Toward the rhodium-catalyzed bis-hydroformylation of 1,3-butadiene to adipic aldehyde

Article abstract of DOI:10.1021/om200334g

The effects of the ligand to metal ratio, temperature, syngas pressure, partial pressures of H₂ and CO, and new ligand structures have been examined on 12 of the most reasonable products resulting from the rhodium-catalyzed low-pressure hydroformylation of 1,3-butadiene. The selectivity for the desired linear dihydroformylation product, 1,6-hexanedial (adipic aldehyde), is essentially independent of all of these reaction parameters, except for ligand structure. However, the reaction parameters do have a substantial effect on the selectivity for the products, resulting from the branched addition of the rhodium hydride to the carbon-carbon double bond. The optimum reaction parameters and ligand have resulted in a so far unprecedented maximum selectivity of 50% for adipic aldehyde.

Full text of DOI:10.1021/om200334g

ARTICLE
pubs.acs.org/Organometallics
Toward the Rhodium-Catalyzed Bis-Hydroformylation of
1,3-Butadiene to Adipic Aldehyde
Stuart E. Smith,† Tobias Rosendahl,† and Peter Hofmann*,†,‡
†Catalysis Research Laboratory (CaRLa), University of Heidelberg, Im Neuenheimer Feld 584, D-69120 Heidelberg, Germany
‡Organisch-Chemisches Institut, University of Heidelberg, Im Neuenheimer Feld 270, D-69120 Heidelberg, Germany
S Supporting Information
b
ABSTRACT:
The effects of the ligand to metal ratio, temperature, syngas pressure, partial pressures of H2 and CO, and new ligand structures have
been examined on 12 of the most reasonable products resulting from the rhodium-catalyzed low-pressure hydroformylation of
1,3-butadiene. The selectivity for the desired linear dihydroformylation product, 1,6-hexanedial (adipic aldehyde), is essentially
independent of all of these reaction parameters, except for ligand structure. However, the reaction parameters do have a substantial
effect on the selectivity for the products, resulting from the branched addition of the rhodium hydride to the carbonÀcarbon double
bond. The optimum reaction parameters and ligand have resulted in a so far unprecedented maximum selectivity of 50% for adipic
aldehyde.
’ INTRODUCTION
1,6-hexanedial (adipic aldehyde; eq 1). This product is an attractive
precursor to the synthesis of other more valuable C6 compounds
(e.g., adipic acid, hexamethylenediamine, ε-caprolactone, and 1,
6-hexanediol).5À10 Of course, the hydroformylation of 1,3-but-
adiene has been investigated by quite a number of research groups
since Roelen’s discovery of Co-catalyzed hydroformylation in
1938,5,6,8À19 but the selectivity for adipic aldehyde has always
been much too low to make a technical process viable. To our
knowledge, the best results with respect to the yield of adipic
aldehyde (up to 31%) were claimed by Ohgomori et al.8 These
authors used Rh4(CO)12 as the Rh source and DIOP as the
ligand. This paper presents our work on the rhodium-catalyzed
low-pressure hydroformylation of 1,3-butadiene to form adipic
aldehyde utilizing a new ligand family.
Hydroformylation (the “oxo reaction”) is one of the largest
homogeneous transition-metal-catalyzed processes operated in-
dustrially. The substrates commonly employed in these pro-
cesses are nonconjugated terminal or internal alkenes. Typically,
linear aldehydes are the desired products,1 which are then sub-
sequently converted to more valuable downstream compounds
(e.g., carboxylic acids, alcohols, ...).1 Despite the importance and
atom economy of the oxo process, the dihydroformylation of
conjugated dienes is not a standard reaction. Arguably, the rea-
son stems from the lack of regioselectivity in these reactions.
For instance, the simplest diene, 1,3-butadiene, typically obtained
from large industrial steam-cracking units, can yield up to 14 diff-
erent aldehyde products.2 These numerous products result from
n and i addition of the intermediate metal hydride to the metal-
coordinated diene, a step that is alternatively often viewed as
double-bond n or i insertion into the metalÀhydride bond.
Frequently, i addition is favored both kinetically and thermody-
namically because the barrier to hydride addition is decreased as a
result of conjugation so that a stable η3-methallyl complex is pro-
duced.3,4 Once this occurs, the formation of the linear dialdehyde
is prevented, provided that rapid double-bond isomerization and
hydroformylation of the terminal position of the carbonÀcarbon
double bond does not occur. If a solution to these problems
of regioselectivity were found, an attractive commercial process
would be the dihydroformylation of 1,3-butadiene to form
Previous work in our group has established that a new family
of chelating P-based bidentate ligands (phosphines, phospho-
nites, phosphites, phosphoramidites) are capable of producing
very high n:i ratios for monoaldehydes from nonconjugated
terminal olefins.20À26 These ligands generally feature a doubly
Received: April 20, 2011
Published: June 16, 2011
r
2011 American Chemical Society
3643
dx.doi.org/10.1021/om200334g Organometallics 2011, 30, 3643–3651
|

Organometallics
ARTICLE
time, a number of products formed when the bisphosphite 13
displayed in eq 2 was used.
The compounds trans-3-pentenal (3), cis-3-pentenal (4), trans-
2-pentenal (10), 4-pentenal (9), and 2-methylbut-3-enal (8) are
transient and are eventually converted to the final products
2-methylbutyraldehyde (1), pentanal (2), 2-ethylbutanedial (5),
2-methylpentanedial (6), and adipic aldehyde (7). The products
3, 4, 8, and 10 likely result from the branched (iso) addition of a
rhodium hydride to the terminal position of 1,3-butadiene,6,8,12,16
producing a π-methallyl complex (Scheme 1). The hydroformy-
lation of this η3-methallyl complex then produces 3 and 4, which
can either undergo an additional hydroformylation to produce 5
and 6 or isomerize to trans-2-pentenal (10). trans-2-Pentenal is
then hydrogenated to form 2; the rhodium-catalyzed hydrogena-
tion of R,β-unsaturated aldehydes is known to be very fast.18,27,28
Products 7 and 9 result from the normal addition of the rhodium
hydride to the carbonÀcarbon double bond forming the rho-
dium alkyl complex (Scheme 1). This intermediate is eventually
converted to 4-pentenal (9), which can undergo a further normal
hydroformylation to yield 7. The efficiency of the latter catalytic
step from 9 toward adipic aldehyde (7) was tested indepen-
dently: Rh hydroformylation of commercially available 4-pente-
nal (9) under standard conditions (viz. Table 5) using the bis-
phosphite displayed in Figure 1 (upper right) yields 94% adipic
aldehyde (7).24 Compound 12 results from the self-condensa-
tion of the desired dialdehyde 7. These competing reaction
pathways follow previously proposed mechanisms.6,29À37 Due
to these numerous reaction products, we sought to systematically
study the effects of the ligand/metal ratio, temperature, and pres-
sure on the selectivity for these products.
Hydroformylation of 1,3-Butadiene at 90 °C Under 40 bar
Syngas with Variable L/M Ratios. Selectivity has previously
been shown to be highly dependent on the L/M ratio.38À42 We
examined the possibility for this dependence by varying the L/M
ratio from 1 to 8. As depicted in Table 1, the selectivity for adipic
aldehyde is not dependent on the L/M ratio. However, the
presence of excess ligand does decrease the rate of conversion of
3 and 4 to 2, 5, and 6. The i:n ratio is defined here as the sum
of aldehyde products 1À6, 8, and 10 divided by the sum of
aldehyde products 7, 9, and 12. While the aldehydes 2À4 and 10
are linear aldehydes, they do not arise from a normal addition of
the rhodium hydride to the carbonÀcarbon double bond but
from a branched 1,4-addition as depicted in Scheme 1.
Figure 1. TTP, typical TTP-derived bisphosphite chelate ligand sys-
tems (entries 11 and 12 in Table 5, vide infra), and Xantphos.
P-functionalized 9,10-C2-bridged 9,10-dihydroanthracene scaf-
fold. They are derived from the parent system Triptyphos (TTP;
Figure 1) and are easily accessible in broad structural variety
by DielsÀAlder reactions of properly prefunctionalized anthra-
cenes. TTP and two representative structures of bisphosphites
are displayed in Figure 1. Their synthesis has been described
elsewhere.21,24À26 Essentially the choice and development of this
new ligand family was based upon the published knowledge
about ligand structures, coordination chemistry, reactivity pat-
terns, and catalytic performance of van Leeuwen’s xantphos
ligand (Figure 1) and its congeners, originally with the inten-
tion to approach a rational ligand design strategy in order to
boost activity, catalyst stability, and in particular n selectivity of
Rh-catalyzed low-pressure hydroformylation of terminal mono-
olefins.20,21 Bisphosphines such as TTP and related C2-bridged
analogues have indeed produced n:i ratios of up to 60:1 and are
considerably faster than xanthphos-based bisphosphines. In par-
ticular, rhodium complexes of bisphosphites such as the ones
shown in Figure 1 have turned out not only to be highly active
(as fast as triphenylphosphine) but also lead to full conversion
to n-aldehydes (n:i above 300:1), practically without side pro-
ducts. Experimental and theoretical (DFT) mechanistic work
has shown that the high regioselectivity in these systems has
both kinetic and thermodynamic origins, which will be reported
separately.
’ RESULTS AND DISCUSSION
Hydroformylation of 1,3-Butadiene at 90 °C and 40 bar
Syngas. Due to our previous experience with 1-alkenes, we fo-
cused our attention on utilizing these ligand types in the hydro-
formylation of 1,3-butadiene. The hydroformylation of 1,3-but-
adiene was initially conducted in toluene under 40 bar of syngas
(1/1 H2/CO) and at 90 °C with a ligand/metal (L/M) ratio of 1.
The reaction was monitored over the course of 5 h. During this
The use of 1.0 equiv of the ligand produces an i:n ratio of
4.40, higher than for all other reactions. This could result from
partial ligand decomposition, a known phenomenon for phos-
phite ligands,43À46 producing a second catalyst which could be
less selective for dialdehyde 7 and be responsible for the slightly
3644
dx.doi.org/10.1021/om200334g |Organometallics 2011, 30, 3643–3651

Organometallics
ARTICLE
Scheme 1. Proposed Mechanism for the Hydroformylation of 1,3-Butadienea
a For trigonal-bipyramidal intermediates only the bis-equatorial chelate coordination modes are displayed.
Table 1. L/M Dependence on the Hydroformylation of 1,3-Butadiene
L/M
i:n
1a
2a
3a
4a
5a
6a
7a
8a
9a
10a
11a
1.0
2.0
3.7
8.1
4.40
3.44
3.69
3.80
7.0
2.5
1.3
0.7
41.4
42.8
14.5
5.3
0.0
0.0
0.0
0.0
6.3
8.7
4.0
1.9
24.3
23.5
13.6
8.4
18.5
22.2
21.1
20.5
1.8
0.1
0.0
0.0
0.0
0.0
0.0
0.3
0.3
0.0
1.8
1.3
0.2
0.2
1.2
0.0
33.2
9.2
44.4
17.1
a Relative percentages of the products. [1,3-butadiene] = 0.18À0.25 M.
higher i:n ratio. Consequently, all subsequent reactions were per-
formed with a slight excess of ligand relative to rhodium.
This unsaturated complex can then undergo β-hydride elimination
to generate trans-2-pentenal, or insert CO to form a rhodium-
acyl complex, yielding5. As the temperature increases, the rate of β-
hydride elimination to trans-2-pentenal (followed by rapid hydro-
genation) increases relative to the rate of CO insertion (Table 2).
At or above 110 °C, it is important to closely monitor the reaction,
as adipic aldehyde undergoes self-condensation to cyclopent-1-
enecarbaldehyde (12). This decomposition product can then be
hydrogenated to yield cyclopentanecarbaldehyde. If the reaction is
not closely monitored and decomposition ensues, then the ob-
served selectivity could be lower than it otherwise actually is.
Hydroformylation of 1,3-Butadiene at 90 °C with Variable
Syngas Pressures. The pressures of syngas (1/1 H2/CO) were
Hydroformylation of 1,3-Butadiene at 40 bar of Syngas
with Variable Temperatures. The selectivity dependence on
temperature was investigated from 70 to 130 °C. The reactions
were conducted with a L/M ratio ranging from 1.0 to 1.4 under 40
bar of syngas in toluene. The selectivity for adipic aldehyde (7) did
not vary appreciably with temperature. However, there is a consistent
trend toward decreasing amounts of 5 and increasing amounts of 2
with increasing temperature. This suggests that 5 and 2 are produced
from a common intermediate. It seems likely that the rhodium
hydride intermediate adds to the carbonÀcarbon double bond of 3/
4, producing an unsaturated rhodium alkyl complex (Scheme 2).
3645
dx.doi.org/10.1021/om200334g |Organometallics 2011, 30, 3643–3651

Organometallics
ARTICLE
Scheme 2. Putative Mechanism for the Formation of 2 and 5 from 3 and/or 4
Table 2. Temperature Dependence on the Hydroformylation of 1,3-Butadiene
T (°C)
time (h)
i:n
1a
2a
5a
6a
7a
8a
10a
12a
L/M
70
90
22
5
3.93
4.40
3.50
3.69
2.9
7.0
38.8
41.4
45.2
45.6
13.1
6.3
24.8
24.3
20.6
17.0
20.1
18.5
22.2
21.1
0.0
1.8
0.0
0.0
0.0
0.3
2.9
0.1
0.1
0.0
0.0
0.2
1.1
1.0
1.2
1.4
110
130
1
4.6
4.4
1
13.6
2.4
a Relative percentages of the products. [1,3-butadiene] = 0.18À0.25 M.
Table 3. Pressure Dependence on the Hydroformylation of 1,3-Butadiene
P (bar)
time (h)
i:n
conversn (%)a
1b
2b
5b
6b
7b
8b
10b
11b
12b
L/M
10
20
40
80
23
5
3.72
3.81
4.40
3.70
94
5.1
2.9
7.0
1.9
56.8
53.4
41.4
36.7
0.6
2.7
11.8
16.6
24.3
24.6
19.8
20.4
18.5
21.2
0.0
0.0
1.8
0.0
2.7
2.6
0.3
0.8
3.0
1.2
0.2
0.0
0.2
0.2
0.0
0.1
1.4
1.4
1.0
1.4
88
5
ND
6.3
5
84
14.8
a Percent conversion to the products. ND = not determined. b Relative percentages of the products. [1,3-butadiene] = 0.18À0.25 M.
varied from 10 to 80 bar (Table 3). The concentrations of 5 and 2
appear to be the species most dependent on pressure. The
concentration of 5 consistently increases with increasing pres-
sure, while the concentration of 2 consistently decreases with
increasing pressure. This observation is also consistent with the
supposition that 5 and 2 are made from a common intermediate,
trans-/cis-3-pentenal (3/4). At increasing pressures, the rate of
CO insertion to yield 5 increases relative to the rate of β-hydride
elimination and is consistent with the competing reactions de-
scribed in Scheme 2.
was varied from 20 to 43 bar while the CO partial pressure was
fixed at 20 bar (Table 4). Similarly, the partial pressure of CO
was varied from 20 to 48 bar while the H2 partial pressure was
fixed at 20 bar. These experiments demonstrate that the rates and
selectivities to adipic aldehyde are not greatly affected by the
partial pressure of either H2 or CO; however, we note that these
experiments are not rigorous kinetic studies, and no claims re-
garding the quantitative order of substrates are made. Frequently,
the reaction rate is inverse order in CO,47 as a result of the need
to dissociate CO from the 18-electron resting state (Scheme 1).
A possible reason for the qualitative independence of the reaction
rate on the partial pressures in our system could result from the
rate-determining step being the insertion of CO to form the
rhodium acyl intermediate. The dissociation of CO would still be
Hydroformylation of 1,3-Butadiene at 110 °C with Vari-
able Partial Pressures of H2 and CO. The partial pressures of
H2 and CO have previously been shown to influence the i:n ratio
for hydroformylation reactions.30,47 The partial pressure of H2
3646
dx.doi.org/10.1021/om200334g |Organometallics 2011, 30, 3643–3651

Organometallics
ARTICLE
Table 4. Partial Pressure Dependence of H2 and CO on the the Hydroformylation of 1,3-Butadiene
a
a
PH
Pco
i:n
conversn (%)b
1c
2c
3c
4c
5c
6c
7c
8c
10c
11c
12c
L/M
2
43
34
20
20
20
20
20
20
31
48
3.47
4.22
3.50
3.54
3.54
86
79
85
82
80
5.1
6.0
5.0
6.8
2.4
44.7
42.9
45.4
41.2
34.1
0.0
0.0
0.0
0.0
6.0
0.0
0.0
0.0
0.0
2.8
5.7
4.8
4.0
5.7
7.6
20.6
17.7
20.2
23.5
21.4
22.1
18.9
21.8
21.9
21.8
0.0
8.4
0.0
0.0
0.0
1.5
0.8
2.5
0.8
3.6
0.0
0.2
0.9
0.0
0.0
0.3
0.2
0.3
0.1
0.2
1.4
1.1
1.2
1.3
1.4
a P = partial pressure in bar. b Percent conversion to the products. c Relative percentages of the products. [1,3-butadiene] = 0.19À0.25 M.
required to form an active intermediate, but CO recoordination
to form the rhodium dicarbonyl alkyl complex would precede
the rate-determining step, resulting in a negligible dependence
on the partial pressure of CO. Since the reaction with H2 would
occur after the rate-determining step, the rate would be zero
order in H2.48 If the concentrations of H2 and CO are involved
in the rate law for the formation of linear and branched products,
the order is the same for each competing reaction and has no
effect on the i:n ratio. Thus, it appears that the two competing
transition states are of the same composition.
We suspect that the dependence of selectivity on ligand archi-
tecture within the ligand family of our bisphosphites is predomi-
nantly a result of variable steric interactions. The two putative
intermediates produced from branched vs normal addition require
substantially differentcoordinationenvironments. Normaladdition
to the coordinated diene results in a linear alkenyl intermediate,
occupying only one coordination site in the square-planar complex
(Scheme 1). However, branched addition produces an η3-co-
ordinated π-methallyl ligand, occupying two coordination sites,
and can thus be highly sensitive to the steric environment around
rhodium. By creating and fine-tuning a sterically bulky environ-
ment at the rhodium center, it should be possible to relatively
destabilize the π-methallyl complex, hindering the formation of
the branched products. However, if the ligand is too sterically
encumbering, then it is also possible that it prevents coordination
of the alkene, inhibiting catalysis. We believe these factors deter-
mine the different selectivities and rates for this ligand class. Our
observation that rather subtle structural changes, which are easily
and broadly accessible from our modular synthetic strategy, can
strongly influence the selectivity for adipic aldehyde, in our eyes
makes ligand design based on a combination of appropriate ex-
perimental and theoretical strategies very attractive.
Hydroformylation of 1,3-Butadiene at 110 °C with New
Ligand Systems. Since the selectivity for adipic aldehyde is
not altered significantly by changing the temperature, pressure,
or partial pressure, we decided to screen a number of different
ligands. All of them (except PPh3, which functioned as a standard)
are members of our new ligand family derived from Triptyphos
(Figure 1) and its congeners (bisphosphines, bisphosphites,
entries 2À12 in Table 5), which had shown exceptional n selec-
tivities for 1-alkene hydroformylation. We wanted to determine
ligand structure effects on the selectivity for the Rh-catalyzed
1,4-bis-hydroformylation of 1,3-butadiene. The reactions were con-
ducted at 110 °C under 40 bar of syngas (1/1 H2/CO) in toluene
by means of a Chemspeed high-throughput robot system (for
details see the Experimental Section). Table 5 demonstrates that
the bis-phosphines tested gave by far the least active catalysts and
produced little, if any, adipic aldehyde. This decrease in reactivity
is consistent with the notion that a more electron-donating li-
gand stabilizes the putative resting state, L2Rh(H)(CO)2.45,49
The increased stability makes it more difficult for CO to disso-
ciate to form the coordinatively unsaturated rhodium complex res-
ponsible for catalysis. Not unexpectedly, the bulky and electron-
rich ligands used for entries 2À4 of Table 5 neither lead to active
catalysts for butadiene hydroformylation nor are they active in
the n-selective hydroformylation of 1-alkenes. Most remarkably,
small changes in the ligand structure of the phosphites can have a
profound impact on the selectivity. For instance, ligand 13 yields
a selectivity for adipic aldehyde of 22.8% (entry 8), but when four
methyl groups are added to the bisphenol backbone, the selec-
tivity increases by 15% to 37.8% (entry 10). Notably, entry 12
with a 9,10-ethano bridge gave close to 50% selectivity to adipic
aldehyde. To the best of our knowledge, this is the highest selec-
tivity for adipic aldehyde reported to date. Oftentimes, the selec-
tivity for adipic aldehyde has been found to be <10%,6,11À13,16,18,19
except for the singular 31% claimed by Ohgomori et al.8
’ CONCLUSION
In conclusion, we have examined the influence of syngas pres-
sure, H2 and CO partial pressures, temperature, and ligand archi-
tecture on the selectivity for adipic aldehyde. The temperature
and pressures have no significant effect on the selectivity, but sub-
stantial changes were observed upon alteration of the ligand
backbone and the chelating P subgroups. Given the rather com-
plicated structures of our chelate ligands and thus of the “reaction
pocket” in which the transformation to either the n-insertion-
based n-alkenyl complex or the i-insertion-based η3-π-methallyl
product is occurring at the Rh center, the energetic differen-
tiation between these two stereoisomeric processes can only
amount to a few kJ/mol, as adipic aldehyde is produced with
some of the bisphosphites. In fact, extensive DFT calculations34
for both complete catalytic cycles of Scheme 1, which will be
reported separately, using the ethano-bridged ligand of Figure 1
without the 4-methyl substituents as a model ligand, predict an
insertion barrier difference ΔΔGq of only 5 kJ/mol in favor of
i insertion. Given the large number of attractive and repulsive long-
range intraligand and intracomplex interactions in these systems, it
seems impossible to even qualitatively identify dominant selectivity
3647
dx.doi.org/10.1021/om200334g |Organometallics 2011, 30, 3643–3651

Organometallics
ARTICLE
Table 5. Ligand Influences on the Hydroformylation of 1,3-Butadiene50
3648
dx.doi.org/10.1021/om200334g |Organometallics 2011, 30, 3643–3651

Organometallics
ARTICLE
Table 5. Continued
a Selectivity for 7. b Percentage conversions to the shown products. c Percentages of the respective product relative to the amount of added 1,3-butadiene.
NA = not applicable. [1,3-butadiene] = 0.23 M. All values given are an average from two identical runs.
and rate-determining features for various, partially quite small
structural modifications of electronically identical motifs: e.g.,
of our bisphosphites. We believe, however, that the small
ΔΔGq values computed by DFT and reflected by the formation
of adipic aldehyde in our actual experiments is indicative of a
good chance for ligand design toward improved 1,4-bishydro-
formylation selectivity.
Importantly, the selectivity for adipic aldehyde has been
reproducibly improved to around 50%. Work in our laboratories
is now directed toward rationally modifying our TTP-type
ligands on the basis of systematic synthetic ligand structure varia-
tions, guided by experimental ligand and Rh complex struc-
ture determinations, in situ spectroscopy, mechanistic studies, Ir
model chemistry, and DFT calculations. This combined approach
will be applied for the development of structural ligand modifica-
tions, which will hopefully increase the selectivity for a catalytic
1,3-butadiene to adipic aldehyde conversion to a range where a
practical application becomes feasible.
((E/Z)-2-butene),2.82(2-methylbut-3-enal),3.06(2-methylbutyraldehyde),
3.28 (4-pentenal), 3.42 (pentanal), 3.49 (trans-3-pentenal), 3.58 (cis-3-
pentenal), 4.30 (trans-2-pentenal), 6.05 (4-vinylcyclohexene), 6.27 (cyclo-
pentanecarbaldehyde), 7.58 (cyclopent-1-enecarbaldehyde), 7.96 (nonane),
9.01 (2-ethylbutanedial), 9.49 (2-methylpentanedial), 11.44 (adipic
aldehyde). Peak assignments were confirmed with samples of commer-
cially available or independently synthesized compounds. GC calibra-
tion curves were constructed for all of the compounds except for 1,3-
butadiene and (E/Z)-2-butene. The amounts reported are estimates
made using the effective carbon numbers (ECN) of 3.8 for 1,3-butadiene
and 3.9 for (E/Z)-2-butene, as described in the literature.51 Hydro-
formylation reactions were conducted using a Premex Reactor or a
ChemSpeed Accelerator SLT 106 high-throughput robot system.
Materials. Unless otherwise noted, reagents were purchased from
commercial suppliers and used without further purification. Toluene,
hexane, THF, and methylene chloride were dried using an MBraun SPS-
800 drying machine and degassed by either three freezeÀpumpÀthaw
cycles or by sparging with Ar for a minimum of 15 min. Deuterated
solvents were purchased from Deutero GmbH or Aldrich. CD2Cl2 was
vacuum-transferred from CaH2 after stirring for at least 8 h at room tem-
perature and was degassed by three freezeÀpumpÀthaw cycles. The bis-
phosphine ligands from Table 5 were a generous gift from Dr. Thomas
Schnetz52 (entries 2À4), and the bisphosphite ligands (entries 5À7 and
9À12) were synthesized and characterized according to the thesis24 of
Dr. Tobias Rosendahl (see the Supporting Information), as was ligand
13, used in Tables 1À4 and as entry 8 in Table 5.
’ EXPERIMENTAL SECTION
General Experimental Details. Unless otherwise noted, all
reactions and manipulations were carried out under an Ar atmosphere
using standard Schlenk and high-vacuum-line techniques or in a Braun
inert-atmosphere glovebox (Ar) at ambient temperature. Glassware was
dried for a minimum of 8 h at 150 °C. Unless otherwise noted, all NMR
spectra were obtained at ambient temperature using a Bruker DPX-
200 MHz spectrometer. All NMR chemical shifts are reported as δ in
parts per million (ppm). 1H NMR spectra were referenced to residual
protiated solvent, and chemical shifts are reported in parts per million
downfield from tetramethylsilane. 31P{1H} NMR spectra are reported
relative to trimethyl phosphate as the external standard. 13C{1H} NMR
spectra were referenced to the solvent. Gas chromatographic analysis
was performed on an Agilent Technologies 6890N Network GC System
equipped with a 30 m BGB-5 column (5% diphenylpolysiloxane, 95%
dimethylpolysiloxane). The He flow rate was kept at 2.0 mL/min. The
column temperature was initially held at 40 °C for 1 min, then ramped
at 4 °C/min to 90 °C, followed by an immediate temperature ramp
of 30 °C/min to 200 °C, and held at this temperature for 5 min.
Typical retention times (min) are as follows: 1.99 (1,3-butadiene), 2.02
Representative Procedure for the Catalytic Hydroformyla-
tion of 1,3-Butadiene Using Ligand 13. (CO)2Rh(acac) (10.1 mg,
0.0391 mmol) and 13 (32.9 mg, 0.0460 mmol) were added to an auto-
clave followed by 15 mL of toluene in the glovebox. The autoclave was
sealed, removed from the glovebox, and charged to 30 bar of syngas
(1/1 H2/CO). The solution was stirredat1000rpm whilebeing heated to
90 °C. Once the solution reached 90 °C, the autoclave was pressurized to
40 bar of syngas and the mixture stirred at this temperature and pressure
for 1 h prior to the addition of 1,3-butadiene. After catalyst preformation,
the temperature and pressure were adjusted to the desired values. In this
experiment, the temperature was increased to 110 °C and the pressure
vented to 40 bar of syngas. 1,3-Butadiene (1.00 M in toluene, 5.0 mL,
5.0 mmol) was added to a Swagelok 304 L SS/DOT-3E 1800 (ordering
number 304 L-HDF2-40) double-ended cylinder in the glovebox, and the
cylinder was sealed under Ar. The cylinder was charged to 69 bar of Ar,
3649
dx.doi.org/10.1021/om200334g |Organometallics 2011, 30, 3643–3651

Organometallics
ARTICLE
attached to the autoclave, and added with this overpressure. An over-
pressure of 20À30 bar of Ar is typically used to add 1,3-butadiene to the
reaction mixture. After the mixture was stirred for 1 h, an aliquot was
removed for immediate GC analysis using the method described below.
For reactions run at 90 °C, the reaction mixtures were typically stirred
for 5 h.
bisphosphite 13 (entry 8 in Table 5), Table S1, giving a more
detailed version of Table 5, Figure S1, showing the possible
aldehyde products resulting from the hydroformylation of 1,3-
butadiene, figures giving calibration curves for compounds 1À10
and 12, and text giving synthetic procedures for the bisphosphite
ligands (entries 5À7 and 9À12 in Table 5). This material is
available free of charge via the Internet at http://pubs.acs.org.
Representative Procedure for the Catalytic Hydroformy-
lation of 1,3-Butadiene under Variable Partial Pressures of
H2 and CO. (CO)2Rh(acac) (10.7 mg, 0.0415 mmol) and 13 (38.8 mg,
0.0543 mmol) were added to an autoclave followed by 15 mL of toluene
in the glovebox. The autoclave was sealed, removed from the glovebox,
and charged to 9 bar of CO. The solution was stirred at 1000 rpm while
being heated to 110 °C. Once the solution reached 110 °C, the CO
pressure was 11 bar and stirring was stopped. Syngas (40 bar, 1/1 H2/
CO) was charged to the autoclave, resulting in a total pressure of 51 bar.
Stirring was commenced, and the catalyst was allowed to preform for 1 h.
1,3-Butadiene (0.94 M in toluene, 5.0 mL, 4.7 mmol) was added to a
Swagelok 304 L SS/DOT-3E 1800 double-ended cylinder in the glove-
box, and the cylinder was sealed under Ar. The cylinder was charged to
68 bar of Ar, attached to the autoclave, and added with this overpressure.
After the mixture was stirred for 1 h, an aliquot was removed for
immediate GC analysis using the method described below.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: ph@oci.uni-heidelberg.de.
’ ACKNOWLEDGMENT
The Catalysis Research Laboratory (CaRLa) is jointly funded
by the University of Heidelberg, BASF SE, and the state of Baden-
W€urttemberg. We gratefully acknowledge support from these
institutions. We thank Dr. Thomas Schnetz for his generous gift of
the bisphosphine ligands (entries 2À4 in Table 5) and Professor
Charles Caseyfor discussionsand valuable adviceduringhisstay as
a CaRLaFellow in Heidelberg. We thankDr. FengLu of ourgroup
for earlier work directed toward the ligand of entry 11 in Table 5.
Procedure for Quantifying the Reaction Products from
the Hydroformylation of 1,3-Butadiene. An aliquot (0.900 mL)
from the reaction mixture was removed and added to a 1.00 mL volu-
metric flask followed by nonane (5.00 μL, 0.0280 mmol) and diluted to
1.00 mL with toluene. The sample was then analyzed by GC and the con-
centration determined from the calibration curve.
’ REFERENCES
(1) Weissermel, K.; Arpe, H.-J. Industrial Organic Chemistry, 4th ed.;
Wiley-VCH: Weinheim, Germany, 2003.
General Method for the High-Throughput Screening
Using the Chemspeed Accelerator SLT 106. Toluene stock
solutions of (CO)2Rh(acac) (5.8 mM), ligand (8.8À30 mM), and a mix-
ture of 1,3-butadiene (0.71 M) and nonane (89 mM, internal standard)
were prepared. The Chemspeed Accelerator programmable robot sys-
tem then combined 2.0 mL of the rhodium stock solution and 2.0 mL of
the ligand stock solution together. Catalyst preformation was achieved
by heating at 90 °C under 40 bar of syngas for 1 h. After catalyst pre-
formation, the solution was cooled to À10 °C and 2.0 mL of the 1,3-
butadiene/nonane stock solution was added. The reaction mixture was
heated to 110 °C under 40 bar of syngas while being vortexed at 800 rpm
for 2 h. Afterward, the reaction mixtures were cooled to À10 °C, vented,
and kept at À10 °C overnight. The following morning the samples were
analyzed by GC using the GC method described above with calibrated
GC response factors for each of the products (relative to nonane). The
reactions were run in duplicate, and the yields reported are the average
from these two runs.
Procedure for Preparing the 1,3-Butadiene Stock Solu-
tion. Dry toluene (172.31 g, 199 mL) was added to a premassed, oven-
dried 250 mL Schlenk flask. 1,3-Butadiene was then bubbled through the
toluene solution for 20 min. The Schlenk flask was remassed, and it was
determined that ca. 11 g of 1,3-butadiene was added. The absolute con-
centration of 1,3-butadiene was determined by syringing 0.600 mL of the
1,3-butadiene stock solution into an NMR tube and adding dichlor-
omethane (32.8 mg, 0.386 mmol), followed by ∼0.3 mL of C6D6. A 1H
NMR spectrum was acquired three times (number of scans 1, dummy
scans 0, delay 60 s), and the average of these values was used as the con-
centration for the 1,3-butadiene solution (0.94 ( 0.03 M). The con-
centration of the 1,3-butadiene stock solution was periodically monitored
in the same manner, and the stock solution was stored in the freezer
at À30 °C.
(2) See Figure S1 in the Supporting Information.
(3) Huo, C.-F.; Li, Y.-W.; Beller, M.; Jiao, H. Organometallics 2005,
24, 3634.
(4) Kaita, S.; Koga, N.; Hou, Z.; Doi, Y.; Wakatsuki, Y. Organome-
tallics 2003, 22, 3077.
(5) Bunel, E. E.; Schwiebert, K. E. Phosphonite ligands, catalyst
compositions and hydroformylation process utilizing same. U.S. Patent
6,362,354, March 26, 2002.
(6) Horiuchi, T.; Ohta, T.; Shirakawa, E.; Nozaki, K.; Takaya, H.
Tetrahedron 1997, 53, 7795.
(7) Kummer, R. Manufacture of 1,6-hexanediol from butadiene. U.S.
Patent 3,947,503, March 30, 1976.
(8) Ohgomori, Y.; Suzuki, N.; Sumitani, N. J. Mol. Catal. A: Chem.
1998, 133, 289.
(9) Packett, D. L. Hydroformylation process for producing 1,6-hexane-
dials. U.S. Patent 5,312,996, May 17, 1994.
(10) Packett, D. L.; Briggs, J. R.; Bryant, D. R.; Phillips, A. G.
Processes for producing 1,6-hexanedials and derivatives. U.S. Patent 5,
892,127, April 6, 1999.
(11) Adkins, H.; Williams, J. L. R. J. Org. Chem. 1952, 17, 980.
(12) Bertozzi, S.; Campigli, N.; Vitulli, G.; Lazzaroni, R.; Salvadori,
P. J. Organomet. Chem. 1995, 487, 41.
(13) Botteghi, C.; Branca, M.; Saba, A. J. Organomet. Chem. 1980,
184, C17.
(14) Bunel, E. E. Hydroformylation of conjugated dienes to alkenals
using phosphonite ligands. U.S. Patent 6,437,192, August 20, 2002.
(15) Burke, P. M.; Garner, J. M.; Tam, W. Process for preparing
aldehyde compounds. U.S. Patent 6,664,427, December 16, 2003.
(16) Fell, B.; Bahrmann, H. J. Mol. Catal. 1977, 2, 211.
(17) Fell, B.; Hermanns, P. Process for the hydroformylation of
1,3-butadiene. U.S. Patent 5,434,312, July 18, 1995.
(18) Fell, B.; Rupilius, W. Tetrahedron Lett. 1969, 10, 2721.
(19) van Leeuwen, P. W. N. M.; Roobeek, C. F. J. Mol. Catal. 1985,
31, 345.
’ ASSOCIATED CONTENT
(20) Ahlers, W.; Paciello, R.; Vogt, D.; Hofmann, P. Ligands for
Pnicogen Chelate Complexes with a Metal of Subgroup VIII and Use of the
Complexes as Catalysts for Hydroformylation, Carbonylation, Hydrocyana-
tion or Hydrogenation. WO 02083695 A1, 2002.
S
Supporting Information. Text giving details of the synth-
b
esis and characterization data of compounds 3À6, 8, and 12 and
3650
dx.doi.org/10.1021/om200334g |Organometallics 2011, 30, 3643–3651

Organometallics
ARTICLE
(21) Ahlers, W.; R€oper, M.; Hofmann, P.; Warth, D. C. M.; Paciello,
R. Phosphor, Arsenic and Antimony Compounds Based Upon Diaryl-
Anellated Bicyclo[2.2.n] Parent Substances and Catalysts Containing Same.
WO 0158589, 2001.
(51) Scanlon, J. T.; Willis, D. E. J. Chromatogr. Sci. 1985, 23, 333.
(52) Schnetz, T.; R€oder, M.; Rominger, F.; Hofmann, P. Dalton
Trans. 2008, 2238. The tert-butyl bisphosphine in Table 5 (entry 4) is
synthesized as described in this paper for the i-Pr and Cy analogues
(entries 2 and 3).
(22) G€othlich, A.; Tensfeldt, M.; Hofmann, P.; Paciello, R.; R€oper,
M. (BASF AG) NAE 278/2000, 2000.
(23) Hofmann, P., Invited Lecture, 14th International Symposium on
Homogeneous Catalysis, Munich, 2004; abstract IL7.
(24) Rosendahl, T. Dissertation, University of Heidelberg, 2007.
Details of ligand syntheses are given in the Supporting Information.
(25) Warth, D. C. M. Diploma Thesis, University of Heidelberg,
1999.
(26) Warth, D. C. M. Dissertation, University of Heidelberg, 2001.
(27) Botteghi, C.; Salomon, C. Tetrahedron Lett. 1974, 15, 4285.
(28) Goetz, R. W.; Orchin, M. J. Am. Chem. Soc. 1963, 85, 2782.
(29) Carbꢀo, J. J.; Maseras, F.; Bo, C.; van Leeuwen, P. W. N. M. J. Am.
Chem. Soc. 2001, 123, 7630.
(30) Casey, C. P.; Martins, S. C.; Fagan, M. A. J. Am. Chem. Soc.
2004, 126, 5585.
(31) Casey, C. P.; Paulsen, E. L.; Beuttenmueller, E. W.; Proft, B. R.;
Matter, B. A.; Powell, D. R. J. Am. Chem. Soc. 1999, 121, 63.
(32) Casey, C. P.; Paulsen, E. L.; Beuttenmueller, E. W.; Proft, B. R.;
Petrovich, L. M.; Matter, B. A.; Powell, D. R. J. Am. Chem. Soc. 1997,
119, 11817.
(33) Casey, C. P.; Whiteker, G. T.; Melville, M. G.; Petrovich, L. M.;
Gavney, J. A.; Powell, D. R. J. Am. Chem. Soc. 1992, 114, 5535.
(34) Hofmann, P.; Deglmann, P., Schmidt, S. Unpublished results;
University of Heidelberg, Heidelberg, Germany. 2006, 2010.
(35) Kamer, P. C. J.; van Leeuwen, P. W. N. M.; Reek, J. N. H. Acc.
Chem. Res. 2001, 34, 895.
(36) van der Veen, L. A.; Kamer, P. C. J.; van Leeuwen, P. W. N. M.
Organometallics 1999, 18, 4765.
(37) van Rooy, A.; Kamer, P. C. J.; van Leeuwen, P. W. N. M.;
Goubitz, K.; Fraanje, J.; Veldman, N.; Spek, A. L. Organometallics 1996,
15, 835.
(38) Buisman, G. J. H.; Kamer, P. C. J.; van Leeuwen, P. W. N. M.
Tetrahedron: Asymmetry 1993, 4, 1625.
(39) Hughes, O. R.; Unruh, J. D. J. Mol. Catal. 1981, 12, 71.
(40) Kranenburg, M.; van der Burgt, Y. E. M.; Kamer, P. C. J.;
van Leeuwen, P. W. N. M.; Goubitz, K.; Fraanje, J. Organometallics 1995,
14, 3081.
(41) Sꢀemeril, D.; Jeunesse, C.; Matt, D.; Toupet, L. Angew. Chem.,
Int. Ed. 2006, 45, 5810.
(42) Yan, Y.; Zhang, X.; Zhang, X. J. Am. Chem. Soc. 2006, 128,
16058.
(43) Diꢀeguez, M.; Pꢁamies, O. Acc. Chem. Res. 2010, 43, 312.
(44) Franco, D. W.; Taube, H. Inorg. Chem. 1978, 17, 571.
(45) Kamer, P. C. J.; Reek, J. N. H.; van Leeuwen, P. W. N. M.
Rhodium Catalyzed Hydroformylation. Catalysis by Metal Complexes;
Kluwer Academic: Dordrecht, The Netherlands, 2002; Vol. 22, pp
35À40.
(46) van Leeuwen, P. W. N. M. Appl. Catal., A 2001, 212, 61.
(47) van Rooy, A.; Orij, E. N.; Kamer, P. C. J.; van Leeuwen, P. W. N.
M. Organometallics 1995, 14, 34.
(48) A reviewer has suggested that another possible mechanism to
account for the qualitative independence of reaction rate and selectivity
on partial pressures could result from the formation of mixtures of
rhodium dimers and/or clusters in solution, in addition to the putative
monomeric catalyst. As the CO partial pressure increases, there would
be a shift in concentration of the rhodium dimers/clusters to the
monomer, compensating for the typically inverse dependence on CO
partial pressure. We do not, at this time, have any data to discount this
alternative mechanism.
(49) van der Veen, L. A.; Boele, M. D. K.; Bregman, F. R.; Kamer,
P. C. J.; van Leeuwen, P. W. N. M.; Goubitz, K.; Fraanje, J.; Schenk, H.;
Bo, C. J. Am. Chem. Soc. 1998, 120, 11616.
(50) For the complete table giving all compounds and their corre-
sponding conversions, see Table S1 in the Supporting Information.
3651
dx.doi.org/10.1021/om200334g |Organometallics 2011, 30, 3643–3651