v‘ P'fi’gfl" if". AME \ t '1. >1 ¥ .T_1.~:‘$“ ‘33:? . .‘gg. _ - ‘ . “(-5: .' ‘6 \ Egg." 7:: '. - ‘65 9%“22‘3' ,_ _.- x. "-v‘pfii‘é‘fi ‘ ' 2‘: «LA! ‘- ' l. «EV 2‘7". .7. ‘ . ~ x. _ '&\ ' '* ’ “big-3 I ’r‘g‘ffl f». t. 'Q‘E‘ 3. '1‘. .‘a ‘ . ‘23:“! ”1:;- 1:')' ,3! 1:5: ”+3, 4"} #12:? {REV}? ’u r"" , - " 3&1}- . . I " '17:.713’ ; 3’ V , ,~':-§.‘.-fl74 .- - " t . .l' l". . It: . {"7r‘!‘""r;£ r' ’ "v; .. . m. "’ 'r: .11 ' ‘ 28’s,? ' an“ m .,. vaum, “‘1’, 4.1;.” ‘-" “I 703/9" Fl}? .J’ll I 1.7! (jam! '4'! 1" . 3,‘ {5'4" 1"": / '- 31 ‘. ;-:.u‘!|-\ ‘- .r. “(#33 4‘ I272 (8% MIC C(‘Hl GAN STAYE HHHHII‘IHW fllIllHrHl 31293 00577 0114 ll! LIBRARY Michigan State UniversitL—L This is to certify that the dissertation entitled CH:"BT:IR I.Tiiil ALKYLATION OF E\ LLITLQS ‘IID I‘IETILLU- LN“ NUS. Clri rimIPTIJH II. RL . IICTIOI OF BRO}? EIHIV: N233 T;i ORuJNuILIILLIT I I“GIII“|‘J.'_'1L fiLfliTa-"RYL ITIUN unh— Uh KLIJUII.) CHI "3R III. LRYLI‘ITION OE LIXL-LIATII'3 MID "If" «LLLfi’III INES .wITIi presentedby iRYL IOL IL " Demetris Vogiazoglou has been accepted towards fulfillment of the requirements for Bq.D. Chemistry degree in WW Major professor 18th Of March 1988 Date MS U is an Affirmative Action/Equal Opportunity [mutilation 0' 1 2771 MSU LIBRARIES =— U" RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. CHAPTER I THE ALKYLATION OF ENOLATES AND METALLOENAMINES CHAPTER II REACTION OF BROMDENAMINES WITH ORGANOMETALLIC REAGENTS. ZEEIALPHA-ARXLATION OF KETONES CHAPTER III ARXLATION OF ENOLATES AND METALLOENAMINES WITH ARXL IODIDES By Demetris ngiazoglou A,DISSERIATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1988 ABSTRACT CHAPTER I THE ALKYLATION OF ENOLATES AND METALLOENAMINES CHAPTER II REACTION OF BRDMDENAMENES WITH ORGANOMETALLIC REAGENTS. THE ALPHAéARXLATION OF KETONES CHAPTER III ARXLATION OF ENOLATES AND METALLOENAMINES WITH ARXL IODIDES By Dimitris VOgiazoglou I. The reaction of enolates and.metalloenamines of ketones and aldehydes with alkylating agents has been investigated. Imines resulting from the condensation of carbonyl compounds with 3-dimethylaminopropylamine have been deprotonated and.alkylated at —7E°to give, after hydrolysis, the corresponding substituted carbonyl compounds in good to excellent yields. 11. .A variety of enamines have been brominated with Brz in the presence of triethylamine at -78° in THF. The resultant bromoenamines are found to react with diarylcuprates to give, after hydrolysis, alpha-arylketones in generally good yields. III. Ketones have been arylated yia their metalloenamines with aryl iodides at refluxing temperatures in cyclohexane to give, after hydrolysis, alpha—arylketones. W3 I..wish to express my heartfelt gratitude to Dr. Michael Rathke for making my stay at MSU a most pleasant experience. I also wish to thank members of the Rathke group for making the lab a fun place to be: Michael, Rick, Paul and Ezzedine. I would also like to thank Lisa, Mark, Nick, John, Jeff, Mike Kondylis, MIke Tornaritis and Ennanuel for creating an environment in which I could feel at home. Appreciation is also extended to Paolo, Luka, Eleni, Filippo, Sofia, Stephanos, Laura, Yiannis, Rania, Potis and D. Dakos for making leisure time fun. " The financial assistance provided by MSU is gratefully acknowledged. Finally, I wish to thank my family, Nikos, Athena, Lori and Vicky without whose love and support this thesis would have been impossible. iii TABLE OF CONTENTS LIST OF TABLES CHAPTER I. THE ALKYLATION OF ENOLATES AND METALLOENAMINES Introduction Results and Discussion Experimental Materials Preparation of 2-Butylcyclohexanone 3-Dimethylaminopropylimdne of Cyclohexanone 3-Dimethylamdncpropylimine of Cyclopentanone 3-Dimethylaminopropylimine of 2-Heptanone 2-sec-Butylcyclohexanone Z-Butylisophorone CHAPTER II. REACTION OF BROMDENAMINES WITH ORGANOMETTALIC REAGENTS. THE ALPHAHARILATION OF KETONES Introduction Results and Discussion Experimental Materials Preparation of 2-Phenylcyclohexanone 2-Phenylcyclopentanone 2-Phenylcycloheptanone 2-Phenyl-3—pentanone 2-Phenylacetophenone 2-Methy1-6-phenylcyclohexanone 22Methyl-2-phenylcyclohexanone 2-(2+Methoxyphenyl)cyclohexanene CHAPTER III. ARYLATION 0? warns AND METALLOENAMINES WITH ARIL IODIDES Introduction Results and.Discussion Experimental Materials Preparation of 2-Phenylacetophenone 2-(4-Methyl)acetophenone 2-(4-Methoxy)acetophenone 2-(1,3,5-Trimethyl)acetophencne 2-Phenylisobutyrophenone 2-Phenylpropiophencne 2-Phenylcyclohexanone 2-Phenyl-3-pentanone BIBLIOGRAPHY iv TABLE 10 LIST OF TABLES Butylation of the Lithium Enolate of Cyclohexanone Comparison of Deprotonation and Butylation of Dimethylhydrazine and Dimethylaminopropylamine Imines of Cyclohexanone Alkylation of Metalloenamines in was Reaction of Branoenamine of Cyclohexanone with PhM Bromination-Phenylation of Enamines with thCuLi Reaction of the Metalloenamine of Acetophenone (2 eq) with PhI (1 eq) Under a Variety of Solvents Arylation of Various Metalloenamines of Acetophenone (2 eq) with PM (1 eq) The Effect of Various Initiators on the Reaction of the n-Butylimine of Acetophenone (2 eq) with PM (1 eq) Reaction between 1 eq of Imine and 1 eq PhI.GI.C Yields Reaction of Metalloenamines with Aryl Iodides. Isolated Yields PAGE 13 15 33 35 SS 56 58 59 60 CHAPTER I THE ALKYLATION OF MLATES AND WW5 INTRODUCTION One of the most important reactions in organic synthesis is carbon-carbon bond formation, by reaction of a carbon electrophile with a carbon nucleophile: C—C A very useful method for generating carbon nucleophies is removal of an acidic proton by a base (eq 1). l . l “'7‘“ +13 --- -—C" + H8 (1) A carbonyl group bonded to the carbon nucleophile can stabilize negative charge by resonance, and hence is a key function for the formation of carbon nucleophiles. Retones and aldehydes are examples of compounds bearing a carbonyl group fruit which stabilized anions can be generated. These anions, formed by deprotonation of a carbon alpha to a carbonyl group by a base, bear most of the negative charge on oxygen and are normally referred to as enolates, as illustrated for cyclohexanone, (QC! 2). o o 0" :B .. In an ideal alkylation method, the enolate is formed and alkylated rapidly, in a high yield, with no side reactions (eq 3). ’ f: O (3 However, besides the above reaction, there are side reactions ) which limit the utility of enolate nucleophiles. The most common side reactions are: l. Condensation: C) (5' B a“: + 6 -- 2. Alkylation at oxygen: 3 6‘: w as 3. Dialkylation due to proton transfer from a monoalkylated product to enolate: + ig-OOO The condensation reaction to produce aldol-type products becomes a serious problem when insoluble bases such as sodium hydride, or weak bases, such as metal alkoxides, are used to generate slowly or as a state of equilibrium conditions.1 However, self-condensation of the enolate with starting material canfibe minimized by employing strong, organic soluble bases. Lithium diisopropylamide (LDA) is the most useful, since it can be easily prepared.from butylithium and diisopropylamine, and.moreover, is sufficiently strong to generate enolates quantitatively (eq 4) . THF BuLi + >-N—< 4. BUH + >—N H 15 min., Li-< (J 0 o . on (4) THF 4- >‘N. —- + >N ‘ L.-< 15 min., 6 H.< (J 0 Another advantage of LDA.is its steric hindrance. Consequently, it preferentially abstracts an alpha proton frat the carbonyl ccmpound, and side products resulting frat attack upon the carbonyl are usually not observed: Alkylation at oxygen is another side reaction. It is known2 that the ratio of carbon versus oxygen alkylation depends upon the alkylating agent, the leaving group, the solvent and the counter ion. For reactive alkylating agents such as MeI, PhCBzx, allyl-x, oxygen alkylation is a minor side reaction especially if lithium enolates are employed.3 Another side reaction that can lower the yield of an alkylation reaction is dialkylation due to proton exchange. It can become a serious problem with the less reactive alkylating agents4 such as i-PrI or BuI, due to the lower rate of alkylation vis-a—vis the higher one of hydrogen exchange (foz, page‘C). we decided to investigate the reaction between enolates and alkylating agents which, as mentioned.above are not very reactive. Our goal was to increase the yield of monoalkylated.product by decreasing the amount of oxygen alkylated.and.dialkylated.compounds. we chose butyl as a representative alkyl group of relatively low reactivity. According to Conia‘r butyl bromide is 380 times less reactive than allyl bromide and 1000 times less reactive than benzyl bromide in reaction with the enolate generated frat 2-methylcyclohexanone and sodium t- amyloxide. RESULTS AND DISCUSSION we started our project by generating the enolate of cyclohexanone and.examining its reaction with a variety of butylating agents. 2- Butylcyclohexanone 1 as well as the other byproducts, 2, 3,.1 (eq 5), were determined.by Gas Liquid Chromatography (GLC). O . 0U + LDA Bu _6 ; : 0 , 15min. THF (5) O OBu O OBU BU / BUBU Bu ‘ \ + + + l 2. §_ 4 Some representative reaction conditions are list‘d in Table 1. In the case of BuCl and BuBr no reaction occurs under our standard conditions. On the contrary, BuI reacts with the enolate of cyclohexane, especially in the presence of hexamethyl phosphoramide (HMPA). However, HMPA promotes the proton exchange leading to undesirable dialkylated products. The fourth butylating agent that we studied.was (BuO)2502. (BuO)2502 is more reactive than BuI. unfortunately, it leads to increased amounts of oxygen-butylated.product. The last butylating agent that we examined was BuOTs. This reagent proved to be less reactive than BuI, besides the fact that it gives low yields of oxygen and dibutylated products. TABLE 1. Butylaticn of the Lithium Enolate of Cyclohexanone in: TE! Butylating‘ Temp. Time HMPA l g 2 i I“ Agent °c hrs 1 eq Yield*% Yield*% Yield*% Yield": BuCl RT 12 Yes 0 O 0 0 Bus; RT 12 Yes 0 0 0 0 BuI RT 12 No 10 0 0 0 BuI RT 12 Yes 37 Trace Trace 4.5 BuI Reflux 1 Yes 33 0.5 2 ‘ 15 (BuO) 2302 Reflux 1 Yes 44 13 5 7.5 BuOTs Reflux 1 Yes 22 0.5 ' l 7.5 *Grc yields. From the above study we concluded that BuI is the best butylating agent for cyclohexanone enolate. Moreover, it is commercially available at low cost, and.it results in low yields of byproducts. However, this method is far fran being ideal. The reaction requires hexamethylphosphoramide (EMPA) -'a cancer suspected agent -, and the yield is only about 40%. The limitation of the number of leaving groups that we could use forced.us to switch our attention to more suitable enolate equivalents. A literature search revealed.that the nitrogen analogs of enolates - referred as metalloenamines - can be generated frat imines and alkylated with various alkylating agents of low reactivitys (eq 6). RNMgBr O EtMgBr C6 H6 A THF, A Imine Metalloenamine( R NMgBr (DEX THF, A CHE. Gas) 15 hrs. th-Butyl Magnesimn as a counterion has little to recannend it, since formation of the metalloenamine requires elevated temperatures and long reaction times. Furthermore, the alkylation step requires vigorous conditions. Lithiated enamines are more reactive than their magnesium analogs. Theycanbemadebytreatingtheiminewithastrongbase suchasBuLi or LDA. However, the deprotonation and alkylation steps still require long reaction times5 and temperatures above ~10°C (eq ‘7) . 10 + U LDA ———PhCHZBr H30 0 (7) H DME ; 250 H ' 435.20 1'3 hrs 3 hrs. Ph (90 °/o) One way7 to enhance the reactivity of lithiated enamines is to use cosolvent additives such as hexamethyl phosphoramide (HMPA) or N, N,N' ,N'-tetramethylethylenediuine (MA) . Unfortunately, nest of these additives are eaqaensive, hamful to the environment and cancer suspected agents. We thought that the rate of formation as well as the reactivity of metalloenamines might be enhanced with an imine which bears an -N(C33)2 group as a chelating agent. These imines can be made quantitatively by refluxing ketones or aldehydes with dimethylaminopropylamine in refluxing benzene for 3-4 hours (eq 8) . O ' N|/\/‘N< . H N/\/\N/ CSHG 2 \ +H 0 (8) A Z 1]. We discovered that these imines are very reactive. They can be lithiated with BuLi or LDA at 48°C in one hour (eq 9). t N‘ <- 4.: N/\/‘N/ N U I \ . BULI or LDA (9) THF, -78 1 hr. The resulted metalloenamine is also very reactive. We found that the metalloenamine of cyclohexanone can be butylated with BuI at -78°C in 15 min, (eq 10) : Na - .5\ N Li 0 Bul 3” 1' 10 o ( ) THF.-78 1 hr. The butylated imine was hydrolyzed in slightly acidic conditions (PH.5-6) at roan temperature (eq 11) . 12 NMN< + O ' Bu +50 Bu RT,1hr. 2 \ The -N(CH3)2 group accelerates lithiation probably by chelating the lithium metal and promoting the formation of monomeric reagents. The rapid.butylation can be explained by presuming that the -N(CH3)2 group chelates lithium generating a more "naked! metalloenamine.3 The parent imine was then compared with the imine of dimethylhydrazine which is considered the most reactive of conventional imines. The dimethylamincpropylamine imine of cyclohexanone is more reactive compared.to dimethylhydrazine imine of cyclohexanone towards deprotonation as well as butylation. (Table 2). 13 Table 2. Comparison of Deprotonation and Butylation of Dimethylhydrasine and Dimethylaminopropylamine Imines of Cyclohexanone. Imine Temp.°C Temp-°C % yield (Deprotonation (Butylation (GLC) Time) Time) . M/vu/ I \ (:3 -78° (1 hr) ~78° (1 hr) 87 w’vw/ \ (:3 0 (1 hr) -78° (1 hr) 90 O ~7a° (1 hr) -‘78° (1 hr) 3 In, 0 (1 hr) -78° (1 hr) 31 w/ (j -7e° (1 hr) 0 (1 hr) 59 " 0 (1 hr) 0 (1 hr) 91 14 The dimethylhydrazine imine fails to react at -78°C, where the dimethylaminopropylamine imine reacts at -78°C to give very good yield of butylated cyclohexanone. We then decided to study the reactions of : a) metalloenamines of various ketones with BuI and b) The metolloenamine of cyclohexanone with other alkylating agents. Our results are sunmarized in Table 3. For saturated symetrical ketones either BuLi or LDA can be used as a base. The choice of bases is however crucial in the case of unsyumetrical ketones . 2-Heptanone, for example, upon deprotonation with BuLi and subsequent alkylation with BuI produced after hydrolysis both of possible regioisomers, where LDA employment at low temperature resulted in the nearly exclusive formation of the metalloenamine from the less hindered side (eq 12) . 1 Base ’ NR ) O O o W .780 AMBU+ W (12) 2’81." BU 43°C T A 3)H30 8 Base L B BuLi 58% I 11% LDA - 71% 0% R: - (C82) 3N(CH3) 2 . 15 TABLE 3. Alkylation of Metalloenamines in THF Starting Alkylating Temp. °C Temp. °C Hydrolysis Produc:" material Base agent (Deprot. time) (Alkyl time) reaction time* % Yield (Isolated) OM BuLi BuI -78° (1hr) -78° (15 min) 1 hr 00 80 lo O'- BuLi Bu]: -7a° (1hr) -78° (1hr) 16 hrs CW 73 I - - 0 e17 W LDA BuI 78° (1hr) '78 (1hr) 1 MW 1 V3 LDA BuI -7e° (1hr) -78° (1hr) 1 hr Y 64 been LDA BuI -7a° ->RT(2hrs) -7a° (2hrs) 3 days Q0 73 O... BuLi i-BuI -78° (1hr) 0° (1hr) 16 hrs o 58 One BuLi W “78" (1hr) -'78° (15 min) 1 hr Q: 89 * The imines were hydrolyzed under slightly acidic conditions (PH~5-6) at roan temperature. ** After hydrolysis. :-(CHz)3M2 in all cases. It is surprising that the imine of isophorone gives the thermodynamic product (eq 13) LINE 0 LiNH NR 1 Bul "GOA o' c: LgA ) o a" 3.221 __..‘7°"‘" '73 (13) >3. 2 m + THF THF 2”'30 Kinetic Thermodynamic 16 When the deprotonation and alkylation of the imine were carried out at -78°C (without warming to room temperature, to prevent equilibration) the same product was observed in lower yields. Therefore it is not likely that the initial site of deprotonation is d' . It is conceivable that the dimethyl group provides enough steric hindrance so that in the case of imines the deprotonation occurs in the a site. The only report in the literature on the imine of isophorone5b involves an alkylation procedure which empoloys Man in THF containign 10% HMPA at refluxing temeratures overnight . The thermodynamic product was also isolated. Another point worth mentioning is the alkylation step temperature . At -78°C proton exchange between alkylated imine and metalloenamine is very slow, and no dialkylated products were observed. At 0°C however, hydrogen exchange becmes a competing reaction leading to dialkylated promote, In conclusion, a method for performing alkylations using imines of dimethylaminopropylamine has been described. Unlike conventional imines these imines can be deprotonated and alkylated with primary and secondary alkyl iodides at low temperatures to give, after hydrolysis, good yields of substituted carbonyl compounds. Other advantages of these imines is their low toxicity compared to dimethylhydrazine imines as well as the inexpensiveness of the starting amine. MIMI!“ THF was distilled frcsn sodium benzophenone just prior to use. The carbonyl compounds were obtained frcm Aldrich and distilled over CaHz. n-Butyl iodide was made from the butyl chloride and Hal in refluxing acetone. 2-Iodobutane was made from 2-butanol, phosphorous and iodide . 17 BuLi (1.6M in hexanes) as well as dimethylaminopropylamine were obtained from Aldrich. All the imines except the imine of propanol were made by the azeotropic method. 8 The imine of isophorone requires long reaction times (3-4 days). The imine of propanal was generated9 at low temperatures using NazSOq as a drying agent and was used inmediately, since it is unstable as a pure liquid. All reactions were carried out under an argon atmosphere. Gas chromatographic analyses were performed on a Hewlett and Packett 5880 chromatograph equipped with a 25 m x 0.25 rum column (stationary phase: fused silica, film thickness: 0.25 m) In MIR spectra were recorded on Hruker Ila-250 and Varian T-60 spoectrcmeters with CDC13 as the solvent. Low resolution mass spectra were recorded with a Finnegan 4000 GS/MS at 70 eV. W A flame-dried 50 ml round-bottom flask equipped with septum inlet and magnetic stirring was maintained under a positive argon pressure and charged with 5 ml of THF. 5 mole (1.02 ml) of the dimethyl aminopropylimine of cyclohexanone was added. The flask was cooled to - 78°C and 3.2 ml (5 mol) of Bull (1.6M) was added. The reaction mixture was stirred for 1 hr at -78°C and 5 nmol (0.58 mol) of BuI was added. The reaction mixture was stirred for 1 hr at -78°C and quenched with H20 at -78°C and 3N HCl was added until the pH~5-6. The organic layer was separated and the water layer was extracted twice with 10 ml CH2C12. The combined organic layers were dried with M9304 and concentrated in vacuo, and the residue was distilled under reduced pressure (bp: 78°C/3.5mm) to give 0.629 (80%) of 2-butylcyclohexanone. 18 WW 13 NMR (00013) 5 1.7 (m, 63), 1.8 (m, 23), 2.2 (s, 6H), 2.4 (t, 23), 2.5 (t, 43), 3.4 (t, 23) MS m/e 182 (3+), 111, 83, 72, 58(CH2N(CH3)2+), 42 IR: 2940, 2250, 1650, 1460, 900, 640 cm'1 I I W 13 NMR (00013) 5 1.7 (m, 23) 2 (t, 43), 2.1 (m, 63), 2.2 (s, 6H), 2.8 (m, 23) MS: m/e 168(M+), 85, 72, 58, (0323(ca3)2+) 48. IR: 2940, 2290. 1460, 900, 640 cm '1 Wm 18 NMR (00013) :8.1-1.5(m, 93), 1,8(t, 23), 1,8(s, 33) 2,2(5, 6H), 2,3(t, 23), 3,2(t, 23). us: 198(M+), 140, 98, 84, 71, 58(0323(033)2+), 42. IR: 2910, 2250, 1650, 1460, 900, 630 curl Wrens 13 ma (M13) 61 (s. 63). 1,8(m, 5H) 2,2(m, SH), 3,3(m, 2H), 6(d, 1H). MS: 223(M+), 151, 136, 72, 58(C32N(CH3)2+). 42 IR: 3160, 2950, 2250, 1630, 1460, 880, cm'1 WM '31“: (CDC13) 0.9-1.2(10, 98) 1,800. SH), 2,2(t, 28), 2,3(t, 13). MS 154(M+), 125, 98, 83, 70, 55. IR 2980, 2300, 1720, 900, 660, cm‘1 We: '3 M (CDC13)= 50.9‘1-203133): 1:9(d: 33): 2,2(m,6H) MS 194(M+), 141, 84, $5, 41. IR: 2960, 2250, 1660, 1640, 1380, 880 cm‘1 CHAPTER II REACTION OF WENAMINES WITH WC REAGENTS. THE ALPHA- ARYLATION OP KETONIS. 19 20 INTRODUCTION The introduction of an aryl group alpha to a carbonyl is a useful step in the synthesis of a variety of biologically and pharmacologically interesting compounds. Examples of these compounds are cephalotoxine 5, clorindione (an anticoagulant, .6) , ketamine (an anesthetic, 1) and ibuprofen (an anti-inflanmatory, fl) 21 Arylated carbonyl compounds are also very useful because they provide a wide range of compounds which have other functional groups alpha to the aromatic ring. For example conversion of the carbonyl to an amine group provides amphetamine compounds.1° (eq 15) (D 11113 IVFEB A” —- AP“ (15) (3Fk710 NaBHacN Our main goal in this project was to study the synthesis of arylated ketones. There are two synthetic approaches toward arylated ketones: a) the reaction of an enolate or its equivalent with electron deficient aryl species, (A) and b) the connection of an “enolate cation," with an electron rich aryl species, (3), (eq 16) . 1 (A) ,/J\s. + ‘N' ‘\\\\\\\ 0 km (16) 22 The most successful example of the first approach is the reaction of enolates with photogenerated aryl radicalsll (eq 17) . 0 N Br MO Rn hv & n! __ r 3' (17) + R'> MeBSiCH éVIgC) _ ———- (23) H9 l2 OSiMea (I X C) . l M63 SICHZOCHZX Allylic a-haloenamines 11 appeared to be interesting haloketone equivalents. Our primary aim was to develop a synthetic methodology for those compounds and then to examine their reaction with electron-rich aryl species. Our studies began by first attemting to react the enamine of cyclohexanone with PhLi in the presence of oxidizing agents such as Cu"2 salts according to the mechanism: 4 +2 +1 ° 4- CU + PhLi Cu 4» Ph + Li [:1 [:3 _ Ph . ‘ _ ‘8 27 ES) .4. Cu |+ Ph 4. +2 (:1 0p: + + PhH +Li H30 0] EN Ph + PhLi -—- 0 GP?) D .0 D O /.|\ /I\ 28 In the presence of Cu"2 salts such as CuClz and Cu (OAc)2 no phenylation was observed. However, in the presence of CuBrz the reaction yielded 38% of 2-phenylcyclohexanone. Most likely, the above mechanism does not operate in the presence of Cu“’2 salts. When CuBrz was used, .the reaction followed a different route: The CuBrz reacted with the enamine to give a bromoiminium salt which in the presence of PhLi gave, after hydrolysis the desired product, (eq 24). One experimental observation which supports the above statement was the formation of a white precipitate, presumably CuBr, when CuBrz was added to the enamine. The other Cu+2 salts gave no precipitate. O N Br O:I N 4- Br THF + 2CuBr + ZCUBI’ 2 RT 1 hr (24) Br- 9 :1 o 2: s 29 0 (r 4w 0 In a separate experiment, bromination of the enamine with CuBr; followed by hydrolysis of the branoiminium salt, yielded only 40% of 2- bromocyclohexanone. The cyclohexanone recovery was 53$ (eq 25) . (:2 q, :o W cm At this point, a literature search revealed that enamines can be + 2 CuBr2 1hr brominated with branine in good yields, to form branoiminium salts which can then be hydrolyzed to give o—branoketones 17 (eq 26) . O- O O O: The authors proposed the above sequence as an useful methodology toward a—bromoketones . RESULTS AND DISCUSSION It was apparent to us that a successful synthesis of a-aryl ketones from a-bromoenamines would depend in part on the availability of the precursor d-bromoenamines themselves. Extension of the reaction between enamines and branine, led to formation of bromoenamines in a high-yield sequence. By mnploying Et3N as a base we managed to obtain brcmoenamines in almost quantitative yields, (eq 27) . (:1 ° C) N Br I + N Br EtSN Br :" 0 = (27) -780 FC -78 -HT 90-95 % 31 This is an improved method for the preparation of bromoenamine compounds. The most ccrmnonly employed method is based18 on the reaction of bromoketones with a secondary amine (eq 28) . Br 0 Br 3:2! It is surprising that allylic bromide 11 are produced in this reaction. One would expect that, in the presence of triethylamine, the branoiminium salt 13, should be converted to a vinylic bromide 15 due to the higher acidity of Ha versus Hb (eq 29) . It is possible, however, that allylic strainl9 plays a more important role than hydrogen acidity. In short, the allylic strain theory suggests that pyrrolidine enamines such as that of 2-methylcyclohexanone mainly exist in the allylic form, 20 1.6, because in the vinylic form 11 there are severe steric interactions between the methyl and the amine a—methylene groups. These steric interactions can be reduced by rotation about the N-nc (spz) bond. However, this will reduce the nitrogen-lone pair interaction with the double bond. On the other hand, the enamine with an allylic methyl . (:3 {N} HE G];- IL 6: O ”2 fi 3 0 ~78 THF \ SEN ‘1 group would not suffer any allylic interactions because of the quasi- axial position of the methyl 1&- One can apply the ease arguments in the case of the two possible braaoenamines of cyclohexanone, especially- since the Van der Weals radiusof brmine (1.9 A) is about the same as the one of methyl (2.0A) .21 [I] U 90% 10% At any rate, no. studies revealed exclusive formation of the allylic branide in at least 90! yield in the cyclohexanone case. Ournextgoalwastostudythereactionofthebraaoenamineof cyclohexanone with a variety of nucleophilic phenyl species (PhM) capounds (eq 30). Our results are summarized in Table 4. 33 TABLE 4. Reaction of Bromoenamine of Cyclohexanone with PhM O 5:3 O O , --78 1) PhM P" _2_. 0 (30) 513 N. THF -23 ..25 3hrs 2) H 5 ‘9 a _ PM 19, Yield ’8 (GLC) PhLi 19 PhCuS(Caa)2 68 PhCu 78 thCuLi 94 It appears that thCuLi is a superior phenylating agent for the bromoenamine of cyclohexanone. PhLi gave only 19% of the product. This low yield could be attributed to a competing Pavorski-type product (eq 31) . That precinct has been isolated fran the reaction of haloenamines O [O] N ' N /\ {Br Ph (31) with bases. 22 34 Since the superiority of thCuLi as a phenylating agent was obvious, we focused our attention on the reaction of thCuLi with various ketones. These results are listed in Table 5. Enamines which exist only in the vinylic form such as that of acetophenone were phenylated in poor yields: (3 E N E :3 UPDZCUU O 1)Br2 ,3Et N A —‘6"'"" Br + PhA’ Ph Ph Ph 2)H 3O '78 19% The regioselectivity of the phenylation was briefly examined using the equilibrium mixture of morpholine enamines of 2-methylcyclohexanone 20 and 21 (eq. 32). Interestingly, thCuLi and PhCu gave different regioisomer major product: O O O O --O O 20 21 17% 68% PhZCULi 54% 18% PhCU 48 % 52% 35 TABLI S. Braintion-Phenylntion of Examines with thCuLi Enamine Organocuprate Product (isolated yield. 1) NO I Ph2"“"1.=0 (55) Ph Ph O (87) Co 9212 Cum. .0, (dam m N 0 3120.114 Ll. N Ph Gnu. I o o (72) 2 / \_./ a, m P“ / N 0 thcuu “)0 (19) 36 In neither case, however, is the ratio of the products related simply to the ratio of starting materials. Presumably, this is the result of the known ability of organocuprates to react with allylic substrates both with and.without allylic rearrangement 23, (eq 33). (33) SN2 8N2 (CH) CuLi so °/. 70% 3 2 (3F13FHJ 71496 za$°° we were unable to identify the major bromoenamine produced in the case of the enamine of 2-methyl cyclohexanone. It is likely that the major isomer is 22 which then couples with thCuLi by an suz reaction and.with PhCu by an Su2' reaction: 1) PhCu Br CH3 1)Ph2CUU ph 1 CH ; C313 Ph + + 2)HO 2)HO 3 3 22 37 In this chapter, a new synthesis of a-phenyl ketones from enamines and phenyl copper reagents in two high-yield steps was described. The sequence appeared to provide a simple method for the phenylation of symmetrical ketones in good yield. 31238110331111. THF was distilled fran sodium benzophenone still. Ether was taken from a freshly opened can. Triethylamine was distilled from calcium hydride before use and stored under argon. Phenyllithium was purchased from Aldrich. Copper (1) iodide was purchased from B. M. Industries and was used without purification. All the enamines except that of acetophenone were prepared by the azeotropic method. 24 The enamine of acetophenone was prepared using TiCl4 as a water scavenger.“ All reactions were carried out under an argon atmosphere. Gas chranatographic analyses were performed on a Hewlett and Packett 5880 chromatograph equipped with a 25m x 0.25 mm column (stationary phase: fused silica, film thickness 0.25 mm) . 18 NMR spectra were recorded on Bruker m—zso and Varian T-60 spectrometers with CDCl3 as the solvent and are reported as part per million in the 8 scale relative to internal MeqSi. Low resolution electron impact mass spectra were obtained with a Finnegan 4000 68/148 at ‘70 ev. WWW Bromine (5 mol, 0.8 g) was added dropwise to a solution of the morpholine enamine of cyclohexanone (5 1:11:01, 0.84 ml) and triethylamine (5.5 mol, 0.77 ml) in THF (5 m1) at -‘78°C. After 10 minutes the cold bath was removed, and the reaction mixture was stirred for 10 additional minutes. 10 ml of ether was added to the flask, and the ammonium salt was removed by filtration. The solution of brcmoenamine was injected into a flask containing 5 mol of thCuLi in 5 ml of THF. 38 Diphenylcopperlithium was prepared by the addition of 5 ml of PhLi (2M, 10 mol) to a'susxaension of CuI (5 mmol, 0.95 g) in rap (5 ml) at -230c. The reaction mixture was stirred for 2 hrs at -23°C and for 1 hr at room . temperature. Cold hydrochloric acid was added (10 mol, 5 m1) and after 6 hrs. the organic layer was separated. The water layer was extracted twice with 10 ml ether, the combined organic layers were dried with K2003, and the ether was removed. Silica-gel chromatography [hexane- ether (60:40)] gave 2-phenylcyclohexanone as a white solid, mp 55-560 (0.76 g, 87% yield). 18 mm (coma) 8 1.5-2.6 (m, 8 a). 3.4-3.8 (m, 1 H), 7.2-7.4 (m, 511). MS M/e 174 (14+, 3), 120(25), 105(100), 91 (8). 77 (64) . In a separate experiment, solvent was removed from the bromoenamine solution and the residue (1.4, eq 29) was examined: 1%! NMR (CD301) 81.6-3:1 (m. 10 H), 3.7 (t, 4 H), 4.9 (m, 2 8). Using a similar procedure, the following compounds were prepared. W. 18 ma (CDC13) 8 1.76-2.6 (m, 6 a), 3.1-3.5 (m. 13), 7.2-7.4 (m, S H). m We 161 (Ml- +1, 12) 160 (27), 104 (100), 91 (17), 78 (19), 77 (15). Wanna. 18 MR (CDc13) 81.2-2.8 (m, 10 a). 3.5-3.8 (m. 1 a), 7.2-7.4 (m, s a). us male 188 (m, 25) 117 (78), 104 (90), 91 (100), 84 (47), 78 (33), 77 (32), 51 (37). - - - . 18 ma (M13) 5 0.8 (t, 3 8), 1.3 (a, 3 u), 2.2 (q, 2 a), 3.6 (q, 1 a), 7.2 (s, s 3). us We 163 (14+ +1, 70), 154 (25), 105 (89), 104 (50), 91 (18), 77 (39), 57 (100). Wanna. 18 mm (@013) 5 4.2 (S, 2 a), 7.2 (S, 5 H), 7.4- 8.2 (m, 5 8), MS M/e 196 (14+, 1), 105 (100), 91 (8), 77 (51), 65 (10). 39 - - - 18 NMR (c0c13) 5 0.9-2.8 (m, 10 H), 3.5-3.9 (m, 1 H), 7.1-7.5 (m, 5 a). MS M/e 188 (14+, 53), 130 (78), H7 (81), 115 (33), 104 (74), 91 (100), 78 (23), 77 (27). - - - . 114 111411 (c0c13) 51.1-2.8 (m, 11 H), 7.2-7.4 (m, s a). 145 14/e 188 (14+, 63), 145 (80), 144 (97), 131 (97), 129 (48), 118 (80), 117 (71), 91 (100), 77 (40). WW. 18 1848 (cuc13) 5 1.6-2.8 (m, 8 a), 3.8 (s, 311), 3.7-4.1 (m, 114), 6.8-7.5 (m, 4a). 148 M/e 204 (14+, 79), 160 (49), 147 (100), 121 (44) 119 (30), 91 (75), 77 (22), 65 (23). owns. 111 ARYLATION OP EWIATES AND MEMBNAMINES WITH ARYL IODIDES 4O 41 INTKJDUCTION In the second chapter we focused our attention on arylation procedures which follow the general scheme, (eq 34). + O O A + Ar- A’ Ar. (34) Another approach towards arylated ketones is the combination of an enolate (or its equivalent) with electron-deficient aryl species, (eq 35). C5- 0 A 1 A: _.;. A»: (35) Unfortunately, reaction between an enolate and aryl halides does not take place (eq 36) . 42 CM 0 A + AfX A» /IK/ Ar (36) 0 °/c However, in the presence of certain bases such as 1781482 in N83, phenylation is observed in moderate yields 25 as illustrated for acetophenone, (eq 37) . 0 4 NaNH 0 2 . 2 Ph)l\ +1 PhBr 5’ Ph/IK/Ph (37) NH3 28 % The mechanism for the above reaction is believed to be addition of the ketone enolate to benzyne, the dehydrohalogenated derivative of branobenzene : Br NH- 43 +NH + Br- 44 Besides the moderate yields, other disadvantages of the benzyne mechanism are the requirement of a large excess of base and the lack of regioselectivity in cases of mono and polysubstituted aryl halides. .. Another source of electron deficient aryl species are arenediazonium salts. These species are known to react with silyl enol ethers to produce diazenes,2'7 (eq 38) OSiMe3 PhN 281:4 N=NPh (38) THF, ~078 61 % However, when pyridine is employed as a solvent, 23 the desired a- arylated ketones can be obtained, (eq 39). 081M931)PhN28F O h Ph /l\3 —- Ph /K/ P (39) / I \ N '1' 72°/ 2) H30 ° The authors proposed that the key step in the above reaction is formation of aryl radicals via decouposition of azo—type addicts: 45 ATNZ + ——- '1" I —. A} +N2+ + N T N N 2 l Ar OS'M ' l 93 . OSIMe3 R 1 + Ar —" R ° Ar 23 24 The yield varies according to the nature of R group in the silyl enol ether 23. When R is aliphatic the yield is quite low (31%), as illustrated for the silyl enol ether of 2 nonanene . 0n the other hand. when R is an aryl group the yield, depending upon the diazonium salt used, is between 58-720. The authors suggested that the resulting radical intemediate, 21, can be stabilized if R is an aryl group (benzylic-type radical). In the aliphatic case, the phenyl radical generated will not be so readily captured. and is eventually trapped by pyridine with the fonation of a substantial amoiint of phenylpyridines. 46 A second arylation procedure which is believed to proceed via a radical mechanism is the reaction of diphenyl iodonium salts with ketone enolates29 (eq 40) . OLi + o 1)Ph IPF ,THF P“ __21.6§_. (40) -78 , 15min, 4. 2)H o 3 42% It was proposed” that electron transfer fran the enolate to the iodonium ion gives radical pairs, which then can react either by radical displacement of the enolate on diphenyliodine or by coupling of the enolate and phenyl radicals: + - o ArIAr 0 4 A’\/'\ 47 One of the most useful arylation procedures was discovered by Bunnett31 He found that aryl halides undergo a light induced reaction with ketone enolates in liquid ammonia or DMSO (eq 41). ._.. O 41 A + PhBr 3 : P“ ( ) hv 86% The proposed radical chain mechanism.was termed.as the Saul reaction, The four following steps are involved: hv , Electron source +PhX > [PhX] + Residue [PhX] T. Ph + x- _ / 0 \-‘ O o ’,”l¥§§ O+ l3h :; ’/,/fl\\\~/,Ph \\ 1’ r’ C) ' \\-— (3 A,” + PhX 2K)“ + [PhX]- 48 The process is a chain reaction. The mechanism of photoionization is not known but an attractive possibility is that a charge transfer complex of nucleophile with substrate undergoes electron transfer from one moiety to the other upon interaction with a photon 32 (eq 42). ._ hv _- . PhX.Nu —-[PhX] + Nu (42) There are other ways to initiate the SRN1 mechanism. Alkali metals, 33 Fe” salts34 can serve as initiators in liquid ammonia. Alkali metals such as potassium dissolve in amonia to furnish primarily potassium cations and solvated electron anions. Combination of solvated electrons with aryl halides produces phenyl halide radical anions (eq 43) which enter the catalytic cycle. The function of the " PhX PhX _. 43 eNH+ [ l ( ) Fe+2 salts is unclear.’ Possibly there is an electron transfer from Fe+2 to Phx, or an iron-mediated electron transfer from nucleophile to Phx to generate the radical anion of phenyl halide. There are also cases, 35 such the reaction of pinacolone enolate with iodobenzene in DMSO, where neither light nor reducing agent is required. A reasonable possibility36 is thermally activated electron transfer from the enolate to iodobenzene (eq 44) . 49 KO- — OK DMSO OK c5 .- + + Phl _. + [Phl] + K By means of canpetition experiments, the relative reactivities of pairs of halobenzenee have been measured. 31 Invariably the order of reactivity is: PhDPhBr>Ph61>PhL Even though Bunnett was the first to study the Saul reaction, he never developed general synthetic procedures based on his results. Semelhack applied the photo-81ml reaction to the arylation of various ketone enolates.” He found that the sail reaction operates efficiently with enolate anions derived frat simple ketones, but dialkylsubstituted ketones, such as diisopropylketone give low yields. Be attributed this to hydrogen atm transfer frat then-carbon to the phenyl radicals: 50 0" +911 .____.. W +PhH O. 'o / e . +Phx—-—./ +Ph+x_ He also compared various solvents for the photo-Saul reaction of pinacolone enolate with brmobenzene. He found that solvent can slow the reaction by donating hydrogen atms to chain-carrying radicals, producing benzene as a byproduct. m almost completely inhibited the photo-81ml process, leading slowly to benzene as the major product. On the other hand no benzene was formed in liquid amonia Smelhack applied the 83141 reaction to the synthesis of a natural product. In his successful cephalotaxine synthesis he emloyed an intramolecular photo-Saul reaction38 (eq 45) . 51 Other methods gave successful ring closure including addition to a transient aryne (15% yield) and coupling via a c-aryl nickel complex (30%) . The photo 81ml reaction gave superior results (94%) . The photoinduced Saul reaction has other advantages to offer. It does not require activation by other substitutents (contrary to aromatic nucleophilic substitution (suAr) reactions). Therefore 31ml reactions occur satisfactorily with simple phenyl halides. The nucleophile invariably occupies the position vacated by the halogen. This contrasts with substitution by the aryne mechanism in which cine substitution often occurs.39 It is not sensitive to steric effects of ortho substituents. Even the photostimulated reaction of l-iodo-2,4,6- triisopropylbenzene with acetone enolate gives a significant amount of substitution product. 4° On the other hand Saul reaction does have some disadvantages. It requires 37 a three-fold excess of enolate in order to obtain satisfactory yields. Moreover, it requires expensive photochemical apparatus . 52 Our intention was to develop a new arylation method.based on the reaction of an enolate or its equivalents with a phenyl halide. It has been known for many years41 that copper (I) salts of heteroatom nucleophiles react under relatively mild conditions with unactivated ary1.halides to give substitution. It is also known42 that copper- promcted.nucleophilic substitution is successful using stable carbanions such as that of sodium diethylmalonate (eq 46). Cul Phl + NaCH(COOEt)2 :- PhCH(COOEt)2 (46) (:11 i ~ 62% The above results encouraged us to examine the reaction of ketone enolates with phenyliodide in the presence of various copper (I) salts. Our assumption was that an oxidative addition followed by a reductive elimination would.yield.the desired.product: () (3L! CNDU ’ ‘ LDA CUI THF 0’0" 0 Ph , PM P" 4 + Cul 4\ l “4 53 RESULTS AND DISCUSSION The enolate of cyclohexanone was prepared and reacted with phenyl iodide in the presence of various copper (I) salts (CuBr.Me28, CuI, Cu(CH3CN)4 PF5, (eq 47) . 0 DU 1) CuA LDA 2) Phl fl. f” (47) THEO 3) H O 3 15 min. 0 % No trace of 2-phenylcyclohexanone was formed at various amounts of copper salts and at temperatures ranging fran -78°C to 66°C. The recovery of PM was almost quantitative in all cases, even when the amount of copper (I) salts and the reaction temperature varied. 54 Our unsuccessful experiments with various copper (I) salts led us to use other enolate equivalents. The metalloenamines - the nitrogen equivalent of enolates - were our first target. We were hoping that in the presence of copper (I) salts the metalloenamines would react with phenyl iodide, (eq 48). LiNBu NBu O 0. Ph/u\ LDA.OV. Ph/\ ”GUI 911k?“ (48) 1 hr, THF 2) Phl .+ 3) Pg 0‘ The yield for the above reaction was 66%. To our surprise, a control experiment without copper (I) yielded 64% of phenylated acetophenone. Therefore, copper (I) was not required for the above reaction. The synthetic aspect of the arylation of metalloenamines was our first goal. Our hope was to develop a method for the arylation of ketones. An ideal method should: a) give high yield of products with minimum side reactiOns b) anploy a 1:1 ratio of the two reagents c) use materials and solvents which are inexpensive. We decided to use acetophenone and phenyl iodide as our model canpounds. One of our first goals was to maximize the yield using different solvents. One difficulty was that in the presence of THF a lot of PM was converted to PhH, presumably by hydrogen abstraction from the methylene units of THF which are next to the oxygen. Therefore we decided to make the metalloenamine in THF, then remove the TEF by applying high vacuum and then introduce various other solvents. Our results for the reaction below are listed in Table 6. 55 TAB“ 6 Reaction of the Metalloenamine of Acetophenone (2 eq) with Phil: (1 eq) Under a Variety of Solvents 1) THF . amine [NB removed 0 u 2 Ph __.6—. 2 Ph’ \ v Ph 4. 1hr 4) H O 3 O . Phl, **% Sglnnt W- PhAPh . % THF 31' 64 0 Pentane RT 18 82 Pentane Reflux 55 33 Hexane 3911““ ' 84 16 Cyclohexane Reflux 33 V 17 our m 0 96 Benzene RT 40 40 *Yield detemined by GLC. MRecovered PhI determined by GLC. 56 From the above table one can see that the solvents that give the best results are hexane and cyclohexane at refluxing temperatures. Our next change was the group attached to the nitrogen on the imine. We examined the reaction of a variety of metalloenamines of acetophenone with PhI. Our results are listed in Table 7. TAB“ 7 Arylation of various letalloenouinea of Acetophenone (2 eq) with par (1 eq) ma 1) THF, amine we 0 2 LDA. THF removed 0 . 1hr 2)Cydohsxane and solvent additive added an Phl. A 6-8 hrs a. 4)H o 3 a Solvent 0 ~- Additives Ph/K’Ph . % pm ’ °/o n-Bu- None 84 16 (C83) 211 (C82) 3- None 80 16 t-Bu- None 45 20 (CR3) 2N- None 0 100 n-Bu- TMEDA (2eq) 24 17 n-Bu- MA (2 eq) 55 0 (CR3)2N(C82)3- TMRDA (2 eq)33 4 (CR3)2N(C82) 3- BMPA (2 eq) 45 0 n-Bu- 12-Crown-4 16 0 *Yield determined by GLC. ** Recovered Phl determined by GLC. 57 The n-butyl imine together with the dimethylamino propyl imine work the best. It is surprising that the dimethylhydrazine imine gave no product. we were hoping that the addition of cosolvents such as hexamethyl phosphoramdde (HMEA) or N,N,N',N'-tatramethylenediamine (TMEDA) as well as 12-crown-4 ether would.improve the yields by solvating the lithium. However, as it can.be seen.from.Table 7 none of the possible additives increased the yield of the phenylated.product. The mass balance of phenyl iodide and product are low. Presumably, all of the above solvent additives donate a hydrogen to the phenyl radical.as shown on page 64. The effect of different initiators on the reaction yield.was our next goal. As it can be seen for Table 8 none of them was successful for the reactions. we were hoping that He and CrClz would.improve the yield.by generating phenyl radicals: .+ Na + Phl _. Na ... [pm]; [Phl]. 1 —— P|1 + l These phenyl radicals would then enter the catalytic cycle. The diphenyliodonium salts could also generate a radical chain mechanism: UNR pr. F /R - "a + 2+ Ph ——,"’Ph+Phl+Ph/u\.+u 58 TABLR 8 The Effect of Various Initiators on the Reaction of the nButylimine of Acetophenone (2 eq) with Rh]: (1 eq) NBu THF LiNBu 1) THF. amine O ZLDA. moved ‘ 2 Ph/u\ o 2Ph* f fiphklh 0 . 11w 2) Cydohexane an! initiator added 3)1PhL£i $8hn 4))4 0+ 3 Iniatiator lip , % Phl, *4) g P!) " . CrClZ (0.1 ep) 75 0 thIPFs (0.1 eq) 15 77 *Yield determined by GLC. 1“Recovered PhI.determined by GLC. Presumably none of the above reagents was able to initiate phenyl radicals more efficiently than the metalloenamine itself. Our last point of investigation was the ratio between the metalloenamine and PM. In all of the above experiments we use one extra equivalent of the metalloenamine. A series of experiments using equivalent amounts of the metalloenamine and the PM was completed. The results listed in Table 9 are disappointing. 59 TABIJ 9 Reaction between 1 eq of Imine and 1 eq Phl . GLC Yields . LiNBu 1) THF. amine 0 N80 1 LDA. THF ed . ”k 0 1 Pit/K remov ”kph 0 2) Solvem- and ' "" additive added 3) 1 Phl. A 3811!: 01-130 i l Addi i O % n - Phl" R Concentrat on/ So ve t t vs" /I\’ p" ' % n-Bu- 0 . 5 Hexane None 23 63 n-Bu- 1 Hexane None 26 53 n-Bu- 5 Cyclohexane None 25 38 n-Bu- 1 Cyclohexane None 40 31 (CH3) 2N (C82) 3- 0 . 5 Hexane None 14 56 (CH3) 2N (CH2) 3- 1 Hexane None 15 68 n-Bu- 1 Hexane 0.leriI 13 25 n-Bu- l Hexane 0 . lequC12 34 22 *Yield determined by GLC. / ** Recovered PhI determined by GLC. We examined the phenylation of a variety of metalloenamines using 2 eq of metalloenamines and 1 eq of ArI. Our results are listed in Table 10 . 60 TABLE 10 Reaction of Metalloenamines with Aryl Iodides. Isolated Yields NBu LiNBu 1)THF. amine o 2 ski? 2LDA.T1~IF 2 a a, removed ‘ R’Krn' o 2 C Cl 11 Ar 0 .1hr ) Y o exane added 3)1Ar1. A 8M 4) H 0+ 3 Imine Aryl iodide Product Yield 8 was 0 m"\ I ”A0 73 N80 o ”A CH3 I “9&0 CH3 46 we... o "A °"” ‘ "O0 6143 04:113 42 was cw o c“, p“ 3 . "A0 on3 on: cw: 41 "°" O ° "N ' "'40 75 A.“ 0- ° F!) "*0 39 we» o O‘ . . 3. use . o W V6 72 61 Regarding the mechanism of this reaction, one can say that the most likely pathways are: a) benzyne mechanism (eq 49) b) direct nucleophilic substitution (eq 50) c) an aromatic substitution by an Saul mechanism (eq 51) d) Metal-halogen exchange followed by coupling (eq 52) . UNBU | N80 Ph/\ «1 —-——ph/l\ + 4» Lil (49) + . UNBu - 0 N80 .4 N80 H30 Ph&+ Phl—op" e ‘Ph/K’ph—eph/K/Ph (50) I LINBU NB“ -3 + Ph/K‘.’ Phl —- Ph , + Phl + Li (51) Phl——-Ph+l N80 _ N80 ”A. + Ph —- ”A,” LiNBU UNBU ~ . P11 P11 + P" -———- Ph/l\/ 62 LINBu N80 . Ph "' (ah/V + Ph' _.Ph)\/Ph 4, Phl 4. N80 HO O - + LINBu Phl N81) 0 N81) ph —.Ph)\/l +PhLi_..P "kph“ H03 (flak/P" (52) One way to exclude the benzyne mechanism was to achieve arylation in a case where no benzyne intemediate can be formed. 2-Iodanesitylene cannot form benzyne. The desired product 25 was obtained in 41% yield (eq 53) . fl 2 LDA ”NB“ ”QPI‘OZ 2 Ph 0 2Ph",\ ' ' (53) THF 1° Hexane, A 1hr Am' - me free 12 hrs _ + 2""50 41% 11‘ 63 Therefore, benzyne mechanism is not operative, at least in this case. For determining which mechanism prevails in an aromatic nucleophilic substitution, the halogen mobility order is a criterion of some value. In most direct aromatic nucleophilic substitution reactions, 43 it is .ArF>>ArCl~ArBr~A.rI. In our studies the reactivity order for the following reaction was ArI> ArBr>ArCl. Therefore one can say that unless the expulsion of the halogen is rate-limiting an addition-elimination mechanism (SNAr) does not operate in this case. 1)1 - 0 N80 LINBU 2 LDA )K/ 2 Ph)\ 0 2 Ph’k‘ x — Ph Ph THF,0 Hexane, A 1hr ' - P110. 5% AmIne free 12 hrs 2, (.55 PhBr 14% Phl 84% One can argue also about the last mechanism (metal-halogen exchange followed by coupling) in the following points: One would expect to see some 1, 4-diketone producted from metalloenamine-iodoimine coupling. However, we were unable to observe any 1,4-diketone or pyrrole derivative. We found that benzophenone inhibits the reaction. If metal-halogen exchange mechanism was operating, benzophenone would have no effect. On the other hand, if a radical mechanism operates the yields are lower because benzophenone quenches the radical anions to produce a relatively stable ketyl which terminates the chain process. Another point is that the equilibrium for the reaction: 64 LiNR ph/\ + Phl .- PhLi + Phi/l lies on the side of the weaker base. 44 Although there are no accurate pk values for the metalloenamine and PhLi, one would expect that the basicity of PhLi should be at least five pR units higher than that of metalloenamines. The low yield of the phenylation when no excess of metalloenamine was used (Table 9) also supports the radical chain mechanism: The phenyl radical is trapped more efficiently by the use of excess metalloenanine. From the above observations one can presume that the mostlikely mechanism that operates is an 8331 mechanism. Light is not required since the yields are the same under light and in the dark. One can explain the fact that the reaction occurs spontaneously in the dark by proposing that the PhI"- is thermally produced by electron transfer from the nucleophile to the PM: (011.54). UNBU N80 .- + “x 1pm -——PHA,+PM Li- Ph1—-—- Ph+l NB“ . N80 + Ph “NP" ”A. (54) UNSU LINE" 911* 1 P" ”NP" UNBU Nan , 65 The phenyl radicals generated in the second step react with the matalloenamine to generate a phenylated nitrogen analog of a ketyl which then can generate more phenyl radicals by an electron-transfer process. In sunmIary, a new phenylation method for ketones was developed. It is based on the reaction of metalloenamines with aryliodides in hydrocarbon solvents and refluxing temperatures. This method has the advantage that it requires neither special apparatus nor exotic reagents, and it is suitable for large scale reactions. On the other hand it has certain disadvantages: a) one equivalent of imine is wasted. It gives good yields for some compounds but it fails to react with others. In spite of these. limits, this method probably ranks among the best at present, for the u-arylation of ketones. mm 8861310! All the imines were prepared by the azeotropic method.8 All amines and ketones were purchased from Aldrich and used without further purification. BuLi was purchased fran Aldrich as 1.6 M solution in hexanes. All the aryl halides were purchased fran Aldrich and used without purification. Diisopropylamine and TH? were distilled from CaHz,2-Iodomesitylene was synthesized from 2-branomesitylene by lithiation45 followed by quenching the organolithium reagent with iodine. 45 Mass, MIR and IR spectra were recorded with the same instrmnents as in the previous chapters. 2W1). BuLi (12.8 ml, 20 mmol was slowly added to a solution of diisopropylamine (2.8 ml, 20 mol) in THF (10 m1) at 0°. After 15 min 20 mol (3.8 ml) of the n-butylimine of acetophenone was added dropwise and the reaction mixture was stirred for 1 hour. THF and diisopropylamine were removed by vacuum distillation to 40°. High 66 vacuum was applied for 30 min and a warm bath (~40°C) was applied to the flask for 10 additional minutes to remove traces of the solvent and diisopropylamine . Argon was introduced and 10 ml of cyclohexane was added. The flask contents were heated slowly with a heating mantle to reflux. Iodobenzene (1.12 ml, 10 mmol) was then added dropwise. The initially light yellow solution became dark red-brown. After 8 hours of reflux, the reflux was imersed in an ice-water bath and 5 ml of water was added followed by 15 ml of 3N acetic acid. Shortly the light yellow solution turned dark red-brown. After 8 hours of reflux an ice-cold bath was applied to the flask and 5 ml of water was added followed by 15 ml of 3N acetic acid. After 4 hours stirring to cmplete the hydrolysis of the imine, the organic layer was separated and the water layer was extracted twice with 25 ml of ether. The combined organic layers were dried with magnesium sulfate, filtered and the solvents were evaporated. Distillation at low temperature (580/ 5 (in) removed the acetophenone and the phenyl iodide. Rugelrohr distillation at llS°/0.5 nan gave a light yellow oil which was crystallized frat hexane to afford 1.43 g of 2- phenylacetophenone (73!), mp 54-560- 18 114a (cn3cu) 5 4.2 (s,za), 7.2 (3,53), 7.4-8.2(m,SH) 14$ m/e relative intensity 196 (14+,1), 105 (100), 91(8), 77 (51), 65(10). With a similar procedure, the following compounds were prepared. W 111114): (c0013) 5 2.2 (3.38). 4.2 (8,28), 7.1(s,4H), 7.3-7.6(m,3H) 7.8-8.2(m,23): MS, m/e(relative intensity) 210 (M-hl) , 105 (100) , 104 (45) e 77 (51) . 67 WW: 18 11143 (coc13) 5 3.79 (s, 3H), 4.1 (s,2H), 6.7-8.1 (m, 9H): MS m/e (relative intensity) 226 (M+,9), 135(11), 121(90), 120(31), 105(100), 77(87). 2:1l‘3‘5zII1mgghglghgngllaggngphgngng; 18 NMR (CDC13) 5 2.1 (3,65), 2.2 (3,38), 4.2(s,2H), 6.8(s,28), 7.1-7.6 (m,33) 7.7-8.1(m,23): MS m/e (relative intensity) 238(11), 133(39), 105(100), 77(33). W 18 111m (coc13)5 1.5 (d,3H), 4.7(q,114), 7.1-7.6(m,8H), 7.9-8.2(m,2H): MS m/e (relative intensity) 210(3), 105 (100) 104 (29) , 88(11) . Wham“; 111 1am (coc13) 51.5(s,68), 6.4- 6.7(m,2H) 7.0-7.6(m.88): MS m/e (relative intensity) 224(M+,l), 160(34), 105(48), 104(100), 91(24), 77(29). W 1H MIR (CDC13) 51.6-2.8(m,8H), 3.4- 3.8(m,1H), 7.1-7.5(m,5H) ms, m/e (relative intensity) 174 (M+,3), 120(25), 105(100), 91(8), 77(64). - - - . 18 ma (coc13) 50.8(t.38), 1.3 (d,3e), 2.3(q,ZH), 3.7(q,lH), 7.2(s.58): MS, m/e (relative intensity) 163(M+1, 70) 154 (25) , 105 (89) p 104 (50) , 91 (18) p 77 (39) , 57 (100) . Emma! 10. 11. 12. 13. 14. 69 Bibliography H.O. House, "Modern Synthetic Reactions." 2nd Edition, W.A. Benjamin Inc., New York, 1972, chapter 9. O.A. Reutov and 11.1... 1(1):th W .45, 1040, (1977). R.L. Augustine, "Carbon-Carbon bond formation, " Marcel Dekkar, New York, 1979. J. Conia, W" 21, 43, (1963). a) G. Stork and 8.1!. DOWd. W 3.5.: 2178, (1963). b) G. Stork and J. Benaim, We. 23. 5938, (1971). H.O. House, W.C. Liang, and P.D. Weeks, M" 3.2, 3102. p.13. Hudrlik and c. N. Wan, My 40, 2963, (1975). J. K. Whitesell and M. A. Whitesell, Synthesis, 517, (1983). A. I. Meyers, G. S. Poindexter and z. Brich, W, A}, 892, (1978). R. F. Borch, M.D. Bernstein and H. D. Durst, W, 9.3, 2897, (1971). R. A. Rossi and J. P. Bunnett, W, 3.8, 3020, (1973). a. c. Brown, a. Nambu and 14. 14. Radio, Won... 91. 6852, (1969). C. B. Sacis and P. L. Pucks, W... 97, 7372, (1975). P. A. Wender, J. M. Brhardt and L. J. Letendre, W” 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 70 a) C. E. Sacks and P. L. Pucks, W, 21, 7372, (1975). b) G. H. Posner and J. J. Sterling, W, 25, 3076, (1973). - A, Hosomi, A. Shirahata, Y. Araki and H. Sakurai, Wm, .46, 4631, (1981). W. Seufert and P. Effenberger, M 11211670, (1979). D. Cantacuzene and M. Tordeux, M" 5.0,, 4807, (1979) ‘ P. W. Hickmott, Tetrahedron, 3.8... 1975, (1982). W. D. Gurowitz and M. A. Joseph, M... 3.2, 3289 (1967). L. Pauling, "The Nature of the Chemical bond," Third Edition, Cornell University Press, Ithaca, N.Y., 1960. J. Szmuszkovicz, M. P. Grostic and J.P.Zieserl. m 3969 (1967). K. Maruyama and Y. Yamamoto, W 9.9,, 8068. a) P. Y. Sollenberger and R. B. Martin, ”Chemistry of the Amino Group," S. Patai, Interscience, New York, 1968. b) J. K. Whitesell and M. A. Whitesell. W313: 517, (1983). W. A. White and H. Weingarten, m 32, 213, (1967). w. w. Leake and R. Levine, W... 31. 1169, (1959). M. E. Garst and D. Lukton, 5m 10, 155, (1980). T. Sakakura, M. Hare and M. Tanaka, M... m 1545, (1985). a) P. A. Manis, DOctoral Thesis, Michigan State University, 1982, p. 49. b) F. M. Beringer, W. J. Daniel, S.A. Galton and G. Rubin, .L_ng_., me: 3.1: 4315 (1966). 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44.‘ 45. 46. 71 F. M. Beringer, S. A. Galton and S. J. Huang, W, 34, 2819, (1962). J. F. Bunnett, W.) 11, 413, (1978). s. Hoz and J. F. Bunnett, Wham 7463, (1977). a, J, Jun abd H, P, Bunnett, W. 92. 7463, (1977). c. Galli and J. r Bunnett, My 42, 3041, (1984). R. G. Scamehorn and J. F. Bunnett, M” 3.8, 1407, (1973). N. Rornblum, R. 8. Michel and R. C. Kerber, W” .88, 5662, (1966). M. P. Smlhack and ‘1‘. Ber-gar. Wu 1.0.2. 7765, (1980). M. P. Semelhack, B. P. Chong, R. D. Stautter, '1'. D. Rogerson, A. Chong and L. D. Jones, W 91, 2507, (1975). J. K. Kim, Ph.D. Dissertaion, University of. California, Santa Cruz, 1970. J. r. Bunnett and s. r Gloor, My .12, 382, (1974). J. Lindley, W 40, 1433, (1984). J. Tsutsune, R. Mantuvala, H. Wakemoto, and '1'. Ritao, my.” 367. (1980). G. P. Briner, J. Miller, M. Liveris and P. G. Lutz, M, 1265, (1954). D. J. Cram, "Fundamentals of Carbonion Chemistry," Academic Press, New York, 1965. H. Gilman and J. W. Morton, W. 8., 286 (1954). Organic Synthesis, 6:, 108 (1987). HICHIGRN STRTE UNIV LIBRRRIE II|I1II|I2III|9II3|IIIIIII SII7 IIIIII IIIIIIIIIIIII