CATALYTIC CARBONYLATION OF AMINES AND DIAMINES AS AN ALTERNATIVE TO PHOSGENE DERIVATIVES: APPLICAT ION TO SYNTHESES OF THE CORE STRUCTURE OF DMP 323 AND DMP 450 A ND OTHER FUNCTIONALIZED UREAS By KEISHA-GAY HYLTON A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2004
Copyright 2004 by Keisha-Gay Hylton
Dedicated to my father Alvest Hylton; he never lived to celebrate any of my achievements but he is never forgotten.
ACKNOWLEDGMENTS A number of special individuals have contributed to my success. I thank my mother, for her never-ending support of my dreams; and my grandmother, for instilling integrity, and for her encouragement. Special thanks go to my husband Nemanja. He is my confidant, my best friend, and the love of my life. I thank him for providing a listening ear when I needed to discuss my reactions; and for his support throughout these 5 years. To my advisor (Dr. Lisa McElwee-White), I express my gratitude for all she has taught me over the last 4 years. She has shaped me into the chemist I am today, and has provided a positive role model for me. I am eternally grateful. I, of course, could never forget to mention my group members. I give special mention to Corey Anthony, for all the free coffee and toaster strudels; and for helping to keep the homesickness at bay. I thank Daniel for all the good gossip and lessons about France. I thank Yue Zhang for carbonylation discussions, and lessons about China. In addition, I would like to thank my other group members for the laughs over the years. iv
TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES.............................................................................................................vi ABSTRACT......................................................................................................................vii CHAPTER 1 INTRODUCTION........................................................................................................1 2 W(CO)6/I2 CATALYZED OXIDATIVE CARBONYLATION OF AMINES AND DIAMINES: APPLICATION TO SYNTHESIS OF THE CORE STRUCTURES OF THE HIV PROTEASE INHIBITORS DMP 323 AND DMP 450.............................22 3 W(CO)6/I2 CATALYZED OXIDATIVE CARBONYLATION OF SECONDARY AMINES AND DIAMINES: ANALOGS OF THE CORE STRUCTURES OF THE HIV PROTEASE INHIBITORS DMP 323 AND DMP 450....................................37 4 APPLICATION OF W(CO)6/I2 CATALYZED OXIDATIVE CARBONYLATION TO THE CARBONYLATION OF AMINO ALCOHOLS........................................46 5 EXPERIMENTAL PROTOCOLS.............................................................................54 LIST OF REFERENCES...................................................................................................80 BIOGRAPHICAL SKETCH.............................................................................................84 v
LIST OF TABLES Table page 1-1 Oxidative carbonylation of primary and secondary amines.....................................19 1-2 Oxidative carbonylation of primary diamines..........................................................20 1-3 Oxidative carbonylation of para-substituted benzylamines.....................................21 2-1 Carbonylation of compounds 24-26.........................................................................25 2-2 Optimization studies for the oxidative carbonylation of 26.....................................26 2-3 Oxidative carbonylation of 35 catalyzed by W(CO)6 under various reaction conditions.................................................................................................................29 2-4 Oxidative carbonylation of 35 using Mo(CO)6 and Cr(CO)6...................................30 2-5 Oxidative carbonylation of 44 under various reaction conditions...........................33 3-1 Study of effect of the presence of the benzyl group on the W(CO)6 /I2 catalyzed carbonylation of 1,4-secondary diamines.................................................................39 3-2 Optimization study of the carbonylation of amines 58, 62 and 65..........................43 3-3 Effects of base on the oxidative carbonylation of N-ethylbenzylamine, 66............44 4-1 Carbonylation of 69 under various reaction conditions...........................................48 4-2 Carbonylation of 78 under various reaction conditions...........................................52 vi
Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy CATALYTIC CARBONYLATION OF AMINES AND DIAMINES AS AN ALTERNATIVE TO PHOSGENE DERIVATIVES: APPLICATION TO SYNTHESES OF THE CORE STRUCTURE OF DMP 323 AND DMP 450 AND OTHER FUNCTIONALIZED UREAS By Keisha-Gay Hylton May 2004 Chair: Lisa McElwee-White Major Department: Chemistry W(CO)6-catalyzed carbonylation provides an alternative to phosgene or phosgene derivatives such as 1,1-carbonyldiimidazole (CDI) for the conversion of amines to ureas. The core structure of the HIV protease inhibitors DMP 323 and DMP 450 has been prepared by catalytic carbonylation of diamine intermediates from the original syntheses. Yields of the ureas from the catalytic reaction are comparable to those previously reported for reaction of the substrates with stoichiometric CDI. Intermolecular urea formation in related acyclic systems was demonstrated by catalytic carbonylation of O-protected phenylalaninol derivatives. W(CO)6-catalyzed carbonylation has been successfully applied to the preparation of analogs of DMP 323 and DMP 450 and to the carbonylation of amino alcohols to selectively form hydroxyl ureas. vii
CHAPTER 1 INTRODUCTION Substituted ureas are interesting compounds because of the broad range of their applications. They have found use in a variety of capacities, for example, as antioxidants in gasoline and have been known to exhibit useful biological activity. Ureas are structural components of drug candidates such as HIV protease inhibitors,1,2 CCK-B receptor antagonists,3,4 and endothelin antagonists.5 In addition, they have found widespread usage as agricultural chemicals, dyes, and intermediates en route to carbamates; and as additives to petroleum compounds and polymers.6 The classical synthetic methodology used to synthesize substituted ureas from amines involves phosgene. Phosgene is useful for the carbonylation of primary and secondary amines (Scheme 1.1). The major drawback of phosgene is that it is a highly toxic and corrosive gas. Because of its toxic nature, it requires special handling. This has discouraged its use in laboratory settings. Cl Cl O RR'NHCl NRR' O 1 RR'NHR'RN NRR' O -HCl-HClwhere R' = H, alkyl, aryl Scheme 1.1 Phosgene production7 and use on an industrial scale raise serious environmental risks and problems connected with the use and storage of large amounts of chlorine, and the transportation and storage of a highly toxic and volatile reagent. Other safer derivatives such as 1,1-carbonylimidazole, triphosgene, and a variety of other reagents 1
2 have been used in the carbonylation of amines to form substituted ureas, and are more common in the laboratory setting. Diphosgene (ClCO2CCl3) behaves in the same manner as phosgene. However, diphosgene is in a liquid state at ambient temperature and pressure, while phosgene is in a gaseous state. Thus, it is much safer and more convenient to transport, store, and use diphosgene. Unfortunately, it decomposes to phosgene and chloroform, and is as toxic and corrosive as its phosgene counterpart. The use of isocyanates (2) is undesirable because of their toxic nature and the need to synthesize them from phosgene (Scheme 1.2). In addition, this method can be used in conversion of primary amines to symmetrical and unsymmetrical ureas. Cl Cl O R'NH2Cl NHR' O 1 -HCl-HClwhere R', R = H, alkyl, arylR'NC O 2 HN NH O R' R RNH2 Scheme 1.2 Triphosgene [bis(trichloromethyl)carbonate, 3], a crystalline solid, is deemed a safe and stable replacement for phosgene and can be handled without special precautions. It is prepared by exhaustive photochlorination of dimethyl carbonate8 (Eq.1-1). H3CO OCH3 O CCl4where R' = H, alkyl, arylCl3CO OCCl3 O 3Cl2,hv(1-1) An activating nucleophile such as triethylamine, pyridine, or dimethylformamide is needed to activate triphosgene. These liberate three molecules of phosgene in situ on reaction with triphosgene. Triphosgene is useful for a variety of conversions, including
3 the carbonylation of amines to ureas. By no means is triphosgene a perfect reagent. It is expensive; and long-term storage or presence of impurities may lead it to form phosgene. The crystalline solid 1,1-carbonyldiimidazole (CDI, 4) has also been used as an alternative to phosgene. Its commercial availability and ease of handling make CDI a desirable alternative to phosgene. An example of CDI as a carbonylation agent is the synthesis of N, N-substituted ureas such as the HIV protease inhibitors DMP 323 and DMP 450.1,2 The mechanism of carbonylation using CDI involves the stepwise displacement of imidazole by the attacking amine, to form the corresponding urea. It has been successfully used for the carbonylation of primary amines (Scheme 1.3). HN NH O HN N O N NNH N N O N N 4 NNH RNH2RNH2R R R Scheme 1.3 On a large scale, the major drawbacks of using CDI (4) are the use of phosgene in its synthesis (Scheme 1.4) and the disposal of the imidazole waste stream generated after the reaction of CDI with amines. In addition, CDI is expensive, and therefore not desirable for large-scale synthesis. Cl Cl O Cl N O 4-HClNHN N N N O N N -HClNHN Scheme 1.4 Similarly, N, N-carbonyldibenzotriazole (5)9 can be used to synthesize unsymmetrical ureas. This is achieved via a one-pot addition of the first amine to form the carbamoylbenzotriazole intermediate 6, that can be further reacted with a second amine to afford the final urea (Scheme 1.5). Both aliphatic and aromatic amines are
4 substrates for the reaction. However, the reaction conditions and yields of 6 are significantly affected by the steric hindrance of the amines used, and a long reaction time of 2 days is required. In addition, reagent 5 is not commercially available, and must be synthesized from benzotriazole and phosgene. RRN NRR' O R'RN Bt O where R' = H, alkyl, aryl,Bt Bt O RR'NH56Bt =NNN RRNH Scheme 1.5 S, S-Dimethyldithiocarbonate (DMDTC, 7), can be used in carbonylation of primary and secondary amines. It is a mild reagent, and structurally similar to phosgene. DMDTC (7) is prepared from methanol, carbon disulfide, and dimethyl sulfate by a two-step sequence.10,11 Its mode of action is shown in Scheme 1.6. Even though the use of dimethyl sulfate requires safety precautions, this method is useful. It is applicable to a variety of substrates, but does not effect the carbonylation of aromatic amines such as aniline.12 RHN NH R O MeS SMe O where R = H, alkyl7+RHN SMe O RNH2RNH2 Scheme 1.6 Various other methods have been used to convert amines to ureas. These include the use of phenyl chloroformate (8) to form substituted ureas from primary amines (Scheme 1.7).
5 PhO Cl O PhO NHR O 8RHN NHR O -HClwhere R = H, alkyl, arylDMSO RNH2RNH2 Scheme 1.7 In this method,13 the byproduct is phenol, which is easily removed by washing with sodium hydroxide. The drawback to this method is the use of DMSO as solvent. DMSO is known to be toxic and a possible carcinogen. Furthermore, it is difficult to remove because of its high boiling point. X3C Cl O X3C NHR O 9 R'HN NHR O -HClwhere R = H, alkyl, aryl X = I, Br, Cl, F DBURNC O 10 RNH2R'NH2 Scheme 1.8 Trihaloacetyl chlorides (9) can also be used to produce symmetrical and unsymmetrical ureas. Trihaloacetyl chloride can also be converted to the stable, easily handled, crystalline trihaloacetimide intermediate 10. This methodology can be applied to both aliphatic and aromatic amines (Scheme 1.8). Monosubstituted ureas can be synthesized with a two-step process using 4-nitrophenyl-N-benzylcarbamate (11).14 The first step involves the reaction of amine with 4-nitrophenyl-N-benzylcarbamate, (11). The second step consists of removing the benzyl moiety to form the desired monosubstituted urea (Scheme 1.9).
6 RNH2NH NH O R 11, Et3NSolvent NH NH2 O R H2, PdCH3COOH Scheme 1.9 This method is applicable to a wide spectrum of amines. It can successfully form monosubstituted ureas from simple amines, sterically hindered amines, water-soluble amines such as amino alcohols, and less nucleophilic amines such as aniline. Primary and secondary amines are both substrates for the reaction. One application of this method is in the derivatization of complex amines such as kanamycin A (12), an amino glycoside antibiotic.15 O HO H2N OH OH O H2N HO O HO O OH OH NH2 NH2 12 Main group elements such as sulfur 16,17 and selenium 18,19 can also serve as catalysts for the carbonylation of amines. The first investigation of selenium-mediated carbonylation reactions with carbon monoxide was performed by Sonoda et al. almost three decades ago.18 Selenium can successfully catalyze oxidative carbonylation of primary aliphatic and aromatic amines to the corresponding N, N-disubstituted ureas. In addition, secondary amines form the corresponding ureas under selenium-mediated oxidative carbonylation reaction conditions (Scheme 1.10). Selenium-catalyzed reactions (Scheme 1.10) can generate high yields of urea under mild conditions.
7 Se + COSeC O 13 RR'NHR'N Se O R 14 14 NC O R 15 RNH2For primary amine, R' = HNH NH O R R RN Se O R' 16 RR'NHN N O R R R' R' RR'NH CO, SeN N O R R R' R' -H2Se Scheme 1.10 The reaction has been postulated to proceed via carbonyl selenide (13), which reacts with an amine to form the carboselenoate intermediate (14). With primary amines, it is postulated that the urea is formed from a nucleophilic attack on an isocyanate (15) generated from the elimination of hydrogen selenide from the carbamoselenoate. The selenium catalyzed carbonylation of secondary amines is believed to occur via a nucleophilic attack on bis(carbamoyl) diselenides (16) (Scheme 1.10). Other selenium-mediated carbonylations involving aromatic and aliphatic nitro compounds in the dual role of oxidant and reagent have been reported20 (Scheme 1.11). Under these reaction conditions, aromatic nitro compounds and secondary aliphatic amines such as piperidine
8 react to form unsymmetrical ureas in moderate to good yields (Scheme 1.11). On the other hand, nitromethane and primary aliphatic amines react only to form symmetrical ureas (Scheme 1.12). Symmetrical ureas are observed in the absence of the nitro compound. In the presence of the nitromethane, however, the yield of symmetrical ureas doubles. NH2 X +HN + 3 COSetolueneN NH O X X = H, Cl, Br, Me+ 2CO2 Scheme 1.11 Se, NEt3toluene+ 2 RNH2R = aromatic, alkylHN NH R R O + 2CO2CH3-NO2 Scheme 1.12 Organometallic methodologies have also been examined as a substitute for phosgene chemistry. A variety of metal centers can be used as catalysts in the presence of CO2 or CO. However, not all transition-metal catalysts have been very useful in the oxidative carbonylation of amines to form ureas. Oxidative carbonylation of primary amines to substituted ureas has been reported for transition-metal catalysts involving Ni,21 Co,22 Mn,23,24 Ru,25 and most commonly Pd.18-19 Of all the transition metals, palladium complexes are the most commonly used in oxidative carbonylation of amines. Both homogenous and supported Pd catalysts have been studied.26 Palladium catalysts are generally more selective for the formation of carbamates than ureas, when alcohols are present with the amines in the reaction mixtures.
9 Palladium (II) salts or complexes have been known to catalyze the formation of carbamates by the carbonylation of aliphatic or aromatic amines and aliphatic alcohols in the presence of CO and oxygen. In addition, hydrochloric acid and a cocatalyst such as FeCl3 or CuCl2 are generally added to the reaction mixtures.27 One such example is shown in Equation 1-2. In the absence of alcohols, urea formation occurs. NH2 +CO++1/2 O2CH3CH2OH HN O O +H2OPd, I-170 C90 bar (1-2) Early work established that N, N-disubstituted ureas could be obtained in good yields by the reaction of aromatic and aliphatic amines in the presence of CO and O2 under mild conditions (70-90C, 1 atm). The presence of PdCl2 or a palladium (II) catalyst, PdCl2L2 (Scheme 1.13) was required. The type of the ligand, (L) had no effect on the efficiency of the reaction. Copper (II) chloride, CuCl2 was used as a promoter/ co-catalyst (Scheme 1.13).28 The PdCl2L2/ CuCl2 catalytic system was found to be most effective for urea synthesis. Reaction conditions were best suited to carbonylation of primary aromatic and aliphatic amines. Carbonylation of aromatic primary amines gave higher yields of the corresponding ureas than carbonylation of aliphatic primary amines. Secondary aromatic and aliphatic amines did not form the corresponding ureas under the reaction conditions. The main products were carbon dioxide, tertiary amines, and unidentified products. 2 RNH2 + CO + 1/2 O2 NH NH O R R + H2OPdCl2L2 / CuCl270-90 CsolventL = PPh3, CH3CN, C6H5NH2, p-ClC6H4NH2, CH3C6H4NH2Solvent: CH3OH, CH3CH2OHR= alkyl, aryl Scheme 1.13
10 In addition, it was found that at more drastic reaction conditions, i.e., higher pressure and temperature, carbamate esters were formed instead of the desired ureas. Based on control experiments where aniline was used as substrate, it was established that in the same temperature range (110 to 150 C), N, N-diphenylurea reacts with methanol to form methyl-N-phenylcarbamate in the equilibrium process shown in Equation 1-3. Giannoccaro postulated that it was likely that carbamates are formed not by direct interaction with aniline, CO, and alcohol; but rather by alcoholysis of the urea (Eq. 1-3). NH NH O Ph Ph + CH3OH110150 CCO/ O2NH O O Ph CH3 + PhNH2 (1-3) Research by Pri-Bar et al.29 indicates that iodine or potassium iodide can be used as a promoter for the carbonylation of primary and secondary amines, instead of a copper (II) salt such as CuCl2. Treatment of primary or secondary amines with carbon monoxide in acetonitrile, in the presence of iodine, potassium carbonate, and a catalytic amount of palladium (II) acetate for 3 hours at 95C afforded diand tetrasubstituted ureas (Eq. 1-4). Iodine and the catalyst are needed for the reaction to occur. Absence of the base results in a lowering of the overall yields. This is believed to be due the partial conversion of the amine to the ammonium salt by the HI byproduct. In the presence of a small quantity of air or oxygen, the reaction results in the formation of an oxamide (18), a double carbonylation product which is formed in addition to the urea (17) (Eq. 1-5). This phenomenon is more prevalent with secondary amine substrates. 2R'RNH2 + CO + I2 Pd(OAc)2, K2CO3CH3CN, 95 CN O N R R R' R' + 2 HI(1-4)R' = alkyl, arylR = H, alkyl
11 2R'RNH2 + CO + O2 Pd(OAc)2, K2CO3CH3CN, 95 CN O N R R R' R' + H2O(1-5)R' = alkyl, arylR = H, alkyl O N N O R' R' R R +1718 More recent work involves the use of palladium (II) iodide (PdI2) in the presence of air, CO, and CO2 to oxidatively carbonylate amines to form symmetrical and unsymmetrical ureas (Eq. 1-6, 1-7). For this catalytic system, primary aliphatic amines are more reactive than electron-rich primary aromatic amines. Ureas were obtained in high yield when primary amines were carbonylated. The postulated mechanism for urea formation is shown in Scheme 1.14. Formation of the disubstituted urea is believed to involve the formation of carbamoylpalladium species I as the key intermediate (anionic iodide ligands are omitted for clarity). 2 RNH2 + CO + (1/2) O2 Pd cat.-H2ORHN NHR O RNH2 + R'NH2+ CO + (1/2) O2 Pd cat.-H2ORHN NRR' O (1-6)(1-7) PdI2 + CO + RNH22 HI + (1/2) O2 IPd NHR O RNH2-[Pd(0) + HI]RHN NHR O -HII2 + H2OPd(0) + I2 PdI2R = alkyl, arylI Scheme 1.14
12 As shown in Scheme 1.14, in the course of the process, RNH2 acts as a nucleophile and substrate; and carbonylation is accompanied by the formation of 2 mol of HI and 1 mol of Pd (0). Reoxidation of Pd(0) under the reaction conditions occurs through oxidative addition of I2, which is generated by oxidation of HI by oxygen. Clearly, in the presence of a basic substrate such as RNH2, an acid-base equilibrium (Eq. 1-8) lowers the concentration of both HI (thus hindering Pd(0) reoxidation) and free RNH2. As a consequence, the overall carbonylation process tends to be inhibited. The equilibrium shown in Equation1-8 is strongly shifted to the right in polar solvents. HI + RNH2RNH3+ + I(1-8 ) The use of coordinating aprotic solvent of low polarity is essential for the carbonylation reaction. In low-polarity aprotic solvents, however, the basicities of amines are significantly reduced with respect to polar solvents.30 Therefore, in a low-polarity solvent, the catalytic cycle can run effectively even in the presence of a basic substrate. The PdI2-catalytic system in the presence of excess CO2 (40 atm) results in an increase in the yields of urea formed from primary aliphatic amines. In the absence of CO2, there was a 16% decrease in the yield of urea. This CO2 effect is related to the fact that substrate basicity is further diminished in the presence of CO2 by formation of a carbamate species (Scheme 1.15). The reaction did not occur in the absence of carbon monoxide, which means that CO2 indeed acts as a promoter, and not as a carbonylating agent. CO2 + RNH2RHN NHR O RNH2CO2-H+RNHCO2I, -CO2-[Pd(0), + I Scheme 1.15
13 Interestingly, secondary amines are unreactive under the above reaction conditions. This is because an isocyanate is a key intermediate for reaction. The isocyanate is formed from intermediate I as shown in Scheme 1.16. When the isocyanate intermediate is reacted with excess secondary amine, trisubstituted ureas result in moderate yields. PdI2 + CO + RNH2RNH2RHN NHR O -HIR = alkyl, aryl NC O R IPd NHR O I Scheme 1.16 Dialkylureas can also be prepared by electrocatalytic carbonylation of aliphatic amines under mild conditions.31 The reaction is achieved in the presence of Pd(II) and Cu(II) cocatalyst at 30C and at atmospheric pressure. Synthesis of ureas using Pd(II)/Cu(II) or Pd(II)/I2 is not selective and side products such as oxamides and isocyanates are generally obtained. Reaction with a Pd catalyst requires the use a stoichiometric oxidant to transform Pd(0) (that results from the reaction) into Pd(II) (the reactive oxidation state of the catalyst). Usually high pressures of CO and O2 are necessary, sometimes in combination with I2.29,32 Alternatively, CO, O2, and copper salts are used.28 2 RNH2 + CO Pd cat.+ 0.4V vs. SCEHN NH O R R + 2H+ + 2e-(1-9) Most recently Chiarotto and Feroci33 found a more selective Pd(II) electrocatalytic process for the carbonylation of aliphatic amines (Eq. 1-9). The corresponding disubstituted ureas were obtained in moderate to good yields. Selectivity was good with little or no formation of oxamide. There is no evidence of isocyanate formation. RNH2 + CH3COMn(CO)5CH3COMn(CO)4NH2R+ CO(1-10 )
14 The study of manganese catalysts for the carbonylation of amines to ureas arose when some anomalies were observed during kinetic studies (Eq. 1-10). Later studies showed that decacarbonyldimanganese and pentacarbonylmethylmanganese can catalyze the carbonylation of primary aliphatic amines to disubstituted ureas.24 On heating primary aliphatic amines to 180-200 under a pressure of CO (106 135 atm), in the presence of Mn2(CO)10 or CH3Mn(CO)5, the corresponding disubstituted ureas were obtained in moderate yields. Small yields of N-alkyl formamides were also obtained, but their formation is believed to be due to a reaction pathway unrelated to the one that produces ureas. This was supported by a control experiment in which N-alkyl formamides were formed in the presence of manganese (II) acetate or in the absence of any manganese compound. Little or no gas absorption was observed during the reaction and hydrogen was detected. On this basis, Equation 1-11 was proposed. Secondary aliphatic amines, however, were unreactive; and aromatic amines such as aniline gave very poor yields of the corresponding diphenylurea. 2 RNH2 + CORHN NHR O R = alkyl, aryl + H2(1-11) Poor yields of N-alkyl formamides were attributed to the fact that the overall reaction occurs in two steps (Scheme 1-17). Calderazzo proposed that since both steps take place in the coordination sphere of the metal, the condition for the formation of the dialkylurea is that the alkyl formamide formed in the first step is retained on the metal, and undergoes a fast nucleophilic attack on the formyl carbon atom by the second molecule of amine. Hydrogen formation in the second step could not be explained at the
15 time.24 Results of the early investigations indicated that an essential prerequisite for successful carbonylation of the amine is the consumption of Mn2(CO)10. RNH2 + CORHN NHR O R = alkyl, aryl H NH O R H NH O R + RNH2+ H2 Scheme 1-17 The Mn2(CO)10-catalyzed reactions of n-butylamine, benzylamine and cyclohexylamine with CO to yield the corresponding ureas were examined by Angelici et al. This is a follow-up to Calderazzos24,34 pioneering work which addressed unanswered questions from the original paper.24 The reaction of Mn2(CO)10 with primary aliphatic amines gives a carbamoyl complex (Eq. 1-12). Under CO pressure, the isolated carbamoyl complex reacted to give the urea (Eq. 1-13). The mechanism is believed to involve the formation of an organic isocyanate intermediate. It was found that for primary amines the yield of the corresponding urea increases with increasing temperature and pressure. The reaction reaches equilibrium after 12 hours. It was previously reported by Calderazzo24,34 that secondary amines were unreactive under the Mn2(CO)10 catalytic conditions. Angelici et al found that secondary amines actually inhibit carbonylation. Mn2(CO)10 + 3 RNH2 cis-Mn(CO)4(NH2R)(CONHR) + RNH3+ + Mn(CO)5-(1-12 ) cis-Mn(CO)4(NH2R)(CONHR) + CO(1-13) NH NH O R R + HMn(CO)5 Diamines such as ethylenediamine and 1,4-diaminobutane were surprisingly unreactive. On the other hand, 1,3-diaminopropane reacted to form a low yield of the corresponding cyclic urea, with the major product being 1,4,5,6-tetrahydropyrimidine
16 (Eq.1-14). The side product is believed to result from the formamide (19) initially formed in the catalytic reaction. Under the conditions of the reaction and on distillation during workup, 19 is known to lose water, to give tetrahydropyrimidine. Tetrahydropyrimidine is also formed in the absence of Mn2(CO)10, but only after an induction period of 20 hours, which is not seen in the presence of Mn2(CO)10. When 1,6-diaminohexane was subjected to the reaction conditions above only polyurea was formed. NH2NH2 HNNH NNH +O 60 % 6%(1-14)Mn2(CO)1012h, 200 C NHNH2 H O 19 Other manganese catalysts such as (-methylcyclopentadienyl) manganese tricarbonyl (20) have been also used in the carbonylation of primary amines to the corresponding disubstuted ureas35 (Eq. 1-15). MnCO OC CO 20 H3 C 2 RNH2 20, hv100-250 h, ArHN NH O R R (1-15)R = n-C4-C9-alkyl, sec-C4H9, cyclo-C6H11, C6H5CH2Other transition metals have been used for carbonylation of amines to ureas, but not commonly. Cobalt complexes, for example, can only catalyze anilines to form the
17 corresponding diaryl ureas (Eq. 1-16). The selectivity for the oxidative carbonylation of amines with cobalt catalysts is not very high. CoNN O O ( 21 ) X NH2 2 21, O2, COCH3OHNH NH O X X (1-16) When N, N-bis(salicylidene)ethylenediaminecobalt(II) (21) is used to carbonylate amines in the presence of methanol, CO (1 atm) and dioxygen (1 atm); the oxidative carbonylation produces a number of side products. Optimization work carried out by Rindone et al. at increased pressure of CO and O2 using 21 eventually obtained a higher selectivity of urea.22 Later work using 21 established that the highest urea selectivity was obtained when para-substituted anilines were used.36 Minor amounts of urea were obtained when ortho-substituted anilines and meta-substituted anilines were used. These observations were attributed to steric effects. Less-expensive nickel catalysts were investigated as an alternative to palladium. The choice of nickel as an alternative was prompted by the ability of Ni-complexes to promote carbonylation and to give stable carbamoyl intermediates. Carbamoyl complexes are considered to be key intermediates in the catalyzed synthesis of ureas and carbamates.21 Formation of low yields of N,N-dialkyl urea resulted when nickel (II) complexes were used to oxidatively carbonylate primary aliphatic amines (Eq. 1-17).
18 2 RNH2 [NiII], O2, CO70 CHN NH O R R (1-17)R = C4H9, cyclo-C6H11, C6H5CH2[NiII] = NiX2(RNH2)4, R = alkyl, X = Br, Cl The study of the reaction mechanism revealed that the amines were carbonylated in a reductive step, during which either oxamide or urea was obtained, depending on the amount of water present. When dioxygen was used as the oxidant, the oxidative step produced water, which promoted urea formation. The oxygen was also responsible for the low yields owing to a side reaction involving amine oxidation. Nickel complexes, unfortunately, have not found a foothold in carbonylation catalysts because of their low reactivity. This is true of a number of complexes of other transition metals; such as platinum, rhodium, and iron. The issues with most transition metal complexes are the low yield of urea and formation of side products. In addition, most transition-metal catalysts cannot carbonylate both primary and secondary aliphatic amines. They are mainly successful in the formation of N, N-diaryl ureas from the primary aromatic amines. More recently, gold has been used to catalyze the carbonylation of amines to form diformamides37 and carbamates.32 Alkyl ureas have been found to be formed in good yield when an Au(I) catalyst such as Au(PPh3)Cl is used.32 Tungsten complexes have been recently used to carbonylate amines to ureas. Studies carried out by McElwee-White et al.,38 found that direct carbonylation of primary amines to N, N-disubstituted ureas could be attained using the tungsten (IV) imido carbonyl complex [(CO)2W(NPh)I2]2 in a stoichiometric reaction (Eq. 1-18).
19 R = H, alkyl, arylRHN NHR O 2 RNH2[(CO)2W(NPh)I2]2(1-18) The reaction is carried out at room temperature under nitrogen, followed by oxidation with air. Unfortunately, this method is not applicable to the carbonylation of secondary amines to form tetrasubstituted ureas because formamides are formed instead. Efforts to improve yields of ureas and the range of amines substrates led to the W(CO)6/I2 catalytic system. In this particular catalytic system, the I2 acts as the oxidant, CO is the carbonyl source, and potassium carbonate acts as the base to regenerate the free amine from the amine salt formed during the course of the reaction. This reaction was applicable to the carbonylation of both primary and secondary aliphatic amines to ureas (Table 1-1).39 However, it fails for aromatic amines.39 2 RR'NHW(CO)6, I2CO, K2CO3 N N O R' R R + 2 HIR' (1-19) Table 1-1. Oxidative carbonylation of primary and secondary amines Rn-Pri-Prn-But-BuPhR'HHHHHYield(%)915384720 R' = REtn-Bui-PrBnPhYield(%)40trace0550 a Isolated yield of urea calculated per equivalent of amine. Values are 5%. b Reaction conditions: amine (7.1 mmol), W(CO)6 (0.14 mmol), I2 (3.5 mmol), K2CO3 (10.7 mmol), CH2Cl2 (21 mL), 70 C, 80 atm CO, 24 h. c The solvent was CH2Cl2 (21 mL) and H2O (3 mL).
20 When compared to the other group VI metal carbonyls (chromium hexacarbonyl and molybdenum hexacarbonyl), tungsten hexacarbonyl works best for the oxidative carbonylation of primary and secondary aliphatic amines to the corresponding ureas. In investigating the scope of this methodology, it was found that primary and secondary diamines could be converted to the corresponding cyclic ureas with ring sizes ranging from five to eight members.40 As ring size increased from five to eight, however, there was an overall decrease in yield. When the Thorpe-Ingold effect was introduced by adding geminal dialkyl substituents to the backbone,41 there was an increase in yield (Table 1-2). H2N(CH2)n+2NH2 HNNH O nW(CO)6CO/ I2/ K2CO3(1-20) Table 1-2. Oxidative carbonylation of primary diamines HNNH O HNNH O HNNH O HNNH O Ph Ph HNNH O HNNH O Bu Bu HNNH O NHHN O NHHN O Bu Et 4051384680trace507038Yield (%)Yield (%)Yield (%) a Isolated yield of urea calculated per equivalent of diamine. Values are 5%. b Reaction conditions: amine (7.1 mmol), W(CO)6 (0.14 mmol), I2 (7.1 mmol), K2CO3 (14.2 mmol), CH2Cl2 (21 mL), 70 C, 80 atm CO, 24 h. c The solvent was CH2Cl2 (21 mL) plus H2O ( 3 mL).
21 One concern for early metal catalysts is functional group compatibility. The W(CO)6/I2 catalytic system can operate in the presence of a variety of functional groups as demonstrated in Eq. 1-21 and Table 1-3.42 Due to the commercial availability and ease of handling of the catalyst, the W(CO)6/I2 catalytic system would be an alternative to phosgene derivatives and main group catalysts for laboratory scale syntheses. In addition, its compatibility with various functional groups makes it a good candidate for carbonylation of complex molecules to the corresponding ureas. The subsequent chapters report the application of W(CO)6/ I2 catalyzed carbonylation to complex substrates. The scope and limitations of the reaction are discussed. NH2 2 W(CO)6, I2CO, K2CO3X NH NH O X X (1-21) Table 1-3. Oxidative carbonylation of para-substituted benzylamines Yield (%)a(CH2Cl2)36041453728Yield (%)a,c(CH2Cl2/H2O)737777708170 Yield (%)a(CH2Cl2)63353047245Yield (%)a,c(CH2Cl2/H2O)553769766814p-H p-CO2Etp-Cl p-COOHp-Br p-CH=CH2p-OMe p-NO2p-SMe p-CNp-CH2OH p-NH2X X a Isolated yield of urea calculated per equivalent of amine. Values are 5%. b Reaction conditions: amine (7.1 mmol), W(CO)6 (0.14 mmol), I2 (3.5 mmol), K2CO3 (10.7 mmol), CH2Cl2 (21 mL), 70 C, 80 atm CO, 24 h. c The solvent was CH2Cl2 (21 mL) plus H2O ( 3 mL). Other conditions are as in footnote b.42
CHAPTER 2 W(CO)6/I2 CATALYZED OXIDATIVE CARBONYLATION OF AMINES AND DIAMINES: APPLICATION TO SYNTHESIS OF THE CORE STRUCTURES OF THE HIV PROTEASE INHIBITORS DMP 323 AND DMP 450. Earlier work demonstrated the catalytic carbonylation of aliphatic amines to ureas using W(CO)6 as the catalyst and I2 as the oxidant.38,39,42,43 Diamine substrates with simple hydrocarbon linkers were converted to cyclic ureas with five to eight-membered rings.40,41 Because the reaction conditions are relatively mild and the functional group tolerance is surprisingly broad, this procedure was a reasonable candidate for use in synthesis of complex targets. As an illustration of its utility, installation of the urea moiety into the core structure of the HIV protease inhibitors DMP 323 and DMP 4501,2,44 by catalytic carbonylation of the corresponding diamine is reported. NN O HO OH Ph Ph OH HO DMP 323DMP 450NN O HO OH Ph Ph NH2 H2N One of the key viral enzymes that have been targeted for the discovery of HIV therapeutics is protease. DuPont Merck Pharmaceuticals first synthesized C2 symmetric diols as protease inhibitors.45 The key feature of DMP 323 (22) and DMP 450 (23) is a C2 symmetric diol which provides the correct binding site configuration for the protease enzyme. A 7-membered cyclic urea moiety provides a scaffold for the diol. The availability of a substantial body of literature on the synthesis of DMP 323 and DMP 450 22
23 derivatives allows direct comparison of the catalytic carbonylation reaction with stoichiometric reaction of the same substrates with phosgene derivatives. In previously reported routes to DMP 323, DMP 450 and related compounds, the urea moiety was installed by reaction of phosgene or a phosgene equivalent with an O-protected diamine diol. In the initial small-scale preparations, a primary diamine was reacted with the phosgene derivative 1,1'-carbonyldiimidazole (CDI),2,45-47 followed by N-alkylation as appropriate (Scheme 2.1). NHCbz H3C O NH HN H3C CH3 HOOH HN NH H3C CH3 SEMOOSEM ab,cOH NCbzNCH3 O H3C OMe O e,fSEM = 2-(Trimethylsilyl)ethoxymethylReagents:(a) i-BuOCOCl, CH3ONHCH3.HCl(b) LiAlH4(c) VCl3(THF)3, Zn-Cu(d) SEMCl(e) cat. Pd(OH)2(f) 1,1'-carbonyldiimidazole(g) PhCH2Br, NaH(h) HCl.dioxane/MeOH g,hN N H3C CH3 HOOH O NH HN H3C CH3 SEMOOSEM d H Cbz Cbz Cbz Cbz Scheme 2.1 The practical route to DMP 450 utilizes phosgene to form the cyclic urea from a secondary diamine.45 Since all of these routes require protection of the diol, extensive protecting group studies have been carried out.45,48 Three of the previously described O-protected diamine diols: acetonide 24,49 MEM ether 252,50 and SEM ether 26,2 were
24 chosen as representative examples bearing cyclic and acyclic protecting groups, respectively. Carbonylation of 24-26 allows comparison of the W(CO)6-catalyzed process to the stoichiometric reactions of the phosgene derivative CDI (Eq 2-1).51 NH2NH2 P1O OP2 HNNH P1O OP2 O W(CO)6/ COI2, K2CO324 P1, P2 =25 P1, P2 = MEM, MEM26 P1, P2 = SEM, SEMC(CH3)2 27 P1, P2 =28 P1, P2 = MEM, MEM29 P1, P2 = SEM, SEMC(CH3)2 (2-1) Compound 24 was prepared using literature procedures (Scheme 2.2). The difficulties associated with converting acetonide 24 to urea 27 have been discussed in the literature.49 Reaction of 24 with CDI in acetonitrile under standard conditions results in a 15% yield of 27, with the low conversion attributed to strain in the bicyclic product (Table 2-1). OO O O OO OO OO H H NN OO H H N N DIBALH2NNMe2 MeOH sec-BuLiTHF, PhCH3NHHN OO N N Ra Ni/ H2NH2NH2 OO 24 Scheme 2.2
25 Table 2-1. Carbonylation of compounds 24-26 aTypical reaction conditions: Diamine 25 (0.200 mmol), W(CO)6 (0.0242 mmol), K2CO3 (0.635 mmol) and I2 (0.239 mmol), solvent (32 mL CH2Cl2 : 8 mL of water), 80 atm CO, 80 C, 18 h. bNot reported. cThis work. dYields are from two-step sequence involving deprotection of the Cbz-protected diamine. Deprotection is assumed to be quantitative for purposes of the table. Diamine Reagent Solvent T (C) % Yield Urea Ref 24 CDI CH3CN NRb 15 45 24 CDI TCE 140 67 45 24 W(CO)6/CO CH2Cl2/H2O 80 38 c 24 W(CO)6/CO CH2Cl2 80 26 c 25 CDI CH2Cl2 rt 62,76d 2,47,50 25 W(CO)6/CO CH2Cl2/H2O 80 42 c 26 CDI CH2Cl2 rt 52,93d 2,47 26 W(CO)6/CO CH2Cl2/H2O 80 75 c Although higher yields of 27 could be obtained by monoacylation of 24 with CDI followed by reflux at high dilution in tetrachloroethane (TCE), there were practical difficulties with this two-step procedure.45 In comparison, catalytic oxidative carbonylation of 24 in the biphasic CH2Cl2/H2O solvent system afforded 27 in 38% yield. As had been observed for the carbonylation of functionalized benzyl amines,42 yields obtained by using the biphasic solvent system were higher than those in CH2Cl2. Efforts to optimize the reaction conditions by varying CO pressure, temperature, concentration and solvent did not result in higher yields of 27. Although the yields of 27 from 24 are modest, results from the catalytic carbonylation compare favorably to those obtained with CDI under typical conditions. Acyclic protecting groups such as MEM and SEM had been previously explored as alternatives that would lessen the strain problems that had plagued preparation of bicyclic acetonide 27 from diamine 24.2,50 Accordingly, MEM ether 25 and SEM ether 26 were chosen as additional test substrates for conversion to the corresponding cyclic ureas via catalytic carbonylation.51
26 Carbonylation of 25 and 26 was carried out under the conditions used for 24, with the exception of substrate concentration, which was optimized for 25 and the same used for 26 (Table 2-2). In comparison to the literature yields of 62 and 76 % for formation of urea 28 from Cbz protected MEM ether 25 and CDI under slightly different conditions, the catalytic carbonylation reaction provided 28 in 42% yield from 25. Promising results were also obtained for SEM ether 26, for which catalytic carbonylation afforded urea 29 in 75% yield, a value intermediate between the reported yields for reaction of 26 with CDI. Upon optimizing, it was found that lowering the pressure of carbon monoxide or lowering the temperature resulted in lower yields of 29. On the other hand increasing the temperature and pressure did not result in an increase in yield. (Table 2-2) Table 2-2. Optimization studies for the oxidative carbonylation of 26 Entry P(CO) / atm T / C % Yield 29a 1 90 102 65 2 90 80 36 3 77 80 75 4 70 80 42 5 60 80 36 6 79 60 39 aTypical reaction conditions: Diamine 26 (0.164 mmol), W(CO)6 (0.007 mmol), K2CO3 (0.427 mmol) and I2 (0.166 mmol), solvent (32 mL CH2Cl2 : 8 mL of water), 77 atm CO, 82 C, 18 h. Although the yields from both the stoichiometric CDI and catalytic carbonylation methods for urea formation varied with the nature of the protecting group on the diol, the overall results from the two methods were comparable. Due to the fact that low yields were obtained when diamine 24 was carbonylated, an investigation was undertaken which examined how the structure affected the yields.
27 Modification of the substrate structure was investigated. This involved changing the stereochemistry of the acetonide to cis. The synthesis of cis acetonide diamine 33 from dimethyl maleate was attempted according to Scheme 2.3. The scheme failed at the conversion of the dihydrazone to dihydrazine. This may be attributed to large steric hindrance encountered in the chelate controlled addition of benzyl lithium. O O O O O O O O HO OH O O O O OO H O H O OO H N H N OO N N NH HN OO N N NH2 NH2 OO 2930313332NMOK2OsO2(OH)4MeO OMe DIBALMe2NNH2[BnLi]Ni/ H2X Scheme 2.3 DMP 323 and DMP 450 are representatives of a class of HIV protease inhibitors that possess a characteristic C2 symmetric diol moiety. The phosgene methodology that is currently employed in the carbonylation of amino alochols to the corresponding hydroxyl ureas does not operate well in the presence of OH functionalities.48 Carbamates are formed in a competing reaction unless the hydroxyl groups are protected. Thus, previous reported syntheses of these compounds invariably involve sequences of protection and deprotection steps. 46
28 W(CO6, I2CO, K2CO371 %2 W(CO6, I2CO, K2CO371 % HO NH2 HN NH O OH HO 3 4 (2-2) In contrast, the W(CO)6/I2 catalytic system had been demonstrated to be tolerant of OH groups in the reaction depicted in Eq. 2-2.42 Conversion to urea 34 occurred in good yield with no evidence of carbamate or carbonate formation. This result suggests that application of the W(CO)6/I2-catalytic system to carbonylation of substrates containing unprotected OH group(s) could be feasible. In the carbonylation of [4-(aminomethyl)phenyl]methanol (Eq 2-2), the OH group is para with respect to the amine so as to eliminate the possibility of an intramolecular reaction occurring. It should be noted, however, that substituents where intramolecular reaction was possible had not been examined before this study. The carbonylation of the diamine diol 35 was carried out under various reaction conditions and in the presence of W(CO)6, Mo(CO)6, and Cr(CO)6 as catalysts. Possibilities include closure to cyclic urea 36 and the oxazolidinones 37 and 38 (Eq 2-3). The formation of five-membered rings in 37 and 38 is kinetically favored over the formation of the seven-membered diol urea 36, but nitrogen nucleophiles are intrinsically more reactive than hydroxyls (Eq 2-2). This competitive effect makes 35 a more interesting substrate than [4-(aminomethyl)phenyl]methanol (Eq 2-2). NH2 NH2 Bn Bn HOOH 3537HNO NHO O O Ph 38HNO NH2OH O Ph Ph M(CO)6, I2CO, K2CO3+ M = W, Mo or Cr(2-3)HNNH HO OH Bn Bn O 36Ph +
29 Table 2-3. Oxidative carbonylation of 35 catalyzed by W(CO)6 under various reaction conditions. Entrya T (C) Catalyst (mol %) % Yield 36 % Yield 37 % Yield 38 1 82 20 0 15 0 2 84 36 0 7 0 3c 64 36 0 8 13 4 64 8 0 13 0 5 25 8 0 9 not observed 6 25 0 0 trace 9 7 -2 20 0 trace trace 8 -8 100 0 trace trace 9d,e -10 100 0 trace trace 10e 25 100 0 19 6 11e 36 100 0 20 4 12e 82 100 0 25 4 13 82 10 0 0 0 a Concentrations were calculated based on moles of diamines used. All reactions were run at 85 atm CO, 1.75 h and equivalents of reactant are as follows; 1 eq. diamine, 1 eq. I2, 3 eq. K2CO3, concentration of 8.2 mM and W(CO)6 was used as catalyst. A biphasic solvent system (H2O: CH2Cl2 in a ratio of 1:4) was used. b Run overnight c Concentration of 4.0 mM used. dCH2Cl2 used as solvent. e2 eq. of I2 was used. Diamine diol 35 was prepared according to the literature in 41-64% yield.52 The carbonylation of 35 produced the monoand bisoxazolidinones 37 and 38 respectively (Table 2-3). Compounds 37 and 38 are formed from the intramolecular carbonylation reaction involving hydroxyl and amine groups. The urea (36) that would be formed from the intramolecular carbonylation of the amines ends was not observed. The yields of 37 and 38 are low and, in cases where the reaction conditions were extremely mild, starting material was isolated. The yields of 37 and 38 were fairly
30 constant in light of changes of reaction variables. It was found that formation of monooxazolidinone 37 occurs in 9% yield even in the absence of catalyst (entry 6). Other group VI catalysts, i.e., Mo(CO)6 and Cr(CO)6 were examined to determine if preference for the carbonylation of amine over alcohols could be exploited, to overcome the closure to the 5-membered ring(s) (Table 2-4). Results are similar to those obtained with W(CO)6. Variation of the concentration was not pursued as two intramolecular reactions are in competition and the 5-membered ring formation is kinetically favored. Table 2-4. Oxidative carbonylation of 35 using Mo(CO)6 and Cr(CO)6. a Concentrations were calculated based on moles of 35 used. All reactions were run at 85 atm CO, 1.75 h and equivalents of reactant are as follows; 1 eq. diamine, 1 eq. I2, 3 eq. K2CO3, concentration of 8.2 mM and 20 mol % catalyst. A biphasic solvent system (H2O: CH2Cl2 in a ratio of 1:4) was used. Entrya Catalyst T/ C % Yield 36 %Yield 37 % Yield 38 1 Mo(CO)6 64 0 3 26 2 Cr(CO)6 64 0 9 0 6.7% HCl/MeOH 88 %HNNH OO O HNNH HOOH O 2436(2-4) In order to possibly explain the lack of mass balance, and to rule out the possibility that 36 initially formed and then decomposed at the reaction conditions that 35 was run, investigation of 36 under carbonylation conditions was undertaken. An authentic sample of 36 was synthesized via the deprotection of the acetonide group of 24 (Eq. 2-4). Control experiments investigating the stability of 36 under the reaction conditions for
31 carbonylation of 35 were carried out and the reaction mixture analyzed by IR. The results of these control experiments are consistent with formation of 39 (Eq. 2-5). It was found that oxidation of the diol moiety of 35 to form 39 did occur under reaction conditions when the reaction was run overnight (18 hours). Shorter reaction periods (1.75 hrs) did not induce decomposition. These experiments rule out the possibility of the formation of the diol urea 36 upon carbonylation of diamine diol 35 for 1.75 hours. It can therefore, be concluded that under the conditions for the reaction of 35, the diol urea would have survived if formed. Thus, the results are consistent with formation of only 37 and 38. This gives credence to the fact that under carbonylation conditions the kinetic issue with respect to ring size is more important than the preference for NH2 over OH. HNNH HOOH O HNNH HOO O 3936(2-5)W(CO)6, I2CO, K2CO3 It is possible that the diol moiety of 35 was reactive under these reaction conditions based on results of the control experiments involving 36. This led to a control study where the simple 1,2 diol 40 was subjected to the same carbonylation conditions. Only dione 41, -hydroxy ketone 42 and starting material 40 were observed (Eq. 2-6). The 5-membered carbonate (43) that would be formed from carbonylation of the diol moiety, was not observed.
32 40HO 41%Yield W(CO)6, I2CO, K2CO3 OH + w. catalystno catalyst1319208SMrecovered4266 OO O O O OH 0042%Yield43%Yield(2-6)O Oxidation of the 1, 2-diol moiety occurs whether or not the catalyst is present. This indicates that the carbonylation reaction would not play a role in the lack of reactivity of a substrate with a 1, 2-diol moiety. It can be concluded that the 1,2 diol moiety becomes oxidized to the corresponding dione and -hydroxy ketone under carbonylation conditions by iodine. To further investigate formation of the oxazolidinone ring system, amino alcohol 44 was subjected to the same reaction conditions as 35 (Table 2-5). Amino alcohol 44 was chosen as it represented half of 35. The study sought to investigate what was necessary for the formation of the oxazolidinone 45. It was found that closure to the oxazolidinone 45 occurs independently of potassium carbonate or the catalyst. At high concentration neither urea 46 nor oxazolidinone 45 is observed. This may be due to oligomerization of 44. This phenomenon is confirmed by the observation of residue that sticks to the column during purification, which is consistent with oligomeric ureas. 44H2N Ph 45HNO O Ph W(CO)6, I2CO, K2CO3 OH +HN Ph NH Ph HO OH 46O (2-6)
33 Table 2-5. Oxidative carbonylation of 44 under various reaction conditions. Entry T/ C 44 (mM) %Yield 45 % Yield 46 1 25 16.0 25 0 2 -4 16.0 25 0 3b 25 16.0 14 0 4 25 99.6 29 0 5c 25 20.0 38 0 a Concentrations were calculated based on moles of amino alcohol used. All reactions were run at 80 atm CO, 1.75 h and equivalents of reactant are as follows; 1 eq. amino alcohol 1 eq. I2, and 3 eq. K2CO3. A biphasic solvent system (H2O: CH2Cl2 in a ratio of 1:4) was used. 8 mol % of W(CO)6 was used in each case. bNo K2CO3 was used. cNo catalyst was used. Oxazolidinone formation occurs in the absence of the catalyst. This indicates that the catalyst is not the only carbonyl source for the oxazolidinone. Potassium carbonate is another possible source of carbonyl in the carbonylation of 44. Use of labeled potassium carbonate, K213CO3 sought to determine if it was the source of carbonyl in the carbonylation of 44. Carbonylation of 44 was run under standard carbonylation conditions except labeled potassium carbonate substituted as base. From the 13C spectrum it was concluded that 13C isotope was not incorporated as the carbonyl peak was not enhanced. This was done by comparing the 13C spectrum of the labelled reaction product with that of the unlabelled reaction product. The peak heights were the same. This conclusively rules out potassium carbonate as the carbonyl source and leads to conclusion that CO gas is the carbonyl source. Intermolecular urea formation in related acyclic systems was probed by catalytic carbonylation of the O-protected amines 4753 and 48. Compound 47 was prepared
34 according to literature procedures.54 Compound 48 was prepared starting from (R)-N-Cbz-phenylalaninol (Scheme 2.4). NHCbz HO NHCbz SEMO SEMClNH2 SEMO 48 5% Pd/C, H268% Scheme 2.4 Both substrates afforded the corresponding ureas in respectable yields (Eq 2-8). The reaction conditions were identical to those used for diamines 24-26, with the exception of concentration. As expected for intermolecular reactions, the maximum yields of urea were obtained at higher substrate concentrations than those used with the diamines, for which oligomerization competes with ring closure. HN NH POOP W(CO)6, CO, I2 K2CO3 O NH2 PO 247 P = MEM48 P = SEM49 P = MEM 82%50 P = SEM 53%(2-8) Other seven membered urea systems have found application in biological chemistry. 1,3,4,7-Tetrahydro-2H-1,3-diazepine (52) found use as an intermediate in the synthesis of cytidine deaminase inhibitors. The inhibition of cytidine deaminase, a catabolic enzyme for the antitumor arabinosyl cytidine is an important objective in cancer chemotherapy. Urea 52 was prepared in the literature by reaction of cis-2,3-diamino butene with carbon sulfide. Cis-2,3-diaminobutene (51) was synthesized according to the literature procedure (Scheme 2.5). 55 Preliminary carbonylation reactions with 51 have not yielded 52.
35 Cl Cl N3 N3 H2N NH2 2 HClNaN3DMSOPPh3HCl40%51 Scheme 2.5 H2N NH2 2 HCl51 W(CO)6, CO, I2 K2CO3NHHN O 52(2-9) In summary, we have established that catalytic carbonylation of amines provides an alternative to phosgene and phosgene derivatives in the preparation of ureas. Use of this chemistry in preparation of the core structure of DMP 323 and DMP 450 provides the first demonstration of catalytic amine carbonylation as synthetic methodology. Yields of the ureas from the catalytic reaction vary with the protecting group on the diol, as do those reported for ring closure with stoichiometric CDI. Overall, the results are roughly comparable. It also have been established that carbonylation of substrates with a 1,2-diol moiety is not possible due to the oxidative side reaction that occurs in the presence of iodine. In addition, the carbonyl source in the carbonylation of -amino alcohols to oxazolidinone is CO gas. An oxazolidin-2-one will form preferentially over a seven membered cyclic urea under W(CO)6/I2 carbonylation conditions; the kinetic issue with respect to ring size is more important than the preference for NH2 over OH. Under carbonylation conditions with potassium carbonate used as base, -amino alcohols preferably form an oxazolidinone over an acyclic urea. These results are not entirely surprising as -amino
36 alcohols and phosgene (or its derivatives) are the starting materials in the classical syntheses of oxazolidin-2-one.56
CHAPTER 3 W(CO)6/I2 CATALYZED OXIDATIVE CARBONYLATION OF SECONDARY AMINES AND DIAMINES: ANALOGS OF THE CORE STRUCTURES OF THE HIV PROTEASE INHIBITORS DMP 323 AND DMP 450. One feature of the HIV protease inhibitors DMP 450 and DMP 323 is the presence of a seven-membered cyclic tetrasubstituted urea. To date, this moiety has been installed by alkylation of the O-protected N, N-disubstituted cyclic urea or direct cyclization of the secondary diamines with phosgene.45 Direct carbonylation of the corresponding secondary diamine with W(CO)6/I2/CO catalytic system could provide a good alternative to these multi-step sequences. Our ultimate target was synthesis of DMP 450 and DMP 323 directly from the corresponding secondary diamines. N N HOOH O NH2 H2N N N HOOH O HO OH DMP 450DMP 323 It has been previously demonstrated that the W(CO)6/I2-catalytic system can be successfully used for the oxidative carbonylation of secondary amines and diamines to the corresponding tetrasubstituted ureas. In initial studies39 using secondary amines, the optimal reaction conditions were found to be: 2 mol % W(CO)6, equimolar amounts of K2CO3 and amine, and 0.5 eq I2. The optimal reaction conditions were room temperature, 60-80 atm CO and CH2Cl2 as solvent. The yields of urea were found to fall 37
38 upon increasing steric bulk about the nitrogen of the secondary amine. Tetrasubstituted cyclic ureas with ring sizes ranging from five to eight can be attained in good to moderate yields upon carbonylation of the corresponding diamine. In addition, the N-substituent on the diamine can be varied.40 Carbonylation of secondary amines and diamines to the corresponding ureas via the W(CO)6/I2/CO-catalytic system has been continuously plagued with low yields and, in a few cases, formamide byproducts. If these problems could be solved, the catalytic reaction would be a useful alternative to phosgene especially for laboratory scale reactions. Our first approach was examination of synthesis of substructures of the HIV protease inhibitors, DMP 323 and DMP 450, under the optimized reaction conditions.39 This model study examined cases in which the benzyl substituents were absent from the two and five positions, reducing the steric hindrance at the N. In addition, the C2 diol was removed to maintain simplicity. N N HOOH O NH2 H2N DMP 450 Earlier work had indicated that carbonylation of N, N-dibenzyl-1,2-ethanediamine produces the corresponding cyclic urea in moderate yield.40 Its longer chain analog
39 N, N-dibenzyl-1,4-butanediamine and its nitro derivative fulfilled the criteria of simplicity and benzyl substituents on the nitrogen. N, N-dibenzyl-1,4-butanediamine and its nitro derivative were synthesized via the diimine using literature procedures.57 NH HN R R W(CO)6, I2 CO N N R R O (3-1)R = H, NO2 Oxidative carbonylation of N, N-dibenzyl-1,4-butanediamine and its nitro derivative were carried out using the W(CO)6/I2-catalytic system. The low conversion to urea observed in the catalytic carbonylation of simple secondary amines occurs as well for the carbonylation of secondary diamines. To that end, the use of TMEDA and N-methylmorpholine as bases was also examined (Table 3-1). Table 3-1. Study of effect of the presence of the benzyl group on the W(CO)6 /I2 catalyzed carbonylation of 1,4-secondary diamines. Entry R Base Yield Urea (%)a 1 NO2 K2CO3 0 2 NO2 TMEDA 0 3 NO2 N-Methylmorpholine 0 4 H K2CO3 0 5 H TMEDA 0 6 H N-Methylmorpholine 0b aReaction conditions: diamine (25 Eq.), W(CO)6 (1 Eq.), I2 (25 Eq.), K2CO3 (50-75 Eq), CH2Cl2 (40 mL), 24h. b Formamide was observed in addition to unreactied starting material Unfortunately, none of these reactions produced the desired urea. It can therefore be concluded that1,4-diamine substrates with benzyl groups on the nitrogens are too
40 sterically hindered to be carbonylated to the corresponding seven-membered cyclic ureas using the W(CO)6/I2-catalytic system. In addition, it can be concluded that TMEDA and N-methylmorpholine are not suitable bases for this transformation. When a less sterically encumbered group such as ethyl (53) is placed on the nitrogen, formation of the tetrasubstituted urea was possible although the preliminary work indicated that the conversion to the corresponding cyclic urea was low (Eq. 3-2). Synthesis of 53 from 26 using the imine methodology is outlined in equation 3-3. This synthetic route is representative. W(CO)6, I2CO, K2CO3N N OO O 7 %NH HN O O O Si O Si O Si O Si 5354(3-2) 1. R'CHO, MgSO42. NaBH4 R' = CH3, R = CH3CH2 35%H2N NH2 O O O Si O Si 26NH HN O O O Si O Si R R 53,(3-3) Our next approach was synthesis of acyclic analogues of the HIV protease inhibitors DMP 323 and DMP 450 under the optimized reaction conditions.39 Portions of the DMP 323 and DMP 450 backbone allow us to take a look at the substituents at the 2-position and on the nitrogen in the absence of ring closure issues. Amines that
41 represented portions of the HIV protease inhibitors DMP 323 and DMP 450 were carbonylated. Amine 55 represented fragment A. Upon carbonylation, however, the corresponding urea was not formed. This is because of the excess steric bulk around the nucleophilic nitrogen. Amine 56 represented fragment B. Here the steric bulk around the nitrogen is reduced. The yields of the corresponding urea 57, were low with the optimal yield at 16%. The optimal reaction conditions were amine (1 eq.), K2CO3 (2 eq.), iodine (1 eq.), room temperature, 18 hours, and 80 atm CO in methylene chloride. Going to a biphasic (methylene chloride/ water) solvent system reduced the yield to 7%. N N HOOH O NH2 H2N A B C NH 5 5 NH 5 6 Secondary amines with methyl on the nitrogen and variance of the substituent at the C2 (fragment C ) were also studied. The hydroxyl group present on the substrate was protected to eliminate its involvement in side reactions. Amine 58 was commercially available. Amines 62 and 65 were synthesized according to Schemes 3.2 and 3.3.
42 NH H OCH3 5 8 1.BH3 THF(1.5 eq.) THF 2.5 hrs2. MEMCl, 18 h, RT 98% N H3C O O HO O N H3C O O MEMO NH H3C MEMO 10% Pd/C, H2 EtOH, 18h 46 %606162 Scheme 3.2 N O O HO O N O O MEMO 1.BH3 THF(1.5 eq.) THF 2.5 hrs2. MEMCl, 18 h, RT 100%NH MEMO 10% Pd/C, H2 EtOH, 18 h 92 %63646 5 Scheme 3.3 Formation of urea 59 from the carbonylation of amine 58 occurred in modest yields. It was observed that as the steric bulk at C2 increased the yield of the urea decreased (Table 3.2, Entries 5, 8, and 11). When the C2 substituent was changed to methyl and benzyl only trace amounts of urea were isolated. Overall, conversions of these secondary amines to the corresponding tetrasubstituted ureas were modest at best.
43 Potassium carbonate had previously been found to be ineffective for deprotonating the amine salt of secondary amines. 39 The search for other bases is an ongoing effort and to that end a model study was undertaken to search for an alternative to potassium carbonate. The commercially available amine 66 was used as representative secondary amine (Eq 3-4). Table 3-2. Optimization study of the carbonylation of amines 58, 62 and 65 a Reaction conditions: amine (14.2 mmol), W(CO)6 (0.28 mmol), I2 (7.1 mmol), K2CO3 (14.2 mmol), CH2Cl2 (40 mL).b Reaction conditions as is except the solvent was (CH2Cl2/ H2O) in a 4:1 ratio. Entry Amine Amine (M) Time/hrs T/ C Yield (%) Urea 1 58 0.3 M 6 25 trace 2 58 0.3 M 6 60 trace 3 58 0.3 M 18 60 7 4 58 0.3 M 18 25 traceb 5 58 0.3 M 18 25 34 6 58 0.3 M 18 80 13 7 58 1.0 M 18 25 16 8 62 0.3 M 18 2 trace 9 65 0.30 M 48 82 trace 10 65 0.13 M 18 82 0 11 65 0.30 M 18 25 trace NH 2 N N O W(CO)6, I2CO, K2CO367(3-4)66 Inorganic and organic bases were examined as alternatives for K2CO3 (Table 3-3). A bulky amine such as diisopropylamine is a poor carbonylation substrate because of the steric hindrance about nitrogen but is a good base. It, however, proved to be a poor base for carbonylation of secondary amines.
44 It had previously been found that aniline could not be carbonylated under the optimized reaction conditions.39 When the aromatic amine was made more basic as in the case of 68, only unreacted starting material was observed. Anisidine 68, like diisopropyl amine, is a good base and could potentially be used as a substitute for potassium carbonate. This is due in part to its lack of reactivity under carbonylation conditions (Eq. 4-5). When it was utilized as a base for the carbonylation of 66, however, no urea was observed. Table 3-3. Effects of base on the oxidative carbonylation of N-ethylbenzylamine, 66 Entry Base Base (equiv) Yield a, b (%) 1 K2CO3 1.0 33 2 K2CO3 1.5 18 3 Na3PO4 1.0 34 4 Na3PO4c 1.0 0 5 diisopropylamine 1.0 0 6 CsCO3 1.0 trace 7 None 0 17 8 68 1.0 0 a Isolated yield of urea calculated per equivalent of amine b Reaction conditions: amine (14.2 mmol), W(CO)6 (0.28 mmol), I2 (7.1 mmol), K2CO3 (14.2 mmol), CH2Cl2 (40 mL), room temperature, 18 h. c Reaction conditions as is except the solvent was (CH2Cl2/ H2O) in a 4:1 ratio. NH68 O N N O 69 O O XW(CO)6, I2CO, K2CO3(4-5) All the bases tried resulted in poorer yields of urea than with K2CO3, with only Na3PO4 (with CH2CH2 as solvent) being comparable. When Na3PO4 was used under biphasic conditions (CH2Cl2/ H2O), no urea was observed. This may be because of a
45 production of HOin the aqueous phase. Hydroxide has previously been demonstrated to cause problems in related carbonylation reactions. The optimal results with one equivalent of K2CO3 is in keeping with earlier optimization studies.39
CHAPTER 4 APPLICATION OF W(CO)6/I2 CATALYZED OXIDATIVE CARBONYLATION TO THE CARBONYLATION OF AMINO ALCOHOLS Phosgene reacts with the amine and alcohol ends of amino alcohols such as 3-aminopropanol and 5-aminopentanol to form isocyanate chloroformates instead of the corresponding urea or carbamates (Eq 4-1).58, 59 CDI reacts with alcohols and amines to form carbonates and ureas respectively.60 HO NH2 nn = 3, 3 -aminopropanoln = 5, 5-aminopentanol phosgeneCN O O Cl O nn = 3, 3-isocyanatopropyl chloroformaten = 5, 5-isocyanatopentyl chloroformate(4-1) W(CO)6 catalyzes oxidative carbonylation of amino alcohols. Work described in Chapter 2, essentially established that an oxazolidin-2-one will form preferentially over a seven-membered cyclic urea as under the carbonylation conditions, the kinetic preference with respect to ring size is more important than the preference for NH2 over OH (Eq. 2-3, Table 2-3). Amino alcohols preferably form an oxazolidinone over the corresponding acyclic urea with carbon monoxide as the carbonyl source (Eq. 2-7, Table 2-5). These results are not entirely surprising as -amino alcohols and phosgene (or its derivatives) are the starting materials in the classical syntheses of oxazolidin-2-one.56 In light of the interesting results obtained for the carbonylation of diamine diol 35 and amino alcohol 44, an investigation into the reaction of various amino alcohols using the W(CO)6/ I2 catalytic system was undertaken. The study examined the 46
47 W(CO)6/ I2 catalyzed oxidative carbonylation of various amino alcohols and the selectivity of W(CO)6/ I2 oxidative carbonylation towards formation of urea when amino alcohols are used as substrates and formation of carbamates is possible. The case to determine whether a seven membered cyclic carbamate or an acyclic carbamate would form over the corresponding acyclic urea was investigated using 4-amino-2-methylbutanol as a representative substrate. The optimal reaction conditions were found to be at a concentration of 4M, 82 C and a reaction time of 8 hours produced good yields of the diol-urea (Entry 5) and overall, the yields ranged from moderate to good (Table 4-1). Compounds 71 and 72 were not observed when products were investigated by NMR and IR. HO NH2 HO HN HN OH O +OHN O W(CO)6, COI2697071HO HN O NH2 O 72 +(4-2) When potassium carbonate was used as base in the carbonylation of 69, urea formation was confirmed by various spectral methods but purification was difficult as potassium iodide was a persistent contaminant. Potassium iodide is a side product of the carbonylation reaction when potassium carbonate is used as base. Its solubility is similar to that of the diol-urea. This makes it difficult to purify the urea by methods such as chromatography or selective extraction. This particular problem occurs when the urea is
48 very polar and was encountered with substrates such as 3-aminopropanol and 4-aminobutanol. Table 4-1. Carbonylation of 69 under various reaction conditions. Entry 69 (M) Time (hr) Temperature ( C) % Yield 70 % Yield 71 % Yield 72 1a,b 4.0 3.5 40 25 0 0 2 4.0 3.5 40 41 0 0 3 4.0 18.0 40 60 0 0 4a 4.0 18.0 82 46 0 0 5 4.0 8.0 82 93 0 0 6 2.0 8.0 82 80 0 0 7 Neat 8.0 82 52 0 0 aTypical reaction conditions: amine (1eq), W(CO)6, (5 mol %), pyridine (1.5 eq) and I2 (0.5 eq), CH2Cl2,, 80 atm CO. bTypical reaction conditions except DMAP used as base. When pyridine is used as base, no purification is necessary as urea 70 is isolated as pure after work up. Workup was carried out by dissolving the crude residue from carbonylation in water, saturated sodium sulfite and sodium bicarbonate. The aqueous solution was washed with methylene chloride to remove excess pyridine and finally washed with ethanol/ chloroform solution to extract the pure urea. To compare the carbonylation results to a phosgene derivative, 4-amino-2-methylbutanol was treated with CDI, at a concentration of 4M. Compounds 70 and 71 were observed with compound 70 as the major component of the product mixture being formed in 70 % yield and 71 being formed in trace amounts (Equation 4-3). There was no evidence for the formation of 72.
49 HO NH2 HO HN HN OH O +OHN O CDITHF, RT18 hours697071HO HN O NH2 O 72 +tracenot observed(4-3)70% The formation of six-membered cyclic carbamate or an acyclic carbamate over the corresponding acyclic urea was investigated using 3-amino-4-phenyl-butanol (70) as a representative substrate. NH2 OH NH2 OH O BH3 THF4.5 hrsHCl(4-4)7374 Substrate 74 was synthesized by reduction of DL--homophenylalanine (73) with BH3THF at 0 C Celsius. Amino alcohol 74 was oxidatively carbonylated using the W(CO)6/ I2 catalytic system (Eq. 4-5) NHNH HO HO O ONH O +NH2 OH (4-5)74757622W(CO)6, COI2ONH OH NH2 O + 77
50 When the amine hydrochloride of 74 was used in the carbonylation in the presence of excess base, the yields were low. This low reactivity is due to lack of stirring which can be attributed to the large amount of solid potassium carbonate that is needed and the low volume of solvent used. Lack of stirring affects the selectivity, where the intramolecular reaction to form 76 is promoted when there is a lower solvent volume even though higher concentrations should have promoted urea formation. This may be explained based on the fact that lack of stirring in the reaction mixture disallows amine-amine interactions which results in urea formation. To alleviate this problem, the base was changed from potassium carbonate to pyridine and free amine was used instead of the amine hydrochloride. The free amine 74 was carbonylated at the optimal reaction condition for acyclic amino alcohols (Table 4-1, entry 5). Conversion of starting material was low (44% was recovered) with 75 being formed in 9 % yield and 76 being formed in 5 % yield. Compound 77 was not observed. There is selectivity for urea formation over carbamate despite the low conversion. NHNH HO HO O ONH O +NH2 OH (4-6)74757622ONH OH NH2 O + 77S O S CH3OH, 70 C4.5M, 18 h To compare the carbonylation results to phosgene derivatives, 74 was treated with DMDTC and CDI respectively. Free amine, 74 was carbonylated using DMDTC, to
51 form 75 and 76 in yields of 6 % and 22% respectively (Eq 4-6). Compound 77 was not observed. The reaction also produced a number of side products of which were identified by TLC analysis. Cyclic carbamate 76 was the major component of the product mixture with urea, 75 being the minor product. When CDI is used as carbonylating agent, 75 and 76 is observed with 76 being the major component of the reaction mixture being formed in 47% yield with 75 being formed in only 4 % yield. Compound 77 was not observed (Equation 4-7). NHNH HO HO O ONH O +NH2 OH (4-7)74757622ONH OH NH2 O + 77THF, RT4.0M, 18 hCDInot observed4 %47% The amino alcohol 3-amino-2,2-dimethylpropanol was used as a representative substrate to investigate the formation of an acyclic urea vs. a six membered cyclic carbamate. This substrate allows explanation of the effect on reactivity when steric bulk is present at the position to the nucleophilic nitrogen. This particular substrate also ties in the Thorpe-Ingold effect where formation of the carbamate is expected to be favored by the presence of the gem-dimethyl subsituents. Because of the steric bulk of the substrate, the reaction does not go to completion and starting material is recovered as a major component of the product mixture with 79 and 80 being minor components.
52 OHHN NH O OH NH2OH HNO O +2W(CO)6, COI2, K2CO3798078(4-8)NH2O NH O OH 81+2 Table 4-2. Carbonylation of 78 under various reaction conditions. Entry 73 (M) Time (hr) Temperature ( C) % Yield a 79 % Yield a 80 % Yield 81 % SM recovered 1 3.0 3 40 10 5 0 82 2 4.5 3 40 10 7 0 80 3 4.5 24 82 24 3 0 65 4 4.5 b 24 82 24 2 0 60 5 4.5 24 125 16 2 0 56 6 4.5 48 82 35 6 0 4 aTypical reaction conditions: amine 73 (1eq), W(CO)6, (5 mol %), K2CO3 (2 eq) and I2 (0.5), solvent (CH2Cl2), 80 atm CO. a Compounds 79 and 80 were isolated as a mixture and yields were calculated based on NMR. bTypical reaction conditions except pyridine used as base. SMstarting material. Carbonylation of 78 results in formation of the urea as the major product. The selectivity for urea formation over cyclic carbamate formation increases with an increase in temperature and is unaffected by concentration Table 4-2, (Entries 1, 2 and 3). There was no evidence for the formation of the acyclic carbamate 81. Harsher reaction conditions such as a longer time or high temperature do not improve the yields of urea Table 4-2, (Entries 5 and 6). Overall, there is a selectivity favoring formation of urea over carbamate. In short, the W(CO)6/I2 methodology can be successfully applied to the carbonylation of amino alcohols. The oxidative carbonylation is selective, forming urea as the major component of the product mixture when six-membered cyclic carbamate is formed as well. The seven-membered cyclic carbamate is not formed under catalytic carbonylation reaction conditions and the corresponding urea is obtained in good yield.
53 However, the corresponding urea is not observed when -amino alcohols are carbonylated. W(CO)6/I2 methodology can used as an alternative to phosgene and its derivatives in the carbonylation of amino alcohols to form the hydroxyl ureas as it is selective for the formation of hydroxyl urea.
CHAPTER 5 EXPERIMENTAL PROTOCOLS General. All experimental procedures were carried out under nitrogen and in oven dried glassware unless otherwise indicated. Carbonylation reactions were carried out in a glass lined Parr reaction vessel. Solvents and reagents were obtained from commercial vendors in the appropriate grade and used without purification except for solvents used in carbonylation reactions that were dried, degassed and distilled. Syntheses of compounds 24,49 25,2,50 26,2 47,2 and 5155 were carried out according to literature procedures. W(CO)6 was purified by chromatography on alumina using hexanes as eluent. All solvents were removed by evaporation using a rotary evaporator unless indicated otherwise. 1H and 13C NMR spectra were recorded on a Varian Gemini 300 MHz spectrometer. IR spectra were recorded on a Perkin-Elmer 1600 FTIR. High-resolution mass spectrometry was performed by the University of Florida analytical services. 54
55 NHHN O OO 27 (4R,5S,6S,7R)-Hexahydro-5,6-O-isopropylidene-4,7-bis-(phenylmethyl)-2H-1,3-diazapin-2-one (27). To a glass-lined 300 mL Parr high pressure vessel containing 18 mL of CH2Cl2 and 3 mL of water, were added diamine 24 (126.7 mg, 0.37 mmol), W(CO)6 (8.8 mg, 0.025 mmol), K2CO3 (164.9 mg, 1.19 mmol) and I2 (100.5 mg, 0.39 mmol). The vessel was then charged with 85 atm CO and heated at 81C overnight. The pressure was released and 15 mL of water was added. The organics were then separated and washed with a saturated solution of Na2SO3 followed by brine. The resulting solution was dried over magnesium sulfate and filtered. The solvent was removed by evaporation and the resulting residue was purified via column chromatography on silica using ether as eluent. Removal of the solvent afforded 27 as a white solid in 38% yield. The product was identified by comparison with literature data.48 IR (neat): vCO 1671 cm-1; MS(LSIMS) [M+H]+ calcd for C22H26N2O3 367.2021; found 367.2012.
56 NHHN MEMO OMEM O 28 (4R,5S,6S,7R)-Hexahydro-5,6-bis[2-methoxyethoxymethoxy]-4,7-bis(phenylmethyl)-2H-1,3-diazepin-2-one (28). To a glass-lined 300 mL Parr high pressure vessel containing 32 mL of CH2Cl2 and 8 mL of water, were added diamine 25 (101.1 mg, 0.200 mmol), W(CO)6 (7.70 mg, 0.0219 mmol), K2CO3 (87.8 mg, 0.635 mmol) and I2 (54.2 mg, 0.213 mmol). The vessel was then charged with 80 atm CO and heated at 82 C overnight. The pressure was released and 15 mL of water was added. The organics were then separated and washed with a saturated solution of Na2SO3 followed by brine. The resulting solution was dried over magnesium sulfate and filtered. The solvent was removed by evaporation and the resulting residue was purified to obtain 28 as a white solid in 49% yield after purification using literature procedures.50 The product was identified by comparison with literature data.50 IR (neat): vCO 1674 cm-1; MS(LSIMS) [M+H]+ calcd for C27H38N2O7 503.2752, found 503.2797.
57 NHHN SEMO OSEM O 29 (4R,5S,6S,7R)-Hexahydro-5,6-bis[2-(trimethylsilyl)ethoxymethoxy]-4,7-bis(phenylmethyl)-2H-1,3-diazepin-2-one (29). To a glass-lined 300 mL Parr high pressure vessel containing 32 mL of CH2Cl2 and 8 mL of water, were added diamine 26 (92.9 mg, 0.164 mmol), W(CO)6 (2.4 mg, 0.0068 mmol), K2CO3 (68.8 mg, 0.497 mmol) and I2 (42.2 mg, 0.166 mmol). The vessel was then charged with 77 atm CO and heated at 82 C overnight. The pressure was released and 15 mL of water was added. The organics were then separated and washed with a saturated solution of Na2SO3 followed by brine. The resulting solution was dried over magnesium sulfate and filtered. The solvent was removed by evaporation and the resulting residue was purified via column chromatography on silica using 2:1 hexanes/ethyl acetate as eluent. Removal of the solvent afforded 29 as a white solid in 75% yield. The product was identified by comparison with literature data.2 IR (neat): vCO 1679 cm-1; (LSIMS) [M+H]+calcd for C31H50N2O5Si2 587.3336, found 587.3355.
58 OO O O 30 Meso-dimethyl tartrate 2,3-acetonide (30). Dimethyl maleate (22.08 g, 153.2 mmol) and citric acid (22.18 g, 114.8 mmol) were dissolved in 153 mL of a 1:1 mixture of tertbutanol and water. Potassium osmate dihydrate (0.05 g, 1.0 mmol) was then added, followed by 4-methyl-morpholine-N-oxide (40.7 mL of a 50% wt. solution in water, 168.5 mmol). The reaction turned bright green and was stirred at room temperature for 3 hours. The tert-butanol was removed by evaporation and the aqueous solution was then acidified with HCl (1.2 N, 184 mL) and extracted with ethyl acetate (3 x 153 mL). The combined organics were dried with sodium sulfate and concentrated to give the pure diol 29 as a white solid (7.97 g, 29 %). To diol 29 (7.83 g, 43.9 mmol) were added p-toluenesulfonic acid monohydrate (84.4 mg, 0.43 mmol), 2,2-dimethoxypropane (33.3 mL, 263.7 mmol) and 55 mL benzene. The reaction mixture was refluxed for 7 hours, then cooled to room temperature and the solvent was removed by evaporation. The resulting residue was triturated with benzene (3 x 50 mL). The resulting red-brown residue was re-dissolved in benzene and the solution washed with NaHCO3 and dried over sodium sulfate. After filtration, solvent was removed from the filtrate by evaporation. The residue was chromatographed on silica with 2:1 hexanes/ethyl acetate as eluent to yield a white solid (7.52g, 78 %). Compound 30 was characterized by comparison with literature data.61
59 NHHN O HOOH 36 (4R, 5S, 6S, 7R)-Hexahydro-5,6-dihydroxy-4,7-bis(phenylmethyl)-2H-1,3-diazapin-2-one (36). (4R,5S,6S,7R)-Hexahydro-5,6-O-isopropylidene-4,7-bis-(phenylmethyl)-2H-1,3-diazapin-2-one (27) (102.1g, 0.299 mmol) was refluxed in 13 mL of 6.7 % HCl/MeOH for 2.75 hours. The reaction mixture was cooled and water and brine were added. Ethyl acetate (15 mL x 4) was used for extraction of the aqueous layers. The combined organics were washed with brine and dried over magnesium sulfate. Solvent was removed via evaporation to afford 36 as a white solid in 88% yield. The product was identified by comparison with literature data.52 IR (CH2Cl2): vCO 1681 cm-1; MS(LSIMS) [M+H]+ calcd for C19H22N2O3 327.1708, found 327.1715.
60 HNO NHO O O HNO NH2OH O 3738 (4R, 5S, 4R, 5S)-Bis(4-benzyloxazolidin-2-one) (37) and 5-(2-Amino-1-hydroxy-3-phenylpropyl)-4-benzyloxazolidinone (38). To a glass-lined 300 mL Parr high pressure vessel containing 30 mL of CH2Cl2 and 7 mL H2O were added (2R, 3S, 4S, 5R)-2,5-diamino-1,6-diphenyl-3, 4-hexandiol, 35 (44.8 mg, 0.149 mmol), W(CO)6 (21.0 mg 0.0598 mmol), K2CO3 (61.6 mg, 0.445 mmol) and I2 (38.5 mg, 0.151 mmol). The vessel was then charged with 85 atm CO and heated at 64C for 1.75 hours. The pressure was released, water was added and the layers were subsequently separated. The organic layer was washed with 15 mL sodium sulfite, 15 mL water, (3 x 10 mL) 1.2 N hydrochloric acid, 15 mL water, and brine respectively. The solvent was dried over magnesium sulfate and removed by evaporation. The resulting residue was purified via column chromatography on silica using 5 % methanol/ CH2Cl2 to afford 37 as a white solid in 6 % yield. The product was identified by comparison with literature data.52 IR (CH2Cl2): vCO 1774 cm-1. The combined acid layers were treated with 3M sodium hydroxide until the pH was in the range of 79. The neutralized layers were washed with (4 x15 mL) CHCl3 and 15 mL water. The solvent was removed by evaporation. The resulting residue was purified by column chromatography on silica using 89:10:1 CHCl3: MeOH: NH4OH to afford 38
61 as a white solid in 19 % yield. 1H NMR (CDCl3) : 2.51-2.89 (m, 4H), 3.17-3.24 (m, 1H), 4.124.28 (m, 2H), 5.02 (m, 1H). 13C NMR (CDCl3): 41.4, 41.8, 53.9, 55.4, 71.6, 84.2, 126.9, 127.5, 128.9, 129.2, 129.3, 136.0, 138.0, 158.1; IR (CH2Cl2): vCO 1766 cm-1; MS(LSIMS) [M+H]+ calcd for C19H23N3O2 327.1708, found 327.1714.
62 O O OH O 4142 Benzil (41) and benzoin (42). To a glass-lined 300 mL Parr high pressure vessel containing 32 mL of CH2Cl2 were added (S, S)-(-)-1,2-diphenyl-1,2-ethanediol ( 95.7 mg, 0.455 mmol), W(CO)6 (15 mg, 0.0450 mmol), K2CO3 (189.0 mg, 1.36 mmol) and I2 (115.8 mg, 0.456 mmol). The vessel was then charged with 85 atm CO and heated at 84 C for 1.75 hours. The pressure was released and the CH2Cl2 layer was filtered to remove solid particles. The resulting residue was rinsed with hot CH2Cl2 which was added to the organics Sodium hydrosulfite was added to the filtrate and the suspension was stirred for 10-15 minutes then filtered. The solvent was removed by evaporation and the resulting residue was purified via column chromatography on silica using CH2Cl2, 2 % methanol: CH2Cl2, and 5 % methanol: CH2Cl2 respectively. Removal of the solvents afforded benzoin and benzil in yields of 20% and 13% respectively. Benzoin was obtained as a white solid and benzil as a yellow solid. The products were identified by comparison with commercially available authentic samples. IR (CH2Cl2): vCO 1681, vOH 3464 cm-1 (benzoin), vCO 1681 cm-1 (benzil).
63 ONH O 45 (4S)-4-Benzyl-1,3-oxazolan-2-one (45). To a glass-lined 300 mL Parr high pressure vessel containing 26 mL of CH2Cl2 and 6.4 mL H2O were added (R)-(+)-2-amino-3-phenyl-1-propanol, 44 (79.4 mg, 0.525 mmol), W(CO)6 (14.7 mg, 0.0420 mmol), K2CO3 (217.6 mg, 1.57 mmol) and I2 (133.2 mg, 0.524 mmol). The vessel was then charged with 85 atm CO and stirred at room temperature for 1.75 hours. The pressure was released and the CH2Cl2 layer was separated. Sodium hydrosulfite was used to wash the organic layer. The resulting solution was dried over magnesium sulfate, the solvent was removed by evaporation and the resulting residue was purified via column chromatography on silica using 5 % methanol: CH2Cl2. Removal of the solvents afforded 45 as a white solid in 25 % yield. The product was identified by comparison with the commercially available authentic sample. IR (CH2Cl2): vCO 1750 cm-1.
64 NH2 O O Si 48 1-Benzyl-2-(2-(trimethylsilyl)ethoxymethoxy)ethylamine (48). A solution of (R)-N-Cbz-phenylalaninol (999.5 mg, 3.54 mmol), [2-(trimethylsilyl)]ethoxymethoxy chloride (SEMCl) (2.40 mL, 13.5 mmol) and 2.5 mL of diisopropylethylamine in 10 mL of dry methylene chloride was stirred at room temperature overnight until TLC indicated that no starting material was present. The reaction mixture was washed once with water. The water layer in turn was washed with 4 x 10 mL of methylene chloride and the combined organics were washed with water (2 x 10 mL). The combined water layers were washed once with 10 mL methylene chloride. The combined organics were dried over magnesium sulfate and the solvent removed by evaporation to yield 1.456 g of the crude SEM ether as a brown oil. To a solution of 496.0 mg (1.19 mmol) of the crude SEM ether in 10 mL EtOH was added 44.2 mg of 5% Pd/C. The reaction mixture was degassed and subsequently filled with hydrogen at 1 atm. The reaction mixture was then stirred for 2.5 hr until TLC showed no starting material. The mixture was filtered and evaporated to dryness to yield 320.5 mg of crude 48. The crude amine was purified by column chromatography on silica eluting with 2/1 hexanes/ethyl acetate/1% triethylamine followed by 1:1 hexanes/ethyl acetate/1% triethylamine to afford 48 as an oil in 68% yield. 1H NMR (CDCl3) 0.05 (s, 9H), 0.97 (t, 2H, J = 8.1 Hz), 2.56-2.63 (m, 1H), 2.81-2.87 (m, 1H), 3.27-3.28 (m, 1H), 3.39-3.44 (m, 1H), 3.57-3.69 (m, 3H), 4.73 (s, 2H), 7.22
65 7.35 (m, 5H); 13C NMR (CDCl3) -1.2, 18.3, 41.0, 52.7, 65.4, 72.8, 95.4, 126.5, 128.7, 129.4, 139.0; MS(LSIMS) [M+H]+ calcd for C15H27NO2Si 282.1889, found 282.1886.
66 HNNH MEMO OMEM O 49 1,3-Bis-[1-(2-methoxyethoxymethoxymethyl)-2-phenylethyl]urea (49). To a glass-lined 300 mL Parr high pressure vessel containing 34 mL of CH2Cl2 and 9 mL of water, were added amine 47 (106.8 mg, 0.379 mmol), W(CO)6 (6.8 mg, 0.019 mmol), K2CO3 (78.8 mg, 0.570 mmol), and I2 (48.2 mg, 0.189 mmol). The vessel was then charged with 80 atm CO and heated at 82 C overnight. The pressure was released and then 15 mL water was added. The organics were separated and washed with a saturated solution of Na2SO3, water and brine. The resulting solution was dried over magnesium sulfate and filtered. The solvent was removed by evaporation and the resulting residue was purified by washing with 2:1 hexanes/ethyl acetate and hexanes to afford 49 as a white solid in 82% yield. IR (neat): vCO 1673 cm-1; 1H NMR (CDCl3) 2.75-2.94 (m, 4H), 3.36 (s, 6H), 3.45-3.54 (m, 8H), 3.67-3.68 (m, 4H), 4.09-4.11 (m, 2H), 4.63-4.71 (q, 4H, J = 6.3 Hz), 5.03-5.06 (d, 2H, J = 8.1 Hz), 7.21-7.29 (m, 10H); 13C NMR (CDCl3) 38.4, 51.2, 59.1, 67.1, 69.8, 72.0, 96.1, 126.4, 128.5, 129.6, 138.6, 157.5; MS(LSIMS) [M+H]+ calcd for C27H40N2O7 505.2914, found 505.2902.
67 HNNH SEMO OSEM O 50 1,3-Bis-[2-phenyl-1-[2-(trimethylsilyl)ethoxymethoxymethyl]-ethyl]urea (50). Following the same procedure used to prepare 49, urea 50 was obtained as an oil in 53% yield after chromatography on silica using 2/1 hexanes/ethyl acetate as eluent. Urea 50 was comprised of isomers in a 3:1 ratio. The spectral data of the major isomer follows. IR (neat): vCO 1673 cm-1; 1H NMR (CDCl3) 0.02 (s, 18H), 0.93 (t, 4H, J = 8.1 Hz), 2.84 (m, 4H), 3.46 (m, 4H), 3.62 (m, 4H), 4.08 (m, 2H), 4.65 (s, 4H), 7.22-7.29 (m, 10H); 13C NMR (CDCl3) -1.4, 18.0, 38.1, 51.3, 64.4, 68.8, 95.3, 126.3, 128.4, 129.3, 138.2, 157.2; MS(LSIMS) [M+H]+ calcd for C31H52N2O5Si2 589.3493, found 589.3475.
68 N N O 57 N,N'-dimethyl-N,N'-bis(1-phenylethyl)urea (57). To a glass-lined 300 mL Parr high pressure vessel containing 7.5 mL of CH2Cl2 were added R-(+)-N--dimethylbenzyl amine, 56 (364.5 mg, 2.69 mmol), W(CO)6 (18.9 mg, 0.053 mmol), K2CO3 (372.6 mg, 2.69 mmol) and I2 (342.4 mg, 1.34 mmol). The vessel was then charged with 75 atm CO and stirred at room temperature overnight. The pressure was released and 15 mL of water was added. The organics were then separated and washed with a saturated solution of Na2SO3, followed by brine. The resulting solution was dried over magnesium sulfate and filtered. The solvent was removed by evaporation and the resulting residue was purified by chromatography on silica using CH2Cl2 and 1:1 hexane: diethyl ether as eluents. Pure 57 was obtained as a clear oil in 13 % yield. 1H NMR (CDCl3) : 1.56 (d, 6H, J = 7.2 Hz), 2.55 (s, 6H), 5.17-5.24 (q, 2H, J = 7.2 Hz), 7.25-7.34 (m, 10H); 13C NMR (CDCl3) : 16.4, 31.4, 55.0, 127.1, 127.4, 128.5, 141.7, 165.6; MS(LSIMS) [M+H]+ calcd for 297.1966, found 297.1972.
69 N N O O O 59 1,3-Bis-(2-methoxyethyl) 1,3-dimethylurea (59). To a glass-lined 300 mL Parr high pressure vessel containing 7.7 mL of CH2Cl2 were added amine 58 (244.0 mg, 2.73 mmol), W(CO)6 (19.6 mg, 0.0558 mmol), K2CO3 (378.8 mg, 2.74 mmol) and I2 ( 347.6 mg, 1.36 mmol). The vessel was then charged with 85 atm CO and stirred at room temperature overnight. The pressure was released. The organics were then separated by hot gravity filtration. Sodium hydrosulfite was added to the filtrate and mixture was stirred until decolorization occurred. The decolorized mixture was filtered. The resulting filtrate was dried over magnesium sulfate and filtered. The solvent was removed by evaporation. The resulting residue was purified by chromatography using 5% methanol/methylene chloride to obtain 6 as a clear oil in 34% yield. IR (neat): vCO 1635cm-1; 1H NMR (CDCl3) 2.81 (s, 6H), 3.27 (m, 10H), 3.47 (m, 4H); 13C NMR (CDCl3) 37.7, 49.9, 58.8, 70.8, 165.2; MS(LSIMS) [M+H]+ calcd for C9H20N2O3 205.1552, found 205.1544.
70 HN O O O 62 [2-(2-Methoxyethoxymethoxy)-1-methylethyl]-methylamine (62). 2-(Ethoxy-carbonylmethylamino)-propionic acid, 60 (2.62 g, 0.01 mol) was dissolved in 4.4 mL of THF. The mixture was cooled to 0 C. To the mixture was slowly added 16.5 mL 1M BH3THF. The reaction was stirred for 2.5 hours at room temperature. Excess hydride was carefully destroyed with a 1:1 mixture of THF: water. The aqueous phase was saturated with potassium carbonate. The THF layer was separated and the aqueous layer was extracted with 4 x 20 mL ether. The combined organics layer were dried over magnesium sulfate. The solvents were removed by evaporation to afford the 2.55 g of crude alcohol, as a clear oil. To 2.55 g of crude alcohol in 31 mL CH2Cl2 were added 1.96 mL MEMCl and 2.99 mL diisopropylethylamine. The reaction was stirred overnight at room temperature. Water was added and the organic phase was washed several times with water. The solution was dried over magnesium sulfate and the solvent removed by evaporation to afford 3.58 g of crude MEM ether 61 as a clear oil. The MEM ether 61 (3.58 g, 0.01 mol) was dissolved in 32 mL absolute ethanol. To the solution was added 3.75 g 10 % Pd/C. The reaction flask was purged and filled with hydrogen (1 atm). The reaction mixture was stirred overnight at room temperature. TLC analysis revealed no starting material remained. The reaction mixture was filtered through Celite to remove the catalyst. The solvent was removed by evaporation and the resulting residue was chromatographed on silica using 5-20% ethanol/ethyl acetate. The desired amine 62 was obtained as a clear oil (1.15 g, 46 %). 1H NMR (CDCl3) 0.95 (d,
71 2H, J = 6.6. Hz), 2.05 (br s, 1H), 2.34 (s, 3H), 2.682.74 (m, 1H), 3.32-3.34 (m, 4H), 3.433.51 (m, 3H), 3.613.64 (m, 2H), 4.66 (s, 2H); 13C NMR (CDCl3) 16.5, 33.7, 54.3, 59.0, 66.9, 71.8, 71.9, 95.7; MS(LSIMS) [M+H]+ calcd for C8H19NO3 178.1443, found 178.1447.
72 NH O O O 65 [1-(2-Methoxyethoxymethoxy)-2-phenylethyl]-methylamine (65). N-methyl-L-phenylalanine, 63 (1.69 g, 5.41 mmol) was dissolved in 2.3 mL of THF. The mixture was cooled to 0 C. To the mixture was slowly added 8.19 mL of 1M BH3THF. The reaction was stirred for 2.5 hours at room temperature. Excess hydride was carefully destroyed with a 1:1 mixture of THF/water. The aqueous phase was saturated with potassium carbonate. The THF layer was separated and the aqueous layer was extracted with 3 x 20 mL ether. The combined organic layers were dried over magnesium sulfate. The solvents were removed by evaporation to reveal 1.63 g of crude alcohol as a clear oil. To 1.63 g of crude alcohol in 15 mL CH2Cl2 were added 0.93 mL MEMCl and 1.44 mL diisopropylethylamine. The reaction was stirred overnight at room temperature. Water was added and the organic phase was washed several times with water. The solution was dried over magnesium sulfate and the solvent removed by evaporation to reveal 2.11 g of MEM ether 64 as a clear oil. The MEM ether 64 (707.7 mg, 1.83 mmol) was dissolved in 5 mL absolute ethanol. To the solution was added 490.4 mg 10 % Pd/C. The reaction flask was purged and filled with hydrogen (1 atm). The reaction mixture was stirred overnight at room temperature. TLC analysis revealed that no starting material remained. The reaction mixture was filtered through Celite to remove the catalyst. The solvent was removed by evaporation to afford 462.4 mg of amine, 65 as a clear oil. 1H NMR (CDCl3) 1.96 (s, 1H), 2.31 (s, 3H), 2.57-2.75 (m, 2H), 3.21 (s, 3H), 3.283.41 (m, 5H), 3.53 (s, 2H), 4.57 (s, 2H), 7.057.13 (m, 5H); 13C NMR (CDCl3) 33.6, 37.3, 58.6, 60.3, 66.5, 68.2, 71.4, 95.4, 125.9,
73 128.1, 128.9, 138.5; MS(LSIMS) [M+H]+ calcd for C14H23NO3 254.1756, found 254.1752.
74 N N O 67 N,N'-dibenzyl-N,N'-diethylurea (67). To a glass-lined 300 mL Parr high pressure vessel containing 9.2 mL of CH2Cl2 were added N-ethylbenzylamine, 66 (440.8 mg, 3.26 mmol), W(CO)6 (22.9 mg, 0.0652 mmol), K2CO3 (450.6 mg, 3.26 mmol) and I2 (413.7 mg, 1.63 mmol). The vessel was then charged with 75 atm CO and stirred at room temperature overnight. The pressure was released and 15 mL of water were added. The organics were then separated and washed with a saturated solution of Na2SO3 followed by brine. The resulting solution was dried over magnesium sulfate and filtered. The solvent was removed by evaporation and the resulting residue was purified by chromatography on silica using CH2Cl2 as eluent. Pure 67 was obtained as a clear oil in 29 % yield. 1H NMR (CDCl3) : 1.11 (t, 6H, J = 7.2 Hz), 3.18 (q, 4H, J = 7.2 Hz), 4.40 (s, 4H), 7.21-7.32(m, 10H); 13C NMR (CDCl3) : 12.8, 42.8, 51.3, 127.1, 127.5, 128.5, 138.4, 165.2; IR (CH2Cl2): vCO 1636 cm-1; MS(LSIMS) [M+H]+ calcd for C19H24N2O 297.1966, found 297.1969.
75 HO HN HN OH O 70 1, 3-Bis-(4-hydroxy-3-methyl-butyl)-urea (70). To a 15 mL glass vial in a multi-compartment Parr high pressure vessel containing 0.45 mL of CH2Cl2, were added 76 (0.20 ml, 1.78 mmol), W(CO)6 (31.3 mg, 0.089 mmol), pyridine (0.22 ml, 2.72 mmol) and I2 (229.0 mg, 0.90 mmol). The vessel was then charged with 80 atm CO and heated at 40 C for 18 hours. The pressure was released and solvent was removed by evaporation. The residue was dissolved in 10 mL saturated sodium bicarbonate, 10 mL water and 15 mL saturated sodium sulfite. The solution was saturated with sodium chloride. The aqueous layers were washed with methylene chloride (2 x 20 mL) then 2:1 chloroform/ ethanol (4 x 30 mL). The chloroform/ ethanol layers were dried with MgSO4 and the solvents removed by evaporation to reveal 77 as a white solid in 41% yield. 1H NMR (CDCl3) : 0.85-0.87 (d, 6H, J = 6.6 Hz), 1.191.25 (m, 2H), 1.48-1.70 (m, 4H), 3.013.13 (m, 4H), 3.323.45 (m, 4H); 13C NMR (CDCl3) : 16.4, 33.0, 33.4, 38.4, 67.4, 161.0; MS (LSIMS) [M+H]+ C11H24N2O3, calcd 233.1865, found 233.1913
76 NH3OH Cl-74 3-Amino-4-phenyl-1-butanol hydrochloride (74). DL--homophenylalanine (5145.3 mg, 28.7 mmol) was added to 11.5 mL THF and the mixture was cooled to 0 C. BH3THF (1M, 43.1 mL, 43.0 mmol) was added dropwise to the suspension. The resulting mixture was stirred at room temperature for 4.5 hours. The mixture was then cooled to 0 C, 30 mL of 3N sodium hydroxide was slowly added and the mixture was stirred at room temperature overnight. The pH of the solution was adjusted to 11 by adding a few pellets of sodium hydroxide. The aqueous phase was saturated with potassium carbonate, the THF phase was separated and the aqueous phase was extracted with (50 mL x 6) diethyl ether. The combined organic layers were dried over magnesium sulfate. The solvents were evaporated and the residue was dissolved in ether. To the ethereal solution was added 16 mL of 2M HCl in ether which resulted in the precipitation of a white solid which was washed with ether and dried to afford 74 in 58% yield. The product was identified by comparison with literature data.62 MS(LSIMS) [M+Cl]+ calcd for C10H16 ClNO calcd 166.1231 found 166.1228.
77 NH O NH OH HO 75 HNO O 76 N,N'-bis(1-benzyl-3-hydroxypropyl)urea (75) and 4-benzyl-1,3-oxazinan-2-one (76). Authentic samples of compounds 75 and 76 were obtained the following way: 3-Amino-4-phenyl-1-butanol hydrochloride (991.4 mg, 6.00 mmol) was dissolved in 1.3 mL methanol. DMDTC (0.38 mL, 3.60 mmol) was added to the solution and the mixture was heated at 60 C for 18 hours. The solvents were evaporated and the residue was chromatographed on silica with ethyl acetate and 5% ethanol/ethyl acetate. Compounds 75 and 76 were both obtained as a white solids in 6 and 22% yield, respectively. For urea 75, 1H NMR (CDCl3) : 1.18-1.26 (m, 2H), 1.74-1.81 (m, 2H), 2.622.79 (m, 4H), 3.373.47 (m, 4H), 4.10 (s, 2H), 4.82 (s, 2H); 13C NMR (CDCl3) : 38.4, 42.0, 48.0, 58.6, 126.6, 128.6, 129.2, 138.2, 159.9; IR (CH2Cl2): vCO cm-1; MS (LSIMS) [M+H]+ calcd for C21H28N2O3 257.2178, found 257.2161. For carbamate 76, 1H NMR (CDCl3) : 1.621.74 (m, 1H), 1.831.91 (m, 1H), 2.692.88 (m, 1H), 2.862.93 (m, 1H), 3.623.72 (m, 1H), 4.104.18 (m, 1H), 4.244.31 (m, 1H), 6.81 (s, 1H), 7.157.33 (m, 5H); 13CNMR (CDCl3) : 26.5, 42.3, 51.8, 65.4, 126.8, 128.6, 129.1, 136.2, 154.5; IR (CH2Cl2): vCO 1710 cm-1; MS (LSIMS) [M+H]+ calcd for C11H13NO2 192.1024, found 192.1020.
78 NH O NH OH HO 75 HNO O 76 N,N'-bis(1-benzyl-3-hydroxypropyl)urea (75) and 4-benzyl-1,3-oxazinan-2-one (76). To a 15 mL glass vial in a multi-compartment Parr high pressure vessel containing 0.16 mL of CH2Cl2 and 0.12 mL H2O were added 74 (124.4 mg, 0.61 mmol), W(CO)6 (11.20 mg, 0.032 mmol), K2CO3 (170.1 mg, 1.23 mmol) and I2 (79.30 mg, 0.31 mmol). The vessel was then charged with 80 atm CO and heated at 40 C for three hours. The pressure was released. Methylene chloride (10 ml) was added and the organic layer was washed with 15 mL saturated sodium sulfite, and 0.12N hydrochloric acid (2 x 10 mL). The solvent was removed by evaporation and the resulting residue was purified via column chromatography on silica using ethyl acetate and 5% ethanol/ ethyl acetate as eluents. Removal of the solvents afforded a mixture of 75 and 76 as a white solid in 6% and 3% yield respectively. The product was identified by comparison with an authentic sample.
79 NH O NH HO OH 79HNO O 80 N,N'-bis(3-hydroxy-2,2-dimethylpropyl)urea (79) and 5,5-dimethyl-1,3-oxazinan-2-one (80). To a 15 mL glass vial in a multi-compartment Parr high pressure vessel containing 0.6 mL of CH2Cl2, were added 3-amino-2,2-dimethylpropanol (136.3 mg, 1.32 mmol), W(CO)6 (23.2 mg, 0.066 mmol), K2CO3 (274.0 mg, 1.98 mmol) and I2 (336.3 mg, 1.32 mmol). The vessel was then charged with 80 atm CO and heated at 40 C for three hours. The pressure was released and 5 mL of hot ethyl acetate and sodium hydrosulfite were added. The mixture was allowed to stir until decolorized. The decolorized solution was filtered and the residue was washed with 10 mL of hot ethyl acetate. The solvent was removed by evaporation and the resulting residue was purified via column chromatography on silica using 10:90 methanol/ methylene chloride as eluent. Removal of the solvent afforded a mixture of 79 and 80 as a white solid in 10% and 5% yield respectively. For urea 79, 1H NMR (CDCl3) : 0.73 (s, 12H), 2.842.86 (d, 4H, 6.3 Hz), 3.023.04 (d, 4H, 6 Hz), 4.594.63 (t, 2H, 6, 6 Hz), 6.006.05 (t, 2H, 6.3, 6.3 Hz), 13C NMR (CDCl3) : 22.3, 36.6, 46.2, 67.6, 159.7; MS (LSIMS) [M+H]+calcd for C11H24N2O3 233.1751, found 233.1750. For carbamate 80, 1H NMR (CDCl3) : 0.96 (s, 6H), 2.88 (s, 2H), 3.80 (s, 2H), 7.12 (br s, 1H); 13C NMR (CDCl3) : 22.1, 27.3, 50.6, 75.1, 152.4; MS (LSIMS) [M+H]+calcd for 130.0868, C6H11NO2 found 130.0867.
LIST OF REFERENCES 1. Lam, P. Y. S.; Jadhav, P. K.; Eyermann, C. J.; Hodge, C. N.; Ru, Y.; Bacheler, L. T.; Meek, J. L.; Otto, M. J.; Rayner, M. M.; Wong, Y. N.; Chang, C. H.; Weber, P. C.; Jackson, D. A.; Sharpe, T. R.; Erickson-Viitanen, S. Science 1994, 263, 380-384. 2. Lam, P. Y. S.; Ru, Y.; Jadhav, P. K.; Aldrich, P. E.; DeLucca, G. V.; Eyermann, C. J.; Chang, C. H.; Emmett, G.; Holler, E. R.; Daneker, W. F.; Li, L. Z.; Confalone, P. N.; McHugh, R. J.; Han, Q.; Li, R. H.; Markwalder, J. A.; Seitz, S. P.; Sharpe, T. R.; Bacheler, L. T.; Rayner, M. M.; Klabe, R. M.; Shum, L. Y.; Winslow, D. L.; Kornhauser, D. M.; Jackson, D. A.; EricksonViitanen, S.; Hodge, C. N. J. Med. Chem. 1996, 39, 3514-3525. 3. Semple, G.; Ryder, H.; Rooker, D. P.; Batt, A. R.; Kendrick, D. A.; Szelke, M.; Ohta, M.; Satoh, M.; Nishida, A.; Akuzawa, S.; Miyata, K. J. Med. Chem. 1997, 40, 331-341. 4. Castro, J. L.; Ball, R. G.; Broughton, H. B.; Russell, M. G. N.; Rathbone, D.; Watt, A. P.; Baker, R.; Chapman, K. L.; Fletcher, A. E.; Patel, S.; Smith, A. J.; Marshall, G. R.; Ryecroft, W.; Matassa, V. G. J. Med. Chem. 1996, 39, 842-849. 5. vonGeldern, T. W.; Kester, J. A.; Bal, R.; WuWong, J. R.; Chiou, W.; Dixon, D. B.; Opgenorth, T. J. J. Med. Chem. 1996, 39, 968-981. 6. Vishnyakova, T. P.; Golubeva, I. A.; Glebova, E. V. Russ. Chem. Rev. (Engl. Transl.) 1985, 54, 249-261. 7. Bigi, F.; Maggi, R.; Sartori, G. Green Chem. 2000, 2, 140-148. 8. Feng, J. H.; Garland, M. Organometallics 1999, 18, 1542-1546. 9. Katritzky, A. R.; Pleynet, D. P. M.; Yang, B. Z. J. Org. Chem. 1997, 62, 4155-4158. 10. Degani, I., Fochi, R., Regondi, V. Synthesis 1980, 375. 11. Degani, I., Fochi, R., Regondi, V. Synthesis 1981, 149. 12. Leung, M. K.; Lai, J. L.; Lau, K. H.; Yu, H. H.; Hsiao, H. J. J. Org. Chem. 1996, 61, 4175-4179. 13. Thavonekham, B. Synthesis 1997, 1189. 80
81 14. Liu, Q.; Luedtke, N. W.; Tor, Y. Tetrahedron Lett. 2001, 42, 1445-1447. 15. Liu, G. W.; Hakimifard, M.; Garland, M. J. Mol. Catal. A-Chem. 2001, 168, 33-37. 16. Franz, R. A.; Applegath, F.; Morriss, F. V.; Baiocchi, F. J. Org. Chem. 1961, 26, 3306-3308. 17. Franz, R. A.; Applegath, F.; Morriss, F. V.; Baiocchi, F.; Bolze, C. J. Org. Chem. 1961, 26, 3309-3312. 18. Sonoda, N.; Yasuhara, T.; Kondo, K.; Ikeda, T.; Tsutsumi, S. J. Am. Chem. Soc. 1971, 93, 6344. 19. Yoshida, T.; Kambe, N.; Murai, S.; Sonoda, N. Tetrahedron Lett. 1986, 27, 3037-3040. 20. Mei, J., Yang, Y., Xue, Y., Lu, S. J. Mol. Catal. A: Chem. 2003, 191, 135-139. 21. Giannoccaro, P.; Nobile, C. F.; Mastrorilli, P.; Ravasio, N. J. Organomet. Chem. 1991, 419, 251-258. 22. Bassoli, A.; Rindone, B.; Tollari, S.; Chioccara, F. J. Mol. Catal. 1990, 60, 41-48. 23. Dombek, B. D.; Angelici, R. J. J. Organomet. Chem. 1977, 134, 203-217. 24. Calderazzo, F. Inorg. Chem. 1965, 4, 293-296. 25. Mulla, S. A. R.; Rode, C. V.; Kelkar, A. A.; Gupte, S. P. J. Mol. Catal. A-Chem. 1997, 122, 103-109. 26. Kanagasabapathy, S.; Gupte, S. P.; Chaudhari, R. V. Ind. Eng. Chem. Res. 1994, 33, 1-6. 27. Cornils, B., Herrmann, W. Applied Homogeneous Catalysis with Organometallic Compounds; 2nd ed.; Wiley-VCH Verlag GmbH: Weinheim, Germany, 2002; Vol. 1. 28. Giannoccaro, P. J. Organomet. Chem. 1987, 336, 271-278. 29. Pri-Bar, I.; Alper, H. Can. J. Chem.-Rev. Can. Chim. 1990, 68, 1544-1547. 30. Pearson, R. G., Vogelsong, D.C. J.Am. Chem. Soc 1958, 80, 1038. 31. Yang, H.-z.; Deng, Y.-q.; Shi, F.; Sima, T.-l. Gaodeng Xuexiao Huaxue Xuebao 2001, 22, 1913-1915. 32. Shi, F.; Deng, Y. Q. Chem. Commun. 2001, 443-444. 33. Chiarotto, I., Feroci, M. J. Org. Chem. 2003, 68, 7137-7139.
82 34. Calderazzo, F. US Patent 3316297, 1967 35. Srivastava, S. C.; Shrimal, A. K.; Srivastava, A. J. Organomet. Chem. 1991, 414, 65-69. 36. Bolzacchini, E.; Meinardi, S.; Orlandi, M.; Rindone, B. J. Mol. Catal. A-Chem. 1996, 111, 281-287. 37. Shi, F.; Deng, Y. Q.; Yang, H. Z.; SiMa, T. L. Chem. Commun. 2001, 345-346. 38. McCusker, J. E.; Abboud, K. A.; McElwee-White, L. Organometallics 1997, 16, 3863-3866. 39. McCusker, J. E.; Qian, F.; McElwee-White, L. J. Mol. Catal. A-Chem. 2000, 159, 11-17. 40. McCusker, J. E.; Grasso, C. A.; Main, A. D.; McElwee-White, L. Org. Lett. 1999, 1, 961-964. 41. Qian, F.; McCusker, J. E.; Zhang, Y.; Main, A. D.; Chlebowski, M.; Kokka, M.; McElwee-White, L. J. Org. Chem. 2002, 67, 4086-4092. 42. McCusker, J. E.; Main, A. D.; Johnson, K. S.; Grasso, C. A.; McElwee-White, L. J. Org. Chem. 2000, 65, 5216-5222. 43. McCusker, J. E.; Logan, J.; McElwee-White, L. Organometallics 1998, 17, 4037-4041. 44. De Lucca, G. V. J. Org. Chem. 1998, 63, 4755-4766. 45. Confalone, P. N.; Waltermire, R. E. In Process Chemistry in the Pharmaceutical Industry; Gadamasetti, K. G., Ed.; Marcel Dekker: New York, 1999, pp 201-219. 46. Nugiel, D. A.; Jacobs, K.; Worley, T.; Patel, M.; Kaltenbach, R. F.; Meyer, D. T.; Jadhav, P. K.; DeLucca, G. V.; Smyser, T. E.; Klabe, R. M.; Bacheler, L. T.; Rayner, M. M.; Seitz, S. P. J. Med. Chem. 1996, 39, 2156-2169. 47. Nugiel, D. A.; Jacobs, K.; Cornelius, L.; Chang, C. H.; Jadhav, P. K.; Holler, E. R.; Klabe, R. M.; Bacheler, L. T.; Cordova, B.; Garber, S.; Reid, C.; Logue, K. A.; GoreyFeret, L. J.; Lam, G. N.; EricksonViitanen, S.; Seitz, S. P. J. Med. Chem. 1997, 40, 1465-1474. 48. Pierce, M. E.; Harris, G. D.; Islam, Q.; Radesca, L. A.; Storace, L.; Waltermire, R. E.; Wat, E.; Jadhav, P. K.; Emmett, G. C. J. Org. Chem. 1996, 61, 444-450. 49. Rossano, L. T.; Lo, Y. S.; Anzalone, L.; Lee, Y. C.; Meloni, D. J.; Moore, J. R.; Gale, T. M.; Arnett, J. F. Tetrahedron Lett. 1995, 36, 4967-4970.
83 50. Hodge, C. N.; Aldrich, P. E.; Bacheler, L. T.; Chang, C. H.; Eyermann, C. J.; Garber, S.; Grubb, M.; Jackson, D. A.; Jadhav, P. K.; Korant, B.; Lam, P. Y. S.; Maurin, M. B.; Meek, J. L.; Otto, M. J.; Rayner, M. M.; Reid, C.; Sharpe, T. R.; Shum, L.; Winslow, D. L.; EricksonViitanen, S. Chem. Biol. 1996, 3, 301-314. 51. Hylton, K.-G.; Main, A. D.; McElwee-White, L. J. Org. Chem. 2003, 68, 1615-1617. 52. Dondoni, A., Perrone, D., Rinaldi,M. J. Org. Chem. 1998, 63, 9252-9264. 53. Herbert, J. M.; Knight, J. G.; Sexton, B. Tetrahedron 1996, 52, 15257-15266. 54. Jadhav, P. K.; Woerner, F. J.; Lam, P. Y. S.; Hodge, C. N.; Eyermann, C. J.; Man, H. W.; Daneker, W. F.; Bacheler, L. T.; Rayner, M. M.; Meek, J. L.; Erickson-Viitanen, S.; Jackson, D. A.; Calabrese, J. C.; Schadt, M.; Chang, C. H. J. Med. Chem. 1998, 41, 1446-1455. 55. Martin, B., Posseme, F., Le Barbier, C., Carreaux, F, Carboni, B., Seiler, N., Moulinoux, J-P., Delcros, J Bioorg. Med. Chem 2002, 10, 2863-2871. 56. Ager, D. J., Prakash, I., Schaad, D.R., Chem. Rev. 1996, 96, 835-875. 57. Samejima, K., Takeda, Y., Kawase, M., Okada, M., Kyogoku, Y. Chem. Pharm. Bull 1984, 32, 34283435. 58. Hayashi, K., Iwakura, Y Makromol. Chem. 1966, 94, 132-142. 59. Kurita, K., Matsumura, T., Iwakura, Y J. Org. Chem. 1976, 41, 2070-2071. 60. Staab, H. A. Angew. Chem. Intern, Ed. Engl. 1962, 1, 351-367. 61. Helgeson, R. C.; Weisman, G. R.; Toner, J. L.; Tarnowski, T. L.; Chao, Y.; Mayer, J. M.; Cram, D. J. J Am Chem Soc 1979, 101, 4928-4941. 62. Ueno, K.; Ogawa, A.; Ohta, Y.; Nomoto, Y.; Takasaki, K.; Kusaka, H.; Yano, H.; Suzuki, C.; Nakanishi, S. WO Patent 2001047931, 2001
BIOGRAPHICAL SKETCH Keisha-Gay Nicole Hylton was born on August 8, 1977 in St. Andrew, Jamaica to Olive and Alvest Hylton. She was curious even at an early age; and at the age of 4 years, she almost burned down the family home while experimenting to see what would happen when a lit light bulb is placed between two pillows. She spent her formative years at George Headley Primary school, and later went on to attend Westwood High School. At Westwood High School, her love of chemistry first took root, at 13 years old. After excelling in chemistry at the Caribbean Examination Council (CXC) exams and A Level examinations, she went on to attend the University of the West Indies, Mona pursuing a major in pure and applied chemistry. In her second year, after working as an intern at Shell Oil Company, West Indies and discovering the lackluster routine of a laboratory technician, she decided to pursue her Ph.D. She entered the University of Florida and pursued her Ph.D. studies, specializing in the areas of organic and organometallic chemistry. Early in her graduate career, she decided to pursue a nontraditional area of chemistry. After graduating, she will attend Levin College of Law at the University of Florida, and will specialize in the area of Intellectual Property Law, with the goal of becoming a Patent Attorney. 84