An expeditious and efficient bromomethylation of thiols: enabling bromomethyl sulfides as useful building blocks

A facile and highly efficient method for the bromomethylation of thiols, using paraformaldehyde and HBr/AcOH, has been developed, which advantageously minimizes the generation of highly toxic byproducts. The preparation of 22 structurally diverse α-bromomethyl sulfides illustrates the chemo-tolerant applicability while bromo-lithium exchange and functionalization sequences, free radical reductions, and additions of the title compounds demonstrate their synthetic utility.

Heteroatom halomethylations 1 have proven to be extremely useful for the generation of valuable synthetic intermediates. 2 Halomethylation of thiols provides synthetically valuable chloromethylated intermediates (chloromethyl suldes), which are typically prepared by condensation with bromochloromethane in basic media, 3 or with HCl and a formaldehyde source (paraformaldehyde, polyoxymethylene, etc.). 4 While chloromethyl suldes have been traditionally used as alkylating reagents, the analogous bromomethyl counterparts offer superior electrophilicity, recognized since the earliest report describing their syntheses using hydrogen bromide and paraformaldehyde, 5 yet they are oen overlooked in this role. Moreover, the reactivity scope of bromomethyl thiol derivatives remains largely unexplored, despite a potentially broader synthetic range (e.g. for the generation of organometallics by metal-halogen exchange). 6 Other methods for the generation of bromomethylated thiol derivatives consist of replacing hydrogen bromide gas with concentrated aqueous hydrobromic acid, along with a formaldehyde source (usually paraformaldehyde), 7 or by using dibromomethane 8 in basic media. 9 Two or three-step procedures consisting of hydroxymethylation followed by substitution have also been developed. 10 A desilylative rearrangement of a-TMS suldes has also been used for the generation of bromomethylsuldes. 11 As part of our interest in the preparation and application of structurally diverse sulfur-based building blocks, 12 we investigated the preparation of benzyl(bromomethyl)sulfane (2a), previously used as an olenation reagent. 13 However, several attempts to prepare 2a through exposure of benzylmercaptan (1a) to paraformaldehyde and hydrobromic acid, 7b led to a ca. 1.5 : 1 mixture of 2a (32%) and bis(benzylthio)methane 3a (21%, Scheme 1a). Iterations of the experiment always delivered important and variable amounts of the dithioacetal by-product 3a. On the other hand, an alternate approach to the bromomethylation of a cyclohexanethiol bromomethyl derivative 2b, using dibromomethane and K 2 CO 3 , resulted in trace amounts of the dithioacetal derivative 3b only (Scheme 1b). 14 HBr/AcOH is a convenient hydrogen bromide source that minimizes exposure to risky set-ups and has been employed as a surrogate to highly corrosive and toxic hydrogen bromide gas in numerous applications. 15 Although this reagent has been used previously in the generation of bromomethyl suldes, installation of the methylene bridge required rst a Spivaloxymethylation of a mercaptan, followed by cleavage by HBr/AcOH. 16 Surprisingly, sequential exposure of thiols 1a or 1b to paraformaldehyde and HBr/AcOH, 17 rapidly delivered bromomethylated derivatives 2a and 2b with outstanding yields (Scheme Scheme 1 Attempts for the bromomethylation of 1a or 1b under (a) acidic or (b) basic media. (c) and (d) A highly efficient and direct approach for thiol bromomethylation (this work). 1). The simple experimental setup and straightforward purication procedure offer methodological utility; in most cases extraction with a low-boiling point hydrocarbon such as pentane or hexanes is sufficient to recover the material in high purity (>95%). 18 Traces of impurities can be easily discarded through bulb-to-bulb vacuum distillation.
The reaction scope was explored with a series of structurally diverse thiols (Table 1). Aliphatic thiols yielded bromomethyl suldes in excellent yields, although lower yielding 2g is attributed to its high volatility. t-Butyl bromomethyl sulde (2i), a sterically challenging and volatile material that has been used as a synthetic equivalent of the methylmercaptan group (-CH 2 -SH), 19 was prepared satisfactorily in 76% yield. Fluorinated bromomethyl 2e has been used for the preparation of uorinated surfactants, 10a,20 which some exhibit antimicrobial activity. As a previous method involves a 2-step sequence involving thiol hydroxymethylation and substitution by PBr 3 , our method directly delivered 2e in 88% yield.
Aromatic substrates (2j-2v) were generally high yielding. For example, (bromomethyl)(phenyl)sulfane (2j), a useful electrophile and precursor to phenylthiomethyl azide 21 and diethyl phenylthiomethane phosphonate, an olenation reagent, 22 can be prepared in nearly quantitative yield. Aryl derivatives (bromomethyl)(4-methylphenyl)sulfane (2k) and (bromomethyl)(4-chlorophenyl)sulfane (2m), used in the preparation of [(p-phenylphenyl)oxy]methyl (POM) protective group, 23 gave 87% and quantitative yields respectively. Comparatively, previously reported methods delivered 2k and 2m in 43% and 75% yield respectively. 11 Anisyl thiol 1r was a challenging substrate, as the bromomethylation was highly exothermic and resulted in a near 1 : 1 mixture of bromome-thylsulde 2r and dithioacetal 3r. The yield of 2r was improved to 4 : 1 ratio, by cooling the reaction mixture to 0 C. However, purication of 2r was also problematic as distillation led to partial decomposition. We speculate that integrity of 2r during preparation and purication is inuenced by the neighbouring methoxy function. On the other hand, 2s-u modest yields are attributed to a decrease in S-nucleophilicity caused by the EWG groups. Interestingly, although thiol 1v bears an EWG at ortho position, methyl 2-((bromomethyl)thio)benzoate 2v was obtained in excellent yield (85%).
NMR analyses of a fresh mixture of paraformaldehyde and HBr/AcOH 24 revealed a mixture consisting mainly of a component with an 1 H-NMR 5.8 ppm signal, correlating to a 13 C-NMR 68.2 ppm signal (HSQC). This species evolves mainly into two different components: one of them being bis(bromomethyl ether) as determined by a signal at 5.7 ppm ( 1 H-NMR), 25 and bis(bromomethoxy) methane (signals at 5.6 ppm and 5.0 ppm). 26 The 5.8 ppm signal is presumed to belong to bromomethanol, 27 which is consumed promptly by the thiol reagent. This is congruent with our observations, since the best results were obtained when the addition sequence consisted in adding the HBr/AcOH mixture to premixed thiol and paraformaldehyde (Scheme 2). Equimolar ratios of paraformaldehyde are enough for complete transformation, avoiding formation of potentially highly-toxic bis(bromomethyl ether). 28 Alkenyl and alkynyl substrates (2w, 2x) were incompatible to this method as the bromomethylation procedure led to complex mixtures. Mercaptans featuring attenuated nucleophilicity such as thioacetic or thiobenzoic acids (2y, 2z), p-nitrothiophenol (2aa), and 2-mercaptopyridine (2bb) were unsuitable for this methodology. Table 1 Thiol bromomethylation with HBr/paraformaldehyde a a Reaction was performed at À20 C. b Reaction was performed at 0 C. c Reaction was performed at 30 C. d Reaction was performed at 40 C.
Attempts to diversify the a-alkyl component, found that exclusively highly reactive aldehydes underwent bromoalkylation with thiols (Table 2). Bromoalkylation yields using aldehydes is evidence that reaction efficiency is strongly dependent on the carbonyl reactivity, as bromo(4-nitrobenzylation) or bromoethylation of 4-methylbenzenethiol (1k) using electrophilic 4-nitrobenzaldehyde or acetaldehyde 29 respectively, feature fair yields compared to the corresponding bromomethylation using paraformaldehyde (cf. entries 1, 3 and 4). Interestingly, thiol nucleophilicities have a larger impact in thiol bromoalkylations using aldehydes compared to bromomethylations with paraformaldehyde, as illustrated with superior reaction efficiency when benzyl mercaptan 1a was used instead of 1k (cf. entries 6, 7 and 9). Thiol bromoalkylation using ketones had no practical use as dithioketal 3k4 was the only product when acetone was used as the carbonyl component (entries 10 and 11) and acetophenone yielded a complex mixture (entry 12).
To illustrate the versatility of bromomethyl suldes as building blocks, we rst carried out a polarity reversal through a halogen-metal exchange approach, a relatively rare procedure for the generation of a-sulfanylmethyl organometallics. [30][31][32] This approach is underdeveloped, probably because of difficulties in synthesizing bromomethylsuldes. 33 Classically, generation of a-sulfanylmethyl organolithiums has been carried out mainly by deprotonation. 34 However, the deprotonation approach has important drawbacks, such as a substitution side-process that generate thiolates or regioselectivity issues when dialkyl suldes are deprotonated. 35 Sequentially exposing (bromomethyl)(cyclohexyl)sulde (2f) or (bromomethyl)(p-tolyl)sulde (2k) to nBuLi, generated nucleophilic organolithiums 4f and 4k, that were quenched by benzaldehyde thus assembling alkylated derivatives 5f and 5k in good yields (Scheme 3a). Using (+)-neomenthanethiol bromomethyl sulde derivative (2h) for the bromo-lithium exchange and benzaldehyde in the electrophilic quench, generated b-hydroxysulde 5h in good yield albeit low diastereoselectivity (ca. 1.4 : 1). This constitutes a novel approach for the application of sulfenyl methyllithium organometallics for the access of b-hydroxysuldes, valuable intermediates or fragments of natural products and biologically relevant compounds, usually prepared under acidic media or free radical oxidative conditions. 36 On the other hand, preparation methods of mixed or unsymmetrical dithioacetals are scarce, 37 some of them displaying selectivity limitations. 38 Similar bromo-lithium exchange/functionalization procedures were also carried out on probes 2f and 2k using diphenyldi-sulde as electrophile, 39 delivering mixed thioacetals 6f and 6k respectively also with good yields (Scheme 3b). The exceptional electrophilicity of bromomethyl suldes 2f and 2k, also enabled the access to mixed thioacetals 6f and 6k by simple exposure to sodium thiophenolate, 40,41 thus demonstrating the versatility of bromomethyl suldes either as electrophiles or nucleophiles aer umpolung.
To our knowledge, bromomethyl suldes 2 have not been exploited for C-C bond construction through free radical chemistry. As far as we know, there is a single reference to an unrealized effort attempting an intramolecular free radical cyclization of an unavailable alkenyl bromomethylsulde. 14,42 Although our method unfortunately was not compatible with the direct preparation of alkenylsulde bromomethyl derivatives (see Table 1), we could demonstrate exceptional reactivity of a-thiomethyl radical 7k (generated from 2k), towards (TMS) 3 SiH reduction, 43 thus generating thioether 8k (Scheme 4a). On the other hand, Et 3 B initiated 44 additions of nucleophilic radicals 40,45 7f and 7k on radical acceptors acrylonitrile and methyl acrylate led to the  Scheme 4 Unprecedented generation of a-thiomethyl free radicals from bromomethyl sulfides and their reduction and addition to acrylonitrile and methyl acrylate.
generation of g-functionalized suldes 9f, 9k, 10f and 10k (Scheme 4b). This constitutes a novel approach for the synthesis of g-sulfanyl butanenitriles and esters, as an alternative to the thiol-ene reaction approach, 46 and establishes bromomethyl suldes as a new entry on the family of monothiomethyl radical sources. 47

Conclusions
Development of synthetic methods based on bromomethyl suldes has been limited mainly by their ambiguous availability, as a consequence of methods lacking selectivity, efficiency, and perilous set-ups. We have developed a simple and high yielding method for thiol bromomethylation, that involves operational simplicity, and minimizes operational risk. The method has broad scope and good functional group tolerance but is unsuitable for low-nucleophilicity mercaptans. The bromomethylating reagent is suspected to be bromomethanol, obtained stoichiometrically and efficiently captured by thiols, thus preventing the formation of undesired toxic species such as bis(bromomethyl)ether. We re-disclosed the applicability of bromomethyl suldes as precursors of lithiated organometallics and performed unprecedented free radical additions, that support the usefulness of these building blocks. More synthetic applications and derivatizations of bromomethyl suldes are currently being developed in our laboratories.

Conflicts of interest
There are no conicts to declare.