Production of 2,2,3,3,4,4,4-Heptafluorobutyl Acetate from Acetic Acid and 2,2,3,3,4,4,4-Heptafluorobutan-1-ol by Batch Reactive Distillation

Abstract

In the present study, a process for the production of 2,2,3,3,4,4,4-heptafluorobutyl acetate (HFBAc) is proposed for the first time. The production process of HFBAc from acetic acid (AAc) and 2,2,3,3,4,4,4-heptafluorobutan-1-ol (HFBol) was carried out at laboratory scale using batch reactive distillation (BRD). The process was conducted at atmospheric pressure in the presence of an acid catalyst, with an excess of AAc relative to HFBol (initial molar ratio of reagents HFBol/AAc is 45/55). During the BRD, the aqueous phase of the distillate was withdrawn from the system, while the organic phase of the distillate was returned as reflux. Since part of AAc is lost along with the aqueous phase of the distillate, a minor excess of AAc is reasonable for maximizing the conversion of the most expensive reagent—HFBol. The losses of AAc and HFBol with the aqueous phase of the distillate were less than 2 mole % and less than 0.5 mole % of the feed, respectively. The purity of HFBAc after BRD was 97.9 wt. %, and the conversion of HFBol exceeded 99 mole % of the feed. The purity of certain product fractions of HFBAc was greater than 99.6 wt. %. The obtained data can be used for industrial technology development to obtain HFBAc.

1. Introduction

Organofluorine compounds possess a wide range of unique properties and are applied in various industries, generating significant interest among researchers [1]. A key aspect of the organofluorine compounds industry is the search for and design of new synthesis methods and production technologies, particularly due to the high cost of such products. Meanwhile, organofluorine substances, including organofluoric esters, remain much less studied than their hydrocarbon analogues.

This study specifically targets the process for the production of 2,2,3,3,4,4,4-heptafluorobutyl acetate (HFBAc). HFBAc (CAS No. 356-06-9) is used in pharmaceutical aerosol compositions to reduce particle adhesion to can walls [2] and in the production of radiation-sensitive resin compositions [3]. As poly(heptafluorobutyl acetate), it acts as a copolymer for anti-corrosive or antibacterial surface treatments [4]. HFBAc also functions as a component of the nonaqueous electrolyte containing fluorine for lithium secondary batteries [5], among others [6,7,8,9,10,11,12].

Currently, HFBAc is an expensive and difficult-to-access reagent, and its cost is significantly higher than that of its potential precursors due to the low level of scientific development of HFBAc technology. Given the increased consumption of HFBAc resulting from the extension of its application areas, including fine organic synthesis, electrochemistry, and electrical engineering, as well as the demand for HFBAc following the advent of new materials based on this compound, there is a necessity to develop an effective method for its industrial production and purification.

The conventional method for the synthesis of esters is based on the esterification reaction of carboxylic acids with alcohols [13]. Another common method is the transesterification of an ester with an alcohol. Since these reactions are reversible, an ester purification step from the reactive mixture is mandatory to obtain a commercial-grade product. As a result, such techniques for ester production are characterized by a large number of process stages and apparatuses, require the use of secondary reagents and the recycling of streams, lead to the formation of waste products (including side products), and are represented by a relatively low conversion per process stage and potentially low yields of the target product.

The synthesis of fluorinated esters is discussed in [14,15,16,17,18,19,20,21,22,23,24,25,26,27]. Of note are the studies [19,20,27], where reactive distillation (RD) is used for ester production. By integrating the reaction and distillation (rectification) in one multifunctional apparatus, RD can effectively overcome a number of problems arising in the conventional technologies, such as conversion limitations for equilibrium chemical reactions, thermodynamic constraints on separation in the form of azeotropes, reduced yields resulting from side reactions, etc. [19,20,28]. In recent research, diverse methods and approaches have been developed that allow the efficient application of RD for solving a wide range of problems, including challenging tasks when the reaction products are medium volatile components [29,30,31,32,33,34,35,36,37]. Such situations are common in fatty acid production processes, particularly when the limiting product composition is an azeotropic mixture [30]. To address this issue, the following techniques are used: the reaction area location in the bottom section of the column, various levels of reagent feeding, external recycling of the bottom product stream, and side withdrawals. Such engineering approaches are discussed in [29,38].

Considering the high efficiency of RD as a method of intensification for processes with thermodynamic constraints and physico-chemical limitations, RD is examined as a potentially effective method for HFBAc production in the present study. Two chemical reactions that possibly underlie the RD process were previously addressed: the transesterification of isopropyl acetate with 2,2,3,3,4,4,4-heptafluorobutan-1-ol (HFBol) [1,39] and the esterification of acetic acid (AAc) and HFBol [40]. Both reactions are catalytic.

During transesterification, side reactions occur, which result in a reduced yield of the target product and complicated HFBAc purification from the reaction mixture, including those in the RD process [1,39].

In the case of esterification of AAc and HFBol (Equation (1)), there are no side reactions (the products are HFBAc and water) [40]. The product yield for the esterification reaction is limited by the chemical equilibrium constant. At $x_{H^{+}}$ = 2 mole % the “equilibrium” conversion of HFBol for the initial molar ratio of HFBol to AAc is as follows: 1 to 9—about 56%, 1 to 3—about 38%, 1 to 2—about 35%, and 1 to 1—about 30% [40]. Thermodynamic constraints on the distillation (rectification) separation of the HFBol–AAc–HFBAc–water reaction system include the presence of binary azeotropes and a three-component heteroazeotrope. The RD process was not examined for this system.

$$C_3{F}_{7}C{H}_{2}OH+C{H}_{3}COOH\leftrightarrow {CH}_{3}COOC{H_{2}C}_3{F}_7+H_{2}O$$

(1)

Thus, the purpose of the present investigation is to integrate the esterification reaction of AAc and HFBol and distillation (rectification) of the reaction mixture in one apparatus to carry out the reactive distillation and to evaluate the prospects of such a process for HFBAc production.

2. Materials and Methods

In this study, for the desired process AAc and HFBol were used as reagents. AAc was used as-is. HFBol was sourced as a reaction mass (component content from 0.60 to 0.90 mass fr.) and was purified in the laboratory. The process was examined in terms of heterogeneous catalysis, with Amberlyst 35 WET (Rohm And Haas France S.A.S., Valbonne, France) serving as the acid catalyst. The use of a solid-phase catalyst is advantageous because it simplifies the separation of the bottom product from the catalyst at the end of the experiment and more closely resembles industrial conditions [1]. Samples of the reaction mixture were quantitatively analyzed using nuclear magnetic resonance (NMR) according to the methodology presented in previous studies [1,39,40]. A Bruker Avance II—300 MHz NMR spectrometer (Bruker Corp., Billerica, MA, USA) was employed to obtain ¹H and ¹⁹F spectra of the samples at frequencies of 300.211 and 282.499 MHz, respectively, using an internal deuterium lock. The relaxation delay is 0.4 s, the pre-scan delay is 8 µs, the dwell time is 1.2 µs, the acquisition time is 0.078 s, and the high power pulse is 4 µs. Tetramethylsilane and trichlorofluoromethane were used as external references, while dimethyl sulfoxide-d6 (DMSO-d6) served as the solvent. Information on the compounds used in this work is presented in

Table 1. Specifications of the compounds used.

Chemical Name	CAS-No	Molar Mass M/g·Mole−1	Supplier	Mass Fraction Purity	Purification in Laboratory	Mass Fraction After Purification (GC 1)
HFBol	375-01-9	200.05	P&M Invest (Moscow, Russia)	0.60–0.90	Heteroazeotropic distillation; distillation	≥0.998
AAc	64-19-7	60.05	EKOS-1 (Moscow, Russia)	0.99	none	-
DMSO-d6	2206-27-1	84.17	Solvex-D (Moscow, Russia)	0.998 atom % D	none	-
Amberlyst 35 WET by Rohm and Haas France S.A.S.; ionic form supplied: H+; weight TOT capacity (H+): >5.2 eq·kg−1

¹ Gas chromatography—flame ionization detector (Agilent 6890N equipped with a Restek RTX-1701 RK12054 capillary column; Agilent Technologies, Inc., Wilmington, DE, USA).

.

In the present study, batch reactive distillation (BRD) was carried out for the production of HFBAc. The BRD was conducted using the distillation column illustrated in Figure 1. This equipment was used by us previously for conducting BRD in the process of transesterification of HFBol with propyl acetate [1]. The column is packed with Fenske helices, 2.5 mm, made of glass. The height of the packed bed and the diameter of the column are 29.0 cm and 2.5 cm, respectively. The mass transfer efficiency is equivalent to 7.5 theoretical separation stages. The reactive area is located at the bottom, Section 1, of the column. Temperature control is maintained through thermometers 2 and 5. Units 10 and 12 are designated for sampling the distillate ($y_i$) and bottom ($x_i$) product, respectively. The collected distillate can be returned to the system via condenser 6. To determine the composition ($x_i$; $y_i$) of the mixture, the samples were analyzed using ¹⁹F and ¹H NMR. The NMR spectra of HFBol and HFBAc were published in our previous paper [39].

The measuring equipment in the study is the same as that used in [1]. The uncertainties (u) in the quantitative analysis of the mixture (u(x; y)) were 0.005 (¹H) and 0.001 (¹⁹F) mole fr. Pressure (P) was measured using a VACUU•VIE extended vacuum meter (VACUUBRAND GMBH + CO KG, Wertheim, Germany) with a standard uncertainty of u(P) = 0.3 kPa. Temperature (T) was measured with mercury thermometers from Thermopribor OJSC (Moscow, Russia): 0···40−110 °C with a standard uncertainty of u(T) = 0.3 °C for the distillate and 0−150 °C with a standard uncertainty of u(T) = 1.0 °C for the bottom. The sample weight (m) was measured with a Mass Comparator MC-1000 (A&D Company Ltd., Tokyo, Japan), which has a standard uncertainty of u(m) = 0.0005 g.

The BRD was operated in two modes similar to article [1]: total reflux mode and fractionation mode on the distillation column illustrated in Figure 1. When BRD is operating in total reflux mode, an aliquot of the distillate lower (organic) phase is sampled from the system through unit 10. In fractionation mode, batch sampling of the distillate is performed by collecting fractions equal in volume to the reflux holdup volume through unit 10. If a second phase is present, the selected distillate is decanted and unit 10 is used as a separatory funnel. To calculate the sampling ratio (SR), the mass of the collected fraction is weighed before taking an aliquot sample.

During the reaction, the number of moles of the substance remains constant (Equation (1)). Thus, the sampling ratio (SR) is determined by Equation (2):

$$SR={\displaystyle \frac{n}{N^0}}$$

(2)

where $N^0$ is the quantity of substance loaded, mole; $n={\displaystyle \frac{m}{\sum x_i{M}_i}}$ is the amount of substance sampled as the fraction, mole; m is the mass of the substance sampled as the fraction, g; $x_i$ is the concentration of component $i$ in the fraction, mole fr.; and $M_i$ is the molar mass of component $i$, g·mole⁻¹.

For the BRD process, the Damköhler numbers ($D_a$) were calculated (Equation (3)). The parameter is the dimensionless number used in chemical engineering to relate the mass exchange rate to the chemical reaction rate.

$$\begin{array}{c}{D}_a={\displaystyle \frac{R_{Mass-exchange}}{R_{chemistry}}}\\ R_{Mass-exchange}={\displaystyle \frac{n_{water}}{\Delta \tau }}\\ R_{chemistry}={\displaystyle \frac{\Delta n_{HFBol}}{\Delta \tau }}\end{array}$$

(3)

where $R_{Mass-exchange}$ is the rate of mass exchange, which is the same as the rate of water withdrawal from the system, mole·min⁻¹; $n_{water}={\displaystyle \frac{m{x}_{water}}{\sum x_i{M}_i}}$ is the amount of water sampled as the fraction, mole; $\Delta \tau$ is the accumulation time of the fraction, min; $R_{chemistry}$ is the rate of water formation, mole·min⁻¹; $M_i$ is the molar mass of component $i$, g·mole⁻¹; m is the mass of the substance sampled as the fraction, g; $x_i$ is the concentration of component $i$ in the fraction, mole fr.; $\Delta n_{HFBol}$ is the change in the amount of HFBol in the system during the accumulation time of the fraction, mole; and $\Delta n_{HFBol}$ is calculated from the material balance of the system, the quantitative analysis of the bottom composition, and gravimetric and quantitative analyses of the distillate phases.

3. Experimental Results and Discussion

3.1. Total Reflux Mode

The BRD process was conducted at atmospheric pressure in total reflux mode. A binary mixture of HFBol and AAc, with the acid catalyst injected into the bottom of the column (Figure 1), was heated to bring the column to operating mode. The operating mode of the column implies the presence of counter-flow of vapor and liquid, formed at the bottom during heating and flowing from the condenser, respectively. Thus, the moment when the column reaches operating mode corresponds to the appearance of reflux (τ = 0). The feed composition is shown in

Table 2. Feed composition.

Component	${\mathit{m}}_{\mathit{i}}$ /g	${\mathit{N}}_{\mathit{i}}$ /Mole	${\mathit{x}}_{\mathit{i}}$ /Mole fr.
AAc	113.0040	1.8818	0.5498
HFBol	308.3300	1.5412	0.4502
Amberlist 35 WET	6.6685	0.0347 *	0.0101 *

* Mole equivalent (mole fr.) of H⁺; u(m) = 0.0005 g.

, and the experimental results are presented in

Table 3. Dependence of bottom temperature ($T_x$), distillate temperature ($T_y$), bottom composition ($x_i$), and composition of lower (organic) phase of distillate ($y_i$) on time (τ) of BRD in total reflux mode at atmospheric pressure P = 101.2 kPa.

τ/min	${\mathit{T}}_{\mathit{x}}$/°C	${\mathit{T}}_{\mathit{y}}$/°C	${\mathit{x}}_{\mathit{i}}$/Mole fr.				${\mathit{y}}_{\mathit{i}}$/Mole fr. (Lower/Organic Phase of Distillate)
			AAc	HFBol	HFBAc	H2O	AAc	HFBol	HFBAc	H2O
Feed			0.5290	0.4710	0	0
0	110.0	100.0	0.6147	0.3788	0.0065	0	0.0100	0.4279	0.0042	0.5579
30	111.0	91.5	0.5960	0.3757	0.0283	0	0.0200	0.4964	0.0054	0.4782
60	110.8	87.5	0.5965	0.3510	0.0525	0	0.0188	0.4560	0.0058	0.5194
120	109.5	87.0	0.5735	0.3412	0.0853	0	0.0198	0.4871	0.0077	0.4854
180	108.8	86.0	0.5398	0.3490	0.1112	0	0.0050	0.4213	0.1733	0.4004

Catalyst—Amberlyst 35 WET; ¹⁹F and ¹H NMR data; the beginning of the experiment (τ = 0)—the appearance of reflux (appeared 30 min after the heat was applied to the bottom); u($T_x$) = 1.0 °C; u($T_y$) = 0.3 °C; u(x, y) = 0.005 mole fr.; u($H^{+}$) = 0.001 mole fr.; u(P) = 0.3 kPa.

and Figure 2.

According to the data presented in Figure 2, the rate of the esterification reaction (Equation (1)) during the BRD process is quite fast. At the moment of reflux appearance (τ = 0 min), the distillate is already heterogeneous and a significant amount of water is present in the lower (organic) phase of the distillate. At the same time, there is no water in the bottom of the column. This may indicate that the rate of mass exchange within the column is much higher than the rate of the reaction. Such a situation favors a shift in chemical equilibrium toward the products, further increasing the reaction rate. This potentially means that the product (the aqueous phase of the distillate) can be withdrawn from the system almost immediately after the column enters operating mode. It should be noted that in this case, the question is the withdrawal of the upper (aqueous) phase of the distillate from the system. The lower (organic) phase of the distillate should be returned to the column as reflux, as it contains reagents and the target product. The analysis of the temperature profile of the column, specifically the change in distillate temperature during the process, is also of interest. The temperature of the distillate at τ = 0 min does not correspond to the temperature of any of the heteroazeotropes of the system; the latter represents the limiting product composition in this case. This means that at the beginning of the experiment (τ = 0 min), the concentration profile of the column has not yet been established. Only after 60 min did the distillate temperature reach 87.5 °C, which is close to the boiling point of the binary heteroazeotrope HFBol–water [1]. It should also be noted that the presence of AAc impurity indicates that the efficiency of the distillation column, with its 7.5 theoretical separation stages, is insufficient to achieve the limiting product compositions. At 180 min into the experiment, HFBAc appeared in the distillate, and the distillate temperature dropped from 87.5 °C to 86 °C. This event indicates that the composition of the distillate shifts toward the ternary heteroazeotpe HFBol–HFBAc–water, which is the lowest boiling point in the system. Thus, the distillate temperature can serve us as a guide for the start of the withdrawal.

3.2. Fractionation Mode

After completing the experiment in total reflux mode, the column containing the reaction mixture was left overnight. The following day, the BRD process was conducted at atmospheric pressure in fractionation mode. A heat load was applied, and the column was brought to operating mode. The operating mode of the column indicated the presence of counter-flow of vapor and liquid, formed in the bottom during and flowing from the condenser, respectively. The moment when the column reaches operating mode corresponds to the appearance of reflux (τ = 0), and batch sampling of the distillate is performed by collecting fractions equal in volume to the reflux holdup volume through unit 10 (Figure 1). The selected distillate is decanted, and the upper (aqueous) phase of the distillate is removed from the system, while the lower (organic) phase is returned to the distillation column as reflux. The BRD process is carried out until the vapor condensate becomes homogeneous. The resulting product is then separated from the catalyst by distillation. The experimental results are presented in

Table 4. Dependence of bottom temperature ($T_x$), distillate temperature ($T_y$), bottom composition ($x_i$), and composition of distillate ($y_i$) on sampling ratio (SR) and time (τ) of BRD in fractionation mode at atmospheric pressure P = 100.8 kPa.

τ/min	$\mathit{S}\mathit{R}$	${\mathit{T}}_{\mathit{x}}$/°C	${\mathit{T}}_{\mathit{y}}$/°C	${\mathit{x}}_{\mathit{i}}$/Mole fr.				${\mathit{y}}_{\mathit{i}}$/Mole fr. (Lower/Organic Phase)				${\mathit{y}}_{\mathit{i}}$/Mole fr. (Upper/Aqueous Phase)
				Aac	HFBol	HFBAc	H2O	Aac	HFBol	HFBAc	H2O	Aac	HFBol	HFBAc	H2O
	Heterogeneous distillate
50	0.0300	106.5	85.3	0.4721	0.2880	0.1536	0.0863	0.0012	0.2748	0.3975	0.3265	0.0005	0.0010	0.0010	0.9975
107	0.0567	107.3	85.1	0.4610	0.2891	0.1766	0.0733	0.0010	0.2710	0.4048	0.3232	0.0004	0.0003	0.0006	0.9987
171	0.0789	108.2	85.1	0.4470	0.2460	0.2155	0.0915	0.0007	0.2221	0.4950	0.2822	0.0006	0.0002	0.0005	0.9987
228	0.1320	108.7	85.2	0.4108	0.2624	0.2412	0.0856	0.0010	0.2149	0.4962	0.2879	0.0008	0	0.0003	0.9989
275	0.2572	109.1	85.2	0.3912	0.2178	0.2260	0.1650	0.0024	0.2120	0.4748	0.3108	0.0116	0.0001	0.0001	0.9882
384	0.2802	108.8	91.0	0.4066	0.1915	0.4019	0	0.0362	0.1934	0.7704	0.0000	0.0255	0.0003	0.0006	0.9736
464	0.3154	108.0	88.5	0.2884	0.1282	0.3358	0.2476	0.0312	0.1112	0.5410	0.3166	0.0266	0.0001	0.0003	0.9730
526	0.3395	107.5	95.5	0.3027	0.1137	0.3399	0.2437	0.0313	0.0893	0.4287	0.4507	0.0427	0.0001	0.0001	0.9571
586	0.3676	108.5	96.5	0.2551	0.1089	0.3910	0.2450	0.0453	0.0862	0.4919	0.3766	0.0584	0.0001	0.0001	0.9414
709	0.4040	108.0	93.0	0.1856	0.0789	0.3983	0.3372	0.0460	0.0620	0.5354	0.3566	0.0685	0.0002	0.0003	0.9310
773	0.4374	106.5	92.5	0.1607	0.0485	0.4485	0.3423	0.0588	0.0488	0.7061	0.1863	0.0856	0.0005	0.0004	0.9135
847	0.4689	108.5	102.5	0.2071	0.0391	0.5846	0.1692	0.1123	0.0345	0.6253	0.2279	0.1262	0.0007	0.0004	0.8727
974	0.4909	108.5	102.7	0.1633	0.0265	0.8102	0	0.1505	0.0118	0.5322	0.3055	0.1619	0.0007	0.0004	0.8370
1114	0.5014	108.5	104.5	0.1262	0.0087	0.7443	0.1208	0.1771	0.0060	0.5865	0.2304	0.1944	0.0009	0	0.8047
1228	0.5824	108.5	105.0	0.1110	0.0025	0.8812	0.0053	0.1878	0.0029	0.7555	0.0538	0.2013	0.0016	0	0.7971
	Homogeneous distillate
1351	0.6296	108.5	105.6	0.0964	0	0.8999	0.0037	0.1885	0.0000	0.7935	0.0180	-	-	-	-
1411	0.6862	108.5	106.4	0.0939	0	0.9030	0.0031	0.1753	0.0011	0.8168	0.0068	-	-	-	-
1481	0.7404	108.8	106.5	0.0801	0	0.9170	0.0029	0.1542	0.0008	0.8410	0.0040	-	-	-	-
1556	0.7945	109.0	106.7	0.0667	0	0.9304	0.0029	0.1487	0.0000	0.8483	0.0030	-	-	-	-
1626	0.8453	109.2	106.8	0.0464	0.0027	0.9481	0.0028	0.1181	0.0018	0.8770	0.0031	-	-	-	-
1702	0.8890	118.0	107.0	0.0279	0.0031	0.9659	0.0031	0.0953	0.0018	0.9008	0.0021	-	-	-	-
1762	0.9369	125.5	107.4	0.0121	0.0006	0.9851	0.0022	0.0557	0	0.9400	0.0043	-	-	-	-

Catalyst—Amberlyst 35 WET; ¹⁹F and ¹H NMR data; the beginning of the experiment (τ = 0)—the appearance of reflux (appeared 30 min after the heat was applied to the bottom); u(m) = 0.0005 g; u($T_x$) = 1.0 °C; u($T_y$) = 0.3 °C; u(x, y) = 0.005 mole fr.; u($H^{+}$) = 0.001 mole fr.; u(P) = 0.3 kPa.

and Figure 3.

According to Figure 3, during fractionation of the reaction mixture, HFBAc is present both in the bottom of the column and in the lower (organic) phase of the distillate. As the upper (aqueous) phase of the distillate is withdrawn from the system, the concentration of HFBAc increases, indicating that the BRD process proceeds favorably and that the proposed method can be successfully applied to the HFBAc production. As water is withdrawn from the system and HFBol is consumed in the reaction (Equation (1)), the distillation boundaries shift, causing the AAc concentration in the distillate to gradually increase. During vapor condensation and subsequent splitting of the distillate, the Aac has distributed between the aqueous and organic phases. This results in a partial loss of Aac with the upper (aqueous) phase of the distillate being withdrawn. Considering that the limiting product composition is a ternary or binary heteroazeotrope of water with ester and/or alcohol, the concentration profile of the upper (aqueous) phase of the distillate indicates either poor column efficiency or an extra excess amount of Aac in the feed stream. Increasing the number of separation stages and/or reducing the excess Aac relative to HFBol in the feed should help minimize Aac losses. Under the considered conditions the losses associated with the withdrawal of the upper (aqueous) phase of the distillate are as follows: for Aac less than 2% mole of the feed and for HFBol less than 0.5% mole of the feed. The targeted HFBAc is completely absent in the upper (aqueous) phase of the distillate, and the conversion of HFBol exceeds 99 mole % of the feed.

At SR > 0.5824, the distillate becomes homogeneous, and HFBol is virtually absent in the system (in trace amounts). There is also almost no water in the system. This indicates that the BRD process has been completed, and the distillation of the mixture in fractionation mode was proceeding. Thus, after the distillate has become homogeneous, catalyst removal can be achieved through standard procedures, such as distillation of the product without reflux or simple filtration. The analysis of the distillation process in fractionation mode (Figure 3, SR from 0.5824 to 0.9369) shows that HFBAc is concentrated at the bottom of the column. The maximum purity of HFBAc in the fractions (bottom of the column) is more than 99.6 wt. %. (98.51 mole % of HFBAc, 1.21 mole % Aac, and traces of water and HFBol).

At the end of all the steps, the resulting distillate and the bottom product, separated from the catalyst by filtration, were combined. The purity of the product (HFBAc) was 97.9 wt. %. (92 mole % of HFBAc, 8 mole % of Aac, and traces of water and HFBol).

For the BRD process, the Damköhler number was calculated. The results of the calculation are presented in Figure 4.

The Damköhler number during the BRD experiment increased from 1.5 to 2 (Figure 4). This indicates that the limiting stage of the BRD process under the considered conditions is the chemical stage. The concentration of the catalyst affects the chemical reaction rate and should be determined by the mass exchange rate and the apparatus capacity. Therefore, it is advisable to maintain the concentration of the catalyst in the reaction area in the amount of mole equivalent H⁺ so that the Damköhler number, which determines the ratio of the rate of water removal from the system to the rate of water formation by chemical reaction, is close to the value of 1 throughout the entire process.

4. Conclusions

As a result of this work, the process for the production of HFBAc from AAc and HFBol by reactive distillation was realized at a laboratory scale. The process is conducted in semi-continuous mode at atmospheric pressure and can potentially be modified for continuous operation. The reaction area is located at the bottom of the distillation column, and Amberlyst 35 WET served as the acid heterogeneous catalyst. The use of a solid-phase or liquid-phase catalyst should not significantly affect the process performance and depends on the design of the reactive distillation. The catalyst concentration is determined by the required chemistry rate and the necessary capacity of the apparatus.

During the BRD, the aqueous phase of the distillate was withdrawn from the system, while the organic phase of the distillate was returned as reflux. Since a portion of AAc is lost along with the aqueous phase of the distillate, a minor excess of AAc is reasonable to maximize the conversion of the most expensive reagent, HFBol. The losses of AAc and HFBol with the aqueous phase of the distillate were less than 2 mole % and less than 0.5 mole % of the feed, respectively. By combining the reaction and distillation processes in one apparatus, the chemical equilibrium of the AAc and HFBol esterification reaction was shifted toward the products, overcoming the thermodynamic constraints on the distillation separation. Overall, the BRD process proved to be quite effective for the production of HFBAc from AAc and HFBol. At an initial molar ratio of reagents HFBol/AAc = 45/55, the purity of HFBAc after BRD was 97.9 wt. %, and the conversion of HFBol exceeded 99 mole % of the feed. The purity of certain product fractions of HFBAc was more than 99.6 wt. %. To increase product purity, it seems promising to reduce the excess of AAc relative to HFBol. Thus, reactive distillation appears to be a promising method for the production of HFBAc.

Future research should focus on phase equilibria and conducting a thermodynamic-topological analysis [42,43] of the structure of the phase equilibrium diagram for the HFBol–AAc–HFBAc–water system, similar to the analysis performed by the authors [44].