UVLKWt Hut): 25¢ per day per item RETURNING LIBRARY MATERIALS: Place in book return to remove charge from circulation records ATTEMPTED SYNTHESES OF 1,u-DIKETONES BY Patrick J. Cowan A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Chemistry 1980 ABSTRACT ATTEMPTED SYNTHESES OF 1,14-DIKETONES BY Patrick J. Cowan Lithium and potassium ketone enolates failed to react with a.- haloketones to give the corresponding 1,ll-diketone. Lithium ketone enolates also did not react with y-halo-a-ketoesters under a variety of conditions. Lithio methyl acetate reacted with lithio chloroacetyl Meldrum's acid to.give methyl-fl-Meldrum's acid acetoacetate in.83$ yield. The same product was termed with lithio bromoacetyl Meldrum's acid in place of lithio chloroacetyl Meldrum's acid. Reaction of lithio methyl acetate with lithio iodoacetyl Meldrum's acid failed to give any alkylation product. Acetyl Meldrum's acid did not react with chloroacetone to give acetonyl acetyl Meldrum's acid. However, acetonyl Meldrum's acid, prepared in 87.u$ yield from chloroacetone and Meldrum's acid, reacted with acetyl chloride to give acetonyl acetyl Meldrum's acid in 99.51 yield. Acetyl chloride, however, was the only acid chloride which reacted with acetonyl Meldrum's acid. All attempts to cleave the Meldrum's acid ring of acetonyl acetyl Meldrum's acid failed. To My Parents, whose love and support made this project possible. 11 ACKNOWLEDGEMENTS The author wishes to express his deep appreciation to Dr. Michael W. Rathke for his inspiration, patience, and. guidance throughout this project. The author further. expresses his appreciation to his fellow students for both the professional interest and companionship which they have extended to him. Finally, the author wishes to thank his parents and family for their continued encouragement, love, and interest throughout this effort. iii LIST OF TABLES TABLE OF CONTENTS CHAPTER I - REACTIONS OF ENOLATES WITH a—CHLOROKETONES . . . . . . Introduction Results . Discussion Experimental I. II. III. Materials . . . . . . . . . . . . . . . . . . . . . Reaction of Lithium Enolates with Chloroacetone . . A. B. E. Preparation of Lithium Diisopropylamide (LDA) . Reaction of Amine-Free Lithio Acetone with Chloroacetone . . . . . . . . . . . . . . . . . Preparation of Dilithium Tetrachlorocuprate (LiZCuClu) Solution . . . . . . . . . . . . Reaction of Lithio Acetone with Chloroacetone in the Presence of 101 L12CuC1u . . . . . . . Reaction of Lithio Acetone with Chloroacetone in the Presence of Triethylborane . . . . . . . Reaction of Potassium Enolates with Chloroacetone . A. B. Standardization of KH . . . . . . . . . . . . . Reaction of Potassio Acetone with Chloroacetone Reaction of Potassio Acetone with Chloroacetone in the Presence of Triethylborane . . . . . . . Inverse Formation of Potassio Acetone and Its Reaction with Chloroacetone in the Presence of Triethylborane . . . . . . . . . . . . . . iv Page vii ll 22 22 23 23 23 24 24 24 25 25 25 26 26 IV. Page Reaction of Ethyl-u-Chloroacetoacetate Derivatives - . with Lithio Acetophenone . . . . . . . . ..... A. Preparation of‘g . . . . . . . . . . . . . B. Product Analysis . . . . . . . . . . C. Reaction of‘g with Lithio Acetophenone . . . D. Reaction of 5 with Amine-Free Lithio Acetophenone . . . . . . . . . . . . E. Reaction of'l with Lithio AcetOphenone F. Reaction of Lithio Acetophenone with 1 Prepared In Situ . . . . . . . . . . . . . G. Reaction of Ethyl-H-Chloroacetoacetate with 2.0 Equivalents of Lithio Acetophenone . . . . . . H. Preparation of Copper II Complex of Ethyl-u-Chloroacetoacetate . . . . . . . . . . I. Reaction of Copper II Ethyl-h-Chloroacetoacetate with Lithio Acetophenone . . . . . . . . . . . J. Preparation of Nickel II Complex of Ethyl-fl-Chloroacetoacetate . . . . . . . . K. Reaction of Nickel II Ethyl-u-Chloroaceto- acetate with Lithio Acetophenone . . . . . Reactions of Haloacetyl Meldrum's Acid with Enolates . . . . . . . . . . . . . . . . . . . . . A. Preparation of Chloroacetyl Meldrum's Acid (11) B. Reaction of 11 with Lithio Acetophenone . C. Reaction 0f.ll with Lithio AcetOphenone, AMine-Free o o o o o o o o o o o oo o o o o o o D. Reaction 0f.ll with Lithio Methyl Acetate . B. Preparation of Bromoacetyl Meldrum's Acid . F. Reaction of Lithio Bromoacetyl Meldrum's Acid with Lithio Methyl Acetate . . . . . . . . . 26 27 27 28 28 29 29 29 3O 3O 3O 31 31 31 32 33 34 34 CHAPTER II - RE Introducti Results . Discussion Experiment I. II. III. IV. BIBLIOGRAPHY . G. Preparation of Lithio Iodoacetyl Meldrum's A01d I O O O O 0 O I O O O O O O O O O O O H. Reaction of Lithio Iodoacetyl Meldrum's Acid with Lithio Methyl Acetate . . . . . . . . . ACTIONS OF MELDRUM'S ACID DERIVATIVES . on a1 Materials . . . . . . . . . . . . . . . Reaction of Acetyl Meldrum' 3 Acid (2") with Chloroacetone . . . . . . . . . . . . . A. Preparation of g3 . . B. Reaction of g5 with Chloroacetone . . . . Reaction of Acetonyl Meldrum' 3 Acid (25) with Acid Chlorides . . . . . . . . . . . . . . A. Preparation of g; . B. Reaction of g; with Acid Chlorides . C. A Study of the Reaction of 25 with Acetyl Chloride . . . . . . . . . . . . . . . Attempts to Cleave the Meldrum's Acid Ring of gl 3 OEt2 . . . B. Reaction of g1 with Trifluoroacetic Acid A. Reaction of gl_with BF vi Page 34 35 36 36 38 45 51 51 52 52 52 52 52 53 54 54 54 55 56 LIST OF TABLES Table Page I Preparation of Acetonyl Meldrum's Acid (22) . . 40 II Reaction of Acetonyl Meldrum's Acid with Acetyl Chloride 41 44 III Reaction of Acetonyl Meldrum's Acid with Benzoyl Chloride . vii CHAPTER I REACTIONS OF ENOLATES WITHtx-CHLOROKETONES Introduction 1,4-Dicarbonyl compounds are important intermediates for the syn- thesis of organic compounds. For example, 1,u-diketones are precursors 3 ’ ll Cyclopentenones , in to pyrroles1, furansz, and cyclopentenones turn, are precursors to prostaglandinsu. The synthesis of 1,3 and 1,5-dicarbonyl compounds can be accom- plished by simple aldol (Eq 1) or conjugate aldol (Eq 2) condensation reactions. 1,4-Dicarbony1 compounds, however, are not amenable to such simple approaches. 0 0 0 0 é H ' H u v R- :0 + x- C-R -—> R-c—c—c—R (1) 0‘ 0 0 0 I E v H H ' R"C=C + C=C-— —R -—) R—C~C-C-—C-—C—R (2) 1,4-Dicarbonyl compounds have been synthesized by the addition of acyl anion equivalents, such as nitro-stabilized carbanions1’3 (Eq 3). lithium dil:bis(phenylthio)methyl]copper5 (Eq ll), and acyl carbonyl- nickelate6 (Eq 5), to a,B-unsaturated carbonyls. — ‘_ u ' TiCl ' ch + c-cncona R cncnzcnzcon ,‘g Rcocuzcnzcoa (3) N02 N02 4Cu0 . o R . 1 + u .9 | | . 200012 . [(PhS)2 c12 RCOCHZCHZCOR (5) Yoshikoshi and coworkers7 used a SnClu catalysed addition of tri- methylsilyl enol ethers to<:,B-unsaturated nitroalkenes (Eq 6) to form 1,u-dicarbony1 compounds. 4 N02 SnCl4 3‘ H20 \ __) RICHZ— c-CR2R3 TMS I’ R4 R2 3 0 (6) R1 A particularly attractive approach, at least conceptually, to 1,4- dicarbonyl compounds is the direct alkylation of enolates with a- haloketones. This reaction has been accomplished with the enolates of diethyl malonate8 (Eq 7) and barbituric acid9 (Eq 8). The reaction of an enolate of a simple ketone with a a-haloketone, however, has never been reported. An alternate approach to 1,H-dicarbony1 compounds is the alkylation of enolates with masked a-haloketones. Miyano1o reacted methallyl iodide with a ketone enolate to obtain the corresponding alkylation product (1) which was converted to a 1,4-dicarbonyl compound (Eq 9). + COZEt O COZEt Na -"< '+ A Etzo 5 W (7) C02Et 0 CO Et Br 2 o ‘3 H N o 0 ,1 CH CO Na >;0 + '/U\ - 3 2 > Y\£\>:0 (8) Br 0 O H 0 111 OK 0 CH2 6 + 9* -*—> O . °2CH3 I co CA 2 3 .1. (9) O 03 °’ ‘ O 7 O Nan4-0904 COZCH3 Stork and Jung11 reacted 3-iodo-2-triethylsilyl-1-propene with eno- lates to give the corresponding alkylation product which was converted to a 1,H-diketone (Eq 10). Jacobson and coworkers12 have developed a procedure to synthesize 1,4-diketones from lithioenamines and 2-methoxyallyl bromide (Eq 11). The inconvenient synthesis required to prepare 2-methoxyallyl bromide represents the major disadvantage of this method. Conventional methods of preparing enol ethers from ketones give only minor amounts of the desired regioisomer (Eq 12). A pyrolytic cracking procedure was employed CH2 312:3 + {/u‘31rt3 .LEfli_4> lll|i:/\\nr/ 0Li I O cu2 1) mCPBA (1°) 0°C \_ 2 n so /M on 7' ° ) 2 4 e o c H 6 11 - -+ \‘N Li C6Hll\‘N CH2 + 0.3—. . Br 0033 o (11) H0 cco H 2 2 _€> HZOITHF o 84% ocn3 //£§§ 0 Hf Br major product x’u\fi CHBOH (12) Br OCH3 ¢¢L‘n Br minor product to synthesize the desired vinyl ether (Eq 13). This procedure is not only time consuming but it is also restricted to relatively small scale preparations of 2-methoxyallyl bromide. H (iPr) NEt —OTs OCH OCH 2 3 3 H3CO OCH3 + ‘ A /g 150°—19o°c " ”' + (13) Br .1 . Br Br 2 3 2:; = 65:21 An additional difficulty was encountered when the sequence was attempted with 3-bromo-2-methoxy-1-butene (prepared from 3-bromo-2,2- dimethoxybutane)13 . Here a mixture of 8N2 and SNZ' products were obtained in an approximate ratio of 11:1 (Eq 1‘1). Also, hydrolysis of the intermediate alkylation product sometimes gave a mixture of the 1,4- diketone and the corresponding furan and pyrrole (Eq 14).13 C H C H 6 11\ N 0 OCH3 s 2 H+ + N —N—9 H 0 > + OCH3 2 Br 0 0’¥§> HN ' C * SN2 (14) \V C H 6 11.”N O Illll H o 5 o + OCH3 2 85+ 6 A simpler method of protecting the carbonyl group of an a-haloketone is conversion of the carbonyl function to the enolate. The appreciably greater acidity of the a-hydrogens (compared to the a1-hydrogens) of a- haloketones is a serious obstacle to direct application of this approach (Eq 15). However,-y-halo-8-ketoesters should give the desired enolate O B O r H major product base probably inert to nucleOphiles (15) Br 0- A Br minor product reactive to nucleophiles because of the increased acidity of the a1-hydrogens (Eq 16). Alkylation of 3 with enolates followed by decarboxylation should give a 1,11-diketone (Eq 17). 0 o More: Br 0 major product 4 OEt ase reactive to nucleophiles (16) Br _ 0 o NOE. Br minor product probably inert to nucleophiles Rl&/R2+ 9111M“ 9C0 RM (17) O Kato1u has reported the reaction of ethyl-u-halo-acetoacetate eno- lates with the enolates of diethyl malonate (Eq 18) and methyl cyanoace- tate (Eq 19). However, the reaction of’a simple ketone enolate (as shown in Eq 20) or ester enolate (as shown in Eq 21) with the enolate of a-y- halo-B-ketoester has never been reported. + C02 Et +C£fj:>(/fi\\w CO Et 0 Na-< 18 (19) CN Et02C Et Cl 13% 0' o" o 0 0 )§ ‘1' MOR a \IWOR (20) c1 0 o o o o 0 RO ZNm + on' 9 WMOR' (21) C1 8 We began our survey of routes to 1,4-diketones with a brief’study of the direct alkylation of ketone enolates with a-chloroacetone (Eq 22). 0' o R o A + .—_> YVK (22) R 1 0 After this brief study, we turned our attention to the reaction of'ketone enolates with y-chloro-B-ketoesters. Results Reactions of Ketone Enolates with g-Chloroacetone An equivalent of chloroacetone was reacted with an amine-free tetrahydrofuran (THF) solution of lithio acetone at -78°C (Eq 23). GLC 0L1 O 0 AK + (K 1’ ”W 2W Cl 2) 15 min, 25°C 7' (23) O analysis established an 11.7% yield of 2,5-hexadione. The reaction was repeated in the presence of 0.1 equivalents of dilithium tetrachloro- cuprate (LiZCuCln) and in the presence of 1.1 equivalents of triethyl- borane. The yields of 2,5-hexadione were found to be 13.9% and 5.61 respectively. Under all conditions the reaction mixture turned red immediately after the chloroacetone was added. Potassio acetone was prepared by addition of one equivalent of potassium hydride to a THF solution of acetone. An equivalent of chloroacetone was added (the reaction mixture turned red) and analysis revealed no 2,5-hexadione had formed. The reaction was repeated in the presence of 1.1 equivalents of triethylborane and analysis revealed a 0.75% yield of 2,5-hexadione. Inverse formation of'potassio acetone, i.e. addition of'acetone to a suspension of KR in THF, followed by reaction with chloroacetone in the presence of 1.1 equivalents of triethylborane gave only trace amounts (<11) of 2,5-hexadione. Reactions of Ketone Enolates with'Y-Halo-B-Ketoesters We next studied reactions of enolates with‘Y-halo-B-ketoester deri- vatives. 'The first substrate chosen was the trimethylsilyl enol ether of ethyl-u-chloroacetoacetate (2) which was prepared from ethyl-fl-chloro- acetoacetate and bis-trimethylsilyl acetamide in 51.51 isolated yield (Eq Z“). 0 O $11463 M83310 0 ' , <24) 2 r/}L\¢/u\bEt + 1 CH3C NSiMe3‘—€> 2 r/4§§’}u\~OEt C1 C1 2. One equivalent of 5 was added to a THF solution of lithio aceto- phenone at -78°C (Eq 25). GLC analysis revealed one new component which Me 310 OLi OSiMe 3 0 ,/L\\ 3 + l/L\\ ———————é> (25) r/£§”/J*3Et Ph “* Ph'\\ c1 6 was identified as the trimethylsilyl enol ether of'acetophenone (6). The reaction was repeated with amine-free lithio acetophenone with the same results. 10 The next substrate chosen was lithio ethyl-ll-chloroacetoacetate (1). The reaction of lithio acetophenone with 1 (Eq 26) failed to give any alkylation product (8) whether 1 was prepared _i_r_1_ situ or the enolates 0L1 0 ‘ 0L1 ' o 0 Ph Mom; + 1.13% —)<—> W (2.) OEt ° 8 (:1 2. were prepared separately and then allowed to react. The copper II and nickel II enolates of ethyl-4-chloroacetoacetate were prepared. Both enolates failed to give any alkylation product when reacted with one equivalent of lithio acetophenone. We next attempted to activate the a-chloroketo compound by incor- porating it with Meldrum's acid (2). The desired compound, chloroacetyl Meldrum's acid (19), was prepared in 99.21 isolated yield by the reaction of chloroacetyl chloride with Meldrum's acid (2) in the presence of 2.0 equivalents of pyridine (Eq 27). The reaction of lithio chloroacetyl Meldrum's acid (11) with lithio acetophenone (Eq 28) failed to give any alkylation product under a variety of conditions. 0 o 0 pyr(2eq) HO 0 )< + ClCHZCOCI ———> \ )< (27) 0 Cl 0 o o 9 _. 10 L10 0 o 0“ ' HO 0 Ph \\ )< +' ”£§> '5‘é9 ‘\\ 0 Cl 0 Ph 0 (28) o o O 1/ 11 We next reacted an ester enolate with 11, One equivalent of lithio methyl acetate was added to a THF solution of 11, .After work-up, methyl- u-Meldrum's acid acetoacetate (12) was isolated in 833 yield. The same 0 CH 0 3 0 0 0 o 0 product was formed with bromoacetyl Meldrum's acid in place of chloro- acetyl Meldrum's acid. Reaction of lithio methyl acetate with lithio iodoacetyl Meldrum's acid (prepared from 11 and sodium iodide, Eq 29) failed to give any alkylation product. . (29) \ o + NaI A/ \ 0 Discussion We began our survey of routes to 1,4-diketones with a brief study of the direct alkylation of ketone enolates with chloroacetone (Eq 30). We 0 O z/J§§> + Cl\“/Ji\\ ‘_—“€>’ \\Tf/’\\\//j1\\ (30) O 12 expected two difficulties with this reaction; proton exchange (Eq 31) and condensation of the ketone enolate with chloroacetone (Eq 32). It is clear that in order to maximize the yield of the 1,4-diketone (Eq 30), the rates of the two competing reactions (Eq 31 and 32) must be decreased as much as possible. 0" o o o” ,//$§:> -+ H‘jz’JL\\ <3__ ‘//JL\\”II F;§£“\q (31) C1 C1 pKa 3 17 pKa ~ 20 0 -——> \ + A <———- M (32) c1 (:1 It has been shown that the use of enolates containing highly covalent metal-oxygen bonds decreases the rate of proton exchange. For example, lithium forms tighter metal-oxygen bonds than sodium or potas- sium, and lithium enolates have been shown to reduce the extent of polyalkylation, which is a result of proton exchange15’16(Eq 33). OM O O 0 ~\\ CHBI , o 1 + + THF . . . (33) I % M a Na 50% 22% 72% 7 N Li 13 while the use of enolates containing covalent metal-oxygen bonds decreases the rate of proton exchange (Eq 33), the use of'an enolate with a non-coordinating metal may decrease the rate of the condensation reaction (Eq 32). Aldol type condensations are thought to occur via a transition state in which the metal functions as a bidentate ligand coordinating to the oxygen atoms17 (Eq 34). The use of a non-coordi- 0 OM 0"";0 A {A‘} A}\ —> M (34) nating metal (i.e. potassium) would lower the stability of the transi- tion state and thus decrease the rate of the condensation reaction (Eq 32). We studied the reaction of chloroacetone with both the lithium and potassium enolates of acetone. Use of the lithium enolate should decrease the rate of proton exchange (Eq 31) while the use of the potassium enolate should decrease the rate of condensation (Eq 32). Chloroacetone reacted with lithio acetone to give an 11.71 yield of 2,5- hexadione. The reaction of potassio acetone with chloroacetone gave no alkylation product. This may be due to an increase in the rate of‘proton exchange between potassio acetone and chloroacetone. Lindert16 has reported that triethylborane serves as an effective additive to decrease polyalkylation in the reaction of the sodium enolate of cyclohexanone with methyl iodide (Eq 35). Later Negishi18 reported 14 CH3 I 50% L—__I22% ONa (35) \3” are“... . . 0.5%——-—1 that addition of triethylborane to potassium enolates also serves to decrease polyalkylation (Eq 36). It is thought that triethylborane cu ~cucn 337 2’ 21: OK (36) \cn2 :zcncn Br ‘v Et3 B, THF 90% functions by coordinating to the ketone enolate giving an enolate with increased selectivity for alkylation verses proton exchange. We consid- ered that addition of triethylborane to the reaction of lithio acetone and chloroacetone would decrease the rate of proton exchange; however, 15 when the reaction was carried out in the presence of 1.1 equivalents of triethylborane, 2,5-hexadione was produced in only 5.61 yield. The fact that triethylborane did not increase the yield of the alkylation product is not surprising in that Negishi has reported that triethylborane does not coordinate to lithium enolates19. Thus, although triethylborane is present in the reaction mixture, the reative species is lithio acetone, not the triethylborane complex 13. The reaction of potassio acetone with O— BEt:3 x/J§§> ‘ L1+ __1_3_ chloroacetone in the presence of 1.1 equivalents of triethylborane gave 2,5-hexadione in 0.751 yield. The potassium enolates used in the experiments described above were prepared by addition of potassium hydride to a THF solution of acetone. This method has the disadvantage that as the potassium hydride is initially added, a small amount of'potassio acetone is in the presence of a large amount of unreacted acetone and a condensation reaction could result (Eq 37). This problem could be avoided if the enolate is formed OK 0 o 2§ + A 9.; MGR (37) by inverse addition, i.e. addition of acetone to a solution of potassium hydride. However, when potassio acetone formed by inverse addition was reacted with chloroacetone in the presence of 1.1 equivalents of tri— ethylborane only a trace amount of 2,5-hexadione was formed. 16 Kochi20 has reported dilithium tetrachlorocuprate (LiZCuClu) cata- lyzes the cross coupling of Grignard reagents and alkyl bromides (Eq 38). 1 L12CUC14 l R - MgX + R -Br -—_TH§__—€? R-R (38) We considered the use of LiZCuClu would also catalyze the reaction of an enolate with a a-chloroketone. However, the reaction of lithio acetone with chloroacetone in the presence of 101 LiZCuC1u gave 2,5-hexadione in essentially the same yield as the reaction carried out in the absence of LiZCuClu. Although triethylborane inhibits proton exchange between acids with similar pKa's (Eq 39), this additive may be unable to inhibit proton 0 0' O O -——————> z’JJ\\T/H 1+ /’/L§§> Ph (41) C1 for this transfer is that the diisopropylamine present from the formation of the lithio acetophenone may react with 5 at the y-carbon and thus prevent any further reaction at this site (Eq 42). In order to prevent e 310 Me3810 0 M 3 0 __> M +1101 (42) f’/¥§>«’JL‘bEt + HN<1Pr>2 \\‘ 0Et Cl N(iPr)2 this reaction, the amine-free enolate of acetophenone was prepared and reacted with 55 however, the trimethylsilyl enol ether of acetophenone was again produced. Kellogg and Troostwijk21 have reported the alkylation of the sodium enolate of ethyl-4-bromoacetoacetate with various sulfur and oxygen nucleophiles (Eq 43). In this reaction the carbonyl function of the a- haloketone is protected as its enolate. We attempted to extend this method to the preparation of 1,4-diketones by reacting the lithium enolate of ethyl-4-chloroacetoacetate with lithio acetophenone (Eq 44); 18 NaO . 1) Rx’ 0 X - 0,3 \ 0 0 MOIST. 2) H 0 7 MOEt (43) Br 2 RX L10 0 0Li Ph 0 0 M + 13$ --—-> W (44) DEC 0131: C1 0 however, the reaction failed to give any alkylation product under a variety of conditions. We next prepared the copper II enolate of ethyl-4-chloroacetoace- tate and reacted it with lithio acetophenone (Eq 45) with the expectation Cl 0 _ 0Li o 0 \ 2 Ph \ *9 C“ 0 \\\0 GE: 2 1... OEt __2 that the copper II enolate would be more reactive than the lithium enolate. However, the reaction failed to give any alkylation product. Use of the nickel II enolate also failed to give any alkylation product (Eq 46). 19 _._ _1 __ C1 7‘ 0Li /O /0 O (46) N1 \0 + 2 Ph/§ '96") “\0 \ __ oat—2 h ' on J2 Chloroacetyl Meldrum's acid ( _1_(_)_) was prepared in essentially quan- titative yield from Meldrum's acid (2) and chloroacetyl chloride (Eq ‘17). O ClCH COCl X + 2 pyr (Zeq) \ 0 . _.____> 0 (47) We suspected that the carbon containing the chlorine in the enolate of chloroacetyl Meldrum's acid (14) would be more susceptible towards nucleophilic attack by enolates than was carbon 4 in the enolate of ethyl-4-chloroacetoacetate (15) because of 14's higher degree of delo- calization. Reaction of fl with an enolate should give _1_6_; which after cleavage of the Meldrum's acid ring by alkoholysis22 followed by decar- boxylation, should give the ‘1,4—diketone 1_Z (Eq 48). However, the 20 0- o 0 0 0 HTS) + /§ R ' (48> . . R on cl 0 ,1” R "‘49 \\H//\\~/Jk:l:fikb "“€> 0 0 ,1/’ A 0 o R O O -CO R O YWK , —-—) N ' OR 0 0 .11 reaction of lithio acetophenone with lithio chloroacetyl Meldrum's acid (18) failed to give any alkylation product under a variety of conditions (80 49). L10 0 0 0 0Li Pb \ 0 + /g 96 0 (49) 01 ’1’, Ph 0 0 C) o 0./+’ 18 We next decided to try reacting an ester enolate with 18. Ester enolates were chosen because of their increased reactivity over ketone enolates towards alkylation. Reaction of lithio methyl acetate with 18 gave methyl-4-Meldrum's acid acetoacetate (_1_2) in 851 yield (Eq 50). O ,/l\\ MeO \ + \ 0 ——> HWY” (50) MeO 0 0 C1 o/j/ . 0* o 0 \ 18 12 21 This product may arise by addition of lithio methyl acetate to the double bond in 1_8_ followed by a Favorskii rearrangement (Eq 51). We changed the MeO Jr (51) MeO - halogen from chlorine to bromine with the expectation that by making the terminal carbon more electrophilic the lithio methyl acetate would alkylate at that position rather than add to the double bond. However, reaction of lithio methyl acetate with lithio bromoacetyl Meldrum's acid also gave _1_2’_ (Eq 52, yield undetermined). 0 ——‘> 12 (52) 22 Finally, we prepared lithio iodoacetyl Meldrum's acid (12) from 18 and sodium iodide (Eq 53). The brown precipitate that formed during the L10 0 L10 0 \\\ C1 0 + N31 ‘——> \ O o 4/ + NaCl (53) o I 0 ,f/ . o 18 reaction was isolated and solubility tests showed the precipitate to be sodium chloride, indicating that 12 had been formed. Reaction of 12 (prior to the separation of NaCl) with lithio methyl acetate failed to give any alkylation product. This may have been due to the high salt concentration present in the reaction mixture. Experimental I. Materials _n-Butyllithium, Aldrich, was purchased as a 1.6 fl hexane solution and standardized by the method of Watson and Easthamzu. The commercially available compounds diisopropyl ketone, diisopropylamine, chloroacetone, acetone, pyridine, acetophenone, and methyl acetate were distilled from calcium hydride and stored under argon. Tetrahydrofuran (THF) was distilled from sodium and benzophenone and stored under argon. Ethyl-u-chloroacetoacetate, chloroacetyl chloride, and bromoacetyl bro- mide were purchased from Aldrich and used without fUrther purification. Triethylborane was purchased from Gallery and handled under argon. Meldrum's acid was prepared from malonic acid, acetic anhydride, and 23 25 . acetone by the method of Davidson and Bernhard . Bis-trimethylsilyl acetamide was prepared from acetamide and chlorotrimethylsilane by the . A 26 method of Kelbe, Finkeiner, and White . Potassium hydride was purchased as a 2h.7$ dispersion in oil. II. Reaction of Lithium Enolates with Chloroacetone A. Preparation of Lithium Diisopropylamide (L051 A 25 ml round-bottomed flask equipped with septum inlet, magnetic stirrer, and.mercury bubbler was flushed with argon, immersed in an ice bath, and charged with 3.13 ml (5.0 mmol) 1.6 g gfibutyllithium solution and 3.2 ml grpentane. Stirring was initiated and 0.70 ml (5.0 mmol) diisopropylamine was added dropwise. The ice bath was removed and the mixture was stirred for five minutes. The solvent was removed under vacuum to yield LDA as a white powder. B. Reaction of Amine-Free Lithio Acetone with Chloroacetone The following procedure is representative of the preparation of an amine-free lithium enolate solution: LDA (5.0 mmol), prepared as previously described, was dissolved in 5.0 ml THF and cooled to 0°C. Acetone (0.37 ml; 5.0 mmol) was added dropwise and stirred for 15 minutes. The ice bath was removed and the solvent and amine were evaporated under vacuum. After most of the solvent and amine had been evaporated, a warm (approximately HOOC) water bath was placed under the reaction flask and the remaining solvent and amine were evaporated, leaving the lithium ketone enolate as a white powder. The enolate was dissolved in 5.0 ml THF and cooled to -78°C in a Dry Ice/acetone bath. Stirring was initiated and 0.fl5 ml (5.0 mmol) chloroacetone was added 24 over a period of approximately 2 minutes. The reaction was stirred for 15 minutes at -78°C, the cooling bath was removed, and the reaction was stirred for an additional 20 minutes. As the reaction mixture warmed, the color of the mixture turned dark red. ngridecane was added as an internal GLC standard. The reaction was quenched with 2.0 ml H201and the organic phase analyzed by GLC (carbowax 20 M on Chromosorb W, column temperature 125°C). 2,5-Hexadione was observed in 11.71 yield. C. Preparation of Dilithium Tetrachlorocuprate (Lieggglnl Solution A 1.0 g THF solution of Li CuClu was prepared from lithium 2 chloride and copper (II)-chloride as described by Koch127. D. Reaction of Lithio Acetone with Chloroacetone in the Presence of 101 LiZCuClu A 1.1 fl THF solution (“.5 ml; 5.0 mmol) of amine-free lithio acetone was prepared at -78°C as previously described. To this solution was added 0.5 ml (0.5 mmol) 1.0 g THF solution of LiZCuClu giving a dark green solution. Chloroacetone (0.“5 ml; 5.0 mmol) was slowly added and the reaction was stirred for 1 hour at -78°C (the color of the reaction mixture was reddish-brown). The cooling bath was removed and the reaction was stirred for an additional 15 minutes. The reaction was worked up and analyzed as previously described. 2,5-Hexadione was observed in 13.91 yield. E. Reaction of Lithio Acetone with Chloroacetone in the Presence of Triethylborane A 1.0 g THF solution (5.0 ml; 5.0 mmol) of amine-free lithio acetone was prepared at 0°C as previously described. Triethylborane (0.77 ml; 5.5 mmol) was added and the reaction was stirred for’10 minutes 25 at 0°C. Chloroacetone (0.“5 ml; 5.0 mmol) was slowly added (reaction mixture turned red) and the reaction was stirred for 15 minutes at 0°C. The cooling ' bath was removed and the reaction was stirred for an additional 15 minutes. ‘gaTetradecane was added as an internal standard. The reaction was quenched with 1.85 ml (5.55 mmol) 3 §_Na0H, cooled to 0°C, and 1.85 ml (5.55 mmol) 301 320 was slowly added. The organic 2 phase was analyzed by GLC. 2,5-Hexadione was observed in 5.61 yield. III. Reaction of Potassium Enolates with Chloroacetone A. Standardization of KB A 25 m1 round-bottomed flask equipped with septum inlet, magnetic stirrer, and a gas buret was charged with 10.0 ml of acetone. KR (1.00 ml; 2n.7$ in oil) was added dropwise and the gas evolution was measured by the gas buret. 1H7.5 ml (6.15 mmol) of gas was evolved per 1.00 ml KH. B. R§§g§1on of Potassio Acetone with Chloroacetone The following procedure is representative of the preparation of potassium enolate solutions: a 25 m1 round-bottomed flask equipped with septum.inlet, magnetic stirrer, and mercury bubbler was flushed with argon, immersed in an ice bath, and charged with 0.37 ml (5.0 mmol) acetone and 5.0 ml THF. Stirring was initiated and 0.81 ml (5.0 mmol) 6.15 5 K8 was slowly added over approximately 1 minute. Chloroacetone (0.“5 ml; 5.0 mmol) was slowly added (the reaction mixture turned dark red) and the reaction was stirred 15 minutes at 0°C. The cooling bath was removed and the reaction was stirred for an additional 15 minutes. GLC analysis showed no 2,5-hexadione had formed. 26 C. Reaction of Potassio Acetone with Chloroacetone in the Presence of Triethylborane A 1.0 11 THF solution (5.0 ml; 5.0 mmol) of potassio acetone was prepared at 0°C as previously described. Triethylborane (0.77 ml; 5.5 mmol) was added and the reaction was stirred for 5 minutes. Chloroace- tone (0.H5 ml; 5.0 mmol) was slowly added and the reaction was stirred for 15 minutes at 0°C. The cooling bath was removed and the reaction was stirred for an additional 15 minutes. 2,5-Hexadione was observed in 0.751 GLC yield. Du Inverse Formation of Potassio Acetone and Its Reaction with Chloroacetone in the Presence of Triethy1borane The procedure previously described was f61lowed except that the potassio acetone was prepared by slow addition of 0.37 ml (5.0 mmol) acetone to a suspension of 0.81 ml (5.0 m1) 211.71 KH in 5.0 m1 THF at 0°C. GLC analysis revealed a trace (<11) of 2,5-hexadione. IV. Reaction of Ethyl-fl-Chloroacetoacetate Derivatives with Lithio Acetophenone A. Preparation of 5 A 250 ml round-bottomed flask equipped with septum inlet, magnetic stirrer, and mercury bubbler was purged with argon, immersed in an ice bath, and charged with 100 ml grpentane and 13.5 ml (100 mmol) ethyl-u-chloroacetoacetate. Bis-trimethylsilyl acetamide (12.8 ml; 50 mmol) was added and the ice bath was removed. A white precipitate formed after approximately 5 minutes. The reaction was stirred for 30 minutes, filtered, and the solvent removed from the filtrate under vacuum. The 27 yellow liquid residue was distilled (TOO-72°C, 6.5 mm mercury) yielding 11.88 g (51.5 mmol; 51.51 yield) of 5. B. Product Analysis GLC analyses were performed on a Varian Model 920 gas chromato- graph equipped with an 8 ft. x 0.25 in. stainless steel column packed with 31 813-30 on' Chromosorb G. 1H NMR spectra were determined on a Varian T-60 using tetramethylsilane as an internal standard. buss spectra were taken with a Finnigan ”000 with INCOS data system. ‘H mm (cnc13): 50.11 (s, 9H), 61.3 (t, 3H); 61.2 (q, 2H); 611.5 (s, an); 55.1 (s, 111) C. Reaction of 5 with Lithio Acetophenone The f61lowing procedure is representative of the preparation of a lithium enolate solution: LDA (5.0 mmol), prepared as previously described, was dissolved in 5.0 ml THF and cooled to -78°C in a Dry Ice/acetone bath. Acetophenone (0.59 ml; 5.0 mmol) was added and the reaction was stirred for 15 minutes after which 1.03 ml (5.0 mmol) §_was added. The reaction was stirred for 5 minutes at -78°C. The cooling bath was removed and the reaction was stirred for 3 hours periodically analyzing aliquots by GLC (column temperature 175°C). GLC analysis revealed one new component. The new component was GLC prepped and 1H NMR analysis revealed the component to be the trimethylsilyl enol ether of acetophenone (é). 28 ‘H am (00013): 50.5 (s, 93); 511.5 (d. 1H); 65.0 (d. 1H); 67.2-7.8 (m, 511) D. Reaction of 5 with Amine-Free Lithio Acetophenone Amine-free lithio acetophenone (5.0 mmol) was prepared from LDA (5.0 mmol) and acetophenone (0.59 ml; 5.0 mmol) as previously described and dissolved in THF (5.0 ml) at -78°C. Compound § (1.03 ml; 5.0 mmol) was added and the reaction was stirred for 5 minutes at -78°C. The cooling bath was removed and the reaction was followed by GLC. Analysis revealed the fOrmation of 6. E. Reaction of 7 with Lithio Acetophenone A 1.0 g THF solution (5.0 ml; 5.0 mmol) of lithio acetophenone was prepared at -78°C as previously described. A THF solution of‘l was prepared as follows: a 25 ml round-bottomed flask equipped with septum inlet, magnetic stirrer, and mercury bubbler was purged with argon, immersed in a Dry Ice/acetone bath, and charged with 5.0 ml THF and 0.68 ml (5.0 mmol) ethyl-fl-chloroacetoacetate. grButyllithium (3.13 ml; 1.6 ,5 in pentane; 5.0 mmol) was added and the reaction was stirred for 15 minutes. The ester enolate solution was transferred to the ketone enolate solution and the reaction was stirred for 15 minutes at -78°C. The cooling bath was removed and the reaction was stirred for 30 minutes. The reaction was quenched with 7.5 ml 2 _11 aqueous HCl solution, the organic phase was separated, dried (MgSOu), filtered, and concentrated under vacuum. Analysis by 1H NMR revealed acetophenone and ethyl-11- chloroacetoacetate. 29 F. Reaction of Lithio Acetophenone with 7 Prepared In Situ LDA (10.0 mmol) was prepared as previously described, dis- solved in TRE (5.0 m1), and cooled to -78°C with a Dry Ice/acetone bath. Acetophenone (0.59 ml; 5.0 mmol) was added and the reaction was stirred for 15 minutes at -78°C. Ethyl-H-chloroacetoacetate (0.68 ml; 5.0 mmol) was added and the reaction was stirred fer 15 mdnutes at -78°C. The cooling bath was removed and the reaction was stirred overnight. Cupric acetate (0.50 g) was added ferming a green precipitate. The reaction mixture was filtered and the filtrate was concentrated under vacuum. Analysis by 1H NMR revealed only starting materials. 0. Reaction of Ethyl-lI-Chloroacetoacetate with 2.0 Equiva- lents of Lithingcetophenone A 1.0 g THE solution (10.0 ml; 10.0 mmol) of lithio aceto- phenone was prepared at -78°C as previously described. Ethyl-H-chloro- acetoacetate (0.68 ml; 5.0 mmol) was added and the reaction was stirred for 15 minutes at -78°C. The cooling bath was removed and the reaction was stirred fOr 3 hours and concentrated under vacuum. The reaction was quenched with 10.0 ml 2 g HCl, extracted with diethyl ether, the organic phase dried (MgSOu), filtered, and concentrated under vacuum. Analysis by 1H NMR revealed only starting materials. B. Preparation of Copper II Complex of Ethyl-u-Chloroaceto- acetate A 100 ml Erlenmeyer flask equipped with a magnetic stirrer was charged with 50 m1 H20 and 2.00 g (10.0 mmol) Cu (0A0)2'820 giving a blue solution. Ethyl-H-chloroacetoacetate (2.70 ml; 20 mmol) was added and the solution was stirred for 15 minutes forming a green precipitate. The 30 mixture was filtered, the green precipitate was washed with diethyl ether, and dried in a vacuum dessicator yielding 2.711 g (7.01 mmol; 70.11 yield) of the copper II complex. I. Reaction of Copper II Ethyl-ll-Chloroacetoacetate with Lithio Acetophenone A 1.0 _11 THE solution (10.0 ml; 10.0 mmol) of lithio aceto- phenone was prepared at -78°C as previously described. Copper II ethyl- N-chloroacetoacetate (1.95 g; 5.0 mmol) was added and the reaction was stirred for 15 minutes at -78°C. The cooling bath was removed and the reaction was stirred for 15 minutes. While the reaction mixture was warming, a brown precipitate formed which was filtered and washed with acetone. The brown precipitate was found to be insoluble in dilute 8280”. J. Preparation of Nickel II Complex of Ethyl-A-Chloroaceto- acetate A 100 ml Erlenmeyer flask equipped with a magnetic stirrer was charged with 50 ml H O and 2.119 g (10.0 mmol) Ni(0Ac)2'11H20. One 2 equivalent (0.“0g; 10.0 mmol) NaOH was added giving a milky solution to which 2.70 ml (20.0 mmol) ethy1-l1-chloroacetoacetate was added. The reaction was stirred for 5 minutes forming a green precipitate which was filtered, washed with diethyl ether, and dried in a vacuum dessicator. K. Reaction of Nickel II Ethyl-u-Chloroacetoacetate with Lithio Acetophenone A 1.0 ! THE solution (10.0 ml; 10.0 mmol) of lithio aceto- phenone was prepared at -78°C as previously described. Nickel II ethyl- u-chloroacetoacetate (1.93 g; 5.0 mmol) was added and the solution was stirred for 15 minutes at -78°C. The cooling bath was removed and the 31 reaction was stirred overnight. The solution was filtered (a green precipitate was present) and the filtrate was concentrated under vacuum. 1H NMR analysis showed the filtrate to be acetophenone. V. Reaction of Chloroacetyl Meldrum's Acid with Enolates A. Preparation of Chloroacetyl Meldrum's Acid (10) A 500 ml three neck round-bottomed flask equipped with a mechanical stirrer, addition funnel, and mercury bubbler was purged with argon and immersed in a Dry Ice/acetone bath. The flask was charged with 36.0 g (250 mmol) Meldrum's acid, “0.5 ml (500 mmol) pyridine, and 250 ml methylene chloride. Stirring was initiated and 19.9 ml (250 mmol) chloroacetyl chloride was added via the addition funnel over a period of 3 hours. The cooling bath was removed and the reaction was stirred for an additional 1.5 hours. The reaction was quenched with 250 ml 2 g 801, the organic phase washed 2 x 100 ml 2‘! HCl then 2 x 100 ml H 0, dried 2 (MgSOu), filtered, and the solvent removed under vacuum giving 50.57 g (297 mmol; 99.21) chloroacetyl Meldrum's acid (slightly yellow). 1B NMR analysis showed 19 to be sufficiently pure to be used without fUrther purification. Chloroactyl Meldrum's Acid 1H NMR (c0013): 61.8 (s, 6H); 511.8 (bs, an); 615.3 (s, 111) B. Reaction of Lithio Chloroacetyl Meldrum's Acid (11) with Lithio Acetophenone LDA (10.0 mmol) was prepared as previously described and dissolved in 120 ml TRF at -78°C. Acetophenone (0.59 ml; 5.0 mmol) was added and the reaction was stirred for 15 minutes. Chloroacetyl 32 Meldrum's acid (1.10 g; 5.0 mmol) was added and the reaction was stirred 15 minutes at -78°C then 6 hours at room temperature. The reaction mixture was quenched with 7.5 ml (15 mol) 2 g HCl, the organic phase was dried (MgSOu), filtered, and concentrated under vacuum giving a brown precipitate which smelled like acetophenone. Analysis of the brown precipitate (1H NMR) showed it to be chloroacetyl Meldrum's acid and acetophenone. C. Reaction of 11 with Lithio Acetophenone, Amine-Free A 1.0 fl THE solution (5.0 ml; 5.0 mmol) of amine-free lithio acetophenone was prepared as previously described. A THE solution of 11 was prepared as follows: gybutyllithium (6.66 ml; 1.5 5; 5.0 mmol) was added to a THE (5.0 ml) solution of chloroacetyl Meldrum's acid (1.10 g; 5.0 mmol) at -78°C. This solution was stirred for 15 minutes at -78°C and the solvent was removed under vacuum to give 11 as a yellow solid. The solid 11 was suspended in 10.0 ml THF at -78°c and the lithio acetophenone solution was added. The cooling bath was removed and the reaction was stirred for 3 hours. The reaction was quenched with 15 ml 2 1! H01, the organic phase was dried (MgSOH), filtered, and concentrated under vacuum yielding a light brown solid. 1H NMR analysis showed the solid to be acetophenone and chloroacetyl Meldrum's acid. The reaction was repeated with the fbllowing changes: after the lithio acetophenone was added to 11 the reaction was stirred 3 hours at -78°C and worked up without allowing the reaction to warm up. Work-up yielded a yellow solid which was recrystallized from CHZClz[n-pentane (50:50) giving a white solid. 13 NMR analysis showed the solid to be chloroacetyl Meldrum's acid. 33 The reaction was repeated again with the following changes: after the two enolates were combined, the reaction was stirred 1 hour at -78°C, the cooling bath was removed, and the reaction was stirred for 1.5 hours. Acidic work-up gave a yellow oil which was shown to be acetophenone and chloroacetyl Meldrum's acid by 1H NMR analysis. D. Reaction of 11 with Lithio Methyl Acetate A 1.0 M THE solution (52.11 mi; 52.11 mmol) of lithio methyl acetate and a 2 _M THE solution (25 ml; 52.9 mmol) of 11 were prepared at -78°C as previously described. The solution of 1_1 was added to the lithio methyl acetate solution over1a period of’approximately 10 minutes. A white precipitate formed when the enolates were combined. The reaction was stirred for 15 minutes at -78°C, the cooling bath was removed, and the reaction was stirred for 3 hours. The reaction was quenched with 75 ml 2.! H01, the organic phase was washed (2 x 50 ml 2'5 HCl; 2 x 50 ml H20), dried (MgSOu), filtered, and the solvent removed under vaccuum giving 11.22 g of a slightly yellow solid. Analysis showed the solid to be 1g. 1 H NMR (CDCl ): 61.8 (s, 6H), 63.3 (d, 2H), 63.5 (s, 2H), 3 6307 (3, 3H): 53-8 (13, 1H) us: m/e 2H3 (M1 - 15) IR (nujol mull): 1710 cm.1 (C=0), 1730 cm.1 (C=0), 1760 CM.1 (C=0) 34 E. Preparation of Bromoacetyl Meldrum's Acid Bromoacetyl Meldrum's acid was prepared from Meldrum's acid (7.20 g; 50 mmol). PYridine (8.10 ml; 100 mmol), and bromoacetyl bromide (11.57 ml; 52.5 mmol) following the procedure for the preparation of chloroacetyl Meldrum's acid. Bromoacetyl Meldrum's acid (7.07 g; 58.11 yield) was isolated as a brown solid. 1H NMR analysis showed the compound to be sufficiently pure to be used.without further purification. Bromoacetyl Meldrum's Acid 1 H NMR (CD01 ): 61.7 (8, 6H), 69.6 (8, 2H), 51u.8 (b3, 1H) 3 E. Reaction of Lithio Bromoacetyl Meldrum's Acid with Lithio Methyl Acetate The procedure for the reaction of 11 with lithio methyl acetate was followed substituting bromoacetyl Meldrum's acid fOr chloroacetyl Meldrum‘s acid. Compound 1g was isolated as a yellow solid. 0. Preparation of Lithio Iodoacetyl Meldrum's Acid To a 0.5 g THE solution (10 ml; 5.0 mmol) of 1_1 at -78°c was added a solution of 0.75 g (5.0 mmol) of NaI in 25 ml THF. The reaction was stirred for 15 minutes at -78°C, the cooling bath was removed, and the reaction was stirred for 1 hour. A light brown precipitate formed as the reaction warmed to room temperature. The reaction was filtered and the filtrate was concentrated under vacuum to give 0.80 g of a yellow solid. 1H NMR analysis of the yellow solid revealed two peaks. 1H NMR (020, external TMS standard): 61.6 (s, 6H), 6fle3 (s, 2H) 35 Solubility tests of the brown precipitate indicated it to be LiCl. The results of the tests are listed below. Solvent Results Acetone Insoluble NHHOH Soluble NHuOH + AgClOu Precipitate forms H. Reaction of Lithio Iodoacetyl Meldrum's Acid with Lithio Methyl Acetate A THE solution (25 ml; 5.0 mmol) of lithio iodoacetyl Meldrum's acid was prepared as described above with the following changes: after the reaction was stirred at room temperature for'1 hour, the solution was cooled to -78°C without removing the solvent. A THE solution (5.0 ml; 5.0 mmol) of lithio methyl acetate was added to the lithio iodoacetyl Meldrum's acid solution and the resulting solution was stirred for 30 minutes at -78°C. The cooling bath was removed and the reaction was stirred for 1 hour. The reaction mixture was filtered and the solvent was removed from the filtrate under vacuum giving an orange solid. 1H NMR analysis revealed two singlets. 1H NMR (D20, external TMS standard): 61.6 (s, 6H), 159.3 (s, 2H) CHAPTER II REACTIONS OF MELDRUM'S ACID DERIVATIVES Introduction The reaction of an enolate with an a-haloketone has been accom- plished using c-haloacetone and the enolates of diethyl malonate8 (Eq 511) and barbituric acid9 (Eq 55). This type of reaction could be used for CO 2Et + co2 Et N (54) a Moo: Et +EK_—_> CH3 002 Na (55) o o o 11:10 the synthesis of 1,u-diketones provided the active methylene compounds were properly substituted. For example, reaction of the enolate of 2- acyl malonate (22) with chloroacetone should give 2-acyl-2-acetonyl malonate (21), which could be decarboxylated to give the desired 1,9- diketone (Eq 56). Alternatively, compounds such as 2-acetonyl malonate could be used to synthesize 1,u-diketones. iEor example, the reaction of the enolate of 2-acetonyl malonate with an acid chloride should also give g1 (Eq 57). 36 37 0 o o co R R' COZR (’lL“ R' 2 o . Cl c02R COZR o o 39 21 o o ' COZR -1 R c1 ; co R o 2 cozR 0 21 In contrast with acyclic malonic esters (pKa 13.7)22 and acetoace- tate esters (pKa 10.7)22, Meldrum's acid (2) reacts with electrophiles (at carbon 5) even in the absence of a strong base because of its remarkably high acidity (pKa N.97)22. 'For example, Meldrum's acid reacts with propionyl chloride in the presence of pyridine to give propionyl Meldrum‘s acid (22) in almost quantitative yield (Eq 58)22 3 0 4 ° pyr(Zeq) 2X + VK ..____> 6 0 C1 CH2C12 2. 0* (58) Because of the exceptionally mild conditions needed to react Mel- drum's acid with electrophiles, we proposed to study the use of‘Meldrum's acid for the synthesis of 1,u-diketones. Both of the methods described 38 above for the synthesis of 1,u-diketones (Eq 561and 57) were studied with Meldrum's acid in place of the malonic ester (Eq 59 and 60). Either o no '° ° ° ‘54 ‘\\ ba (59) R’J%;[:u\o + r/u\\ -__ff:> R 0 °”1/’ Cl 0 °22 0 0 ° 0* ,u\ + base pol“ R ——-> 22 (6°) Cl -—- 0 method should give the same product, 22, which, after cleavage of the Meldrum's acid ring by alkoholysis, followed by decarboxylation, should give the desired 1,9-diketone 23 (Eq 61). ' R on o COZR -co 0 22 a R 2 0 0 Results Acetyl Meldrum's acid (g1) was reacted with chloroacetone in the presence of one equivalent of pyridine (Eq 62). GLC analysis revealed that the chloroacetone remained unchanged after u hours. HO 0 0 0 pyr 25°C + f/u\\ ’ No Reaction (62) 0 0 X C1 CHZCl2 24 4 hrs \/ 39 We next decided to study the synthesis of 1,9-diketones via the route shown in equation 60. For our first attempt to synthesize acetonyl Meldrum's acid (22), Meldrum's acid (2) was reacted with chloroacetone in the presence of one equivalent of pyridine. The reaction mixture was filtered after 2“ hours and a solid was isolated. 1H NMR analysis showed the solid to be N-acetonyl pyridinium chloride (29; Eq 63). none formed 0 O 0 pyr, 25°C C1 0 ’+/’ N‘\ 24 hrs 0 0 | 25 / 26 Results of related experiments to synthesize acetonyl Meldrum's acid (25) are shown in Table I. The best yield of 22 (87.91) was obtained with 1.1 equivalents of triethylamine in diethyl ether. We next attempted to synthesize acetonyl acetyl Meldrum's acid (g1) from acetonyl Meldrum's acid (_2_5; Eq 611). Results of experiments to synthesize 21 are shown in Table II. The best yield of 21 (99.51) was obtained with 2.0 equivalents of pyridine in methylene chloride. base /“\ fl , o .22 + C1 X (64) 0 40 Table 1. Preparation of Acetonyl Meldrum's Acid (£2) 0 base _9_ + A 1.0! > E c1 Conditions Solvent Base(equiva1ents) (0°C, Time) Yield(1) THE pyridine (1.0) 25°, 21 hr N.R.a’b cn2c12 pyridine (1.0) 25°, 21 n5 N.R. MeOH NaOH (1.0) 25°, 2 hr; 10° , 8 hr 67.1 032012 Et3N (1.0) 25°, 18 hr 25 THE EtZN (1.0) 25°, 12 hr 56.1 cac13 8131 (1.0) 61°C, 12 hr N.R. 2120 3131 (1.0) 25°, 21 hr 78.1 THE 2131 (1.0) 25°, 21 hr 62.9d THE 2131 (1.1) 25°, 21 hr 77.1 EtZO Et3N (1.1) 25°, 21 hr 87.1 aN.R. = No Reaction bIsolated‘gg cReflux temperature d1.1 equivalents of chloroacetone used 41 Table II. Reaction of Acetonyl Meldrum's Acid with Acetyl Chloride 0 base 4% 21- 25 + 1.0 M anditions Solvent Base(equiva1ents) ( C, Time) Yield (1) CH2012 pyridine (1.0) 0°, 10 min; 25°,10 min (2.63)3 032012 pyridine (1.0) 0°, 1 hr; 25°, 1 hr 92.5 Et20 pyridine (1.0) 0°, 1 hr; 25°, 1 hr N.R.b 3620 Et3N (1.0) 0°, 1 hr; 25°, 1 hr (0.61 g)° 0112012 pyridine (2.0) 0°, 15 min 99.5d a1H NMR revealed 3:1 g1:g§ bN.R. = No Reaction O1H NMR revealed 132.21?§§ d1.1 equivalents of acetylchloride used 42 During our study to optimize the yield of 21, we noticed that a small amount of Q was present in the reaction mixture after acidic work- up. We suspected that 21 hydrolysed to g; in the presence of water. To check this assumption, one equivalent of pyridine was added to a chloro- form-d solution of 25 at 0°C. The solution was stirred for 15 minutes and 1.1 equivalents of acetyl chloride were added. Aliquots were periodically removed from the reaction mixture and analyzed by 1H NMR. Analysis showed a 1:1 ratio of glsgg after 1 hour. A second equivalent of pyridine was added and 1H NMR analysis revealed a as the only component present. Deuterium oxide (D20) was added to the NMR sample and analysis indicated the presence of appreciable amounts of g§:g (Eq 65). Armed with this knowledge, we f0und the best yield of 21 (99.51) was obtained when the organic phase of the reaction mixture was immediately separated from the aqueous phase after quenching. 020 Yro (6 ) 7 4%} 5 2— 0 0 0’1/ We next attempted to prepare acetonyl benzoyl Meldrum's acid (28) from acetonyl Meldrum's acid (g2) (Eq 66). As shown by the results in Table III, we were unable to synthesize g§.under a variety of’conditions. We also attempted to react .35 with crotonyl chloride (Eq 67), isobutyryl chloride (Eq 68), and pivaloyl chloride (Eq 69). However, all three acid chlorides failed to react under the conditions shown in equation 67. 43 -. 0 pyr(2eq) (66) 25 + Ph’uK ‘——""“2 -—- 01 CDc13 0°c, 10 min . 0 _2_5_ + M ————> (67) c1 0 0 O 0 )K (68) _____;; - .22 + Cl 0 0 0 0 O 0 0* (69) .32. + c1 -—————5 0 0 0 We next attempted to cleave the Meldrum's acid ring of 21 to give the 1,1-diketone,gg (Eq 70). One equivalent of BE3'OEt2 was added to a + 0 31- ROH ‘7 0 + +CO2 (70) 002R 44 Table III. Reaction of Acetonyl Meldrum's Acid with Benzoyl Chloride 25 + z/u\~ lffifi___> 28 Ph CI 1.0 11 — anditions Solvent Base(equiva1ents) ( C, Time) Yield(1) 0112012 pyridine (1.0) 0°, 1 hr; 25°, 1 hr N.R.a 0112012 Et3N (1.0) 0°, 1 hr; 25°, 1 hr N.R. £620 Et3N (1.0) 0°, 1 hr; 25°, 1 hr N.R. 0112012 pyridine (2.0) 0°, 10 min; 25°, 10 min N.R. 00013 PnN(CH3)2 (1.0) 0°, 30 min N.R. 00013 (iPr)2NEt (1.0) 0°, 30 min N.R. aNo Reaction 45 chloroform-d solution of 21. Water (1.7 equivalents) was added and the reaction was stirred overnight. 1H NMR analysis of the reaction mixture revealed the) presence of acetone and 2,5-dimethyl furan (Eq 71). No 1) BF3 OEt2 0 0001 27 3 __ 9 /u\ + A (71) 2) H20 0 reaction occured when the reaction time was shortened to 30 minutes. When the amount of BF3'OEt-2 added was decreased to 0.1 equivalents, analysis revealed unreacted _2_'_7_ and a small amount of acetonyl Meldrum's acid (22). Finally, when the reaction was repeated with trifluoroacetic acid (instead of BE3°OEt2), analysis revealed acetonyl Meldrum's acid (£2) as the only product. No acetonyl acetyl Meldrum's acid (a) was present in the reaction mixture. Discussion Acetonyl Meldrum's acid (_2_11) failed to react with chloroacetone after 11 hours at room temperature (Eq 72). The acidity constant of _2_11 is no 0 0 Y0 A pyr(1.0eq> '1' i \/ x (72) 0 0* Cl 24 /\ 7 £2. 46 surely less than the pKa of Meldrum's acid (2; pKéVS); thus the anion of 21 is probably an extremely weak nucleophile and unable to react with chloroacetone. Meldrum's acid (2) itself, however, reacts smoothly with chloro- acetone to give acetonyl Meldrum's acid (Q) in good yield provided triethylamine is used as base (Eq 73). Acetonyl Meldrum's acid (22) was _ O 0 O O K + A Et3N(1.1eq)\ 0 7 O 0 01 21:20 0 (73) 0 0 9 _2_5 not formed when pyridine was used as the base; instead N-acetonyl pyridinium chloride (gg) was isolatedx The failure of the reaction under 0 01" +N |\ £2 3. these conditions is probably due to the lower basicity of pyridine (pKa 5.23) compared to triethylamine (pKa 10.7). Pyridine reacts reversibly with Meldrum's acid (pKa 5.2, Eq 711), thus significant amounts of pyridine are available for reaction with chloroacetone to form 2_6_. Triethylamine, on the other hand, reacts with Meldrum's acid (2) almost irreversibly (Eq 75) and enolate 39 is the only nucleophile present in solution for reaction with chloroacetone (Eq 76). 47 0 0 .0 / X + I -——-> 0 I (71) 0 \N 5‘ 0 +11 0 0 n 9 .311 + ———> 75 g + 2:311 e- 10 + Et3NH ( ) 0 g + A -—-—> g._5_ (76) 01 Acetonyl acetyl Meldrum's acid (21) was formed in nearly quantita- tive yield by reaction of acetonyl Meldrum's acid (25) with acetyl chloride (1.1 equivalents) and pyridine (2.0 equivalents; Eq 77). The O -----> ——- (77) 25 + CH Cl -—- Cl 2 2 reaction was repeated with chloroform-d as the solvent so the reaction could be fOllowed by 1H NMR. The reaction was initiated with only one equivalent of pyridine present. 1H NMR analysis revealed a 1:1 ratio of 2_7_:2_5_. A second equivalent of pyridine was then added and analysis 48 showed mainly 2_7 with only a trace amount of 2_5_ present in the reaction mixture. The second equivalent of pyridine is probably needed as an 1101 trap. A few drops of 020 were added to the NMR sample and analysis revealed a considerable amount of _2_7_ had hydrolysed to 25-d (Eq 78). D 1) pyr \ .gz + 020 0001 ~+7 0 (78) 3 0 0’12’ 0 25-d Here the second equivalent of pyridine present from the original reaction mixture probably causes a base catalysed hydrolysis of _2_1. Armed with this knowledge, we found the best yields of _21 (99.51) were obtained by quenching the reaction mixture with 2 _M HCl and immediately separating the organic phase. Benzoyl chloride (31), crotonyl chloride (_3_2), isobutyryl chloride (33), and pivaloyl chloride (31) all failed to react with acetonyl 0 0 0 XOR Ph 01 //fl/“01 \\T/u01 01 31 32 33 34 Meldrum's acid. The failure of these acid chlorides to react with acetonyl Meldrum's acid is possibly due to steric factors. In the acylation of acetonyl Meldrum's acid a quaternary carbon is formed (Eq 79) and as the acid chloride becomes bulkier, the reaction is less likely to occur. 49 O 0 o 0 j\ 25 + R/IK01 __~,base R $0 (79) 0 0 Finally we attempted to cleave the Meldrum's acid ring of _21 to 23 complete a synthesis of a 1,4-diketone. Pihlaja and Ketola have reported the base catalysed decomposition of disubstituted Meldrum's acid derivatives (Eq 80). Base catalysed hydrolysis of 21, however, led 0 0 {)7} 011 ' 0 R - R R {gox "L—H: 1(1ng -—-> _Ozz—R + A + C02 (80) O to cleavage of the acetyl group (Eq 78), not the Meldrum's acid ring. This result is not surprising since attack of the base at the acetyl function would lead to the formation of the stable anion _3_§_ (Eq 81). Z]- Pyr (81) DO 3 SO Yonemitsu22 has reported the acid hydrolysis of 5-substituted Mel- drum's acid derivatives (Eq 82). To hydrolyse a we chose boron trifluoride as the acid because of its high Lewis acidity. 0 ' 0 O R OH R R X ——-—> \/"\OR' + /u\ +002 (82) Compound 21 gave acetone and 2,5-dimethyl furan as products after reacting with one equivalent of BE3'0Et2 and water for 24 hours. These products can be explained by the acid hydrolysis of the Meldrum's acid ring (Eq 83) followed by cyclization of the intermediate diketone 3§_(Eq 31). (}-8F -1 o 110 ,3 3 o . BF3 °Et2 \ “1 00211 .21 320 7 2) ——> (83) 0 ° 1 2.9. \0 “'6 0—11 To (1 0 5 fl“ 4 A (81) '0 O H 51 If the amount of BF3°0Et2 used is decreased to 0.1 equivalents, compound 11 remains largely unreacted. Thus, it appears that a full equivalent of BEB'OEt is needed to hydrolyse the Meldrum's acid ring. 2 Reaction of.g1 with trifluoroacetic acid in the presence of water gave exclusive formation of acetonyl Meldrum's acid (2;) after 24 hours (Eq 85). Thus it appears that the mode of hydrolysis which compound g1 O CF3C02H \ O 27 ’ O (85) — 0 25 will undergo, i.e. equation 83 verses equation 85, depends on the acid used. This brief preliminary study indicates the use of a protic acid will hydrolyse the acetyl function whereas the use of an aprotic Lewis acid will hydrolyse the Meldrum's acid ring. In conclusion, the synthesis of 1,1-diketones from the reaction of Meldrum's acid with c-haloketones and acid chlorides is not promising. However, in reacting chloroacetone with Meldrum's acid a-y-keto ester was formed (Table I). If conditions could be found to hydrolyse the Meldrum's acid ring, the reactions of'antl-haloketone with Meldrum's acid may be useful as a general synthetic approach to y-keto esters. Experimental I. Materials Meldrum's acid, chloroacetone, pyridine, triethylamine, diethyl ether, and THE were prepared or obtained as described in Chapter I. N,N- Diisopropylethylamine, N,N-dimethylaniline, and all acid chlorides were 52 purchased from Aldrich and used without further purification. Boron trifluoride etherate was obtained from Baker and distilled prior to use. Trifluoroacetic acid was purchased from Alfa. Acetyl Meldrum's acid was obtained from Rob Tirpak who prepared it from Meldrum's acid and acetyl chloride according to Yonemitsu's procedurezz II. Reaction of Acetyl Meldrum's Acid (24) with Chloroacetone A. Preparation of Acetyl Meldrum's Acid Acetyl Meldrum's acid was obtained from Rob Tirpak of this laboratory and used without further purification. B. Reaction of 24 with Chloroacetone A 25 ml round-bottomed flask equipped with septum inlet, magnetic stirrer, and mercury bubbler was purged with argon and charged with 1.86 g (10.0 mmol) acetyl Meldrum's acid and 10 ml CHZClZ. (0.80 ml; 10 mmol) was added and the reaction was stirred for 15 minutes. Pyridine Chloroacetone (0.80 ml; 10 mmol) was added and the reaction was followed by GLC analysis (column temperature 110°C). No chloroacetone reacted after 4 hours. III. Reaction of Acetonyl Meldrum's Acid with Acid Chlorides A. Preparation of Acetonyl Meldrum's Acid (25) A 25 ml round-bottomed flask equipped with septum inlet, magnetic stirrer, reflux condensor, and mercury bubbler was purged with argon and charged with 1.114 g (10.0 mmol) Meldrum's acid and 5.0 ml anhydrous diethyl ether. Triethylamine (1.53 ml; 11.0 mmol) was added giving a milky suspension which was stirred for 10 minutes. While 53 stirring two layers formed; the bottom layer was clear and the top layer was milky white. Chloroacetone (0.80 ml; 10 mmol) was slowly added and the reaction mixture began to reflux. A white precipitate formed approximately one minute after the chloroacetone was added. The reaction was stirred overnight, quenched with 5 ml H20, and filtered. The white precipitate isolated was washed with pentane and dried giving 1.75 g (8.74 mmol; 87.4 1 yield) of 22. (Results of additional attempts to synthesize g; are summarized in Table I). 1 H NMR(CDC1): 61.9 (s, 6H), 62.3 (s, 3H), 53.3 (d, 2H), 63.9 (t, 1H) 3 B. Reaction of 25 with Acid Chlorides The reaction of 25 with acetyl chloride in CHZCl2 is represen- tative: a 50 ml round-bottomed flask equipped with septum inlet, magnetic stirrer, and mercury bubbler was purged with argon and charged with 2.00 g (10.0 mmol) 22 and 10 ml CHZCIZ. mmol) was added and the reaction mixture was cooled to 0°C. Acetyl Pyridine (1.61 ml; 20.0 chloride (0.79 ml; 11.0 mmol) was added dropwise giving a clear yellow solution. The reaction mixture was stirred 15 minutes at 0°C, quenched with 10 ml 2 g HCl, the organic phase immediately separated, dried (MgSOu), and concentrated giving 2.41 g (99.5 mmol; 99.51) of a slightly yellow solid which was shown to be g1. ‘11 NMR (00013): a 1.7 (s, 3H), 51.9 (s, 3H), 6 2.2 (s, 3H). 6 2.3 5 (s, 3H), 5 3.6 (s, 28) 54 C. A Study of the Reaction of 25 with Acetyl Chloride A 25 ml round-bottomed flask equipped with septum inlet, magnetic stirrer, and mercury bubbler was purged with argon and charged with 1.00 g (2.5 mmol) g; and 2.5 ml CDC13. The reaction mixture was cooled to 0°C and one equivalent of pyridine (0.20 ml; 2.5 mmol) was added. Acetyl chloride (1.98 ml; 2.75 mmol) was added and the reaction was stirred at 0°C. Aliquots were periodically removed and analyzed by 1H NMR. After 1 hour at 0°C, analysis revealed a 1:1 ratio of 2112;. After 1 hour at 0°C, a second equivalent of pyridine (0.20 ml; 2.5 mmol) was added. Analysis (1H NMR) revealed 21 as the only component present in the reaction mixture (besides pyridine). Deuterium oxide (approxi- mately 3 drops) was added to the NMR sample and analysis indicated the presence of appreciable amounts of 25-d. IV. Attempts to Cleave the Meldrum's Acid Ring of 27 A. Reaction of 27 with BF3'Q§£2 A 5 ml round-bottomed flask equipped with septum inlet, magne- tic stirrer, and mercury bubbler was purged with argon and charged with 0.242 g (1.00 mmol) 21 and 1 ml CDCl One equivalent of BF ’OEta‘2 (0.123 2' 3 ml; 1.00 mmol) was added and the reaction was stirred for 5 minutes. Water (0.03 111; 1.7 mmol) was added and the reaction was stirred overnight. Analysis of the organic phase indicated the presence of acetone and 2,5-dimethylfuran. No a was present in the reaction mixture. 55 ‘H NMR (00013): 52.2 (acetone), 51.1 (s, 6H), 36.1 (bs, 2h) The reaction was repeated using 0.1 equivalents of BE3'0Et2. Analy- sis (1H NMR) revealed unreacted g1 and a small amount of 2;. B. Reaction of 27 with Trifluoroacetic Acid A small amount of 21 was dissolved in CDCl in an NMR tube. 3 Approximately 3 drops of trifluoroacetic acid and 6 drops of D20 was added to the NMR tube, the mixture was shaken, and analyzed by 1H NMR. Analysis revealed no reaction had occurred. Approximately 6 drops of acetone-d6 was added to the NMR tube, the tube was shaken and allowed to stand overnight at room temperature. 1H NMR analysis revealed g; as the only Meldrum's acid derivative present in the reaction mixture. 55 1H NMR (CDC13): 62.2 (acetone), 154.1 (s, 6H), «56.1 (bs, 2H) The reaction was repeated using 0.1 equivalents of BE3‘OEt2. Analy- sis (1H NMR) revealed unreacted g1 and a small amount of gg. B. Reaction of 27 with Trifluoroacetic Acid A small amount of 21 was dissolved in CDCl3 in an NMR tube. Approximately 3 drops of trifluoroacetic acid and 6 drops of 020 was added to the NMR tube, the mixture was shaken, and analyzed by 1H NMR. Analysis revealed no reaction had occurred. Approximately 6 drops of acetone-d6 was added to the NMR tube, the tube was shaken and allowed to stand overnight at room temperature. 1H NMR analysis revealed 22 as the only Meldrum's acid derivative present in the reaction mixture. B IBLIOGRAPHY 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. BIBLIOGRAPHY G. Baltazzi and L.H. Krimen, Chem. Rev., Q}, 511 (1963). P. Basshard and C.R. Eugster, Adv. Het. Chem., 11, 377 (1976). J.E. MoMurray and J. Melton, J. Amer. Chem. Soc., 23, 5309 (1971). T.L. Ho, Synth. Comm., 1, 265 (1974). T. Mukaiyama, K. Narasaka, and F. Furusato, J. Amer. Chem. Soc., 23, 8611 (1972). E.J. Corey and L.S. Hegelous, J. Amer. Chem. Soc., g1, 4926 (1969). M. Miyashita, T. Yanami, and A. Yoshikoshi, J. Amer. Chem. Soc., 213.. 1679 (1976). Gault and Saloman, C.R. Acad. Sci., 187, 755 (1922). A.W. Dox and B. Houston, J. Amer. Chem. $091,,flp, 252 (1924). M. Miyano and C.R. Dorn, J. Org. Chem., 31, 268 (1972). G. Stork and M.E. Jung, J. Amer. Chem. Soc., 29, 3682 (1974). R.M. Jacobson, R.A. Raths, and J.H. McDonald III, J. Org. Chem., 12, 2545 (1977). R.M. Jacobson, A. Abbaspour, and G.P. Lahm, J. Org. Chem., 43, 4650 (1978). T. Kato, H. Kimura, and K. Tanji, Chem. Pharm. Bull., g9, 3880 (1978). G. Stork, P. Rosen, N. Goldman, R.V. Coombs, and J. Tsuji, J. Amer. Chem. Soc., Q1, 275 (1965). M.W. Rathke and A. Lindert, Syn. Comm., 8, 9 (1978). J.E. Dubois and M. Dubois, Tet. Lett., 13, 4215 (1967). E. Negishi and M.J. Adacavage, Tet. Lett., 845 (1979). E. Negishi 31.‘§1., J. Chem. Soc. Perkin II, 1225 (1978). 56 20. 21. 22. 23. 57 M. Tamura and J. Kochi, §1231 303 (1971). R.M. Kellogg and C.B. Trootwijk, J. Chem. Soc. Chem. Comm., (1977). 932 Y. Oikawa, K. Sugano, and O. Yonemitso, J. Org. Chem., 13, 2087 (1978). a) H. McNab, Chem. Soc. Rev.7, 345 (1978). b) X. Pihlaja and J. Ketola, Finn. Chem. Letters, 123 (1976).