Scalable (Enantioselective) Syntheses of Novel 3-Methylated Analogs of Pazinaclone, (S)-PD172938 and Related Biologically Relevant Isoindolinones

Herein, we report the application of an efficient and practical K2CO3 promoted cascade reaction of 2-acetylbenzonitrile in the synthesis of novel 3-methylated analogs of Pazinaclone and PD172938, belonging to isoindolinones heterocyclic class bearing a tetrasubstituted stereocenter. Organocatalytic asymmetric synthesis of the key intermediate and its transformation into highly enantioenriched 3-methylated analog of (S)-PD172938 was also developed. These achievements can be of particular interest also for medicinal chemistry, since the methyl group is a very useful structural modification in the rational design of new and more effective bioactive compounds.


Introduction
The importance of methyl groups in biologically active molecules has been recently highlighted, showing that more than 80% of small-molecule drugs contain at least one methyl group bound to a carbon atom [1][2][3][4][5][6][7][8][9][10][11][12][13]. The introduction of a methyl group is used to optimize many properties of a drug candidate, and it is often associate in drug discovery to the so-called magic methyl effect. This strategy can have a favorable effect on solubility or selectivity against off-targets or can allow one to convert an agonist into an antagonist or a partial antagonist into a negative allosteric modulator [5]. It can induce a conformational change up to 590-fold boosts in potency [1]. For example, in the PLD2 inhibitor 2, the new methyl group improves potency from 11,800 to 20 nM (Figure 1) but adds very little lipophilicity (∆clogP = 0.36) [1,12]. The introduction of the methyl group has the advantage that only a small change of the physical properties of a molecule (∆MW = 14 gmol −1 , ∆clogP = 0.5) is obtained when compared to the trifluoromethyl group (∆MW = 68 g mol −1 and ∆clogP = 0.9), avoiding the formation of too lipophilic molecules and consequently the violation of Lipinsky's rules [13]. In contrast to the CF3 (trifluoromethyl group), which has received much attention from the synthetic community in the past decade, the incorporation of a methyl group represents a significant challenge in many contexts, especially when it must be installed on an existing tertiary stereocenter. Therefore, synthetic protocols that can incorporate the methyl group at specific points in a structure-activity relationship (SAR) program are of In contrast to the CF 3 (trifluoromethyl group), which has received much attention from the synthetic community in the past decade, the incorporation of a methyl group represents a significant challenge in many contexts, especially when it must be installed on an existing tertiary stereocenter. Therefore, synthetic protocols that can incorporate the methyl group at specific points in a structure-activity relationship (SAR) program are of high value to medicinal chemistry and to the pharmaceutical industry [2].
Over the past few years, the isoindolinone ring system has emerged as a valuable pharmacophore, exhibiting a wide range of therapeutic activities including antimicrobial, antioxidant, antifungal, anti-Parkinson, anti-inflammatory, antipsychotic, antihypertensive, anesthetic, vasodilatory, anxiolytic and antiviral activities [14,15]. In particular, isoindolinones 3 [16][17][18] and 4 (Pazinaclone) [19,20]   In contrast to the CF3 (trifluoromethyl group), which has received much attention from the synthetic community in the past decade, the incorporation of a methyl group represents a significant challenge in many contexts, especially when it must be installed on an existing tertiary stereocenter. Therefore, synthetic protocols that can incorporate the methyl group at specific points in a structure-activity relationship (SAR) program are of high value to medicinal chemistry and to the pharmaceutical industry [2].
Over the past few years, the isoindolinone ring system has emerged as a valuable pharmacophore, exhibiting a wide range of therapeutic activities including antimicrobial, antioxidant, antifungal, anti-Parkinson, anti-inflammatory, antipsychotic, antihypertensive, anesthetic, vasodilatory, anxiolytic and antiviral activities [14,15]. In particular, isoindolinones 3 [16][17][18] and 4 (Pazinaclone) [19,20]   The incorporation of a methyl group installed directly on the stereocenter of the isoindolinone ring can be of particularly importance for the effect on the binding activity and affinity to the receptors. However, the construction of the isoindolinone ring with a tetrasubstituted carbon stereocenter is a major challenge, especially when this has to be asymmetrically performed [14,23,24]. To our knowledge there are no examples in literature of 3-methylated analogs of the compounds described in Figure 2 and very few protocols on the synthesis of 3,3-disubstituted isoindolinones introducing a methyl group on the C-3 position have been reported [14,[25][26][27][28][29][30][31][32][33][34]. For instance, as described by Heibel in 2005, an asymmetric synthesis of 3,3-disubstituted isoindolinones involved the use of readily available (+) or (−)-trans-2-(α-cumyl) cyclohexanol (TCC) as the chiral auxiliary after the treatment with NaHMDS at temperatures ranging between −100 °C and −115 °C leading to good diastereoselectivity [25]. Later Cramer disclosed a chiral CpRh(III)-catalyzed C-H functionalization to directly provide isoindolinones bearing a tetrasubstituted stereocenters with high enantioselectivity [26]. In 2017, Liu utilized a Rh (III)-catalyzed [4 + 1] cyclization reaction of benzamides with propargyl alcohols to prepare isoindolinones bearing a quaternary carbon (with ketones in lateral chains) under external oxidant-free The incorporation of a methyl group installed directly on the stereocenter of the isoindolinone ring can be of particularly importance for the effect on the binding activity and affinity to the receptors. However, the construction of the isoindolinone ring with a tetrasubstituted carbon stereocenter is a major challenge, especially when this has to be asymmetrically performed [14,23,24]. To our knowledge there are no examples in literature of 3-methylated analogs of the compounds described in Figure 2 and very few protocols on the synthesis of 3,3-disubstituted isoindolinones introducing a methyl group on the C-3 position have been reported [14,[25][26][27][28][29][30][31][32][33][34]. For instance, as described by Heibel in 2005, an asymmetric synthesis of 3,3-disubstituted isoindolinones involved the use of readily available (+) or (−)-trans-2-(α-cumyl) cyclohexanol (TCC) as the chiral auxiliary after the treatment with NaHMDS at temperatures ranging between −100 • C and −115 • C leading to good diastereoselectivity [25]. Later Cramer disclosed a chiral CpRh(III)-catalyzed C-H functionalization to directly provide isoindolinones bearing a tetrasubstituted stereocenters with high enantioselectivity [26]. In 2017, Liu utilized a Rh (III)-catalyzed [4 + 1] cyclization reaction of benzamides with propargyl alcohols to prepare isoindolinones bearing a quaternary carbon (with ketones in lateral chains) under external oxidant-free conditions [27]. Recently, a nickel-catalyzed reductive dicarbofunctionalization of 1,1-disubstituted enamides with inactivated alkyl iodides was found effective to access N-benzyl-3,3-dialkylsubstituted isoindolinones frameworks in the racemic and asymmetric version [28]. In addition, Watson developed tandem aza-Heck-Suzuki and aza-Heck-carbonylation reactions of O-phenyl hydroxamic ethers to a variety of complex chiral γ-lactams, using highly toxic carbon monoxide gas [29]. However, all these methodologies have several drawbacks, such as harsh reaction conditions, low scale set up and, most importantly, they are not suitable for the synthesis of 3-methylated analogs of the bioactive compounds of Figure 2. In this context, we report a convenient approach to 3,3-disubstituted isoindolinones via cascade reactions of readily available 2-acylbenzonitriles with several nucleophiles promoted by cheap and environmentally friendly K 2 CO 3 [30][31][32]. Asymmetric versions of these reactions have also been developed [33].
Based on our continuing research interest in the synthesis of isoindolinones [31][32][33][34][35], we report the application of our efficient and practical K 2 CO 3 promoted a cascade reaction to the first synthesis of 3-methylated analogs of Pazinaclone, PD172938 and related structures (Scheme 1). An efficient asymmetric synthesis of 3-methylated analog of (S)-PD172938 was also developed based on the combination of the asymmetric-phase transfer catalysis process and a recrystallization step.
have several drawbacks, such as harsh reaction conditions, low scale set up and, mo importantly, they are not suitable for the synthesis of 3-methylated analogs of the bioa tive compounds of Figure 2. In this context, we report a convenient approach to 3,3-disu stituted isoindolinones via cascade reactions of readily available 2-acylbenzonitriles wi several nucleophiles promoted by cheap and environmentally friendly K2CO3 [30-3 Asymmetric versions of these reactions have also been developed [33].
Based on our continuing research interest in the synthesis of isoindolinones [31-3 we report the application of our efficient and practical K2CO3 promoted a cascade reactio to the first synthesis of 3-methylated analogs of Pazinaclone, PD172938 and related stru tures (Scheme 1). An efficient asymmetric synthesis of 3-methylated analog of (S PD172938 was also developed based on the combination of the asymmetric-phase transf catalysis process and a recrystallization step. Scheme 1. This work: new isoindolinones bearing a methyl group installed in 3-position.

Synthesis of Novel Racemic 3-Methylated Analogs of Bioactive Isoindolinones
The first part of the investigation focused on the large-scale synthesis of 3-substitut isoindolinone 7, which was conveniently obtained in the presence of the K2CO3-promot cascade reaction of readily available 2-acetylbenzonitrile 6 and dimethylmalonate in ac tonitrile (Scheme 2). This compound proved to be particularly convenient in the synthes of a series of novel analogs of bioactive 3-substituted isoindolinones, based on the tran formation of 7 into the monocarboxylic derivative 10. For this purpose, several metho were investigated. Decarboxylation of 7 under reflux in 6M HCl solution led to compl mixtures of products, while the treatment of 7 with a LiCl/H2O/DMSO mixture at 130 ° Scheme 1. This work: new isoindolinones bearing a methyl group installed in 3-position.

Synthesis of Novel Racemic 3-Methylated Analogs of Bioactive Isoindolinones
The first part of the investigation focused on the large-scale synthesis of 3-substituted isoindolinone 7, which was conveniently obtained in the presence of the K 2 CO 3 -promoted cascade reaction of readily available 2-acetylbenzonitrile 6 and dimethylmalonate in acetonitrile (Scheme 2). This compound proved to be particularly convenient in the synthesis of a series of novel analogs of bioactive 3-substituted isoindolinones, based on the transformation of 7 into the monocarboxylic derivative 10. For this purpose, several methods were investigated. Decarboxylation of 7 under reflux in 6M HCl solution led to complex mixtures of products, while the treatment of 7 with a LiCl/H 2 O/DMSO mixture at 130 • C, a method known as Krapcho decarboxylation [36,37], allowed for obtaining the ester 8 in a 70% yield. However, a higher yield of 10 was achieved in a two-step process comprising a saponification followed by simple heating of the dicarboxylic acid 9 in CH 3 CN (Scheme 2).
The key intermediate 10 was, then, easily converted in the target isoindolinones according to the strategies described in Scheme 3. Amide 12 derived from N-methylpiperazine 11 was easily obtained by EDC/HOBt activation of the carboxylic group. In the case of 1-piperidone ketal 14, Yamada coupling using DEPC (diethylphosphoryl cyanide) was more efficient [38]. a method known as Krapcho decarboxylation [36,37], allowed for obtaining the ester 8 in a 70% yield. However, a higher yield of 10 was achieved in a two-step process comprising a saponification followed by simple heating of the dicarboxylic acid 9 in CH3CN (Scheme 2).

Scheme 2. Synthesis of key intermediate 10.
The key intermediate 10 was, then, easily converted in the target isoindolinones according to the strategies described in Scheme 3. Amide 12 derived from N-methylpiperazine 11 was easily obtained by EDC/HOBt activation of the carboxylic group. In the case of 1-piperidone ketal 14, Yamada coupling using DEPC (diethylphosphoryl cyanide) was more efficient [38].
NH arylation of the lactams 12 and 15 required several efforts too (Scheme 3). In contrast with arylation of 3-mono-substituted isoindolinones [22,30], phenylation of the lactam 12 was performed using a ligand-less cross coupling reaction based on the Cu(I)-catalyzed arylation of amides [39] in a very good 81% yield, allowing for access to the 3methylated analog 13 of a hypnotic sedative drug ( Figure 2). In the synthesis of 3-methyl-Pazinaclone 16, the palladium-catalyzed Buchwald-Hartwig reaction of the lactam 15 in the presence of Xantphos [40] was necessary with the less reactive 1,8-dichloronaphthiridine, while the use of BINAP as ligand or Cu(I)-catalyzed arylation reactions were not effective. The key intermediate 10 was, then, easily converted in the target isoindolinones according to the strategies described in Scheme 3. Amide 12 derived from N-methylpiperazine 11 was easily obtained by EDC/HOBt activation of the carboxylic group. In the case of 1-piperidone ketal 14, Yamada coupling using DEPC (diethylphosphoryl cyanide) was more efficient [38].
NH arylation of the lactams 12 and 15 required several efforts too (Scheme 3). In contrast with arylation of 3-mono-substituted isoindolinones [22,30], phenylation of the lactam 12 was performed using a ligand-less cross coupling reaction based on the Cu(I)-catalyzed arylation of amides [39] in a very good 81% yield, allowing for access to the 3methylated analog 13 of a hypnotic sedative drug (Figure 2). In the synthesis of 3-methyl-Pazinaclone 16, the palladium-catalyzed Buchwald-Hartwig reaction of the lactam 15 in the presence of Xantphos [40] was necessary with the less reactive 1,8-dichloronaphthiridine, while the use of BINAP as ligand or Cu(I)-catalyzed arylation reactions were not effective. NH arylation of the lactams 12 and 15 required several efforts too (Scheme 3). In contrast with arylation of 3-mono-substituted isoindolinones [22,30], phenylation of the lactam 12 was performed using a ligand-less cross coupling reaction based on the Cu(I)catalyzed arylation of amides [39] in a very good 81% yield, allowing for access to the 3-methylated analog 13 of a hypnotic sedative drug ( Figure 2). In the synthesis of 3-methyl-Pazinaclone 16, the palladium-catalyzed Buchwald-Hartwig reaction of the lactam 15 in the presence of Xantphos [40] was necessary with the less reactive 1,8-dichloronaphthiridine, while the use of BINAP as ligand or Cu(I)-catalyzed arylation reactions were not effective.
The target compound 20, 3-methyl-PD172938, was obtained by two different pathways as described in Scheme 4: First of all, ester 9 was chemoselectively reduced by LiBH 4 in quantitative yield. Subsequent mesylation of the alcohol 17 and displacement of the mesyl group with the 3,4-dimethylphenylpiperazine led to 3-methyl-PD172938 20 in a high overall yield. For this purpose, it has to be noted that the sequential one-pot mesylation/piperazine displacement gave sluggish results (addition of piperazine at the end of mesylation reaction). Purification of 18 on silica gel led to decomposition products, while purification on Florisil afforded 18 in a good 73% yield. However, the best results were obtained treating the crude product First of all, ester 9 was chemoselectively reduced by LiBH4 in quantitative yield. Subsequent mesylation of the alcohol 17 and displacement of the mesyl group with the 3,4dimethylphenylpiperazine led to 3-methyl-PD172938 20 in a high overall yield. For this purpose, it has to be noted that the sequential one-pot mesylation/piperazine displacement gave sluggish results (addition of piperazine at the end of mesylation reaction). Purification of 18 on silica gel led to decomposition products, while purification on Florisil afforded 18 in a good 73% yield. However, the best results were obtained treating the crude product 18 with 3,4-dimethylphenylpiperazine at 50 °C in CHCl3, leading to the target 3-methyl-PD172938 20 with 80% overall yield for two consecutive steps. Alternatively, compound 20 can be afforded with similar level of efficiency by reductive amination of the aldehyde 19 with NaBH4 in the presence of 3,4-dimethylphenylpiperazine. The aldehyde 19 is readily accessible in quantitative yield by Dess-Martin oxidation of the alcohol 17, while the reduction of 8 in 19 was not attempted since DIBAL-H is not selective in the presence of amides [41].

Asymmetric Synthesis of 3-Methylated Analog of (S)-PD172938
Then, the asymmetric synthesis 3-methylated analog of (S)-PD172938 was developed. For direct applications in medicinal chemistry programs, the ability to selectively construct a new stereocenter with enantiocontrol is of great importance. Since convenient methods for the asymmetric synthesis of the key intermediate 8 and related compounds bearing a tetrasubstituted stereocenter are not available, we considered the previously investigated asymmetric cascade reaction of readily available 2-acetylbenzonitrile and dimethylmalonate for the direct access to enantioenriched 7 (Schemes 5 and 6) [33]. The development of an effective asymmetric version of this cascade reaction proved to be particularly challenging. A large number of chiral neutral bifunctional organocatalysts and chiral ammonium salts were tested under a range of conditions [33]. The best results were obtained in the presence of the chiral bifunctional ammonium salt 21-derived form (S,S)-1,2-cyclohexanediamine and K2CO3 as the inorganic base in CH2Cl2 under phase transfer conditions (Schemes 5 and 6), leading to moderate enantioselectivity [33]. The possibility to easily synthesize catalyst 21 in high scale [42] allowed us to conveniently scale up the cascade process to 1.72 mmol of 2-acetylbenzonitrile. Despite the moderate enantioselectivity also achieved on a large scale, the advantage of the entire process relies on a very effective and reproducible process of enantio-enrichment by crystallization, increasing the ee of 7 up to 94% in an acceptable overall yield (Scheme 6). In addition, the chiral ammonium salt 21 was recovered after the chromatography in 75% yield and reused with similar level of enantioselectivity, with the possibility to improve the efficiency of the process.

Asymmetric Synthesis of 3-Methylated Analog of (S)-PD172938
Then, the asymmetric synthesis 3-methylated analog of (S)-PD172938 was developed. For direct applications in medicinal chemistry programs, the ability to selectively construct a new stereocenter with enantiocontrol is of great importance. Since convenient methods for the asymmetric synthesis of the key intermediate 8 and related compounds bearing a tetrasubstituted stereocenter are not available, we considered the previously investigated asymmetric cascade reaction of readily available 2-acetylbenzonitrile and dimethylmalonate for the direct access to enantioenriched 7 (Schemes 5 and 6) [33]. The development of an effective asymmetric version of this cascade reaction proved to be particularly challenging. A large number of chiral neutral bifunctional organocatalysts and chiral ammonium salts were tested under a range of conditions [33]. The best results were obtained in the presence of the chiral bifunctional ammonium salt 21-derived form (S,S)-1,2-cyclohexanediamine and K 2 CO 3 as the inorganic base in CH 2 Cl 2 under phase transfer conditions (Schemes 5 and 6), leading to moderate enantioselectivity [33]. The possibility to easily synthesize catalyst 21 in high scale [42] allowed us to conveniently scale up the cascade process to 1.72 mmol of 2-acetylbenzonitrile. Despite the moderate enantioselectivity also achieved on a large scale, the advantage of the entire process relies on a very effective and reproducible process of enantio-enrichment by crystallization, increasing the ee of 7 up to 94% in an acceptable overall yield (Scheme 6). In addition, the chiral ammonium salt 21 was recovered after the chromatography in 75% yield and reused with similar level of enantioselectivity, with the possibility to improve the efficiency of the process. The absolute configuration of compound (+)-7 was previously established to be (R) by VCD measurements when (S,S)-21 is used [33]. This is an important finding since in the present investigation, we used (R,R)-21 to synthesize 7 with (S) configuration. The knowledge of absolute configuration allowed us to propose a plausible transition state for the enantioselective determining step (Scheme 5). A combination of both ionic and hydrogen bond interactions with 21 should favor the intramolecular aza-Michael reaction from the re face of prochiral intermediate c leading to 7 (Scheme 5). The mechanism of the reaction is proposed based on previous experimental [31,33] and theoretical studies [32]. After deprotonation of dimethylmalonate, the resulting carboanion leads to carbonyl addition to ketone, followed by cyclization at cyano group and formation of the cyclic imidate b, which will give then rearrangement to the final product 7. the enantioselective determining step (Scheme 5). A combination of both ionic and hydrogen bond interactions with 21 should favor the intramolecular aza-Michael reaction from the re face of prochiral intermediate c leading to 7 (Scheme 5). The mechanism of the reaction is proposed based on previous experimental [31,33] and theoretical studies [32]. After deprotonation of dimethylmalonate, the resulting carboanion leads to carbonyl addition to ketone, followed by cyclization at cyano group and formation of the cyclic imidate b, which will give then rearrangement to the final product 7. Scheme 5. Proposed mechanism for the base-catalyzed cascade reaction of 2-acetylbenzonitrile and dimethylmalonate.
The synthesis of (S)-10 was successfully achieved by the two-step decarboxylation of (S)-7 without decrease of the enantiomeric purity (Scheme 6). Krapcho decarboxylation was firstly attempted because it directly leads to the ester 8 in high yield. However, the product was recovered with significant racemization (34% ee) probably due to reversible ring opening/ring closing via the retro-Michael reaction eventually occurring at the high temperature. The target compound (S)-20, 3-methylated analog of (S)-PD172938, was obtained by the two different pathways as previously described for racemate synthesis: via the mesyl group displacement in (S)-18 or reductive amination of aldehyde (S)-19, without loss of enantiopurity in both the cases (94% ee) and acceptable overall yields, as described in Scheme 6.

General Information
Unless otherwise noted, all chemicals, reagents and solvents for the performed reactions are commercially available and were used without further purification. 2-Acetylbenzonitrile was purchased from Fluorochem. 1,8-Dichloronaphthiridine was prepared according to the literature [43,44]. Catalyst 21 was synthesized according to the procedure described in ref. 42. All the reactions were monitored by thin layer chromatography (TLC) on precoated silica gel plates (0.25 mm) and visualized by fluorescence quenching at 254 nm. Flash chromatography was carried out using silica gel 60 (70-230 mesh, Merck, Darmstadt, Germany). Yields are given for isolated products showing one spot on a TLC plate, Scheme 6. Enantioselective synthesis of (S)-20, the 3-methylated analog of (S)-PD172938.
The synthesis of (S)-10 was successfully achieved by the two-step decarboxylation of (S)-7 without decrease of the enantiomeric purity (Scheme 6). Krapcho decarboxylation was firstly attempted because it directly leads to the ester 8 in high yield. However, the product was recovered with significant racemization (34% ee) probably due to reversible ring opening/ring closing via the retro-Michael reaction eventually occurring at the high temperature. The target compound (S)-20, 3-methylated analog of (S)-PD172938, was obtained by the two different pathways as previously described for racemate synthesis: via the mesyl group displacement in (S)-18 or reductive amination of aldehyde (S)-19, without loss of enantiopurity in both the cases (94% ee) and acceptable overall yields, as described in Scheme 6.

General Information
Unless otherwise noted, all chemicals, reagents and solvents for the performed reactions are commercially available and were used without further purification. 2-Acetylbenzonitrile was purchased from Fluorochem. 1,8-Dichloronaphthiridine was prepared according to the literature [43,44]. Catalyst 21 was synthesized according to the procedure described in [42]. All the reactions were monitored by thin layer chromatography (TLC) on precoated silica gel plates (0.25 mm) and visualized by fluorescence quenching at 254 nm. Flash chromatography was carried out using silica gel 60 (70-230 mesh, Merck, Darmstadt, Germany). Yields are given for isolated products showing one spot on a TLC plate, and no impurities were detectable in the NMR spectrum. The NMR spectra were recorded on Bruker DRX 600, 400 and 300 MHz spectrometers (600 MHz, 1 H, 150 MHz, 13 C; 400 MHz, 1 H, 100.6 MHz, 13 C; 300 MHz, 1 H, 75.5 MHz, 13 C; 250 MHz, 1 H, 62.5 MHz, 13 C). Internal reference was set to the residual solvent signals (δ H 7.26 ppm, δ C 77.16 ppm for CDCl 3 ). The 13 C NMR spectra were recorded under broadband proton decoupling. Spectra are reported only for unknown compounds. The following abbreviations are used to indicate the multiplicity in NMR spectra: s-singlet, d-doublet, t-triplet, q-quartet, dd-doublet of doublets, m-multiplet and brs-broad signal. Coupling constants (J) are quoted in Hertz. High-resolution mass spectra (HRMS) were acquired using a Bruker SolariX XR Fourier transform ion cyclotron resonance mass spectrometer (Bruker Daltonik GmbH, Bremen, Germany) equipped with a 7T refrigerated actively shielded superconducting magnet. For ionization of the samples, electrospray ionization (ESI) or MALDI was applied. Enantiomeric excess of chiral compounds was determined by HPLC using chiral column.

2-(1-Methyl-3-Oxoisoindolin-1-yl) Malonic Acid (9)
Isoindolinone 7 (400 mg, 1.44 mmol) was dissolved in CH 2 Cl 2 (12 mL, 0.12M), and 2M NaOH in MeOH (6.5 mL) was added with stirring of the mixture at room temperature for 24 h. The solvent was removed under reduced pressure, and then the white solid was solubilized in water (2.5 mL) and washed with ethyl acetate. The aqueous layer was then acidified with 3N HCl and extracted with ethyl acetate 5 times, giving a white foam. Yield: 97% (350 mg). Used in the next reaction without further purification. IR (