence of
counterions on the chemical shifts of cationic complexes [18]. In the
present case, the addition of NaBF4 to a solution of [Rh(DIPAMP)
(methyl succinate)] has no influence on the 31P NMR signals
(compare spectra a and b, Fig. 6). Therefore we assume no strong
interaction of the anion with the investigated catalyst-substrate
complex.
However, the pH of the solutions strongly influences the chem-
ical shifts. Spectra aec, d and e, f and g refer to solutions which have
comparable pH values, respectively [19]. Consistently, the major
signals in the spectra of each group have almost the same shift. The
left major signal is shifted to lower fields with increasing acid
concentration. The same shift is observed for both minor signals.
However, the shift of the right major signal changes irregularly.
It has to be mentioned that the minor catalyst-substrate
complex is never observed in the spectra of solutions obtained by
dissolving single crystals of [Rh(DIPAMP)(methyl succinate)]
(spectra a, b, e, f, Fig. 6). This provides evidence that the intercon-
version between the catalyst-substrate complexes is kinetically
hampered. This behavior, which is at variance with what observed
with - and -dehydroamino acid derivatives whose catalyst-
substrate adducts do interconvert, can be explained by the ter-
dentate substrate coordination.
Under basic conditions, the major catalyst-substrate complex
precipitates from the solution. Its crystal structure has been
previously reported [3]. This explains the higher relative amount of
the minor diastereomer observed in spectrum c compared to
spectrum d. Because both the J(31P, 103Rh) and J(31P, 31P) couplings
do not vary with pH, we assume that the same complex type is
observed in all spectra, namely type 2 (Scheme 1). The drift of the
chemical shifts with pH is attributed to the chemical environment
of those complexes.
a
b
3.6. Structure of the catalyst-substrate complexes in solution
As the 31P NMR spectra of the in situ generated major substrate-
complex and of the dissolved single crystals of [Rh(DIPAMP)
(methyl succinate)] differ, we set out to investigate the complex
a
b
Fig. 8. Section of 13C NMR and DEPT-NMR measurements (20e60 ppm) of a solution of [Rh(DIPAMP)(MeOH)2]BF4 and itaconic acid (1:5) in CD3OD.

1766
T. Schmidt et al. / Journal of Organometallic Chemistry 696 (2011) 1760e1767
a
b
Fig. 9. Sections of 13C NMR and DEPT-NMR measurements (20e60 ppm) of a solution of single crystals of [Rh(DIPAMP)(methyl succinate)] in CD3OD.
formation in more detail. One problem to be addressed concerns
the mode of substrate coordination in solution: according to Brown
and coworkers it takes place through the CeC double bond [16],
whereas our findings show that coordination involves the forma-
tion of a RheC bond (and the consequent formation of a CH3 group).
Such binding mode has been confirmed by X-ray analysis [3]. In this
section we discuss only selected NMR data. For extensive NMR data,
please refer to the Supplementary information.
by dissolving such crystals in CD3OD [21] shows the presence of
a vinylidenic CH2 unit (Fig. 5 Supplementary information). When
the complexes are prepared in situ in CD2Cl2 NMR investigations
indicate the presence of a methyl group (Figs. 6e8 Supplementary
information).
Because no H/D exchange takes place when single crystals of [Rh
(DIPAMP) (methyl succinate)], isolated from undeuterated methanol,
are dissolved in CD3OD, complex formation must be irreversible.
Fig. 7 shows a comparison between the 1H NMR spectra of
a freshly prepared solution of [Rh(DIPAMP)(MeOH)2]BF4 and ita-
conic acid (1:5) in CD3OD and a solution of single crystals of the
complex [Rh(DIPAMP)(methyl succinate)] in CD3OD. As integral
reference the more downfield shifted methoxy signal of the
DIPAMP ligand was used.
While in case of the in situ generated catalyst-substrate complex
(spectrum a, Fig. 7) the integral of the CH2 group is 2,26 (which
corresponds more to a methylene than a methyl group), in case of
the dissolved crystals of the isolated substrate complex (spectrum
b, Fig. 7) an integral of 2.97 (methyl group) is found.
4. Conclusions
Itaconic acid forms two very stable, colorless Rh(III) alkyl com-
plexes (type 2) with [Rh(DIPAMP)(MeOH)2]BF4 at room tempera-
ture, that are probably diastereomers. The ratio of the formed
complexes varies with time and substrate concentration, unlike
similar known complexes of the Rh-DIPAMP system with dehy-
droamino acid derivatives, which can be attributed to type 1.
Furthermore, esterification of the
b-carboxylic group of the
substrate with the solvent methanol was observed. The chemical
shift of the complex signals in the 31P NMR spectra is influenced by
acidic and basic additives but not by the BFꢀ4 anion concentration.
The structure of the isolated catalyst-substrate complex [Rh
(DIPAMP)(methyl succinate)] as shown by X-ray analysis is retained
in solution, independent of the mode of sample preparation (in situ
formation vs. dissolution of single crystals of preformed complex).
An H/D exchange in CD3OD is responsable for the dependence of the
metheyl group signal shape on the mode of sample preparation.
These findings are confirmed by the measurement of 13C- and
DEPT-NMR spectra (Fig. 8 [20]). Again, the in situ generated
complex provides a CH2 signal which is, however, a triplet. The
spectra of the dissolved single crystals of [Rh(DIPAMP)(methyl
succinate)] give a CH3 signal (Fig. 9). Obviously, the chemical shifts
of both signals are almost identical as is also true for the discussed
signals in the 1H NMR.
The presence of different groups (either eCH2e or eCH3) seems
to be inconsistent, as the single crystals of [Rh(DIPAMP)(methyl
succinate)] were isolated from the in situ generated solution.
This however can be explained by an H/D exchange of the
carboxylic protons with the deuterated solvent, followed by a D-
transfer to the methylene group of the coordinated substrate upon
formation of the Rh(III) alkyl complex. The detected CH2 unit is in
fact a CH2D group. This explanation is consistent with the triplet
pattern observed in the 13C- and DEPT-spectra recorded after in situ
complex generation. The complex [Rh(DIPAMP)(methyl succinate)]
was isolated from undeuterated methanol, thus explaining the
presence of a CH3 group.
Acknowledgments
Prof. Piet van Leeuwen and Dr. Detlef Selent are gratefully
acknowledged for fruitful discussions. We also thank Cornelia
Pribbenow for her skilled technical support.
The authors thank the Graduiertenkolleg 1213 “Neue Methoden
für Nachhaltigkeit in Katalyse und Technik” for financial support.
Appendix A
Further evidence for deuterium incorporation from CD3OD is
provided by isolation of single crystals of [Rh(DIPAMP)(methyl
succinate)] from a solution of [Rh(DIPAMP)(CD3OD)2]BF4 and ita-
conic acid in CD3OD: the 1H NMR spectrum of the solution obtained
Supplementary material CCDC 690298 (2) contains the supple-
mentary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.

T. Schmidt et al. / Journal of Organometallic Chemistry 696 (2011) 1760e1767
1767
[8] Since the substrate in the described Rh(III) alkyl complex contains a CH3 group
instead of the former vinylidenic CH2 group we use methyl succinate instead of
itaconate. A reasonable mechanism for the formation of this complex is
proposed in Ref. [3].
[9] (a) A. Salzer, Laboratory Techniques of Organometallic Chemistry (Chapter 2).
in: W.A. Herrmann (Ed.), Synthetic Methods of Organometallic and Inorganic
Chemistry, S. 8-28. Georg Thieme Verlag, Stuttgart, 1996;
Appendix A. Supplementary information
Supplementary data related to this article can be found online at
doi:10.1016/j.jorganchem.2010.12.020.
References
(b) S. Herzog, J. Dehnert, K. Lühder, An expedient method for preparative
procedures in an inert atmosphere. in: H.B. Jonassen, A. Weissberger (Eds.),
Technique Of Inorganic Chemistry, vol. VII. John Wiley & Sons, New York,
1968, pp. 119e149.
[1] (a) K.F.W. Hekking, L. Lefort, A.H.M. de Vries, F.L. van Delft, H.E. Schoemaker,
J.G. de Vries, F.P.J.T. Rutjes, Adv. Synth. Catal. 350 (2008) 85e94;
(b) M.T. Reetz, J.-A. Ma, R. Goddard, Angew. Chem. 117 (2005) 416e419;
Angew. Chem. Int. Ed. 44 (2005) 412e415;
[10] C.R. Landis, J. Halpern, J. Am. Chem. Soc. 109 (1987) 1746e1754.
[11] A. Preetz, W. Baumann, C. Fischer, H.-J. Drexler, T. Schmidt, R. Thede, D. Heller,
Organometallics 28 (2009) 3673e3677.
(c) U. Berens, M.J. Burk, A. Gerlach, W. Hems, Angew. Chem. 112 (2000)
2057e2060;
Angew. Chem. Int. Ed. 39 (2000) 1981e1984.
[12] Color fading, a phenomenon also reported by Brown et al. [16] was over in
a few minutes: the NMR spectra were recorded after complete color change.
[13] Due to the limited solubility of the substrate in methanol higher substrate-to-
catalyst ratios could not be investigated. In an experiment were a 100-fold
substrate excess was used, about half of the substrate, which was dissolved by
heating, recrystallized immediately with cooling of the sample.
[14] T. Kawabata, T. Mizugaki, K. Ebitani, K. Kaneda, Tetrahedron Lett. 44 (2003)
9205e9208.
[15] J. Otera, N. Dan-oh, H. Nozaki, J. Org. Chem. 56 (1991) 5307e5311.
[16] J.M. Brown, D. Parker, J. Org. Chem. 47 (1982) 2722e2730.
[17] This effect may be caused by an increasing stability of the aryl-bridged dimers
with decreasing temperature. A further possible explanation is the distur-
bance of the thermodynamic equilibrium between the solvent complex and
the above mentioned dimers due to the fact that the observed catalyst-
substrate complex is mainly formed from the solvent complex.
[18] (a) P.G. Anil Kumar, P.S. Pregosin, T.M. Schmid, G. Consiglio, Magn. Reson.
Chem. 42 (2004) 759e800;
[2] (a) P.P. Constantinidis, J. Han, S.S. Davis, Pharmaceut. Res. 23 (2006) 243e255;
(b) J. Goldstein, S.D. Silberstein, J.R. Saper, A.H. Elkind, T.R. Smith, R.M. Gallagher,
J.-P. Battikha, H. Hoffman, J. Baggish, Headache 45 (2005) 973e982.
[3] T. Schmidt, H.-J. Drexler, J. Sun, Z. Dai, W. Baumann, A. Preetz, D. Heller, Adv.
Synth. Catal. 351 (2009) 750e754.
[4] W.C. Christophel, B.D. Vineyard, J. Am. Chem. Soc. 101 (1979) 4406e4408.
[5] (a) J.M. Brown, P.A. Chaloner, J. Chem. Soc. Chem. Commun. (1978) 321e322;
(b) J.M. Brown, P.A. Chaloner, Tetrahedron Lett. 21 (1978) 1877e1880;
(c) J.M. Brown, P.A. Chaloner, J. Chem. Soc. Chem. Commun. (1979) 613e615;
(d) A.S.C. Chan, J.J. Pluth, J. Halpern, Inorg. Chim. Acta 37 (1979) L477eL479;
(e) J.M. Brown, P.A. Chaloner, J. Am. Chem. Soc. 102 (1980) 3040e3048;
(f) B. McCulloch, J. Halpern, M.R. Thomas, C.R. Landis, Organometallics 9
(1990) 1392e1395;
(g) H.-J. Drexler, S. Zhang, A. Sun, A. Spannenberg, A. Arrietta, A. Preetz,
D. Heller, Tetrahedron: Asymmetry 15 (2004) 2139e2150;
(b) P.G. Anil Kumar, P.S. Pregosin, M. Vallet, G. Bernadinelli, R.F. Jazzar,
F. Viton, E.P. Kündig, Organometallics 23 (2004) 5410e5418;
(h) T. Schmidt, W. Baumann, H.-J. Drexler, A. Arrieta, H. Buschmann, D. Heller,
Organometallics 24 (2005) 3842e3848;
(c) J.-S. Song, D.J. Szalda, R.M. Bullock, Organometallics 20 (2001) 3337e3346;
(d) R. Romeo, N. Nastasi, L.M. Scolaro, R.M. Plutino, A. Albinati, A. Macchioni,
Inorg. Chem. 37 (1998) 5460e5466.
(i) H.-J. Drexler, W. Baumann, T. Schmidt, S. Zhang, A. Sun, A. Spannenberg,
C. Fischer, H. Buschmann, D. Heller, Angew. Chem. 117 (2005) 1208e1212;
Angew. Chem. Int. Ed. 44 (2005) 1184e1188.
[19] Upon formation of [Rh(DIPAMP)(methyl succinate) from itaconic acid and [Rh
(DIPAMP)(MeOH)2]BF4, 1 equivalent HBF4 is released (for details refer to
Ref. [3]). Because HBF4 is a strong acid and NEt3 is a strong base, we assume
that an equimolar solution of the two species should be neutral.
[20] The correlation was carried out via 1He13C measurements.
[21] The X-ray analysis of the complex isolated from CD3OD shows that it is the
same catalyst-substrate complex for which the X-ray structure had been
previously reported in Ref. [3]. Unfortunately, it was impossible to conclude
from the data whether an H or D atom had been transferred to the vinylidenic
CH2 group.
[6] T. Schmidt, Z. Dai, H.-J. Drexler, W. Baumann, C. Jäger, D. Pfeifer, D. Heller,
Chem. Eur. J. 14 (2008) 4469e4471.
[7] The X-ray structure of the complex surprisingly revealed that the substrate is
not coordinated via the CeC double bond and one carboxylate oxygen, but
instead via the two carboxylate oxygens and the quaternary carbon [C(2)]
atom. Formally the acidic functions of the substrate had been completely
deprotonated and the methylene group had been already converted into
a methyl group via H-transfer. The tetrafluoroborate anion was no longer
present in the crystal as the counterion. Thus, the complex represents
a neutral Rh(III) alkyl complex, compare Reference [3].