- Email: [email protected]
Catalyst dependent mechanistic paths in the reactions of ethyl diazoacetate with β-keto esters
Tetrahedron Letters,Vol.30,No.31,pp Printed in Great Britain
4093-4096,1989 0040-4039/89 $3.00 + .OO Maxwell Perqamon Macmillan plc
Andrew C. L.ottes, John A. Landgrebe*,and Kristin Larsen Departmentof Chemistry.Universityof Kansas Lawrence,Kansas 66045 ition of ethyl diazoacetatiin the Presence of B-keto esters Summy: The catalyzeddecanpos producesregioisanericenol ethers and the diester fran formal carbenoid insertioninto the a-C-H bond. The productdistributionis very catalystdependent. We have reportedpreviouslythat the CuCl or Rh,(CAc),catalyzed-sition
of ethyl
diazoacetatein the Presence of various alkyl or phenyl substitutedacyclic and cyclic kstones leads to en01 ethers with understandableregiochemistry(and where applicable,stereochemistry)as the only 1:l products,and that the product distributionwas independentof We now report observationsthat for various f_+ketoesters, not only is the
cata1yst.l
regiochemistry of en01 ether form&ion catalystdependent,but in the presenceof CuCl, significantamountsof a new 1:l productare observed.An example is shown below for ethyl acetoacetate (1) with a CuCl catalyst, and caq%ar& to the results previouslyreported under identicalconditionsfor methyl benzyl ketone (41.'
N&g?
&C02Et
CO&
c q
2@+61%)
1
+
/1\
3(394496) m2J9
(=,a
Ph
ic
N&!HCO.$t c&l
+
4 Ph
Ph
with exclusive (E)stereochani stry (confirmedby suitableNOE experiments)consistentwith the mechanisticmodel developedpreviousIn both cases, enoletbers 3 and 6 are fe
1y.l Snewhat surprisingis the observationthat 8-keto ester 1 produces none of the lesssubstitutedenol ether, but insteadforms diester 2. The results for various cyclic B-keto esters are swmarized in Table I.
Whatseelns
clear is that the use of Rh,(CAc),produces far less of the diester-typeproduct than when of en01 ether formation favors the lessCuCl is used. In addition, the regioselactivity substitutedproduct when CuCl is used, and with the exceptionof the 7?nanberedring, the more substitutedproduct when Eh,(oAc), is errrployed.Although the more substitutedenol ether is mre thermodynamically stable, suitablecontrol experimentson the less-substituted
4093
4094
en01 ether from 2-carbethoxycyclopentanone have clearly shown that no equilibrationof the regioisomersoccurs under the reaction conditions in the presence of either catalyst.2 Isolatedyields of the canbined1:l adductswere in the range 40-55%.3 5%ble I. Product
Distributionforcyclic B+?toEstersa
&?J:+
Ef
&;;&Z
n
CuCl Rh,(OAc),
CuCl Rh2(oAc),
CuCl Rh,mAc),
5
12-13
2
36-38
57
49-51
6
42-43
4-5
18-19
83-85
38-39
11-13
7
64-65
10-11
5-6
13-15
31-33
74-76
41
a Expressedas a percentof all I:1 adducts;resultsrepresentat least three runs; a 7-fold excess of ketone served as the solvent. In contrast,note that for the catalyzedreaction of ethyl diazoacetatewith amethyl and 2-phenylcycloalkanones sunmr ized in Table II, no diester has ever been observed,the resultsare identicalfor both CuCl and Rh,(C&c),.and the less-substituted en01 ether always dcminates.l
Bble II. EMi Ether Ring Size
Substituent
Distributionfor 2-myland Mor*SubstitutedProd.
2-Phehylcycloalkammesa Less-Substituted Prod.
Me
5
12.5-16.7
83.3-87.5
Me Ph
6
7.7- 7.1
92.3-92.9
5
43.5
56.5
Ph
6
40
60
a Expressedas a percentof all 1:l adducts;for liquid ketones,a 7-fold excess sexed as solvent:for solid ketones,a 6-fold excess was dissolvedin Ccl,; resultsare for at least three runs per ketone and are identicalfor CuCl and Rh,(oAc),.
4095
While the reasons why substantialamounts of diester form from the a-ethowoarbonyl ketones in Table I in the presence of CuCl but not Rh,(OAc), reamin speculative.related differences in the chanioal behavior of carbanoids derived fran diazo esters have been noted.4 It is of special interestthat because our observationsare contrary to the known effectivenessof rhadium(II)carboxylatesin pronmtingC-H insertion,4-7the latter process is probablynot involved. In fact, we suspect that diester 9 arises frcancyolopropanol8 producedfranthe additionof thecopper carbenoidto the enolof 7.
7
8
9
such a path is made plausibleby our observationthat silyl cy~lopropylether 11 prepared fran 10 by the method of Reissig and Hirsch,' was convertedto diester 12 in an isolated yield of 71% after heating for 24 h at 60 'C in ethanol.' Regioselectiveopening of the intermediate oyclopropan -01 under electrophilic conditionshas scme precedent,10-14
11
12
Althoughwe have no unified explanationfor the interestingdifferencesin regioselectivity of en01 ether formationas a functionof catalystfor the various a-ethoxycarbonyl ketones in Table I, it s8ems inescapablethat the catalyst must be present during the 1,4sigmatropicshift of hydrcgen. These interestingmechanisticquestionsrEznainunder active investigation.
Acknouledgaawt: Supported by the Universityof Kansas General Research Fund; a Phillips PetroleumCcinpany Fellowshipfor A.C.L.; an NSF REU Fellowshipfor K.L. Ref-andNotes
this issue. 1. L&ties, A. C.; Landgrebe.J. A.; Larsen,K. Previousccmnunication, 2. Equilibrationdoes ozcu.rin benzene with AlCl, and leads to approxinW.ely80-85% of the
4096
IKmFswtiruuted 3.
en01 ethez at 78 OC.
Structuresof newly prepared canpoundswere identifiedby a combinationof lH and 13C NMR, IX, and MS (includingan exact mass).
4.
(a) Taber. D. F.; Saleh, S. A.: Korsmeyer.R. W. J. 0~. CYmn. 1980, 45, 4699; (b) Tabsr, D. F.; Petty, E. l-i. J. Org. c1clan. 1982, 47, 4808; These articles note a preferencefor addition by copper carbenoidsversus insertionby rhodium carbenoidsderived from the nearly identicaldiazoesters shown below.
Rh2aw,
(R=Me) 5.
Doyle, M. P. Qmn.
6.
Wenkert. E.; Davis, L. L.; Mylari, B. L.; Solarron, M, F.; da Silva, R. R.; St&man, S.;
Rev. 1906, 86, 919.
Warnet,R. J.; Ceccherelli,P.; Curini,M.; Pellicciari,R. J. Org. f&s??.1982, 47, 3242. 7.
(a) Taber, D. F.; Rnamn, K, J. Am. C&a. SGXY,1983, 105, 5935; (b) Taber. D. F.; Ruckle, R. E., Jr. r&M&z-on
L&t.,
1985, 26. 3059.
8.
Reissig.H.-U.; Hirsch,E. dnww. &art., Inter. Ed. Engl.
9.
A trace of 2.3-diethoxy~~nyleycloh~~~~
1900.
29,
813.
was also noted. Treatmentof 11 with 1 M
tetraethylamnonimfluoride in THF resulted in a l.l:l mixture of diester 12 and 2,3diethoxycarbonylcycloheptanone. 10. (a) DePuy, C. H.; Breitbeil,F.
W.;
DeBruin,K. R. J. Am. C&R.
Scic. 1966. 88, 3347; (b)
DePuy, C. H.; Breitbeil,F. W. J. Am. C&m. Ss. 1963, 85, 2176. 11. Wharton, P. S.; Bair, T. I. J. Org. I%YII.1966. 31, 2400. 12. Davis, Br. R.; Woodgate.P. D. J. Mom. Sac., G%n. Cermun. 1966, 65. 1965. 570. 13. Wenkert.E.: Kariv. E. 3. Cha. Soz.. Chem. Cprrm~sn. 14. Girard,C; Conic, J. M. Tet&dm
(Received in USA 25 April 1989)
Lett. 1974, 3333.
4093-4096,1989 0040-4039/89 $3.00 + .OO Maxwell Perqamon Macmillan plc
Andrew C. L.ottes, John A. Landgrebe*,and Kristin Larsen Departmentof Chemistry.Universityof Kansas Lawrence,Kansas 66045 ition of ethyl diazoacetatiin the Presence of B-keto esters Summy: The catalyzeddecanpos producesregioisanericenol ethers and the diester fran formal carbenoid insertioninto the a-C-H bond. The productdistributionis very catalystdependent. We have reportedpreviouslythat the CuCl or Rh,(CAc),catalyzed-sition
of ethyl
diazoacetatein the Presence of various alkyl or phenyl substitutedacyclic and cyclic kstones leads to en01 ethers with understandableregiochemistry(and where applicable,stereochemistry)as the only 1:l products,and that the product distributionwas independentof We now report observationsthat for various f_+ketoesters, not only is the
cata1yst.l
regiochemistry of en01 ether form&ion catalystdependent,but in the presenceof CuCl, significantamountsof a new 1:l productare observed.An example is shown below for ethyl acetoacetate (1) with a CuCl catalyst, and caq%ar& to the results previouslyreported under identicalconditionsfor methyl benzyl ketone (41.'
N&g?
&C02Et
CO&
c q
2@+61%)
1
+
/1\
3(394496) m2J9
(=,a
Ph
ic
N&!HCO.$t c&l
+
4 Ph
Ph
with exclusive (E)stereochani stry (confirmedby suitableNOE experiments)consistentwith the mechanisticmodel developedpreviousIn both cases, enoletbers 3 and 6 are fe
1y.l Snewhat surprisingis the observationthat 8-keto ester 1 produces none of the lesssubstitutedenol ether, but insteadforms diester 2. The results for various cyclic B-keto esters are swmarized in Table I.
Whatseelns
clear is that the use of Rh,(CAc),produces far less of the diester-typeproduct than when of en01 ether formation favors the lessCuCl is used. In addition, the regioselactivity substitutedproduct when CuCl is used, and with the exceptionof the 7?nanberedring, the more substitutedproduct when Eh,(oAc), is errrployed.Although the more substitutedenol ether is mre thermodynamically stable, suitablecontrol experimentson the less-substituted
4093
4094
en01 ether from 2-carbethoxycyclopentanone have clearly shown that no equilibrationof the regioisomersoccurs under the reaction conditions in the presence of either catalyst.2 Isolatedyields of the canbined1:l adductswere in the range 40-55%.3 5%ble I. Product
Distributionforcyclic B+?toEstersa
&?J:+
Ef
&;;&Z
n
CuCl Rh,(OAc),
CuCl Rh2(oAc),
CuCl Rh,mAc),
5
12-13
2
36-38
57
49-51
6
42-43
4-5
18-19
83-85
38-39
11-13
7
64-65
10-11
5-6
13-15
31-33
74-76
41
a Expressedas a percentof all I:1 adducts;resultsrepresentat least three runs; a 7-fold excess of ketone served as the solvent. In contrast,note that for the catalyzedreaction of ethyl diazoacetatewith amethyl and 2-phenylcycloalkanones sunmr ized in Table II, no diester has ever been observed,the resultsare identicalfor both CuCl and Rh,(C&c),.and the less-substituted en01 ether always dcminates.l
Bble II. EMi Ether Ring Size
Substituent
Distributionfor 2-myland Mor*SubstitutedProd.
2-Phehylcycloalkammesa Less-Substituted Prod.
Me
5
12.5-16.7
83.3-87.5
Me Ph
6
7.7- 7.1
92.3-92.9
5
43.5
56.5
Ph
6
40
60
a Expressedas a percentof all 1:l adducts;for liquid ketones,a 7-fold excess sexed as solvent:for solid ketones,a 6-fold excess was dissolvedin Ccl,; resultsare for at least three runs per ketone and are identicalfor CuCl and Rh,(oAc),.
4095
While the reasons why substantialamounts of diester form from the a-ethowoarbonyl ketones in Table I in the presence of CuCl but not Rh,(OAc), reamin speculative.related differences in the chanioal behavior of carbanoids derived fran diazo esters have been noted.4 It is of special interestthat because our observationsare contrary to the known effectivenessof rhadium(II)carboxylatesin pronmtingC-H insertion,4-7the latter process is probablynot involved. In fact, we suspect that diester 9 arises frcancyolopropanol8 producedfranthe additionof thecopper carbenoidto the enolof 7.
7
8
9
such a path is made plausibleby our observationthat silyl cy~lopropylether 11 prepared fran 10 by the method of Reissig and Hirsch,' was convertedto diester 12 in an isolated yield of 71% after heating for 24 h at 60 'C in ethanol.' Regioselectiveopening of the intermediate oyclopropan -01 under electrophilic conditionshas scme precedent,10-14
11
12
Althoughwe have no unified explanationfor the interestingdifferencesin regioselectivity of en01 ether formationas a functionof catalystfor the various a-ethoxycarbonyl ketones in Table I, it s8ems inescapablethat the catalyst must be present during the 1,4sigmatropicshift of hydrcgen. These interestingmechanisticquestionsrEznainunder active investigation.
Acknouledgaawt: Supported by the Universityof Kansas General Research Fund; a Phillips PetroleumCcinpany Fellowshipfor A.C.L.; an NSF REU Fellowshipfor K.L. Ref-andNotes
this issue. 1. L&ties, A. C.; Landgrebe.J. A.; Larsen,K. Previousccmnunication, 2. Equilibrationdoes ozcu.rin benzene with AlCl, and leads to approxinW.ely80-85% of the
4096
IKmFswtiruuted 3.
en01 ethez at 78 OC.
Structuresof newly prepared canpoundswere identifiedby a combinationof lH and 13C NMR, IX, and MS (includingan exact mass).
4.
(a) Taber. D. F.; Saleh, S. A.: Korsmeyer.R. W. J. 0~. CYmn. 1980, 45, 4699; (b) Tabsr, D. F.; Petty, E. l-i. J. Org. c1clan. 1982, 47, 4808; These articles note a preferencefor addition by copper carbenoidsversus insertionby rhodium carbenoidsderived from the nearly identicaldiazoesters shown below.
Rh2aw,
(R=Me) 5.
Doyle, M. P. Qmn.
6.
Wenkert. E.; Davis, L. L.; Mylari, B. L.; Solarron, M, F.; da Silva, R. R.; St&man, S.;
Rev. 1906, 86, 919.
Warnet,R. J.; Ceccherelli,P.; Curini,M.; Pellicciari,R. J. Org. f&s??.1982, 47, 3242. 7.
(a) Taber, D. F.; Rnamn, K, J. Am. C&a. SGXY,1983, 105, 5935; (b) Taber. D. F.; Ruckle, R. E., Jr. r&M&z-on
L&t.,
1985, 26. 3059.
8.
Reissig.H.-U.; Hirsch,E. dnww. &art., Inter. Ed. Engl.
9.
A trace of 2.3-diethoxy~~nyleycloh~~~~
1900.
29,
813.
was also noted. Treatmentof 11 with 1 M
tetraethylamnonimfluoride in THF resulted in a l.l:l mixture of diester 12 and 2,3diethoxycarbonylcycloheptanone. 10. (a) DePuy, C. H.; Breitbeil,F.
W.;
DeBruin,K. R. J. Am. C&R.
Scic. 1966. 88, 3347; (b)
DePuy, C. H.; Breitbeil,F. W. J. Am. C&m. Ss. 1963, 85, 2176. 11. Wharton, P. S.; Bair, T. I. J. Org. I%YII.1966. 31, 2400. 12. Davis, Br. R.; Woodgate.P. D. J. Mom. Sac., G%n. Cermun. 1966, 65. 1965. 570. 13. Wenkert.E.: Kariv. E. 3. Cha. Soz.. Chem. Cprrm~sn. 14. Girard,C; Conic, J. M. Tet&dm
(Received in USA 25 April 1989)
Lett. 1974, 3333.