Green synthesis of glycerol 1,3-bromo- and iodohydrins under solvent-free conditions

ABSTRACT New eco-friendly conditions for Finkelstein reaction employing homogeneous media or alumina-supported reagents, both solventless, were utilized in the bromohydrins and iodohydrins synthesis from 1,3-dichlorohydrin (1,3-DCH), obtained from glycerol. Thus, the trans-halogenation of glycerol-1,3-dichlorohydrin (1) in homogeneous media, using NaI, NaBr salts and organic additives (TBAI, TBAB), led to the synthesis of four different glycerol halohydrins (2a, 2b, 3a, 3b) in regular to excellent conversions (43–96%) and good selectivities to 3a and 2b. Already, the alumina-supported reagents such as KI, KBr, NaBr were used for glycerol-1,3-dichlorohydrin trans-halogenation, in the absence of solvent, producing halohydrin mixtures in high conversions (77–99%) and very good selectivity (82%) to 1,3-diiodo-2-propanol (2b). GRAPHICAL ABSTRACT

The global crude glycerol production forecast for 2020 is 4 billion liters (6). This triol exhibits enormous applicability in the chemical, pharmaceutical, food, cosmetic industries. Furthermore, it has been used as a building block and solvent inorganic synthesis (1)(2)(3)(4)(5). Biologically it is used as source energy and carbon for many microorganisms, to produce various chemicals, as cryopreservation agent of different biological materials, as an osmoregulatory agent and also used in the production of bioplastics (7,8).
The dichlorohydrins (DCH's) are low-cost glycerol derivatives with important industrial applications, being used mainly for the manufacture of epoxy resins, acting as a substrate and solvent during the process. The 1,3-diclorohydrin (1,3-DCH) is largely obtained from propene; however, sustainable synthesis in one step from glycerol using HCl catalyzed by acetic acid and Cl 2 at high temperatures are already consolidated (9)(10)(11)(12)(13)(14). On the other hand, the brominated and iodinated derivatives of glycerol are mainly applied as intermediates in fine chemistry, whereas they exhibit higher reactivity than DCHs. Despite the higher cost, brominated and iodinated intermediates considerably increase the rate in bimolecular nucleophilic substitution reaction (S N 2) when compared with the chlorinated intermediates, since the dissociation energy of the carbon-chloro bond is 83.7 kcal/mol, higher than 72.1 and 57.6 kcal/ mol for C-bromine and C-iodine, respectively (15,16).
In nature, iodohydrins and bromohydrins were found in the essential oils of marine algae of the genus Asparagopsis taxiformis, A. armata and Falkenbergia rufolanosa (17,18). Synthetically, 1,3-dibromohydrins are normally produced from the corresponding epibromohydrin (19) or from the halogenation of propene (20) or yet from the treatment of glycerol with bromhidric acid (HBr) or bromine (Br 2 ), in the presence of red phosphorous, at high temperatures (21,22). The glycerol bromohydrins can be also obtained from different substrates, via nucleophilic substitution in various solvents and moderate yields (23)(24)(25).
On the other hand, attempts to synthesize the glycerol iodohydrins in a single step, by treating the glycerol with excess hydriodic acid, result in the synthesis of isopropyl iodide, due to the exhaustive iodination of glycerol (26). Consequently, some glycerol monoiodohydrins are generally obtained through the mildest Finkelstein reaction with different electrophilic species (27)(28)(29)(30)(31)(32)(33). Specifically, in the case of 1,3-diiodohydrin, there is a few reports in the literature to the production of this important synthon, which employs the homogeneous medium, in the presence of solvent.
In contrast to the glycerol chloro-and bromohydrins, the glycerol iodohydrins have been applied in various pharmaceutical preparations because for their antiseptic, mucolytic and fungicide properties (34,35). Some glycerol iodohydrins were used in the synthesis of cardiolipins (36,37), as key intermediates of a promising class of drugs in treating type 2 diabetes (19,38), and applied in organic synthesis as a building block to produce other chemicals (39,40).
The Finkelstein reaction is widely used in organic synthesis (41)(42)(43)(44)(45)(46)(47)(48)(49)(50). It consists in the displacement of a leaving group under a C-sp 3 (Cl, Br, I, OTs, OMs, etc.) by the action of a metallic halogenated nucleophile mainly acetone as solvent, where the new organic or inorganic metallic salt is insoluble (51,52). Although the reaction has been considered to evolve via a mechanism S N 2, it is an equilibrium process. Kinetic studies ratify that the reversibility of the reaction is not observed during transhalogenation due to the difference in the solubility between the salts involved in the process (53,54). For example, NaI is highly soluble in acetone and when this is reacted with an alkyl chloride the NaCl formed is insoluble in acetone and precipitate, thus there is the equilibrium displacement in favour of the desired products (Le Châtelier's principle) (55).
In our continuing interest in adding value to glycerolsurpluses originated from biodiesel using sustainable and eco-friendly (8,9) routes, now we desire to describe the development of green processes for obtaining bromohydrins and iodohydrins from glycerol-biodiesel. Thus, the reactivity of metal halides in the presence or absence of additives (tetrabutylammonium bromide (TBAB)) and tetrabutylammonium iodide (TBAI) with glycerol-1,3-dichlorohydrin 1, in homogeneous medium solventless was investigated. Furthermore, the reactivity of 1, with different alumina-supported metal halides, in solvent-free conditions was also studied.
By the analysis of Table 1, it can be observed that 1 was converted in the mixture of glycerol iodohydrins 2a+2b (77:23) in 43% yield by treatment with NaI (Entry 1). Initially, a yellowish homogeneous mixture was formed and with the evolution of time the mixture became brownish with the presence of NaCl precipitate. Interestingly, the 1,3-DCH acted as the solvent and reagent of the reaction, favouring an adequate sodium iodide dissociation. In this manner, as the reaction evolved the action of 1,3-DCH as the solvent decreased, leading to lowest NaI dissociation, which can explain the low conversion. In order to obtain better conversions, we decided to evaluate the use of TBAI as an additive that could promote greater ion iodide availability (Entry 2). Gratifyingly, the use of TBAI provided 1,3-DCH in 96% conversion as a iodohydrin mixture 2a+2b (28:72).
The synthesis of 2a and/or 2b using KI as source iodide anion was also experienced in the absence of the solvent. However, low conversion of 2a+2b was obtained (3% without TBAI addition and 11% with TBAI addition). We believe that the lower trend to the KI dissociation can explain the poor conversion observed.
The bromohydrin synthesis 3a+3b was also tried using NaBr at 80°C (Entry 3). In this case, there was not a good conversion and selectivity to the desired bromohydrins, probably due to the lower nucleophilicity and dissociation constant of the sodium bromide despite the heating. Aiming to develop other eco-friendly route for the selective iodination of 1,3-DCH, we propose the use of metallic salts supported on alumina as a source of halogenation, in the absence of the solvent. In recent years, there is great interest in the support of metallic halides on alumina and its use in the production of biodiesel, due to the basic species present in this solid reagent. Xie (58) and Evangelista (59) independently have used KI/Al 2 O 3 as an efficient basic catalyst for biodiesel production. X-ray diffraction measurements have evidenced the presence of K 3 AlI 6 and K 2 O in significant concentrations, in addition to residual KI, as an active species in this catalyst (58,59). We imagine that the presence of iodine in the species K 3 AlI 6 could alternatively to react with the glycerol 1,3-DCH and provide the desired iodohydrins 2a and 2b, since only the use of KI salt solventless did not exhibit good conversion, as shown previously. Then, we employed KI/Al 2 O 3 reagent, at room temperature for the production of 2a and 2b; however, a low conversion of 1 was observed (Entry 4). Thus, we chose to perform the iodination with the same reagent at higher temperature (Entry 5). Gratifyingly, 99% of consumption of 1 was observed, providing an iodohydrin mixture (2a:2b; 9:91). It is important to mention that when the same 1,3-DCH iodination was processed at 100°C, no product or reagent was obtained. On the other hand, the use of a shorter reaction time led to lower conversions and selectivity (Entries 6, 7). Besides, the use of 1.0 equivalent of KI/Al 2 O 3 led to 90% conversion and 2a:2b; 92:8 selectivity.
Similarly, when the classical halogenating agent used in Finkelstein reaction was supported on alumina (NaI/ Al 2 O 3 ) and reacted with 1 by 3 days, at room temperature, a regular conversion of 1 (76%) was observed with higher selectivity to 2b (Entry 8). Aiming at a greater conversion and selectivity, we processed the reaction at 80°C, however, no mass of product or reagent was recuperated. Possibly, in this temperature there was the alkoxides formation promoted by the action of basic species, causing the impregnation of the substrates and product irreversibly in the reagent (NaI/Al 2 O 3 ). A posterior study about the structural arrangement of this alumina-supported reagent will be necessary to rationalize the results obtained (59).
Reactions using KBr/Al 2 O 3 and NaBr/Al 2 O 3 and 1 were also investigated at room temperature (Entries 9 and 10). In this case, only a conversion of 26% and 48% to bromohydrins was observed, respectively. On the other hand, the use of 80°C led to recuperation of traces of a complex mixture of products to both bromination agents, probably, due to the action of the basic species present in the solid reagent. In view thereof, the use of a lower temperature (50°C) was experimented and as expected, a better conversion was observed (Entries 11 and 12).

Preparation of the alumina-supported reagents
The alumina-supported reagents were prepared by the incipient wet impregnation method, where 15.0 ml of the aqueous solution of KI, NaI, KBr or NaBr (35% p/v) was reacted with 5.0 g basic alumina oxide for 2 h under stirring and refluxing. After the mixture was dried in the oven for 16 h at 120°C, the solid reagent was calcined in a muffle at 500°C for 3 h, and after cooling, it was used as the reactant for the trans-halogenation reactions. It is important to mention that this reagent can be recycled and reused (59).
3.3. General procedure to iodohydrins 2a and 2b synthesis in the homogeneous medium 1,3-Dichloro-2-propanol (1) (1.0 mmol, 0.129 g, 0.1 mL) and metal iodide or tetrabutylammonium iodide (2.5 mmol) were added to a round bottom flask. Next, a Liebig condenser and a pressure equalizer were adapted to the system and the reaction medium was kept under strong magnetic stirring at 80°C for 3 days. After 24 h a brownish mixture with the presence of NaCl precipitate was formed. After the mixture was cooled into room temperature, the solid sodium chloride was filtered and washed with dichloromethane until it became white. Then, dichloromethane was removed, and the crude reaction was extracted with diethyl ether (3 × 40 mL) and the combined extracts were washed with the sodium thiosulfate-concentrated solution to remove the iodine. Next, the phase organic was dried over anhydrous Na 2 SO 4 and evaporated under reduced pressure. When TBAI was used, this organic salt was removed by precipitation with dichloromethane and ethyl acetate (1:1) and removed by filtration. 3.4. General procedure to bromohydrins 3a and 3b synthesis in the homogeneous medium 1,3-Dichloro-2-propanol (1) (1.0 mmol, 0.129 g, 0.1 mL) and metal bromide (2.5 mmol) were added to a round bottom flask. Next, a Liebig's condenser and a pressure equalizer were adapted to the system and the reaction medium was kept under strong magnetic stirring at 80°C for 3 days. After 24 h a brownish mixture was formed with the presence of metal chloride precipitate. After the mixture was cooled to room temperature, the metal chloride was filtered and washed with dichloromethane until it became white. Then, dichloromethane was removed under reduced pressure and the crude yellowish liquid consisted of a mixture of 1,3-dibromo-2-propanol (3b) and 1-chloro-3-bromo-2-propanol (3a), whose ratio was determined by the gas chromatography, see table.
1 3.5. General procedure to iodohydrins 2a and 2b synthesis using solid-supported reagents 1,3-Dichloro-2-propanol (1) (1.0 mmol, 0.129 g, 0.1 mL) and iodide potassium supported alumina or iodide sodium supported alumina (3.0 g) were added to a round bottom flask without magnetic stirring at 80°C for 3 days. After this period, dichloromethane (50 mL) was added to the reaction flask and the heterogeneous mixture was formed after strong magnetic stirring, aiming at the extraction of the iodohydrins mixture. After simple filtration, the brownish filtered obtained was worked up similarly to the reaction conducted in homogeneous medium.
3.6. General procedure to bromohydrins 3a and 3b synthesis using solid-supported reagents 1,3-Dichloro-2-propanol (1) (1.0 mmol, 0.129 g, 0.1 mL) and potassium bromide supported on alumina (KBr/ Al 2 O 3 ) or bromide sodium-supported alumina (3.0 g) were added to a round bottom flask without magnetic stirring at 80°C for 3 days. After this period, dichloromethane (50 mL) was added to the reaction flask under strong magnetic stirring for the extraction of the bromohydrin mixture. After the solids were retired by simple filtration and the yellow organic phase was worked up similarly to the reaction processed in the homogeneous medium.

Conclusions
A green synthesis to glycerol bromohydrins (3a,3b) and glycerol iodohydrins (2a,2b) from bioglycerol was developed under solvent-free conditions. We report a cleaner method for the halohydrin synthesis (2a,b and 3a,b), in high conversions and selectivities, employing reactions in homogeneous media or alumina-supported reagents, both solventless. To the best of our knowledge, this is the first time that KI supported on alumina, a reagent usually used as a basic catalyst in transesterification reactions (58,59), was employed as an efficient iodination agent in the solvent-free Finkelstein reaction.

Disclosure statement
No potential conflict of interest was reported by the authors.

Funding
We thank CAPES and CNPq for fellowships for some authors.

Notes on contributor
Priscila Faustino dos Santos (MSc) is currently attending Federal University of Rio de Janeiro for her Doctor's degree in organic synthesis.
Sara Raposo Benfica da Silva (MSc) is currently attending Federal University of Rio de Janeiro for her Doctor's degree in organic synthesis.
Fernanda Priscila Nascimento Rodrigues da Silva (MSc) is currently attending Federal University of Rio de Janeiro for her Doctor's degree in organic synthesis.
Jeronimo da Silva Costa is Ph.D. in Organic Chemistry (Organic Synthesis) from Federal University of Rio de Janeiro, currently is an associate professor at the Federal Institute of Education, Science and Technology of Rio de Janeiro.
Jane Sayuri Inada is a graduate in chemistry from Federal Institute of Education, Science and Technology of Rio de Janeiro, currently act as Quality Control Analyst at Johnson & Johnson Company, São Paulo, Brazil.
Vera Lúcia Patrocinio Pereira is Ph.D. in Organic Chemistry (Organic Synthesis) from Federal University of Rio de Janeiro. Currently is an associate professor at the Federal University of Rio de Janeiro, where heads for 23 years, the Laboratory of Stereoselective Synthesis of Bioactive Substances of the Federal University of Rio de Janeiro. In this Laboratory, the bioglycerol is used for the sustainable production of small molecules with high added value. Additionally, nitroalkane chemistry is intensively studied and used in the enantioand diastereoselective synthesis of natural products, of bioactive substances, as well as, in the synthesis of enantiomerically pure synthetic intermediates. The Professor Pereira, V.L.P. has supervised more than thirty master's and doctoral students, published more than 23 scientific articles and accomplished the deposit of four patents in the field of glycero-chemistry.