close

Вход

Забыли?

вход по аккаунту

?

Methods of Asymmetric SynthesisЧEnantioselective Catalytic Hydrogenation.

код для вставкиСкачать
ANGEWANDTE CHEMIE
In ternational
V O L U M E 1 0 . N U M B E R 12
D E C E M B E R 1971
P A G E S 871-948
Methods of Asymmetric Synthesis-Enantioselective Catalytic
Hydrogenation
By Yoshiharu Izumi“’
It is proposed that asymmetric syntheses be divided into enantioselective and diastereoselectiue
syntheses. The enantioselectiue hydrogenations discussed in the present progress report were
catalyzed by Raney nickel that had previously been treated with solutions of optically active
compounds. Relationships exist between the enantioselectivity of the catalyst and the structure
of the chiral compound used to rnodifv it.
1. Introduction
described as “asymmetric” if it lacks symmetry elements at
least of the second kind.
“Asymmetric synthesis” is still one of the most undeveloped
fields of organic chemistry, though most natural products
are asymmetric and the enantiomer that occurs in nature
is in many cases the only one that acts as a drug, nutrient,
or flavoring agent. The development of asymmetric syntheses is of particular interest to branches of industry that
produce synthetic natural products such as amino acids.
Nakazaki“] has referred to the need for a new definition of
“asymmetric reactions” based on the concept of chirality[*].
Together with Nakazaki, therefore, the present author proposes that “asymmetric syntheses” be divided into enantioselective syntheses and diastereoselective syntheses. In this
new classification, most “inductive asymmetric syntheses”
are diastereoselective syntheses, while “asymmetric catalytic syntheses”, “absolute asymmetric syntheses”, and
most enzymatic syntheses should be classified as enantioselective syntheses.
Studies on asymmetric syntheses are of great importance
not only scientifically but also preparatively. They may
even help, for example, to elucidate the mode of action of
enzymes. Improved knowledge of asymmetric syntheses
should also provide an insight into unsolved problems in
connection with the origin of life, as well as an answer e. g.
to the question why all proteins that are important to life
consist exclusively of L-amino acids.
We have been concerned in recent years with catalysts for
asymmetric hydrogenation. The results are reported briefly
in the present article. However, let us first examine the
term “asymmetric synthesis”.
2. Definitions
The description “asymmetric” is used in two ways. On the
one hand, “asymmetric synthesis”, is understood to be a
reaction in which the enantiomers of a chiral molecule are
formed in unequal quantities. On the other, a molecule is
[*I
Prof. Dr. Yoshiharu Izumi
Institute for Protein Research, Osaka University,
Kita-ku, Osaka (Japan)
Angew. Chem. internal. Edit. 1 Vol. 10 (1971)
1 N o . 12
2.1. Diastereoselective Synthesis
If a molecule contains a center of chirality and a center of
prochirality, the molecule can be divided by a plane in
such a way that the parts on either side of the plane are diastereotopically related to each other. If a reaction results in
the conversion of the center of prochirality into a new
center of chirality, the reagent may attack from either side
of this plane, with the result that diastereoisomers are formedL3’.The formation of the diastereoisomers in unequal
quantities in such a reaction may be due to differences in
the strength with which a catalyst is adsorbed on the two
sides of the “diastereotopically reacting” plane of the substrate or to differences in the rate at which a reactant attacks
on the two sides of the plane.
We shall refer to reactions of this type as diastereoselective
syntheses. For example, in the hydrogenation of 3-ethyl-6benzylidene-3-methyl-2,5-dioxopiperazine( I ) [41,a cata871
lyst is adsorbed less strongly on the side of the dioxopiperazine ring (0)on which the ethyl group projects
(reverse side in Fig. 1)than on the other side. Hydrogenation thus leads mainly to the product in which the ethyl
and benzyl groups lie on the same side of the piperazine
ring ( I ) (cis-configuration).
A “diastereotopic reaction plane” (P) can also be drawn in
the reaction of phenylglyoxyl ester (2), which is shown in
Figure 2 in the preferred conformation, with methylmagnesium iodide[51.In the reaction, the Grignard reagent will
preferentially attack the phenylglyoxyl ester from the side
opposite the largest substituent L, with preferential formation of the product in which L and the methyl group lie on
opposite sides. of the plane (P) (trans-conformation).
1
2.2. Enantioselective Synthesis
Achiral molecules containing a center of prochirality can
be divided by a plane in such a way that the two sides of the
plane are enantiotopically related to each other. The attack
of a reactant from one side leads to the S enantiomer of the
new chiral molecule, while attack from the other side leads
to the R enanti~rner’~].
An achiral reactant or an achiral
catalyst approaches both sides of the “enantiotopic plane”
at the same rate and so yields the racemate. An enantioselective reactant or an enantioselective catalyst, on the
other hand, can differentiate between the two sides of the
“enantiotopic plane”; the rates of attack on the two sides
of the plane are different, and the R and Scornpounds are
formed in unequal quantities. The difference between the
quantities of enantiomers formed depends on the enantioselectivity of the reagent or of the catalyst. A synthesis
carried out with an enantioselective reagent or catalyst
will be referred to as an “enantioselective synthesis”.
I
Fig. 1. Diastereoselective hydrogenation on the side of the diastereotopic plane (0)favored by the configuration. x, =existing center of
chirality, y , =induced center of chirality.
Fig. 3. Enantioselectivehydrogenation on the enantiotopic plane (Q).
In enantioselective synthesis, the enantioselective action
of the catalyst or of the reagent causes the new center of
chirality to be formed mainly in the R or in the S form.
Examples of such reactions are the enantioselectivehydrogenation described in the present progress report and reactions with chiral Grignard reagents”].
L
I
MgI
Fig. 2. Diastereoselective synthesis on the side of the diastereotopic
plane (P) favored by the conformation.@,@ and@denote the largest,
middle, and smallest substituents respectively. x2 =existing center of
chirality, y2 =induced center of chirality.
For example, the enantioselective, modified Raney nickel
catalyst (see Section 4 er seq.) has two functions; firstly, it
catalyzes hydrogenations in the same way as ordinary
Raney nickel, and secondly it distinguishes between the
two sides of the enantiotopic plane of the substrate. The
hydrogenation is due to the metal of the catalyst, and the
differentiation between the two sides to the chiral compound adsorbed on the catalyst.
3. survey
In this type of reaction, the inequality of the quantities of
the cis- and trans-compounds in the product is proportional
to the diastereoselectivity of the catalyst or of the reagent.
With racemic starting materials, neither a catalyst nor a
reactant can induce the preferential formation of one
enantiomeric form of the new center of chirality. This can
be achieved, however, if optically active substrates are
used. If the ligands on the existing centers of chirality x1
and x2 are constant, the degree to which xI and x2 influence
the stereochemistry of the new centers ofchirality y1 and y2
depends only on the diastereoselectivity of the catalyst or
of the reagent161. “Inductive asymmetric synthesis” is thus
a special case of diastereoselective synthesis.
8 72
The first enantioselective synthesis with a heterogeneous
catalyst was achieved by our research group in 1956 with
a silk-palladium catalystL8- lo]. This catalyst enantioselectively hydrogenated derivatives of oximes and oxazolones (Table 1, Part 1).
During our investigations on the silk-palladium catalyst
and the modified Raney nickel catalyst, Isoda et al.“
published a report on enantioselective hydrogenations
with Raney nickel. They investigated the hydrogenation of
oxazolone and a-oxoglutaric acid derivatives with Raney
nickel that had been treated with amino acids (Table 1,
Part 2).
Angew. Chem. internat. Edi!. 1 Vol. 10 (1971) / No. 12
1.
-CO-CH-(CH&-
[
LOOR
( I ? ) , R=C,H,;
-CHOH-CH-(CH
[
2)6-
LOOR
I.
(19), R = C 2 H s ; (20), R=CH,
(18),R=CH,
co
Table 1. Examples of enantioselective hydrogenations. The italicized figures in brackets in the “substrate” and
“product” columns refer to the formulas reproduced here. @-Ni means Raney nickel.)
Catalyst
Modifying
reagent
pH
T(“C)
Hydrogen- Subation (“C) strate
silk/Pd
Product
(3)
(4)
(5)
(7)
(6)
(6)
(8)
(9)
[a]D (“)
Opt. Yield
(%) [a1
+2.25 [b]
+9.25 [c]
i 1 2 . 5 [c]
+8.75 [d]
6.1
26.3
35.6
Ref.
-
_ _ _ _ _ _ ~ ~ _ _ _ _
~
R-Ni
(S)-tyrosine
(S)-gluramic acid
(S)-leucine
(S)-phenyialanine
~~
-12
-0.3
-0.1
-0.25
-0.1
-0.2
36
7
3
6
2.5
5
(10)
(6)
~
R-Ni
R-Ni
(S)-mandelic acid
(-)-ephedrine
L-( +)-threo-Ip-nitrophenyl
2-aminopropane1,3-dioI
(2R, 3R)-tartaric
acid
(S)-glutamic acid
(2R, 3R)-tartaric acid
(S)-glutamic acid
[a]
[b]
[c]
[d]
[el
(10)
(11)
(4)
(11)
(11)
(11)
(7)
(lo)
(3)
(10)
-
20
20
70
70
(12)
(12)
(13)
(13)
-0.13
+O.ll
0.5
0.4
1451
~451
9.5
20
70
(12)
(13)
-0.10
0.4
1451
5.0
5.1
Sl
0
20
20
70
60
42
(12)
(14)
(13)
(15A)
(UB)
-2.02
-0.024
-0.82
8.1
0.96
2.6
c451
(19)
+3.12[e]
+3.75
+1.42
+6.30
-0.465
5.0
(16)
(17)
(18)
(18)
(21)
(12)
(20)
(20)
(22)
(13)
1451
1451
~461
~461
1461
~461
~461
The optical yield is defined as ( [ a ] , (fo~nd))/([a]~(lit.))x 100 1%].
In 2N HCI.
In water.
Inether.
[a], values in this column measured at 20°C.
Angew. Chem. internal. Edif. / Vol. 10 (1971) / N o . 12
873
After the investigation of silk-noble metal catalysts[’3-18],
we turned in 1963 to the enantioselective hydrogenation of
methyl acetoacetate with modified Raney nickel. Since
then, we have continued our systematic studies on this
catalyst[” -401.
Petrov et al. also investigated modified Raney nickel catalysts (Table I, Part 3)[451.Plate et al.[461recently reported
the enantioselectivityofmodified Raney nickel in the hydrogenation of poly-S-0x0 esters and poly-P-diketones (Table
1, Part 4).
In 1964, Padgett et ~ l . [ ~reported
‘]
the enantioselectivity of
a palladium catalyst on a specially prepared silica gel
support (Table 2). The silica gel had been precipitated from
sodium silicate with hydrogen chloride in the presence of
optically active alkaloids. Beamer et a1[42-441investigated
.
the similar enantioselective hydrogenation of a-methylcinnamic acid and a-acetaminocinnamic acid with palladium catalysts on polyamino acid supports.
Soviet chemists explained the steric effect of a reagent that
allows the asymmetric hydrogenation of the substrate on
the basis of Balandin’s presentation of the multiplet theory
of the catalystr471.They assume further that the modifying
reagent and the substrate are combined in the form of a
chelate.
The mechanism proposed by Beamer et al. for the enantioselective hydrogenation is based on the idea that the cata-
Klabnovskii and Petrov suggested that enantioselective
hydrogenation on modified Raney nickel proceeds as
shown in Figure 5[481.They assume that the adsorption of
the substrate is determined by the stability of the catalyst-
Table 2. Enantioselectivity of a palladium catalyst on an optically active support. a-Methylcinnamic acid is
reduced to x-methylphenylpropionic acid, and a-acetylaminocinnamic acid to phenylalanine.
~~
Substrate
a-Methylcinnamic acid
Opt Yield
(%) [a1
Catalyst
Pd-QN 0.5 SG [b]
Pd-QDN 0.5 SG [b]
Pd-CN 0.5 SG [b]
Pd-CDN 0.5 SG [b]
Pd- Poly-(S)-leucine[c]
+0.87
3.21
~411
- 3.20
1.66
1.74
3.25
1.94
1.18
- 2.6
- 2.5
5.4
5.16
1411
~411
r411
~421
[441
-1.124
4.15
-2.9
6.0
1431
+0.388
+0.245
1.43
0.90
+ 2.06
+ 0.46
0.95
4.25
1431
[441
lyst has two kinds of adsorption sites, which differ in their
affinity for the two sides of the enantiotopic plane of the
substrate. The chiral structure of the polyamino acid used
as the support causes the two types of adsorption sites to
be formed on the surface of the catalyst in unequal quantities. According to Beamer’s view, the enantioselectivity
of the catalyst is due to the surface structure of the silica
gel, which has been modified in such a way by the precipitation in the presence of optically active alkaloids that the
silica gel, like the polyamino acids, can now act as an optically active support for the metal. Ion affinity,dipole interaction, hydrogen bonding, and van der Waals forces are
named as playing an important part in the adsorption of
the substrate on the catalyst (Fig. 4).
+
Q
H-‘<H
HOOC,C ,CH3
I
w
H
Fig. 4. Course of the enantioselective hydrogenation of cr-methylcinnamic acid with palladium on poly-(S)-leucine as the catalyst (after
a suggestion by Bearner et al.) [42--441.
874
Ref
+ 0.45
+ 0.47
+-0.536
0.88
Pd-Poly-:,-henzyl-(S)glutamic acid [c]
Pd- Poly-B-benzyl-(S)aspartic acid [c]
Pd-Poly-(S)-valine [c]
a-Acet ylaminocinnamic acid
[ a I D (“1
Opt. Yield
(%) [a1
Fig. 5. Enantioselective hydrogenation of ethyl acetoacetate with modified Raney nickel as the catalyst (after a suggestion by Klabnomkii) 1481.
substrate bond ; this bond will be influenced by the chiral
structure of the modifying reagent. According to Plate
et aL1461,the same processes underlie the enantioselective
hydrogenation of poly-$oxo esters or poly-P-diketones.
4. Development of the Modified Raney Nickel
Catalyst
In our investigations on enantioselective catalysts, we
started by considering the course of enzymatic reactions.
We simply assumed at first that the enantioselectivity of
enzymatic reactions was due to the asymmetric environment of the protein molecule, and that the chiral product
was formed in a “chiral mou1d”just as a waffle is formed in
a waffle iron.
Angew. Chem. internat. Edit.
Val. 10 (1971) / No. 12
Silk fibroin has a rigid structure, contains no amino acid
residues containing sulfur, and can coordinate metal irons.
If silk fibroin is boiled with aqueous palladium chloride
solution and the silk-palladium complex subsequently
hydrogenated, one obtains a palladium catalyst in which
the metal is embedded in an asymmetric environment. We
achieved the first enantioselective hydrogenation with this
catalyst (see Section 3, Table 1).Unfortunately, it was found
after a few months that the enantioselectivity of the silkpalladium catalyst is very strongly dependent on the type
of silk, and that the results were accordingly not reproducible.
While investigating the reproducibility of the enantioselectivity of the silk-palladium catalyst, we found that the
hydrogenation activities of silk-palladium, silk-platinum,
and silk-rhodium catalysts are much weaker for the C=O
double bond than for other double bonds. To establish
whether this behavior was attributable to the silk protein,
Raney nickel was treated with aqueous gelatin solution,
and the activity of the resulting catalyst for the carbonyl
group was tested[49! It can be seen from Table 3 that the
hydrogenation activity of the catalyst for the carbonyl
group, as expected, was greatly decreased by this treatment.
At that time we assumed that the active sites on the hydrogenation catalyst behaved differently toward C=O, NO,,
C=C, etc. To explain this behavior (see Table 3), the concept of competitive inhibition was applied to these reactions. We assumed that the sites of the catalyst that are
active toward the carbonyl group lose their activity when
occupied by the carbonyl group of the modifying reagent,
which evidently has a greater affinity for these active sites
than the carbonyl group of the substrate.
The hydrogenation activities of Raney nickel catalysts
modified with amino acids and with chelating agents were
also tested on carbonyl, nitro, and C=C double bonds.
The results are shown in Tables 3 and 4.
Below IO’C, the carbonyl group undergoes practically no
hydrogenation in the presence of the modified Raney
Table 3. Hydrogenation activity of modified Raney nickel [49].
Modifying
reagent
-
Hydrogenation activity [a] compared to
2-Butanone
Allyl
Nitroalcohol
benzene
at 1 0 T at 60°C
at 10°C at 10°C at 40°C
2.0
Gelatine [b]
0.5
(S)-Glutamic acid 0
0
Glycine
0
Glycylglycme
Sodium hydrogen
(S)-glutamate
Disodiumdihydrogen ethylenediamine
tetraacetate
0.5
0
Succinic acid
Sodium acetate
2
0.5
Diacetyldioxime
Et hylenediamine
hydrochloride
0
55
15
1.5
3
2
4
3-3.5
6
5.5
3
34
38
36
35
45
9
16
2
8
32
45
Table 4. Hydrogenation activity of a Raney nickel modified with
glutamic acid [49] (R-Ni means Raney nickel.)
Substrate
Hydrogenation activity [a]
at 10°C
at 60 C
R-Ni
R-Ni+Glu
R-Ni
R-Ni + Glu
Acetone
2-Butanone
Acetophenone
Cyclohexanone
Allyl alcohol
Cinnamic acid
Maleic acid
Ethyl acrylate
Diethyl maleate
Mesityl oxide [el
Cinnamaldehyde
[a]
[b]
[c]
[d]
[el
5.7
2
8
1.8
34
16
41
100
100
28
12.5
0
0
0
0
55
20
120
24.3 [b]
8
2
22
7
34
16
63
100 [c]
100 [d]
14
0.5
See footnote [a] in Table 3.
Reduction at 80°C.
Taken up in 3 min.
Taken up in 4 min.
The C O losses in the products were 11
6
14
1
and 3% respectively.
nickel, and the uptake of hydrogen begins only above 60°C.
The hydrogenation activity towards other double bonds,
on the other hand, remains unchanged or is even improved
by the modification of the catalyst.
From the above results, we expected that the chirality of an
amino acid must take effect if a Raney nickel modified with
this amino acid is used as a catalyst in hydrogenations
above 60°C. The Raney nickel catalyst with which the first
enantioselective hydrogenation was achieved had been
modified with an aqueous solution of (S)-glutamicacid.
We have developed the following view on the active sites
in modified Raney nickel catalysts. The hydrogenation can
take place at any site on the catalyst on which the substrate
is adsorbed. The active sites are not specificfor the various
functional groups such as C=O, NO,, or C=C. In our
view, the active sites on the catalyst surface, e. g. dislocations, are capable of functioning even before contact with
hydrogen ; they initiate the activation of the hydrogen like
initiators in a chain reaction.
The position of the modifying reagent on the catalyst surface is determined by the affinity of the reagent for the catalyst on the one hand and the repulsive forces between the
reagent molecules on the other. The poisoning of the catalyst can be explained as a special case of modification, i. e.
poisoning is observed when the repulsive forces between
the modifying reagent and the substrate are stronger than
the affinity of the substrate for the catalyst.
Since modification of a catalyst often promotes the hydrogenation of one substrate and inhibits the hydrogenation
of another (see Tables 3 and 4), the term “modification”, is
more apt than “poisoning”.
37
25
-
50
52
[a] The figure given under “hydrogenation activity” is the percentage
of the calculated quantity of H, taken up after a certain time (after 30,
5, and 10 minutes respectively) in the cases of C=O, C=C, and NO,
groups.
[b] Only half of the calculated quantity of H, was taken up.
Angew. Chem. internat. Edit. 1 Vol. 10 (1971) N o . I 2
5. Preparation of Modified Raney Nickel
Catalysts and Determination of the
Enantioselectivity
For the investigations discussed here, we prepared Raney
nickel from an alloy containing 40% of Ni and 60% of Al.
An aqueous solution of the modifying reagent was adjusted
875
to a predetermined pH value and a predetermined temperature. The freshly prepared Raney nickel was then added,
and the mixture was allowed to stand at the same temperature for 1.5 h with occasional shaking. The modifying solution was decanted off, and the catalyst was washed once
with water and twice with methanol.
To determine the enantioselectivity of the catalyst, 17 g of
methyl acetoacetate was hydrogenated in the presence of
the modified Raney nickel catalyst obtained from 1.5 g of
alloy. The reaction was allowed to proceed in a shaking
autoclave at 60°C under an initial hydrogen pressure of
80- 100kg/cm2. After distillation under reduced pressure
(b. p. 61 -62 "C/12 torr), the optical rotation of the undiluted methyl 3-hydroxybutyrate was measured. The
result was taken as the specific rotation, since the specific
gravity of the product is approximately 1.0. According to
Leuene and
[aJ;' of methyl (R)-(-)-3-hydroxybutyrate is - 20.9 '.
The reliability of the results of the enantioselective synthesis improves with increasing purification of the product
and with increasing care in the measurement of the optical
rotation. Methyl acetoacetate was used as the substrate in
most cases, since its hydrogenation product has the following advantages. It can be distilled (with no change in the
optical purity) ; impurities can be easily detected by gas
chromatography; its optical rotation is high (see above),
and can be measured without dilution of the liquid.
6. Dependence of the Enantioselectivityof the
Catalyst on the Modification Conditions
The enantioselectivityof the catalyst is strongly dependent
on the temperature and the pH value of the modifying
solution. The optimum pH value is 5.0 in the case of acidic
modifying reagents and the isoelectric point in the case of
monoamino acids (Fig. 6) :
12.0
10.0
8.0
-
6.0
8
u
L.0
20
I i
0
/1BL171
I
,
20
LO
I
60
80
100
1("C)
Fig. 7. Enantioselectivity of modified Raney nickel as a function of the
temperature of the modifying solution. Substrate: methyl acetoacetate;
product : methyl 3-hydroxybutyrate. The right-hand ordinate is valid
for the following reagents:
-U-0- (+)-erythro-2-methyltartaric acid [27] (modified at pH = 5.0
to 5.2)
-0-0(ZS,3S)-tartaric acid [27] (modified at pH=5.0-5.2)
The left-hand ordinate is valid for the following reagents:
-A-A- (+)-2-methylglutamic acid [29] (modified at p H = 5.2)
-0-0(S)-valine [20] (modified at pH value of the isoelectric point)
-0-0- (S)-glutamic acid [23] (modified at pH = 5.2)
perature dependence is more complicated, however, when
the catalyst is modified with amino acids (Fig. 7).
On modification of the catalyst with amino acids such as
glutamic acid and leucine, the enantioselectivity changes
direction[511with rising temperature (cf. also Table 6).
7. Conditions for an Enantioselective
Modifying Reagent
To give useful enantioselectivity, the modifying reagent
must satisfy the following conditions:
1. The modifying reagent must be an optically active compound. Experiments with mixtures of (2R, 3R)- and (2S,
3 S)-tartaric acid showed that the enantioselectivity of the
catalyst varies linearly with the mixing ratio of the modifying
2
Ip)uII
+'
6
8
1
0
1
2
PH
Fig. 6. Enantioselectivity of Raney nickel modified with (S)-malic acid
and (S)-aspartic acid [36] as a function of the pH value of the modifying
solution. Substrate : methyl acetoacetate; product : methyl 3-hydroxybutyrate.
-0-0- (S)-malic acid
-0-0- (S)-aspartic acid
On modification with hydroxy acids, the enantioselectivity
of the catalyst increases with rising temperature. The tem876
2. If the molecule contains several centers of chirality, it is
desirable that their enantioselective effects do not cancel
one another out. Part 1 of Table 5 shows the enantioselectivities of some Raney nickel catalysts modified with reagents containing two centers of chirality. (For further exampies see [lo, 2 4 , 2 6 , 3 9 1.)
3. It is desirable that the center of chirality carry one
carboxyl group and one amino or hydroxyl group. Part 2
of Table 5 shows the enantioselectivity of Raney nickel
catalysts modified with a-substituted succinic acids. A
hydroxyl group or an amino group attached to the asymmetric carbon atom gives the catalyst a higher selectivity
than a methyl group. As can be seen in Part 3 of Table 5, the
amino acids make the catalyst more strongly enantioselective than the corresponding amino alcohols.
Angew. Chem. internat. Edit. 1 VoI. 10 (1971) No. 12
Table 5. Examples of enantioselective hydrogenations with Raney nickel modified with compounds of the type
R-CR' X-R" at 0°C and at the pH indicated. Substrate: methyl acetoacetate; product: methyl 3-hydroxybutvrate.
R
X
R"
R'
Abs. Conf.
CP
pH
c,
S
n-C,H,CH(CHJ
NH,
H
COOH
n-C,H,CH(CH,)
OH
H
COOH
iso-C,H,CH(CH,)
NH,
H
COOH
iso-C3H,CH(CH,)
OH
H
COOH
R
s
s
S
R
S
S
S
R
s
s
S
R
S
R
s
s
[u]D (")
la1
Ref
-1.95
-2.13
[38]
[38]
.2.9
6.12
6.10
2.9
2.9
5.77
5.97
2.9
2.9
+o.a
pq
-0.22
-1.81
-2.42
+0.67
+0.22
-2.18
-2.50
+0.17
-0.20
[38]
[381
[38]
[38]
[38]
[38]
[381
1381
[38]
[26]
[26]
HOOCCH,
HOOCCH,
HOOCCH,
NH2
OH
CH,
H
H
H
COOH
COOH
COOH
S
S
S
5.3
5.0
5.0
-1.9
+4.15
+0.20
HOOCCH,CH,
HOOCCH,CH,
HOOCCH,CH,
HOOCCH,CH,
HOOCCHOH
HOOCCHOH
HOOCCHOCOCH,
HOOCCHOH
H,C,OOCCHOH
NH2
NHCH,CH,CN
NHCOCsH,
N(CH3)2
OH
OCOC,H,
OCOCH,
OH
OH
H
H
H
H
H
H
H
H
H
COOH
COOH
COOH
COOH
COOH
COOH
COOH
COOH
COOC,H,
S
5.1
4.7
5.55
5.01
5.0
5.2
5.1
5.0
5.0
-3.51
-0.10
- 0.67
-0.75
- 6.0
-2.15
- 0.66
8.6 [c]
-0.55 [c]
HOOCCH,CH,
HOOCCH,CH,
HOOCCH,CH,
HOOCCH,
HOOCCH,
C2H5
NH,
NH,
NH,
NH2
NH2
NH,
NH2
H
CH3
C,H,CH,
H
CH3
H
CH,
COOH
COOH
COOH
COOH
COOH
COOH
COOH
5.1
6.1
-3.51
-1.92
+0.5
-1.9
f0.4
-1.46
-0.66
C2H5
S
S
S
R
R
R
R
R
R
R
R
s
s
S
Cdl
R
7.0
5.3
7.2
6.0
6.0
S
S
S
S
pi1
+
[iq
[22]
[231
pi1
pi]
pi1
[3l]
[a] Values measured at 25°C in Part 1 of the Table only; otherwise no data available
[b] See Section 11.
[c] Modification at 80°C.
[d] Dextrorotatory form.
Part 4 of Table 5 shows that substituents on the amino,
hydroxyl, or carboxyl group of the modifying reagent
reduce the enantioselectivity of the catalyst.
4. A hydrogen atom on the asymmetric carbon atom is
essential for the success of the modification. If this hydrogen atom is replaced by other residues, the enantioselectivity of the catalyst decreases (Table 5, Part 5).
8. Relation Between the Structure of the Modifying
m-Substituted Acids and the Enantioselectivity
of the Catalyst
8.1. Factors that Determine the Direction of the
Enantioselectivity
Table 6. Direction of the enantioselectivity in hydrogenations with
modified Raney nickel as a function of the absolute configuration on
the a-carbon atom of the modifying reagent (modification at about
pH = 5). Substrate: methyl acetoacetate; product : methyl 3-hydroxybutyrate.
Modifying
reagent
Arg, Asp, Ile, Lys,
Met, Phe [b], Trp,
Val or C,-alkylamino acids
Glu. Leu
Hydroxy acids [c]
The direction of the enantioselectivity of the catalyst is
determined by the absolute configuration of the center of
asymmetry C, of the modifying reagent, which has the
general structure RCHX-COOH. The substituent X also
Angew. Chem. internat. Edit. J Vol. I0
(1971)J No. 12
Abs. Conf.
Direction of
Enantioselectivity
Modification at
0°C
100°C
D
L
L
D
D
L
L
L
D
D
D
L
D
L
D
L
D
D
D
L
L
L
[a] See Section 11.
[b] The result published in reference [IS] will be reported shortly.
[c] That is, the hydroxy acids corresponding to the above-mentioned
amino acids.
877
influences the direction of the enantioselectivity. Table 5
contains examples in which the direction of the enantioselectivity is reversed when the amino group is replaced by
the hydroxyl group. With catalysts modified with L-hydroxy acids, methyl acetoacetate is hydrogenated to methyl
(S)-(+)-3-hydroxybutyrate ;this is not the case on modification with a number of P-substituted hydroxy acids. On the
other hand, catalysts modified with L-amino acids generally
lead to the (R)-(-)-ester (see Table 6).
treated with or-hydroxy acids substituted in this way exhibits stronger selectivity. From these observations, to-
8.2. Factors that Influence the Degree of
Enantioselectivity
The degree of enantioselectivity of the catalyst is determined by the structure of the substituent R. Our investigations on this effect of R lead to the following picture:
1. As described in Section 7 under paragraph 2, a second
center of chirality in the substituent has a strong influence
of the enantioselectivity of the catalyst. The absolute configuration on the P-carbon atom of an or-hydroxy acid is
often of greater importance to the enantioselectivity of the
catalyst than the absolute configuration on the or-carbon
atom.
2. As can be seen from Figure 8, with amino acids as the
modifying reagents, the enantioselectivity of the catalyst
increases with increasing chain length. On modification
with hydroxy acids, the enantioselectivity of the catalyst
decreases with increasing chain length. Figure 9 shows that
secondary or tertiary alkyl groups in the modifying reagent
influence the selectivity of the catalyst more strongly than
primary alkyl groups having the same number of carbon
atoms.
3. As is shown in Table I , hydroxyl, carboxyl, and thiol
groups on the P-carbon atom of the modifying reagent
strongly influence the enantioselectivity of the catalyst. If
the reagent is an amino acid, the substituents not only
reduce the selectivity but often reverse its direction, as can
be seen, e.g. in the examples of modification with serine,
threonine, and cysteine. On the other hand, a catalyst
H-C-H
R = H-C-H
H3C-F-H
HIC-C-CHl
CH,
CH,
I
m
H
CH,
Fig. 9. Enantioseiectivity of modified Raney nickel as a function of the
substituents on the modifying reagent [28]. Substrate: methyl acetoacetate; product: methyl 3-hydroxybutyrate.
-A-A- X=NH, (modified at p H = 6 (isoelectric point) and 0°C)
-0-0- X = OH (modified at pH = 5.0 and 0°C)
Table 7. Influence of the 0-substituent of the modifying reagent of the
type RR'CH-CHX-COOH
on the enantioselectivity of the Raney
nickel catalyst. Modification at 0°C and the indicated pH value. Substrate: methyl acetoacetate; product : methyl 3-hydroxybutyrate.
R
R
H
H
X
Abs. conf.
Ce
PH
S
H
H
H
H
H
H
H
H
CH3
CH3
CH3
H
CH3
COOH
COOH
SO,H
SO,H
OH
SH
CH3
CH3
OH
CH3
OH
COOH
OH
COOH
OH
s
S
s
s
R
R
S
R
S
S
s rbf
[cl
S [d]
[el
S [d]
[el
S [d]
6.0
5.0
6.05
5.3
5.0
5.2
5.2
6.6
7.8
6.0
2.9
6
6
2.9
2.9
5.5
5.2
5.0
[aI~(o)
+0.45 to
-0.1 [a]
+ 1.45
- 1.05
- 1.9
+4.15
- 0.2
t0.1
+ 0.2
+ 0.05
-2.31
+ 1.05
0 to +0.25
-0.05 to +0.6
+1.30
+ 0.50
+0.95
- 0.50
+ 6.0
Ref.
P I
[40al
I281
[281
1311
P I
P I
c361
1201
POI
~ 9 1
1211
P21
[22]
P61
1261
~ 4 b1
0
[40 b1
~ 7 1
[a] See Section 11.
[b] Threonine
[c] aflo-Threonine
[d] rhreo-Form.
re] erythro-Form.
Table 8. Enantioselectivity in hydrogenations with Raney nickel modified with dipeptide at 0°C [32]. Substrate: methyl acetoacetate; product : methyl 3-hydroxybutyrate.
Modification
I
I
,
t
I
Fig. 8. Enantioselectivity of modified Raney nickel as a function of the
substituents on the modifying reagent [28]. Substrate: methyl acetoacetate; product : methyl 3-hydroxybutyrate.
-A-A- X = N H 2 (modified at the pH value of the isoelectric point
and 0°C)
-0-0X = OH (modified at pH = 5.0 and 0°C)
878
[.Id0
Reagent
PH
(S)-Leucine
(S)-Leucyl-glycme
Glycyl-(S)-leucine
(S)-Leucyl-(S)-leucine
(S)-Leucyl-(R)-leucine
(S)-Aspartjc acid
Glycyl-(S)-asparticacid
7.32
5.72
5.70
5.85
5.80
4.65
-
-1.10
+ 1.12
- 0.41
- 0.42
+ 1.20
- 1.10
- 1.47
Angew. Chem. internar. Edii. J Voi. 10 (1971) / No. I 2
gether with results mentioned in Section 7 under paragraph
2, it is understandable that the strongest enantioselectivity
is obtained by modification with threo-tartaric acid [(2S,
3 S,l.
4. If the amino acid or a-hydroxy acid with which the catalyst is modified contains a sulfonic acid group, the enantioselectivity is decreased (Table 7).
pared modified catalyst was used for the hydrogenation of
methyl acetoacetate, recovered, and used again. The [.ID
values of the methyl 3-hydroxybutyrates obtained in experiments with fresh catalysts and with catalysts that had
been recovered once, twice, three times, four times, and
five times were -6.22, -6.30, -6.09, -6.01, -6.25, and
- 6.19” respectively. The modifying reagent was (2S,3 S)tartaric acid.
5. If dipeptides are used instead of amino acids for modification of the catalyst, the relation between structure and
enantioselectivity becomes more complicated. As can be
seen from Table 8, the direction of the enantioselectivity is
generally determined by the center of chirality of the Cterminal amino acid.
9. Durability of the Catalyst
The modified Raney nickel retains its enantioselectivity for
a very long time[27’.In a series of experiments, freshly pre-
10. Substrates for Hydrogenations with
Modified Raney Nickel
The best substrates for modified Raney nickel are p-0x0
esters and P-diketones. Monoketones are also enantioselectively hydrogenated to secondary alcohols, but the
optical yields are low, as can be seen from Table 9.
Table 9. Examples of enantioselective hydrogenations with modified Raney nickel. The italicized numbers in
brackets in the “substrate” column refer to the formulas reproduced above, or, in the case of numbers lower than
(23), to the formulas given’with Table 1 [30].
Cpd.
(23)
(23)
(14)
(24)
(25)
(26)
(10)
(10)
(12)
(27)
(28)
(29)
(30)
(31)
(32)
(33)
Substrate
Amount
(g)
Alloy
Amount (g)
14
10
18
15
5
7
10
10
20
22
14
12
18
5
13
10
1.5
1.5
1.5
1.5
0.5
1.5
1.5
1.5
1.5
1.5
1.o
I .o
1.5
1.5
1.5
1.o
Catalyst
Modification
Reagent [a]
pH
T (“C)
W
G
G
W
W
G
W
W
W
W
W
G
W
W
G
W
5.10
5.05
5.16
50
5.0
5.1
5.2
5.0
5.1
5.1
5.0
5.12
5.0
5.0
5.0
5.2
100
0
0
0
0
0
100
100
0
0
0
0
0
0
0
0
[uID
(“)
+1.0
-0.11
t0.93
-0.06
+0.26
0
0
0
-2.97
-2.28
-0.76
t1.90
-14.83
-1.15
0
-0.40
Product
Abs.
Conf.
s
R
R
R
S(L)
-
R(D)
R(D)
R(D)
R(D)
-
R(D)
Opt.
Yield [b]
1.7
08
2.1
0.5
2.5
0
0
0
18
16
5
8.3
-~
0
3.1
[a] W =(2R, 3R)-tartaric acid, G =(S)-glutamic acid
[b] See footnote [a] in Table 1.
Angew. Chem. internat Edit. 1 Vol. I0 (1971) / N o . 12
879
11. Dependence of Enantioselectivityof Catalyst
on Hydrogenation Conditions
The hydrogenation temperature has no influence on the
enantioselectivity of the catalyst above 60°C. Below 60°C,
however, the catalysts exhibit practically no hydrogenation
activity and practically no enantioselectivity.
-301
\
Fig. 10. Effect of water on the enantioselectivity of a modified Raney
nickel catalyst. The quantities ofwater are based on 17 g of the substrate
methyl acetoacetate [4Oa].
-0-0(S)-glutamic acid (modified at pH = 5 and 0°C)
-A-A- (S)-valine
-0-0(S)-alanine (both modified at the pH of the isoelectric point
and 0°C)
Any trace of impurities, particularly carboxylic acids,
either in the substrate or in the solvent, affects the enantioselectivity of the catalyst[40! This effect is more pronounced
with amino acids as the modifying reagents than with
hydroxy acids. In the former case, as can be seen from
Figure 10, the direction of the selectivity may even be
reversed in the presence of water.
12. Hypothesis on the Course of the
Enantioselective Hydrogenation
As was mentioned in Section 2.2, the enantioselectivehydrogenation is made possible by a preceding enantioselective
adsorption of the substrate on the catalyst surface. To explain the enantioselectiveadsorption, the adsorption process is broken down into three steps, as shown in Figure 11.
From the fact that P-diketones and $-ox0 esters give the
best results in enantioselective hydrogenation with modified Raney nickel, it can be concluded that adsorption of
the carbonyl groups on the catalyst surface to form a
chelate structure plays an important part in the hydrogenation. Thus the Raney nickel catalyst itself has the ability to
adsorb carbonyl compounds selectively in the direction of
the C=O axis.
The modification of the catalyst with an optically active
compound decides whether the catalyst attacks the substrate from the re side or the si side[31.Since the z axis of the
substrate is fixed by the catalyst surface, the modification
of the catalyst must produce the selectivity for the x and y
axes of the substrate. There are two possible explanations
for the production of the enantioselectivity on modification: 1.As proposed by B e a w ~ e r [there
~ ~ ] , are two types of
880
Fig. 11. Enantioselectiveadsorptionofaketoneonthesurfaceofamoditied Raney nickel catalyst (schematic). The z axis is perpendicular to the
surface of the catalyst.
sites on the untreated catalyst, one of which produces the
S-form and the other the R-form, and one type is blocked
or poisoned as a result of the modification, 2. The enantioselectivesite is formed by the combined effect of the catalyst
surface and the modifying reagent.
Experimental results support the second possibility. If the
enantioselectivity of the catalyst were due to the first process, a double modification, first with (S)- and then with (R)or first with (R)- and then with (S)-glutamic acid, should
deactivate both types of enantioselective sites on the catalyst. However, the enantioselectivityof the catalyst depends
only on the last modifying reagent
It is clear from the facts described above that the selectivity
is produced by the modifying reagent adsorbed on the catalyst. The adsorbed reagent is in adsorption equilibrium
with its solution.
In the present review, it has been possible to describe only
the most important results obtained with Raney nickel
catalysts, mainly modified at pH= 5 and 0°C. For further
results and detailed discussions, the reader is referred to
the original literature.
We are grateful to Dr. S. Akabori for his valuable suggestions,
and we would also like to thank Professor M . Nakazaki of the
Faculty of Engineering Science, Osaka University, for discussions on the new classification of “asymmetric synthesis”.
We also thank our many colleagues for their constant efforts
and their assistance.
Received: July 28,1970 [A 847 IE]
German version : Angew. Chem. 83,956 (1971)
[1] M . Nakazaki: Shape and symmetry of organic moiecules (Bunshi
no Katachi to Taisho). Nankodo, Tokyo 1965.
[Z] R. S. Cahn, C. K . Ingold, and K Prelog, Angew. Chem. 78, 413
(1966); Angew. Chem. internat. Edit. 5, 385 (1966).
Angew. Chem. internat. Edit.
Vol. I0 (1971) / N o . 12
[3] The reader is referred here to the nomenclature of the “re”- and “si”attack by a) K. R. Hanson, J. Amer. Chem. SOC.88, 2731 (1965); cf. also
b) H. J . Bestmann and I . Tiimijskozi, Tetrahedron 24, 3299 (1968).
141 7: Zkenaka and K. Matsumoto, J. Chem. SOC.Japan, Pure Chem.
Sec. (Nippon Kagaku Zasshi) 73,112 (1952).
[5] A. McKenzie, J. Chem. SOC.85, 1249 (1904).
[6] For the influence of variable ligands on x 1 and x2 on the ratio of
the diastereoisomers formed, cf. the stereochemical structural model :
E. Ruch and I. Ugi, Topics in Stereochem. 4, 99 (1969); also further
references therein.
[7] H. S. Mosher and E. La Combe, J.Amer. Chem. SOC.72,3994(1950);
cf. also [3b].
[8] S. Akabori, Y. Izumi, Y. Fujii, and S. Sakurai, Nature 178,323 (1956).
1281 Y. Izumi, 7: Tanabe, S. Yajima, and M . Imaida, Bull. Chem. SOC.
Japan 41,941 (1968).
[29] Y. Izumi, K . Matsunaga, S. Tatsumi, and M . Imaida, Bull. Chem.
SOC.Japan 41,2515 (1968).
[30] Y. Izumi, M . Imaida, 7: Harada, 7: Tanabe, S. Yajima, and 7:Ninomiya, Bull. Chem. SOC.Japan 42, 241 (1969).
[31] Y. Zzumi, S . Tatsumi, and M . Imaida, Bull. Chem. SOC.Japan 42,
2373 (1969).
[32] Y. Izumi, S. Tatsumi, M . Imaida, and K. Okubo, Bull. Chem. SOC.
Japan 43,566 (1970).
[33] Y. Izumi and 7: Ninomiya, Bull. Chem. SOC.Japan 43,579 (1970).
[34] I: Izumi, H. Takizawa, K . Nakagawa, R. Imamura, M . Imaida, 7:
Ninomiya, and S . Yajima, Bull. Chem. SOC.Japan 43, 1792 (1970).
[9] S. Akabori, Y. Izumi, I: Fujii, and S. Sakurai, J. Chem. SOC.Japan,
Pure Chem. Sec. (Nippon Kagaku Zasshi) 77, 1374 (1956).
[lo] S. Akabori, Y. Izumi, and Y. Fujii, J. Chem. SOC.Japan, Pure Chem.
Sec. (Nippon Kagaku Zasshi) 78,886 (1957).
[ll] 7: isoda, A . fchikawa, and 7: Shimamoto, J. Sci. Res. Inst. (Riken
hokoku) 34, 134 (1958).
[35] 7: Tanabe, 7: Ninomiya, and Y Izumi, Bull. Chem. SOC.Japan 43,
2276 (1970).
1361 Y. Izumi, S. Yajima, K . Okubo, and K. K. Babieusky, Bull. Chem.
SOC.Japan 44,1416 (1971).
[37] Y. Izumi, 7: Harada, 7: Tanabe, and K. Okuda, Bull. Chem. SOC.
Japan 44, 1418 (1971).
[I21 7: isoda, A. lchikawa, and 7: Shimamoto, J. Sci. Res. Inst. (Riken
hokoku) 34, 143 (1958).
[38] Y. Izumi and K . Ukubo, Bull. Chem. SOC.Japan 44, 1330 (1971).
(391 7: Haruda, M . Imaida, and Y. Izumi, Bull. Chem. SOC.Japan 44,
1419 (1971).
[4Oa] F. Higashi, 7: Ninomiya, and Y. Izumi, Bull. Chem. SOC.Japan
44, 1333 (1971).
[40b] 7: Harada and Y. Izumi, Bull. Chem. SOC.Japan, in press.
[41] R. E. Padgett jr. and R. L. Beamer, J. Pharm. Sci. 53,689 (1964).
[42] R. L. Beamer, C.S. Fickfing, and J . H. Ewing, J. Pharm. Sci. 56,
1029 (1967).
[43] R. L. Beamer, R. H. Betding, and C.S. Fickling, J. Pharm. Sci. 58,
1142 (1969).
[44] R. L. Beamer, R. H. Belding, and C.S. Fickling, J. Pharm. Sci. 58,
1419 (1969).
[45] Yu. I . Petrou, E . I . Klabnouskii, and A. A. Balandin, Kinetika i
Kataliz 8, 814 (1967).
1461 N . A. Plate, C. L. Dauidoua,E. D.Alieua, and V B. Kargin, Vysokomol. Soedin. B 11,495 (1969).
[47] A. A. Balandin, Advan. Catalysis 10,96 (1958).
[48] E . 1. Ktabnouskii and Yu. I . Petrou, Dokl. Akad. Nauk SSSR 173,
1125 (1967).
[49] H. Fukawa, I: Izumi, S. Komatsu, and S. Akabori, Bull. Chem. SOC.
Japan 35,1703 (1962).
[SO] P. A. Leuene and H. L. Halter, J. Biol. Chem. 65, 51 (1925).
1511 Note on the concept of enantioselectivity. In the discussion of
enantioselectivity, two factors must be considered: (i) the numerical
value of the optical rotation of the enantioselectively hydrogenated
product and (ii) the sign + or - associated with the absolute configuration of the product. Whereas the first factor is important in the discussion of the optical yield, the second factor indicates the “direction of the
enantioselectivity”.
1131 Y. Izumi, Bull. Chem. SOC.Japan 32,932 (1959).
[14] Y. izumi, Bull. Chem. SOC.Japan 32,936 (1959).
[IS] Y. Izumi, Bull. Chem. SOC.Japan 32,942 (1959).
[I61 Y. Izumi, A. Akamatsu, and S. Akabori, Bull. Chem. SOC.Japan 34,
1067 (1961).
1171 Y. Zzumi, A. Akamatsu, and S. Akabori, Bull. Chem. SOC.Japan 34,
1302 (1961).
[is]
A. Akamatsu, I! Izumi, and S. Akabori, Bull. Chem. SOC.Japan 35,
1706 (1962).
1191 Y. Izumi, M . Imaida, H. Fukawa, and S.Akabori, Bull. Chem. SOC.
Japan 36.21 (1963).
[20] Y. Izumi, M . Imaida, H. Fukawa, and S. Akabori, Bull. Chem. SOC.
Japan 36, 155 (1963).
[21] S. Tatsumi, M . Imaida, Y. Fukuda, Y. Izumi, and S . Akabori, Bull.
Chem. SOC.Japan 37,846 (1964).
[22] Y. Izumi, S. Akabori, H. Fukawa, S. Tatsumi, M . Zmaida, Y. Fukuda,
and S. K O ~ ~ I Proc.
S O , 3 rd Int. Congr. on Catalysis, Amsterdam, Vol. 2,
1364 (1964).
[23] Y. Izumi, S. Tatsumi, M . Imaida, Y. Fukuda, and S. Akabori, Bull.
Chem. SOC.Japan 38,1206 (1965).
[24] Y. Izumi, S. Tatsumi, M . Imaida, Y. Fukuda, and S. Akabori, Bull.
Chem. SOC.Japan 39,361 (1966).
[25] Y. Izumi, S. Tatsumi, and M . Imaida, Bull. Chem. SOC.Japan 39,
1087 (1966).
11261 Y. Izumi, S. Tatsumi, and M . Imaida, Bull. Chem. SOC.Japan 39,
2223 (1966).
[27] S. Tatsumi, Bull. Chem. SOC.Japan 41,408 (1968).
Angew. Chem. internat. Edit. J Vol. 10 (1971)
1 No. 12
881
Документ
Категория
Без категории
Просмотров
0
Размер файла
870 Кб
Теги
asymmetric, synthesisчenantioselective, catalytic, method, hydrogenation
1/--страниц
Пожаловаться на содержимое документа