- Email: [email protected]
The asymmetric dihydroxylation of cis-allylic and homoallylic alcohols
Tetmhx&on Leuers, Vol. 35. No. 6. pp. 843-846, 1994 msevk Scim Ltd RhtdiIlGrruBrit8ill oCI40439M 36.00400
0040-4039(93)E0355-N
The Asymmetric Dihydroxylation of cis-Allylic and Homoallylic Alcohols Michael S. VanNieuwenhze and K. Barry Sharpless Department OfChemtWy Scripps Research Institute 10666 North Torrey Pines Road Lalolla, CA 92037
Abstract: The asymmetric dihydroxylation of several ck allylic alcohols is reported. The reactions proceed with moderate to good enantioselectivity. The enhanced enantioselection observed relatitie to other cis-olefin classes is believed to be dependent on the presence of the free hydroxyl group. A cis-hommllylic alcohol also gave a useful level of enantiose1ectionin the AD. The osmium tetroxide-cinchona
alkaloid catalyst system has proven to be very
efficient for the asymmetric dihydroxylation The cis-disubstituted
and tetrasubstituted
classes to dihydroxylate
(AD) of most classes of prochiral o1efins.I classes have been the most difficult olefin
with high enantioselectivity.
Recently, it was discovered that
some tetrasubstituted olefins afford high levels of enantioselectlvity ln the AD reaction.* In our continuing studies on the scope of the AD process we have found that cis-allylic and homoallylic alcohols are also useful substrates.
Dihydrcquinidine WQW
Dihydmqtinine (DHQ)
Several cis-allylic alcohols having different steric and electronic environments were studied.
The experiments were carried out using the commercially available AD
mixes which utilize the (DHQD)z-PHAL R = DHQD) and (DHQ)2-PHAL
11,4-bis-9-O-dihydroquinidinephthakinel
[1,4-bis-9-0-dihydroquininephthalazine]
DHQ) ligands.
In a few cases, the results were compared
dihydroxylation
experiments
dihydroquinidinel
to those
(la,
(lb,R = obtained
in
employing the DHQD-IND 1(9-O-indolinylcarbamoyl)-
(2, R = DHQD) ligand which was previously shown to be the ligand
of choice for the catalytic asymmetric
dihydroxylation
843
of cis-olefln substrates.3
The
844
reactions were performed at Ooc in the t-&OH-Hfl
solvent system (I : 1) at a substrate
concentration of 0.1 M. The results of these experiments are summarized in Table 1. Fair
to good
enantioselectivities
were
achieved
in most
cases
with
the
phthalazine l&and system. Previous investigations in this laboratory had shown this Table 1. Enantiomerk excesses obtained in the AD of cis-allyfic alcoholswith various chirai ligands.
exceqes: a) Chinal HPLG analyds of dorIved ViMethods for deismJnirg wwtMwdc acetate (ChiWP OE.2.5% IPA in hexanes, 0.5 mUmin). b) Chid HPLC adysb of derived lrkaba {Chid& OJ, 10% IPA In hesane*, 1.0 rnL/mtn). c) ‘H HMR w~~Iyaia (4.00 MHz, CDCy of ~~(R)-M~A astar. d) Chid HPIS of ~~~~~ (Chidcei*OD, 1% IPA in hexanes, 0.5 mUmin). e) ‘H NMR arwiysis (400 MHz, C,D,j of Irk (I+MTPA ester.
ligand to be very poor for simple hydrocarbon cis-olefins, especially those bearing alkyl groups at each terminus.~ We believe that hydrogen bonding by the free hydroxyl to an 0x0 group on osmium may be responsible for the enhanced enantioselection
seen with
this special class of cis-olefins. Investigations are underway to identify the origin of this phenomenon. hypothesis.
The final entry in Table 1 also appears to support the hydrogen bondingThe enantioselectivity
(DHQD)z-PHAL]
displayed by this substrate is modest [54% ee with
yet surprising considering that the environment near the olefin unit is
very nearly symmetric. 10 The catalyst is able to differentiate
the olefm &mini
though the hydroxyl group is somewhat remote from the reaction center.
even
845
Further support for the beneficial effect of the terminal hydroxyl group was obtained when AD experiments were carried out on the corresponding methyl ethers of selected sub&rat&. The data obtained from these experiments are summa&ed
in Table
2. *When compared to the data for the corresponding alcohols in Table 1, a dramatic drop in’ena&oseiectivity
is e*t
in eacihcase.
Table 2 AD data of methyl &her derivatives and cis+-methylstyrene with (DHQD)zPHAL % Enantiomeric Excess
Substrate
I
-woBn
Major Product
I
23
(ZS,3R)”
--I tdemalakrdstermlningenontkmricex-: a) Chiml HPLC of datfved trls-mathyfether (Chlralodo OB 2.59b IPA in hrmcnes, 0.5 mlfmln). b) Chiral HPLC of derived daetato (Chbb& OJ. 10% IPA in hexanos. 1.O mL&nin). c)‘% NMR (400 MHz, C&t) of bisMTPA ester d) Data taken from reference 3. An
attractive feature of this methodology is that the -de
be crystalline
trio1products tend to
and thus present the opportunity for enantiomeric enrichment.
example, the enantiomeric purity
For
of the trio1 obtained from the AD of cis-2-hexen-l-01
increased from 74% to 96% after a single reaystallization.~~ In summary, we have demonstrated that &-ally&
alcohols hold considerable
promise as substrates in the AD reaction. Current efforts are underway to determine the role of the free hydroxyl group and to examine the possibility that other properly placed
acidic
functional
groups
may
also
have
beneficial
effects
on
the
enantioselectivity. Acknowledgment
. Financial support was provided by the National Institutes of
Health (GM 28384). Support for MSV was provided by the National Institutes of Health (GM 15299-01) in the form of a postdoctoral fellowship.
846
References 1.
2. 3. 4.
5. 6. 7. 8.
a) Jacobsen, E.N.; Ma&o, I.; Mungall, W. M.; Schroder, G.; Sharpless, K.B. J. Am. Chem. Sot. 19$8,120,1968. b) Ogino, Y.; Chen, H.; Kwong, H. -L.; Sharpless, K. 8. Tetrahedron Lett. 1991,32,3965. c) Sharpless, K. B.; Amberg, W.; BeRer, M.; Chen, H.; Hartung, Y.; Kawanami, Y.; Ltibben, D.; Manoury, E.; Ogino, Y,; Shibata, T.; Ukita, 1. Org. ciretn. 1991,56,4585. d) Sharpless, K. B.; Amberg, W.; Bennani, Y. L.; Crispino, G. A.; Hartung, J. ; Jeong, K. -S.; Kwong, H. -L.; Morikawa, K.; Wang, Z. -M.; Xu, D.; Zhang, X. -L. 1. Org. Cheer.. 1992,57,2769. Morikawa, K.; Park, J.; Andersson, P. G.; Hashiyama, T.; Sharpless, K. B. J. Am. lzfletn. sot. 1993,115,8463. Wang, L.; Sharpless, K. B. 1. Am. Chem. Sot. 1992,114,7568. The absolute configuration was proven by comparing the optical rotation of the triacetate with that of an authentic sample obtained via the following sequence :
The absolute configuration was established by comparison to commercially available L-(-)-eryfhro-l-C-phenylglycerol. The absolute stereostructure was established by comparison to an authentic sample prepared via an asymmetric epoxidation/Payne rearrangement sequence starting from commercially available frans-2-hexen-l-01. The absolute configuration was assigned by application of the AD mnemonic. The absolute configuration was established by comparing optical rotations of a sample prepared via benzoylation of the AD product obtained from cis-3-hexen-l01 using (DHQDh-PHAL with that of an authentic sample prepared via the following sequence: Ph
1) AD-mix-~. MeSO#4H, t&OH : I+& 0% 2) PhCHO. pTsOH C&la rdlux
9. 10. 11. 12. 13.
-b
OM
l)DEAD.Ph#,THF PhCO$i 2) pTsOH, MeOH 3) BzCI. TEA, CH&
c OBZ
The best results with this ligand are obtained when one of the olefin substituents is a phenyl group. See reference 3. Replacement of the terminal -0I-l group by a hydrogen would give a nonprochiral olefin and, therefore, a meso dial in the AD reaction. The absolute stereostructure was assigned by comparison of the f&-methyl ether with an authentic sample made via the sequence cited in reference 4. The absolute configuration was assigned by application of the AD mnemonic. The AD reaction was run on a 2.0 g scale and proceeded in 80% overall yield. The crude trio1 product was recrystallized from ethyl acetatez hexanes (1 : 11. [a]~$= +20.9o (c = 1.9, EtOH); mp 80-82oC.
(Received in USA 1 November 1993; accepted 29 November 1993)
0040-4039(93)E0355-N
The Asymmetric Dihydroxylation of cis-Allylic and Homoallylic Alcohols Michael S. VanNieuwenhze and K. Barry Sharpless Department OfChemtWy Scripps Research Institute 10666 North Torrey Pines Road Lalolla, CA 92037
Abstract: The asymmetric dihydroxylation of several ck allylic alcohols is reported. The reactions proceed with moderate to good enantioselectivity. The enhanced enantioselection observed relatitie to other cis-olefin classes is believed to be dependent on the presence of the free hydroxyl group. A cis-hommllylic alcohol also gave a useful level of enantiose1ectionin the AD. The osmium tetroxide-cinchona
alkaloid catalyst system has proven to be very
efficient for the asymmetric dihydroxylation The cis-disubstituted
and tetrasubstituted
classes to dihydroxylate
(AD) of most classes of prochiral o1efins.I classes have been the most difficult olefin
with high enantioselectivity.
Recently, it was discovered that
some tetrasubstituted olefins afford high levels of enantioselectlvity ln the AD reaction.* In our continuing studies on the scope of the AD process we have found that cis-allylic and homoallylic alcohols are also useful substrates.
Dihydrcquinidine WQW
Dihydmqtinine (DHQ)
Several cis-allylic alcohols having different steric and electronic environments were studied.
The experiments were carried out using the commercially available AD
mixes which utilize the (DHQD)z-PHAL R = DHQD) and (DHQ)2-PHAL
11,4-bis-9-O-dihydroquinidinephthakinel
[1,4-bis-9-0-dihydroquininephthalazine]
DHQ) ligands.
In a few cases, the results were compared
dihydroxylation
experiments
dihydroquinidinel
to those
(la,
(lb,R = obtained
in
employing the DHQD-IND 1(9-O-indolinylcarbamoyl)-
(2, R = DHQD) ligand which was previously shown to be the ligand
of choice for the catalytic asymmetric
dihydroxylation
843
of cis-olefln substrates.3
The
844
reactions were performed at Ooc in the t-&OH-Hfl
solvent system (I : 1) at a substrate
concentration of 0.1 M. The results of these experiments are summarized in Table 1. Fair
to good
enantioselectivities
were
achieved
in most
cases
with
the
phthalazine l&and system. Previous investigations in this laboratory had shown this Table 1. Enantiomerk excesses obtained in the AD of cis-allyfic alcoholswith various chirai ligands.
exceqes: a) Chinal HPLG analyds of dorIved ViMethods for deismJnirg wwtMwdc acetate (ChiWP OE.2.5% IPA in hexanes, 0.5 mUmin). b) Chid HPLC adysb of derived lrkaba {Chid& OJ, 10% IPA In hesane*, 1.0 rnL/mtn). c) ‘H HMR w~~Iyaia (4.00 MHz, CDCy of ~~(R)-M~A astar. d) Chid HPIS of ~~~~~ (Chidcei*OD, 1% IPA in hexanes, 0.5 mUmin). e) ‘H NMR arwiysis (400 MHz, C,D,j of Irk (I+MTPA ester.
ligand to be very poor for simple hydrocarbon cis-olefins, especially those bearing alkyl groups at each terminus.~ We believe that hydrogen bonding by the free hydroxyl to an 0x0 group on osmium may be responsible for the enhanced enantioselection
seen with
this special class of cis-olefins. Investigations are underway to identify the origin of this phenomenon. hypothesis.
The final entry in Table 1 also appears to support the hydrogen bondingThe enantioselectivity
(DHQD)z-PHAL]
displayed by this substrate is modest [54% ee with
yet surprising considering that the environment near the olefin unit is
very nearly symmetric. 10 The catalyst is able to differentiate
the olefm &mini
though the hydroxyl group is somewhat remote from the reaction center.
even
845
Further support for the beneficial effect of the terminal hydroxyl group was obtained when AD experiments were carried out on the corresponding methyl ethers of selected sub&rat&. The data obtained from these experiments are summa&ed
in Table
2. *When compared to the data for the corresponding alcohols in Table 1, a dramatic drop in’ena&oseiectivity
is e*t
in eacihcase.
Table 2 AD data of methyl &her derivatives and cis+-methylstyrene with (DHQD)zPHAL % Enantiomeric Excess
Substrate
I
-woBn
Major Product
I
23
(ZS,3R)”
--I tdemalakrdstermlningenontkmricex-: a) Chiml HPLC of datfved trls-mathyfether (Chlralodo OB 2.59b IPA in hrmcnes, 0.5 mlfmln). b) Chiral HPLC of derived daetato (Chbb& OJ. 10% IPA in hexanos. 1.O mL&nin). c)‘% NMR (400 MHz, C&t) of bisMTPA ester d) Data taken from reference 3. An
attractive feature of this methodology is that the -de
be crystalline
trio1products tend to
and thus present the opportunity for enantiomeric enrichment.
example, the enantiomeric purity
For
of the trio1 obtained from the AD of cis-2-hexen-l-01
increased from 74% to 96% after a single reaystallization.~~ In summary, we have demonstrated that &-ally&
alcohols hold considerable
promise as substrates in the AD reaction. Current efforts are underway to determine the role of the free hydroxyl group and to examine the possibility that other properly placed
acidic
functional
groups
may
also
have
beneficial
effects
on
the
enantioselectivity. Acknowledgment
. Financial support was provided by the National Institutes of
Health (GM 28384). Support for MSV was provided by the National Institutes of Health (GM 15299-01) in the form of a postdoctoral fellowship.
846
References 1.
2. 3. 4.
5. 6. 7. 8.
a) Jacobsen, E.N.; Ma&o, I.; Mungall, W. M.; Schroder, G.; Sharpless, K.B. J. Am. Chem. Sot. 19$8,120,1968. b) Ogino, Y.; Chen, H.; Kwong, H. -L.; Sharpless, K. 8. Tetrahedron Lett. 1991,32,3965. c) Sharpless, K. B.; Amberg, W.; BeRer, M.; Chen, H.; Hartung, Y.; Kawanami, Y.; Ltibben, D.; Manoury, E.; Ogino, Y,; Shibata, T.; Ukita, 1. Org. ciretn. 1991,56,4585. d) Sharpless, K. B.; Amberg, W.; Bennani, Y. L.; Crispino, G. A.; Hartung, J. ; Jeong, K. -S.; Kwong, H. -L.; Morikawa, K.; Wang, Z. -M.; Xu, D.; Zhang, X. -L. 1. Org. Cheer.. 1992,57,2769. Morikawa, K.; Park, J.; Andersson, P. G.; Hashiyama, T.; Sharpless, K. B. J. Am. lzfletn. sot. 1993,115,8463. Wang, L.; Sharpless, K. B. 1. Am. Chem. Sot. 1992,114,7568. The absolute configuration was proven by comparing the optical rotation of the triacetate with that of an authentic sample obtained via the following sequence :
The absolute configuration was established by comparison to commercially available L-(-)-eryfhro-l-C-phenylglycerol. The absolute stereostructure was established by comparison to an authentic sample prepared via an asymmetric epoxidation/Payne rearrangement sequence starting from commercially available frans-2-hexen-l-01. The absolute configuration was assigned by application of the AD mnemonic. The absolute configuration was established by comparing optical rotations of a sample prepared via benzoylation of the AD product obtained from cis-3-hexen-l01 using (DHQDh-PHAL with that of an authentic sample prepared via the following sequence: Ph
1) AD-mix-~. MeSO#4H, t&OH : I+& 0% 2) PhCHO. pTsOH C&la rdlux
9. 10. 11. 12. 13.
-b
OM
l)DEAD.Ph#,THF PhCO$i 2) pTsOH, MeOH 3) BzCI. TEA, CH&
c OBZ
The best results with this ligand are obtained when one of the olefin substituents is a phenyl group. See reference 3. Replacement of the terminal -0I-l group by a hydrogen would give a nonprochiral olefin and, therefore, a meso dial in the AD reaction. The absolute stereostructure was assigned by comparison of the f&-methyl ether with an authentic sample made via the sequence cited in reference 4. The absolute configuration was assigned by application of the AD mnemonic. The AD reaction was run on a 2.0 g scale and proceeded in 80% overall yield. The crude trio1 product was recrystallized from ethyl acetatez hexanes (1 : 11. [a]~$= +20.9o (c = 1.9, EtOH); mp 80-82oC.
(Received in USA 1 November 1993; accepted 29 November 1993)