Hydrogenation of (E)-2-methyl-2-butenoic acid over cinchona-modified Pd catalyst in the presence of achiral amines: Solvent and modifier effect

Article abstract of DOI:10.1016/j.catcom.2013.11.029

The effect of the solvent, modifier structure and concentration on the enantioselective hydrogenation of (E)-2-methyl-2-butenoic acid over Pd/Al ₂O₃ modified by cinchona alkaloids was influenced by the addition of achiral amines to t

Full text of DOI:10.1016/j.catcom.2013.11.029

Catalysis Communications 46 (2014) 113–117
Contents lists available at ScienceDirect
Catalysis Communications
journal homepage: www.elsevier.com/locate/catcom
Short Communication
Hydrogenation of (E)-2-methyl-2-butenoic acid over cinchona-modified
Pd catalyst in the presence of achiral amines: Solvent and
☆
modifier effect
Zsolt Makra a, György Szőllősi b,
⁎
a
Department of Organic Chemistry, University of Szeged, Dóm tér 8, Szeged H-6720, Hungary
MTA-SZTE Stereochemistry Research Group, Dóm tér 8, Szeged H-6720, Hungary
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 19 June 2013
Received in revised form 12 October 2013
Accepted 27 November 2013
Available online 3 December 2013
The effect of the solvent, modifier structure and concentration on the enantioselective hydrogenation of (E)-2-
methyl-2-butenoic acid over Pd/Al2O3 modified by cinchona alkaloids was influenced by the addition of achiral
amines to the reaction slurry. The solvent dependence in the presence of achiral amines showed that additives
are involved in the rate determinant step of the reaction, whereas the effect of the dilution pointed to the presence
of the acid dimers in the intermediate responsible for enantioselection. The dependence of the enantioselectivity on
the cinchonidine concentration in the presence of amines and results obtained using cinchona derivatives and
mixtures thereof were in line with our earlier assumptions related to the participation of the amine additive in
the formation of the surface intermediate. Based on these and previously published results possible structures of
this intermediate are suggested.
Keywords:
Unsaturated acid
Amine additive
Cinchona alkaloid
Enantioselective hydrogenation
Palladium
© 2013 Elsevier B.V. All rights reserved.
Solvent effect
1. Introduction
it was suggested that the additive participates in the formation of the
surface intermediate responsible for the chiral induction. However,
Enantioselective hydrogenations are among the most useful catalytic
methods for preparing optically pure chiral intermediates [1]. A large
variety of chiral metal complexes have been developed and applied in
asymmetric hydrogenations [2]. Recent trends of applying green and
sustainable processes stimulated the development of heterogeneous
catalytic systems [3]. Efficient asymmetric heterogeneous catalysts for
hydrogenations were obtained by chiral surface modification of metal
particles [3–5]. Among these Pd modified by cinchona alkaloids were
found efficient for the enantioselective hydrogenation of unsaturated
carboxylic acids [6–9]. High enantioselectivities were obtained in the
hydrogenation of α-substituted cinnamic acids when achiral primary
amine additives were used [8–10].
In contrast, the hydrogenation of aliphatic acids resulted in moderate
optical purities, thus studies aimed at improving the stereocontrol in
these hydrogenations are of great importance. Previous studies showed
that achiral amines are also efficient in improving the enantioselectivities
in the hydrogenation of aliphatic carboxylic acids over Pd modified by
cinchonidine (CD) [11–14]. In our precious studies we have investigated
the effect of the amine structure and the influence of the amines on the
H2 pressure and temperature dependence [13,14]. Based on these results
further studies are needed to ascertain on the role of the additive
and for the rational tuning of the catalytic system to obtain better
stereocontrol of the reaction. Accordingly, we continued these studies in-
vestigating the effect of the achiral amines in various solvents and using
different modifiers, earlier studied only in the absence of additives
[15–19].
2. Experimental
Commercial 5% Pd/Al2O3 catalyst (Engelhard, 40692; [14,16]) was
used as received. CD, cinchonine (CN), quinine (QN), benzylamine (BA),
N-methylbenzylamine (MBA), (E)-2-methyl-2-butenoic acid (tiglic acid,
TA) from Aldrich and H2 gas (Linde AG, 99.999%) were used without pu-
rification. Cinchonidine methyl ether (CDM) and cinchonidine phenyl
carbamate (CDPC) were prepared by known procedures [20,21]. Analyt-
ical grade solvents were used (in brackets the normalized empirical
solvent polarity parameters, ENT [22]): cyclohexane (0.006), mesitylene
(0.068), toluene (0.099), benzene (0.111), ethyl acetate (0.228), 2-
butanone (0.327), acetone (0.355), 2-propanol (0.546), hexanol (0.559),
methanol (0.762), water (1) and acetic acid (0.648).
Hydrogenations were carried out as earlier described using stainless
steel autoclaves equipped with pressure transmitter for recording the
pressure [13,14]. Typical reaction conditions were: 15 mg Pd/Al2O3,
10 cm3 solvent, 0.05 mmol modifier, 1 mmol TA and 1 mmol amine,
☆
Dedicated to Professor Mihály Bartók on the occasion of his 80th birthday.
Corresponding author. Tel.: +36 62 544514; fax: +36 62 544200.
E-mail address: szollosi@chem.u-szeged.hu (G. Szőllősi).
⁎
1566-7367/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.catcom.2013.11.029

114
Z. Makra, G. Szőllősi / Catalysis Communications 46 (2014) 113–117
5 MPa H2 pressure, magnetic agitation (1000 rpm), 297 K, 1–2 h.
Products were identified by GC–MSD analysis; quantitative analysis
was carried out by GC–FID following washings with 10% aqueous
HCl to remove the amine additive and the modifier. Initial rates (Ri;
mmol h−1 g−1) were calculated from the recorded H2 pressure de-
crease up to 20 mol% conversion, as in previous studies [14]. Complete
conversions were reached in each reaction. Enantioselectivities (enan-
tiomeric excess, ee (%)) were calculated as: ee (%) = 100 × |[(S)-
MBu] − [(R)-MBu]| / ([(S)-MBu] + [(R)-MBu]); where [(S)-MBu]
and [(R)-MBu] are the concentrations of the 2-methylbutyric acid
enantiomers. The reproducibility of the ee was 2%, whereas the Ri
values were reproducible within 20 mmol h−1 g−1, as ascertained
from results of experiments repeated three times. The ee values corre-
sponding to linear behavior with mixtures of cinchona alkaloids were
calculated: eecalc (%) = (Ri1 × ee1 + Ri2 × ee2) / (Ri1 + Ri2); where
Ri1 and Ri2 are the initial rates and ee1 and ee2 are the ees obtained
with the individual modifiers.
3. Results and discussions
3.1. Solvent dependence in the presence of amine additives
Hydrogenation of TA using CD as modifier results in the excess
formation of (S)-MBu (Scheme 1). In the hydrogenation of aliphatic
unsaturated acids higher ees are obtained in apolar solvents than
polar solvents [15,17]. Here we studied the effect of two achiral amines,
i.e. BA and MBA, on the TA hydrogenation in various solvents (Fig. 1)
and the results were compared with those obtained in the absence of
amine. Results are plotted as a function of ENT values of the solvent,
which provide a sensitive characterization of solvent polarity and has
proved to give satisfactory correlations in several solvent-dependent
processes [22].
Interestingly similar ees were obtained in apolar and polar solvents
having ENT up to 0.4 (i.e. in toluene, ethyl acetate or acetone), even in
the presence of amine additives only slight decrease in the ee values
(within the limit of error) were observed up to 0.4 ENT . The lower ee in
cyclohexane, (ENT 0.006) in the presence of both amines may be
accounted for the low solubility of the acid–amine salt in this solvent.
Significant and almost linear ee decrease as a function of ENT was detect-
ed in polar protic solvents. The lowest ees were obtained in water and
acetic acid, both strong H-bond donors. Generally the use of BA or
MBA increased the ee in all solvents (except acetic acid) as compared
with the absence of additive. The Ri increased with the ENT value of the
solvent, however, lower increase in presence of amine additives were
obtained. These observations suggested that H-bond donor solvents
(i.e. protic solvents) perturbed the efficient H-bonding between the
modifier and the acid. The smaller variations of the initial rate by ENT in
presence of amines as compared with the reactions in the absence of
achiral additives (see the slope of the curves between 0.1 and 0.4 ETN
values) indicate that the amine is involved in the rate determinant
step of the reaction; otherwise the slope of the curves would be close
with that observed in the absence of additives.
Fig. 1. Correlation between the solvent ENT value and the ee (A) or initial H2 uptake rate
(B) obtained in the enantioselective hydrogenation of TA using CD modifier. Reaction
conditions: see Experimental section; in the absence of amine (♦); in the presence of
BA ( ) or MBA ( ).
complex responsible for enantiodifferentiation in the hydrogenation
of aliphatic acids is still not well established. Formation of acid-CD
complexes in 2/1 or higher ratios were suggested [15,17,26], however,
1–1 complexes were also suggested [18]. It is known that the acid
monomer–dimer equilibrium and the acid–tertiary amine interactions
are affected by the nature of the solvent and dilution [27,28]. According-
ly, we studied the dilution effect on the hydrogenation of TA in two
solvents, i.e. toluene and methanol (Fig. 2).
Under our experimental conditions even at the lowest tiglic acid
initial concentration associated dimer species should be in excess in
apolar solvents, whereas in polar protic solvents the quantity of the
monomer may equal that of the dimer, as shown in previous studies
using acetic acid [27] or an α,β-unsaturated acid [26]. Nearly constant
ees were obtained in toluene except at the lowest acid concentration,
where the ee decreased significantly. Based on the decrease of the ee
on increasing the toluene amount, i.e. by possible slightly shift of the
Although, in the hydrogenation of cinnamic acid derivatives forma-
tion of acid-modifier 1–1 complex is accepted [23–25], the surface
N
8
9
OH
H
COOH
COOH
COOH
+H2, Pd/Al2O3
H
H3C
H3C
H3C
S
R
+
+CD, solvent
CH3
CH3
CH3
(S)-MBu
major product
TA
(R)-MBu
minor product
N
(CD)
Scheme 1. Scheme of the enantioselective hydrogenation of TA over CD-modified Pd.

Z. Makra, G. Szőllősi / Catalysis Communications 46 (2014) 113–117
115
Fig. 2. The ee obtained in the enantioselective hydrogenation of TA as a function of solvent
amount in toluene (closed symbols) and methanol (open symbols): without amine (♦,◊),
in the presence of BA ( ) or MBA ( ). Reaction conditions: see Experimental section.
Fig. 4. Effect of CD concentration (logarithmic scale) on the ee of TA hydrogenation in
,
,
BA
absence of amine (♦), using BA ( ) or MBA ( ) and on the Δee ( : eewith
−
amine
eewithout
;
: eewith MBA − eewithout amine). Reaction conditions: see Experimental
section; solvent: 10 cm3 toluene.
dimer–monomer acid equilibrium in the liquid phase, one may assume
that the decrease is caused by formation of CD–acid monomer interme-
diates, even if the liquid phase dimer/monomer ratio may be altered on
the surface. Although confirmation of the composition of the surface
species are still needed, according to these results it is probable that
the high ee in apolar solvents is due to the interactions of acid dimers
with the modifier (see Fig. 3), as shown in previous studies [16,17,26].
In methanol in the absence of additives the ee decrease was small
even at the highest dilution. It is known that solvents with hydrogen-
bond donor ability may interact with amine–carboxylic acid adducts
[28]. Thus, formation of an associated CD–TA–HOMe surface complex
incorporating methanol even under the most concentrated conditions,
which is kept even under the most diluted conditions is plausible
(see Fig. 3). Such an intermediate may explain both the low ee in protic
solvents and the only slightly decreasing ee values obtained by increas-
ing the dilution.
Table 1
Effect of the amine additives on the hydrogenation of TA over Pd modified by cinchona
alkaloids.a
Entry
Modifier
Additive
Ri/t (min)b
ee (%) (config.)
Δee (%)c
1
2
3
4
5
6
7
8
CD
CD
CD
QN
QN
QN
CDM
CDM
CDM
CDPC
CDPC
CDPC
CN
–
260/30
190/45
239/30
287/30
233/45
249/40
548/15
136/30
222/20
291/35
142/45
186/40
270/20
226/30
249/25
48 (S)
62 (S)
63 (S)
15 (S)
10 (S)
14 (S)
25 (S)
28 (S)
33 (S)
16 (S)
17 (S)
18 (S)
39 (R)
48 (R)
51 (R)
–
In the presence of amine additives similar decrease in the ee was ob-
served at the highest dilution in both toluene and methanol. According-
ly, we suppose that in both types of solvents the additives act similarly,
however these results could not confirm the involvement of the achiral
amines in the surface intermediate responsible for enantioselection, as
suggested previously.
BA
MBA
–
14
15
–
BA
MBA
–
−5
−1
–
BA
MBA
–
3
8
9
10
11
12
13
14
15
–
3.2. Effect of the modifier concentration and structure
BA
MBA
–
1
2
Further we attempted to ascertain on the involvement of the achiral
amines in the enantiodifferentiating step by studying the effect of the
CD concentration ([CD]) and the modifier structure. Such studies have
been reported without using achiral amine additives [15–17] and in a
recent report using epicinchonidine [29]. The effect of [CD] is presented
in Fig. 4.
–
CN
CN
BA
MBA
9
12
a
Reaction conditions: see Experimental section, solvent: 10 cm3 toluene.
b
c
Initial H2 uptake rate (mmol h−1 g−1)/reaction time.
Δee = eewith amine − eewithout amine
.
in aprotic solvents
in protic solvents
O
O
O
O
Solv
H
H
O
O
H
dilution
H
O
O
H
H
N
H
N
H
N
O
O
O
R
R
R
S
S
S
N
N
N
Fig. 3. Probable intermediate structures in the absence of amine additives in aprotic and protic solvents.

116
Z. Makra, G. Szőllősi / Catalysis Communications 46 (2014) 113–117
presence of either C6- or C9–O-substituents the beneficial effect of the
additives was partially or completely lost. Taking into consideration
the difference in the adsorption strengths of the aliphatic and cinnamic
acid derivatives and the adsorption mode of the cinchona alkaloids over
Pd [29], these observations can be rationalized by the steric hindrance of
the substituents on formation of intermediates incorporating the addi-
tive, supporting the participation of the amine additive in the reacting
surface intermediate.
In the hydrogenation of TA over catalyst modified by 1/1 mixtures
of cinchona alkaloids (Table 2) the experimentally determined ees
deviated from the calculated values corresponding to linear behavior.
This non-linear phenomenon has its origin in different adsorption
strengths of the intermediates formed with the adsorbed cinchona alka-
loids [4,24,30,31]. The increased deviation from the linear behavior in
the presence of amines should be the consequence of further increase
in the difference in the adsorption strength of the intermediates in
the presence of the additives. This supports the assumption that the
amine participate in the formation of the surface intermediate responsi-
ble for enantioselection.
Table 2
Results obtained using cinchona alkaloid 1/1 mixtures as modifiers.a
Entry
Modifier mixture
Additive
Ri
ee (%)
eecalc (%)b
ΔLee (%)c
1
2
3
4
5
6
7
CD + CN
CD + CN
CD + CN
QN + CN
QN + CN
CDM + CN
CDM + CN
–
230
82
13 (S)
20 (S)
21 (S)
35 (R)
47 (R)
29 (R)
45 (R)
4 (S)
2 (S)
4 (S)
11 (R)
19 (R)
4 (S)
9
18
17
24
28
33
35
BA
MBA
–
182
214
173
371
246
MBA
–
MBA
10 (R)
a
Reaction conditions: see Table 1, 0.05 mmol modifier 1/1 mixture.
ee corresponding to linear behavior.
ee deviation from the linear behavior.
b
c
Higher increases in the ee in the presence of amines were obtained
at low [CD] and under these conditions ([CD] 0.5 mM) the ee increase
obtained by using MBA was higher than with BA (see Δee, Fig. 4).
A possible explanation of the smaller effect of both achiral amines on
the ee at high [CD], i.e. by increasing the modifier surface coverage,
can be explained by hindered adsorption of the additive on the surface.
This supports the suggestion that the amines influence the surface reac-
tion adsorbed on the surface, i.e. possibly by participating in the
formation of the intermediate responsible for the enantioselection.
Results obtained over Pd modified by other cinchona derivatives are
shown in Table 1.
The ees obtained with both CD and CN were strongly increased by
the amine additives. In contrast the ee obtained using QN decreased in
the presence of BA. Transforming the C9–OH to methyl ether or phenyl
carbamate resulted in smaller ee increase as obtained with CD, though
the latter derivative bears a H-bond donor moiety. Accordingly, in the
4. Conclusions
We have disclosed results obtained in the hydrogenation of (E)-2-
methyl-2-butenoic acid over Pd modified by cinchona alkaloids
supporting the involvement of the amine in the surface reaction. The
solvent dependence of the ee in the presence of amines indicated that
these additives are involved in the rate determinant step of the reaction.
Dilution effects were in agreement with the participation of acid dimers
in reactions carried out in aprotic solvents, whereas protic solvents
interfere in the modifier–acid interaction. The influence of amines on
R
H
O
R
O
N
H
O
O
N
H
H
H
N
O
O
dilution
O
N
R
H
H
N
R
O
S
R
H
S
N
N
R
O
H
R
N
N
O
H
H
dilution
H
H
O
O
O
O
N
R
H
S
O
N
R
H
H
N
N
O
R
S
N
Fig. 5. Possible structures of the intermediates in the presence of amine additives in aprotic solvents.

Z. Makra, G. Szőllősi / Catalysis Communications 46 (2014) 113–117
117
[2] H.-U. Blaser, C. Malan, B. Pugin, F. Spindler, H. Steiner, M. Studer, Adv. Synth. Catal.
345 (2003) 103–151.
[3] In: K. Ding, Y. Uozumi (Eds.), Handbook of Asymmetric Heterogeneous Catalysis,
Wiley-VCH, Weinheim, 2008.
[4] T. Mallat, E. Orglmeister, A. Baiker, Chem. Rev. 107 (2007) 4863–4890.
[5] M. Bartók, Chem. Rev. 110 (2010) 1663–1705.
[6] K. Borszeky, T. Mallat, A. Baiker, Tetrahedron Asymmetry 8 (1997) 3745–3753.
[7] Gy Szőllősi, S. Niwa, T. Hanaoka, F. Mizukami, J. Mol. Catal. A Chem. 230 (2005) 91–95.
[8] Gy. Szőllősi, B. Hermán, K. Felföldi, F. Fülöp, M. Bartók, Adv. Synth. Catal. 350 (2008)
2804–2814.
the effect of modifier concentration showed the involvement of the
adsorbed additive in the reaction possibly by participating in the forma-
tion of the intermediate responsible for the enantioselection. These
suggestions were confirmed by ees obtained over catalyst modified
by cinchona alkaloid derivatives and deviations from the non-linear
behavior obtained using modifier 1/1 mixtures.
Based on these results combined with published data, i.e. the effect
of the amine amount [14] and the structure of the intermediate in the
absence of amine additives [26], we suggest structures shown in Fig. 5
as possible surface intermediate complexes formed in the presence of
amine additives in aprotic solvents. According to this suggestion the
possible modifier/acid/amine 1/2/2 adducts are partially transformed
in 1/1/1 surface complexes by dilution resulting in weaker stereocontrol
and lower ee. As was shown in Fig. 3 in protic solvents the involvement
of the solvent in these interactions is probable resulting in a complex
hydrogen-bonding network. Elucidation of the structure and composi-
tion of such surface species needs further studies. Although these
suggestions should be further confirmed by various methods, these re-
sults represent a novel step in tailoring this catalytic system to obtain
high ees in the hydrogenation of aliphatic carboxylic acids.
[9] T. Sugimura, T. Uchida, J. Watanabe, T. Kubota, Y. Okamoto, T. Misaki, T. Okuyama,
J. Catal. 262 (2009) 57–64.
[10] Gy Szőllősi, B. Hermán, E. Szabados, F. Fülöp, M. Bartók, J. Mol. Catal. A Chem. 333
(2010) 28–36.
[11] Gy. Szőllősi, T. Hanaoka, S. Niwa, F. Mizukami, M. Bartók, J. Catal. 231 (2005) 480–483.
[12] Gy Szőllősi, K. Balázsik, M. Bartók, Appl. Catal. A Gen. 319 (2007) 193–201.
[13] Gy Szőllősi, Zs. Makra, M. Bartók, React. Kinet. Catal. Lett. 96 (2009) 319–325.
[14] Zs Makra, Gy Szőllősi, M. Bartók, Catal. Today 181 (2012) 56–61.
[15] K. Borszeky, T. Mallat, A. Baiker, Catal. Lett. 41 (1996) 199–202.
[16] K. Borszeky, T. Bürgi, Z. Zhaohui, T. Mallat, T. Mallat, A. Baiker, J. Catal. 187 (1999)
160–166.
[17] I. Kun, B. Török, K. Felföldi, M. Bartók, Appl. Catal. A Gen. 203 (2000) 71–79.
[18] R. Bisignani, S. Franceschini, O. Piccolo, A. Vaccari, J. Mol. Catal. A Chem. 232 (2005)
161–164.
[19] S. Tan, J.R. Monnier, C.T. Williams, Top. Catal. 55 (2012) 512–517.
[20] C. Exner, A. Pfaltz, M. Studer, H.-U. Blaser, Adv. Synth. Catal. 345 (2003) 1253–1260.
[21] G. Uccello-Barretta, S. Bardoni, F. Balzano, P. Salvadori, Tetrahedron Asymmetry 12
(2001) 2019–2023.
[22] C. Reichardt, Solvents and Solvent Effects in Organic Chemistry, Third edition
Wiley-VCH, Weinheim, 2003.
Acknowledgments
[23] Y. Nitta, A. Shibata, Chem. Lett. 27 (1998) 161–162.
[24] Gy Szőllősi, I. Busygin, B. Hermán, R. Leino, I. Bucsi, D.Yu. Murzin, F. Fülöp, M. Bartók,
ACS Catal. 1 (2011) 1316–1326.
[25] Gy. Szőllősi, Catal. Lett. 142 (2012) 345–351.
[26] D.M. Meier, A. Urakawa, N. Turrà, H. Rüegger, A. Baiker, J. Phys. Chem. A 112 (2008)
6150–6158.
[27] G.M. Barrow, E.A. Yerger, J. Am. Chem. Soc. 76 (1954) 5248–5249.
[28] G.M. Barrow, E.A. Yerger, J. Am. Chem. Soc. 76 (1954) 5211–5216.
[29] E. Schmidt, C. Bucher, G. Santarossa, T. Mallat, R. Gilmour, A. Baiker, J. Catal. 289 (2012)
238–248.
Financial support by OTKA Grant K 72065 is highly appreciated.
The research was supported by TÁMOP 4.2.2A-11/1/KONV-2012-0047
and TÁMOP 4.2.4.A/2-11-1-2012-0001 “National Excellence Program”
projects.
References
[1] T. Ohkuma, M. Kitamura, R. Noyori, Asymmetric hydrogenation, in: I. Ojima (Ed.),
Catalytic Asymmetric Synthesis, 2nd ed., Wiley-VCH, New York, 2000, p. 1, (Chap. 1).
[30] T. Sugimura, H. Ogawa, Chem. Lett. 39 (2010) 232–233.
[31] Gy Szőllősi, B. Hermán, F. Fülöp, M. Bartók, J. Catal. 276 (2010) 259–267.