Correction: Synthetic routes for a variety of halogenated (chiral) acetic acids from diethyl malonate

Correction for ‘Synthetic routes for a variety of halogenated (chiral) acetic acids from diethyl malonate’ by Manuel R. Mazenauer et al., RSC Adv., 2017, 7, 55434–55440.


Introduction
Small chiral molecules are excellent prototype candidates for the investigation of molecular parity violation, 1,2 the determination of absolute congurations by new direct methods based on Coulomb explosion imaging, [3][4][5][6] and the study of frequency dependent chiroptical properties. 7,8 Due to the small number of atoms, they are also well suited to high-quality ab initio calculations 9-12 that are relevant for a detailed understanding of the spectroscopic investigations of chiral molecules. [13][14][15] The synthesis of chiral acetic acid with all three hydrogen isotopes at the chiral C-atom and known handedness was truly pioneering work for the investigation of enzyme reaction mechanisms back in 1969. [16][17][18] Here we focus on chiral halogenated acetic acids, 19 CXYZCOOH, which are of general interest in the elds mentioned above. They serve as precursors to halomethanes, specically CHBrClF, [20][21][22][23] CHBrFI, 24 and isotopically chiral CHDXY. [25][26][27][28] CHBrClF (7) itself has been subject to very early attempts to measure parity violation effects in molecules. 29 Since the prediction and experimental verication of the weak interaction in particle physics in around 1956/57, 30,31 it became immediately clear that this interaction also has consequences in chemistry: for an isolated enantiomer, racemization following R ¼ S leads to a tiny heat of reaction, D r H , in the order of 10 À15 kJ mol À1 (ref. 32 and 33) which can be related to the parity violating energy difference, D pv E, between both enantiomers according to D pv E z D r H , thus leading to slightly different absorption spectra of isolated R and S molecules. Attempts to measure these effects have failed until today, [32][33][34] but molecular parity violation is still an active eld of research.
The synthesis of these halomethanes usually starts from perhalogenated ethene (see Scheme 1). 21,23 This reactant is not very well suited for the synthesis of a large variety of halogenated acetic acids since it has a low boiling point and is subject to producing undesirable byproducts. Furthermore, due to the CFC(chlorouorocarbon)-ban in 1989, its commercial availability is very limited. We therefore examined a synthetic pathway starting from an easily available diethyl malonate. [35][36][37] We explored the reaction scheme (Scheme 2) for the synthesis of a whole series of halogenated chiral (and also achiral) acetic acids with high yields and purity. 38 Some of these lead directly to the corresponding chiral halomethane upon decarboxylation. 39,40 Results and discussion First we focus on CHBrClF (7) and its corresponding acetic acid (5d). A complete multistep synthesis for 7 was described by Doyle and Vogl 22,39 and optimized by Beil et al. 21 a few years later. Both research groups 21,22,39 used the same experimental procedure as shown in Scheme 1 but instead of using 1,2-dichloro-1,2-diuoroethylene (1) they started with chlorotriuoroethylene with different corresponding reaction conditions for some steps. Our modied synthetic route (Scheme 1) has some advantages compared to the past approaches that use chloro-triuoroethene as the starting material, namely 1 has a higher boiling point of about 19 C (À27 C for chlorotriuoroethene). Furthermore, the symmetric ethene 1 leads to fewer byproducts than when using chlorotriuoroethene. The overall yields, however, do not differ much.
In this work we present pathways to obtain chiral haloacetic acids starting from diethyl malonate 8. In comparison to past procedures starting from perhalogenated ethene [20][21][22] for the synthesis of chiral halogenated acetic acids as just described in Scheme 1, the procedure advocated here has several advantages: (1) the reactant 8 is a common chemical reagent, cheap and easily accessible; (2) due to its CH acidity at the a-carbon atom it is a perfect starting material for halogenation reactions and (3) only a small number of reaction steps are needed. This paves the way to synthesize different dihalogenated diethyl malonates and, through decarboxylation, the corresponding halogenated acetic acids. Scheme 2 shows the halogenation reactions explored to synthesize various combinations of brominated, chlorinated and uorinated diethyl malonates together with the main reaction details. The yields from our reaction in Scheme 2 are summarized and compared to previous work in Table 1.
The halogenation of 8 with bromine leads to diethyl bromomalonate (9a) (experimental conditions are detailed in Scheme 2); the same reaction has been studied in the past by irradiating with light in the presence of Br 2 . 36 Following Scheme 2, diethyl chloromalonate (9b) was obtained from 9a by replacing bromine with chlorine (99% yield), whereas diethyl uoromalonate (9c) was inaccessible with usual mild uorination agents. Fluorination is oen performed with HF and triethylamine, 41,42 which we did not try to reproduce in our laboratory since 9c is commercially available. 9a and 9b have also been obtained from 8 in the presence of MgX 2 (X ¼ Br, Cl, and I) under microwave irradiation. 43 We tested different methods to obtain 9b directly from 8, but these attempts resulted in poor yields and mainly the formation of diethyl dichloromalonate. However, the two step procedure from 8 to 9b via 9a is an efficient alternative.
When the acidic hydrogen in 9a is substituted with chlorine, we obtain the diethyl bromochloromalonate (10a, 97% yield). When optimizing the procedure described by Read and McMath, 35 Zimmer et al. 37 started from 8 and added bromine to synthesise 9a and added SO 2 Cl 2 to synthesize 10a. Recently, 10a was obtained from the diazo compound of diethyl malonate with a yield of about 90% (ref. 44) in the presence of a Cucatalyst. Another way to synthesize 10a is by the bromination of 9b (100% yield). To obtain diethyl bromouoromalonate (10b), we uorinated 9a (86% yield) or brominated 9c (82% yield). Two routes are possible to synthesize diethyl chloro-uoromalonate (10c), either by uorinating 9b (73% yield) or chlorinating 9c (86% yield). Thus, we successfully explored at least two different routes to synthesize diethyl dihalogenmalonates which enabled us to choose the pathway with the higher yield (see Table 1) and the most suitable reaction conditions. Scheme 3 summarizes the synthetic pathways from a dihalogenated diethylmalonate to the corresponding dihalo acetic acid upon introducing D or H. Decarboxylation to the deuterated acetic acids 5a-c was successfully performed for deutero bromochloroacetic acid 5a with a yield of 67%. Deutero bro-mouoroacetic acid (5b, 51% yield) and deutero chlorouoroacetic acid (5c, 49% yield) were also obtained. Scheme 4 depicts the complete synthetic pathway from 8 to chiral halomethane 7. The decarboxylation of 10c to the bro-mochlorouoroacetic acid 5d (Scheme 4) requires two more reaction steps compared to the decarboxylation when D or H are inserted (see Scheme 2). Firstly, 10c was saponied specically to the monoethylester potassium salt. The Hunsdiecker decarboxylation 54 was carried out with the corresponding silver salt (11c), which was obtained in a second step using a silver loaded ion exchange resin. Secondly, the resulting silver salt was then decarboxylated to obtain bromochlorouoroacetate as product. The overall yield of these two steps was approximately 26% and the purity of the crude product was better than 80%. The decarboxylation was carried out with up to 20 mmol (z6 g) of silver salt. Thirdly, a simple saponication of bromochloro-uoroacetate (4) with NaOH results in 5d with a yield of 99%. Table 1 summarizes our results with respect to recent synthetic attempts to obtain monohalogenated (9), and dihalogenated (10) malonates, as well as the corresponding chiral halogenated acetic acids.
Based on the chiral halogenated acetic acids, the corresponding halomethanes were, in principle, accessible. The required decarboxylation procedure was described in the literature for the example of the bromochlorouoroacetic acid 5d to the corresponding halomethane CHBrClF (7). 39 We have repeated this reaction here with a yield of 48% and a purity of 99% compared to a 50 to 70% yield reported by Doyle and Vogl. 39 Although the overall yield of this 9 step synthesis is only approximately 12% (which is strongly affected by the 26% yield of the Hunsdiecker decarboxylation Table 1 The yields (in %) for various intermediates and target compounds described in Schemes 2 and 3 compared to the literature (n.s.: not specified, mw.: microwave irradiation and l. The chiral acetic acids (5a-d) were obtained via the unilateral decarboxylation of the halogenated diethylmalonates (10a-c). For 5b and 5c the decarboxylation was carried out without NaOD to obtain Z ¼ H.

Scheme 4
The synthesis pathway of this work to obtain 7 from 8 over a 9 step route. The steps from 8 to 10c are a branch of Scheme 2.
from 11c to 5d), it turned out that the 11c to 5d step formed the reactant 10c in a byreaction, therefore, 10c could be reused again.

General conditions
A ow or an atmosphere of dry nitrogen was used with a Schlenk-apparatus for all water-and/or air-sensitive reactions.
For the same purpose, the glassware was dried with a heat gun. Quantitative and qualitative characterization was performed using GC-MS and NMR (500 MHz). For the gas chromatography with mass spectrometry an InertCap 5MS/NP capillary column was used and the signals were recorded as the total ion current of the ions between m/z 50 and 500. The temperature method started as isothermal at 40 C for 1 minute and was increased to 150 C by 15 C per minute. At a rate of 40 C per minute, the oven was heated to 300 C and was kept at this temperature for 1 minute. This resulted in a method length of 11 minutes and 45 seconds. 1 H, 13 C and 19 F NMR were recorded in parts per million and the data were reported as chemical shis (d) in parts per million (ppm), multiplicity, coupling constants (J) in Hz and integrals. The 13 C NMR spectra were 1 H decoupled.
Commercial reagents were used with the given purity of the supplier. The purication of the intermediates was done using column chromatography if necessary. The given yields are the 1 H NMR purity corrected values. For comparison we list the 1 H NMR and GC-MS purities.

Diethyl chloromalonate 9b
The reaction apparatus was heated under vacuum and ushed with nitrogen before usage. Diethyl bromomalonate 9a (6.9 g, 98%, 28.2 mmol) was dissolved in dimethylsulfoxide (25 ml, anhydrous). N-Chlorosuccinimide (6.7 g, 99%, 49.0 mmol) was added and the mixture was stirred at room temperature overnight. Aer the addition of saturated ammonium chloride solution (20 ml) and water (50 ml), the mixture was extracted with ethyl acetate (3 Â 20 ml). The combined organic layers were washed with saturated NaCl solution (3 Â 20 ml

Diethyl bromochloromalonate 10a via bromination
For the preparation of the sodium hypobromite solution, sodium hydroxide (3.0 g, 85%, 63.8 mmol) was dissolved in water (50 ml) and cooled to 0 C. Bromine (3.3 g, 99%, 20.3 mmol) was added to the ice cold mixture and stirred for 30 min. In a separate ask diethyl chloromalonate 9b (3.4 g, 99%, 17.3 mmol) was dissolved in acetone (10 ml) and acetic acid (5 ml) and cooled to 0 C. The prepared sodium hypobromite solution was added dropwise to 9b. The reaction mixture was stirred at RT overnight. The reaction mixture was added to saturated NaHCO 3 solution (20 ml) and extracted with dichloromethane (3 Â 25 ml). The combined organic layers were washed with water (2 Â 20 ml) and with saturated NaHCO 3 solution (2 Â 20 ml), dried over Na 2 SO 4 and were then completely evaporated. The total yield of the light brown liquid was 4.8 g (99.9%) with a purity of 98.5% according to 1 1F). GC-MS t R 6.68 min.
H 2 SO 4 -solution (20 ml, 10%) was added and the reaction mixture was extracted with diethyl ether (3 Â 20 ml). The combined organic layers were dried over Na 2 SO 4 and evaporated to dryness. The yield of the light brown liquid was 0.75 g (75.7%) with a purity of 95.1% according to 1 H NMR (95.5% GC-MS). The degree of deuteration was $89% and it was determined using the abundance ratio of m/z 172 and 173. 1

Ethyl bromochlorouoroacetate 4
Bromine (3.8 g, 99%, 23.7 mmol) was dissolved in CCl 4 (40 ml). At 0 C, silver monoethyl chlorouoromalonate 11c (6.0 g, 98%, 20.2 mmol) was added portion-wise (about 1.5 g per hour) while the reaction mixture was prevented from humidifying. Aer 1.5 h a continuous evolution of CO 2 started. When all of the reactant was added the mixture it was allowed to warm to room temperature and was then held at that temperature overnight. The precipitate was ltered and washed with CCl 4 (20 ml). Bromine was removed with NaHSO 3 solution (8%, 3 Â 10 ml) and washed with saturated NaHCO 3 (3 Â 10 ml). Aer drying with Na 2 SO 4 the solvent was completely evaporated. The yield of the light brown liquid was 1.39 g (26.1%) with a purity of 82.6% according to 1

Bromochlorouoromethane 7
Bromochlorouoroacetic acid 5d (9.0 g, 74%, 34.7 mmol) was dissolved in CHCl 3 (30 ml). Strychnin (11.6 g, 98%, 34.0 mmol) was dissolved in CHCl 3 (80 ml) and added to the acid at 0 C over 30 minutes. Aer 1 h the CHCl 3 was evaporated and the strychnin salt was obtained quantitatively as a brownish solid. A portion of the strychnine bromochlorouoroacetate (5.2 g, 99%, 8.76 mmol) was suspended in ethylene glycol (10 ml). The apparatus, which consisted of an Anschuetz head, Liebig condenser, collecting ask and cooling trap, was heated and ushed with nitrogen. The collecting ask and cooling trap were cooled with a mixture of acetone and dry ice to À78 C. The suspension was heated to 80 C and slowly increased to 100 C as the gas formation also increased. Aer 45 min the gas formation stopped and the suspension was stirred at 140 C for 1 h. The yield of the colorless liquid was 0.6 g (46.2%) with a purity of 95.2% according to 1

Conclusion
In conclusion, we have presented an efficient and versatile pathway to obtain a large variety of chiral haloacetic acids from a common reagent which is easily commercially available.
Further decarboxylation provides, in principle, access to the corresponding halomethanes. Currently, this has been explored only for the trihalogenated chiral acetic acids with success. Furthermore, the ability of obtaining the different haloacetic acids from a single reagent is another advantage. Our reagent 8 is also more easy to use compared to a differently halogenated ethene. The drawbacks of halogenated ethenes are clearly the low boiling points and their decreasing availability due to the CFC (chlorouorocarbon) ban. CHBrClF 7 has been obtained starting from 8 by exploring the full reaction pathway (Scheme 4) described here. Chiral acetic acid has been obtained as a racemic mixture, however, it can be derivatized for enantioseparation (e.g. by preparative gas-chromatography 55 or by diastereomeric resolution 19 ) along the reaction pathways shown in Schemes 3 and 4. This will then also remove minor impurities which may be present in the target compounds.

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