EDTA-assisted hydrothermal synthesis of cubic SrF₂ particles and their catalytic performance for the pyrolysis of 1-chloro-1,1-difluoroethane to vinylidene fluoride

Abstract

Uniform, free-standing and cubic SrF₂ microparticles were successfully fabricated by a facile hydrothermal method with ethylenediaminetetraacetic acid (EDTA) as the chelating agent. The influences of preparation conditions, such as the pH value, amount of EDTA and hydrothermal time, on the formation of SrF₂ crystals were investigated. The formation mechanism of cubic SrF₂ particles was proposed based on the experimental results. Following calcination in air at 500 °C, SrF₂ particles were evaluated as the catalyst for the pyrolysis of 1-chloro-1,1-difluoroethane (HCFC-142b, CH₃CClF₂) to vinylidene fluoride (VDF, CH₂=CF₂) at 350 °C and a space velocity of 600 h⁻¹. The results indicate that SrF₂ cubes exhibit high catalytic activity with a HCFC-142b conversion of about 70% and a selectivity to VDF of 80–87%. No significant deactivation was observed within the time on stream of 30 h. With the reaction temperature increased to 450 °C, the conversion of HCFC-142b is close to 94%, while the selectivity to VDF remains almost unchanged. Although the SrF₂ catalyst prepared by the conventional precipitation method also shows high conversion, its selectivity to VDF is only around 50–70%. We suggest that the surface acidity and specific surface area play major roles in the catalytic performance. Compared with the temperatures for industrial manufacture of VDF of 650–700 °C, the SrF₂ catalysts provide a promising pathway to produce VDF at much lower temperatures.

---

Introduction

Alkaline earth fluorides are important photonic materials because of their exclusive and diverse features.^1,2 SrF₂ is one of the typical earth fluorides. Nanomaterials of SrF₂ with various morphologies and sizes have been successfully prepared by various methods, majorly for their luminescence. Doped with trivalent rare earth elements, such as Yb³⁺, Er³⁺, Tb³⁺ and so on, they function as satisfactory up-conversion luminescent phosphors.^3–7

In addition, homogeneous and monodispersed SrF₂ nanocrystals were obtained in a water, ethanol and oleic acid system.⁸ Thermal decomposition of trifluoroacetate in a mixture of oleic acid, oleylamine and 1-octadecene also leads to the formation of SrF₂ nanocrystals.⁹ The concentration of trifluoroacetate, decomposition temperature and decomposition time play major roles in the shapes of SrF₂ crystals (from one dimension to three dimensions) and their corresponding luminescence performance. Hydrothermal synthesis was confirmed to be an effective route for the preparation of hollow micro-spheres, solid microspheres and hexagonal flakes of SrF₂ and SrF₂:Ln³⁺ (Ln = Eu, Ce, Tb).¹⁰ In addition, in the presence of different surfactants or chelating agents, micro and cubic SrF₂ nanocrystals were achieved.¹¹ However, there are very few reports on the application of SrF₂ particles as catalysts, except as the promoter of catalysts for oxidative coupling of methane.^12–14

Vinylidene fluoride (VDF, CH₂=CF₂)^15,16 is one of the most important fluorinated alkenes and monomers and is applied in various fields, such as the feedstock for the production of polyvinylidene fluoride (PVDF).^17,18 In addition, it is also copolymerized with other monomers (such as hexafluoropropylene and vinylidene chloride) for the preparation of various co-polymers.^19,20 It is also the key component of fluorine-containing resins and fluorine-containing rubbers.^21–23 Due to its excellent performance,²⁴ the demand for VDF is expected to increase. At present, thermal dehydrochlorination of HCFC-142b (1-chloro-1,1-difluoroethane, CH₃CClF₂) is the main method to manufacture VDF in the PVDF industry, although other routes such as co-pyrolysis of CHF₃ and CH₄ were reported.^25–27 The pyrolysis of HCFC-142b takes place through reactions (1) and (2),^15,28 leading to the formation of VDF and VCF (CH₂=CClF) at elevated temperatures. Industrially, the pyrolysis temperatures are usually between 650 °C and 700 °C. A HCFC-142b conversion of about 80–100% and a selectivity to VDF of about 85–95% are achieved under optimized conditions. However, the production suffers from significant coke deposition in the tubular reactors and has to be shut down for the removal of coke periodically (usually every two weeks).

CH₃CClF₂(HCFC-142b) → CH₂=CF₂(VDF) + HCl (1)

CH₃CClF₂(HCFC-142b) → CH₂=CClF(VCF) + HF (2)

In addition, high reaction temperatures also lead to high energy consumption of the pyrolysis process. Furthermore, elevated temperatures usually result in significant amounts of by-products. Therefore, a catalyst is highly desired and it is an effective and environmentally friendly route for the pyrolysis of HCFC-142b and production of VDF. However, following dehydrochlorination, HCl is formed as a major by-product which poses a significant challenge for the survival of catalysts. To survive in a highly corrosive HCl atmosphere, the use of metal chlorides, metal fluorides, metal oxides, activated carbon (AC), and metal salts impregnated on activated carbon as catalysts was attempted. Herbert et al.²⁹ adopted NiCl₂ as the catalyst for HCFC-142b dehydrochlorination, achieving 80% conversion of HCFC-142b and 100% selectivity to VDF. R. Geetha et al.³⁰ reported that BaCl₂/AC catalyst suppresses side reactions significantly on HCFC-142b dehydrochlorination. Wang et al.³¹ suggested metal chlorides (FeCl₃, CuCl₂, and NiCl₂) impregnated on activated carbon as catalysts, achieving 10–90% conversion of HCFC-142b and 60–90% selectivity to VDF at 400–630 °C. Unfortunately, the stability of the catalyst is far from industrial application. In our previous work,³² we found that N-doped carbon materials (NAC) are efficient catalysts in dehydrochlorination of HCFC-142b. Unfortunately, the NAC catalysts are difficult to be recovered after deactivation.

In the present work, uniform, free-standing and cubic SrF₂ particles were prepared by a hydrothermal method with ethylenediaminetetraacetic acid (EDTA) as the chelating agent. The effect of preparation conditions on the formation of SrF₂ cubic particles was investigated. Following calcination in air at 500 °C, SrF₂ cubic particles were evaluated as the catalyst for pyrolysis of HCFC-142b for the first time. For comparison, a SrF₂ catalyst prepared by the precipitation method was also included. SrF₂ prepared by the hydrothermal method presents satisfactory performance for the pyrolysis of HCFC-142b. Hence, it provides a promising pathway to produce VDF through a low-temperature and energy-saving process (catalytic dehydrochlorination of HCFC-142b).

Experimental

Reagents and materials

Strontium chloride hexahydrate (≥99.5%), ethylenediaminetetraacetic acid, potassium hydroxide (≥95%), ammonium fluoroborate (≥99.5%) and ammonium fluoride (≥98%) were purchased from Aladdin Industrial Corporation (Shanghai, China). Ethanol (≥99.5%) and calcium chloride anhydrous were obtained from Sinopharm Chemical Reagent Co. LTD in China. HCFC-142b (99.8%), from Juhua Group Corporation (Quzhou, China), was adopted as the feed gas without further purification.

Preparation of SrF₂ cubic particles

Hydrothermal method. A series of SrF₂ samples were prepared by a facile hydrothermal method. Stoichiometric amounts of EDTA and NH₄BF₄ were applied as raw materials without further purification. In a typical process, desired amounts of SrCl₂·6H₂O and the chelating agent, EDTA, were first dissolved in deionized water under vigorous stirring for 0.5 h to obtain solution A. Then, the fluorine source, NH₄BF₄, was added into solution A with constant stirring for a few minutes to form a transparent solution B. Subsequently, solution B was transferred into a 50 mL Teflon-lined autoclave, which was sealed and maintained at 160 °C for 8–48 h. Following cooling down to room temperature, the obtained emulsion mixture was centrifuged. The precipitates were washed and centrifuged three times with a mixed solution of ethanol and water and then dried at 80 °C for 12 h. The sample is denoted as SrF₂-H.

Precipitation method. For comparison, SrF₂ was also synthesized by precipitation. 0.1 mol SrCl₂·6H₂O was dissolved in 200 mL deionized water and stoichiometric amounts of solid NH₄F were introduced stepwise. Following vigorous stirring for 2 h, the solution was filtered using a Buchner funnel with a vacuum pump. Then the obtained solid was dried at 80 °C for 12 h. The sample is denoted as SrF₂-P.

Catalytic activity evaluation

The performance of the catalysts was evaluated using a fixed-bed reactor (nickel tube with an i.d. of 22 mm). The system was first purged with nitrogen to remove water vapour and air before each evaluation experiment at reaction temperatures. Then the gas phase of HCFC-142b with a GHSV (gas hourly space velocity) of 600 h⁻¹, mixed with equivalent flow of nitrogen, was passed through the reactor. The effluent gas from the reactor passed into a scrubber containing about 1 M KOH solution (450 mL) to remove the acid gases of HCl and HF, followed by composition analysis with a Jie Dao GC-1690 gas chromatograph equipped with a thermal conductivity detector (TCD).

Catalyst characterization

For morphology investigation, SEM (scanning electron microscopy) images were obtained from a FESEM, Hitachi S-4700 at an accelerating voltage of 15 kV, equipped with an X-ray energy spectrometer (EDS). The X-ray diffraction (XRD) patterns of the catalysts were recorded on a Kratos AXIS Ultra DLD analytical instrument. A monochromatic Al K radiation source (1486.6 eV) with an analyzer pass energy of 80 eV was operated at 3 mA and 15 kV. X-ray photoelectron spectroscopy (XPS) measurements of the catalysts were performed using a Thermo ESCALAB 250XI, with monochromatised Al Kα X-ray as the excitation source (24.2 W), with an analyser pass energy of 187.85 eV for survey scans and 46.95 eV for detailed elemental scans. In order to subtract the surface charging effect, binding energies were referenced to C1s binding energy of carbon, taken to be 284.6 eV. The XPS spectra were analysed by the XPS peak software. TEM (transmission electron microscopy) was adopted to further explore the microstructure of the catalysts using a JEOL 2100F transmission electron microscope at an accelerating voltage of 200 kV. The surface area and total pore volume of the catalysts were measured by N₂ adsorption–desorption at −196 °C with a Quantachrome autosorb automated gas sorption system from USA. Prior to the measurement, the catalyst samples were degassed at 200 °C for 6 h under vacuum. Thermogravimetric analysis (TGA) was performed on a TA TGA Q500 instrument. The samples were heated in platinum crucibles up to 800 °C from room temperature with a heating rate of 10 °C min⁻¹ in an atmosphere of 30 mL min⁻¹ air. NH₃-TPD was carried out in a self-made instrument and a thermal conductivity detector (TCD) was used for detecting the desorption of NH₃.

Results and discussion

Synthesis, morphology, and structure of SrF₂ particles prepared by the hydrothermal method

Effect of the pH value. SrF₂ particles were first prepared under hydrothermal conditions with pH values of 5, 7, 9 and 11, respectively. During the preparation procedure, the hydrothermal temperature was kept at 160 °C for 24 and EDTA was used as the chelating reagent (EDTA : Sr²⁺ = 1 : 1). The phase structures of synthesized SrF₂ samples by the hydrothermal route were examined by XRD. Fig. 1 presents their XRD patterns.

Fig. 1 XRD patterns of the as-prepared SrF₂ samples with EDTA as the chelating agent (EDTA : Sr²⁺ = 1 : 1) at 160 °C for 24 h. During the hydrothermal synthesis, pH values were controlled at (1) pH = 5, (2) pH = 7, (3) pH = 9, and (4) pH = 11, respectively. The standard profile of SrF₂ (PDF# 88-2294) was included as a reference.

As indicated in Fig. 1, the XRD patterns of SrF₂ obtained from hydrothermal synthesis agree well with that of the standard SrF₂ pattern (PDF# 88-2294, space group: Fm3m (225)), indicating that the samples obtained at different pH values possess the cubic phase of the SrF₂ crystal. No other impurities were identified. The strong, sharp diffraction peaks imply that the as-obtained samples are highly crystallized. In addition, with the variation of the pH value, nearly the same intensities based on the crystal diffraction peaks of (111), (220) and (311) of all the SrF₂ samples indicate that the samples are composed of the same primary particles.

Although no significant difference was observed from the XRD patterns, the pH value of the solution during hydrothermal synthesis plays an important role in the size and the morphology of the synthesized samples. The pH dependent morphology evolution of SrF₂ was investigated by SEM as illustrated in Fig. 2. With the pH value of 5, the SrF₂ sample is mainly composed of aggregated particles (Fig. 2a). The cubic particles start to form when the pH value of the hydrothermal solution is higher than 7 (Fig. 2b). With the further increase of the pH value to 9, regular and well-defined cubic particles with an average size of 133.75 nm are obtained (Fig. 2c and e₄). As revealed by the HRTEM images in Fig. 2e₁ and e₂, clear and solid SrF₂ crystals are identified. As displayed in Fig. 2e₃, the inter-planar distances between the adjacent lattice fringes are determined to be 0.335 and 0.290 nm, respectively, which are indexed to the d-spacing values of the (111) and (100) planes in the cubic SrF₂ crystal. Clearly, more facets of (111) and (100) are exposed over these samples. The corresponding selected area electron diffraction (SAED) pattern suggests the poly-crystalline nature of the SrF₂ cubic particles, indicating that the particles are not composed of individual crystals (Fig. 2e₃). The diffraction dots can be indexed to (220), (311) and (420), corresponding to the (110), (311) and (210) planes of the SrF₂ crystal, respectively. It is also proved by TEM images (Fig. 2e₁ and e₂) that cubic SrF₂ is actually a multi-layer structure. The chemistry involved in the SrF₂ formation process is illustrated in reactions (3)–(7).^33,34 As demonstrated in reaction (3), the hydrolysis of BF₄⁻ releases F⁻ which then combines with Sr²⁺ leading to the formation of SrF₂ (reaction 4). According to the SEM images, when the pH value is lower than 9, aggregates of cubic particles are obtained. In particular, with the pH value of 5, no cubic SrF₂ is derived. Clearly, in addition to the above two reactions, EDTA also plays an important role in the formation of SrF₂ particles. It is well accepted that EDTA (H₄Y for short) exhibits ionization equilibrium in the solution as expressed in reaction (5).³⁵ When the pH value is low, the equilibrium shifts toward left for reaction (5) and majorly it is in the form of H₄Y, which is only slightly soluble in water. Therefore, with a low pH value, EDTA shows weak chelating ability (reactions 6 and 7).³⁶ As displayed in reaction (3), the hydrolysis of BF₄⁻ releases H⁺ and it leads to the decrease in the pH value to around 3 after the hydrothermal process. Hence, low pH values are not favourable for the formation of cubic SrF₂ particles.

[BF₄]⁻ + 3H₂O → 4F⁻ + 3H⁺ + H₃BO₃ (3)

Sr²⁺ + 2F⁻ → SrF₂ (4)

H₄Y ⇋ [H_xY]^x−4 + (4 − x)H⁺ (5)

[H_xY]^x−4 + Sr²⁺ → [H_xY]^x−4 − Sr²⁺ (6)

[H_xY]^x−4 − Sr²⁺ + 2F⁻ → SrF₂ + [H_xY]^x−4 (7)

Fig. 2 SEM images of SrF₂ samples prepared at different pH values: (a) pH = 5, (b) pH = 7, (c) pH = 9 and (d) pH = 11; (e₁) and (e₂) TEM and inset showing the corresponding SAED pattern; (e₃) HRTEM image of the SrF₂ sample prepared at pH = 9; (e₄) width distribution determined from the SEM image in (c).

Effect of the EDTA to Sr²⁺ ratio on the particle structure and morphology. To understand the role of EDTA in the particle structure and morphology of the samples, additional experiments were carried out by adjusting the molar ratio of EDTA to Sr²⁺ with hydrothermal treatment at a pH of 9 and 160 °C for 24 h.

Fig. 3 illustrates the representative SEM images of the as-prepared SrF₂ samples with different molar ratios of EDTA to Sr²⁺ at 160 °C for 24 h. Clearly, the morphology and size of SrF₂ particles significantly change with the addition of EDTA. In the absence of EDTA in solution, the morphology with irregular and mixed-size SrF₂ particles is derived (Fig. 3a). Following the introduction of EDTA, the morphologies and sizes differ dramatically. In the presence of a relatively small amount of EDTA (with an EDTA to Sr²⁺ ratio of 0.5 : 1), aggregates of fine powders were formed (Fig. 3b). Hence, the introduction of EDTA with an EDTA to Sr²⁺ ratio of 0.5 : 1 is not favourable for the formation of SrF₂ cubic particles. On further increasing the ratio to 1 : 1.5, relatively small, ordered and uniform SrF₂ cubes were fabricated (as shown in Fig. 2c). This indicates that the chelating agent, EDTA, plays an important role in the formation of uniform SrF₂ cubes and prevents the particles from aggregating. However, with the molar ratio of EDTA to Sr²⁺ of 2 : 1, the excessive EDTA seems to facilitate the growth of SrF₂. The size of the cubes increases to about 2 μm (Fig. 3d). In addition, with the ratio of 2 : 1, it is disclosed that SrF₂ cubes are attached to SrF₂ sheets. This reinforces the role of EDTA in the formation of SrF₂ cubes which is consistent with the results mentioned previously.

Fig. 3 SEM images of the SrF₂ samples obtained with different molar ratios of EDTA to Sr²⁺: (a) 0 : 1, (b) 0.5 : 1, (c) 1.5 : 1, and (d) 2 : 1.

Based on the above results, we suggest that the formation of [H_xY]^x−4via dissociation of EDTA is inhibited by H⁺ released via the hydrolysis of BF₄⁻ (reaction 3) with low EDTA to Sr²⁺ ratios. Consequently, it is not favourable for EDTA to function as a chelating agent (reactions 6 and 7). However, with excessive amounts of EDTA, in addition to the chelating ability, it also serves as the layer growth agent. Hence, the EDTA to Sr²⁺ ratio of 1 : 1 is the most favourable for the formation of SrF₂ microcubes.

Effect of the hydrothermal time. To further elucidate the possible growth mechanism of the cubic SrF₂ particles, the morphology and size of SrF₂ particles were investigated following hydrothermal treatment at various times at a pH of 9, a temperature of 160 °C and a molar ratio of EDTA to Sr²⁺ of 1 : 1. These experiments facilitate the disclosure of the formation and development of SrF₂ particles.

Fig. 4 displays the SEM images of the samples obtained after hydrothermal treatment at different times. At the beginning (30 minutes of reaction, Fig. 4a), a large number of irregular nanoparticles with a diameter of 40–50 nm are found. After 45 minutes (Fig. 4b), large amounts of cubic SrF₂ particles start to appear. The average length of the cube is about 60 nm. In addition, the surface of the cubic SrF₂ particles is rough and covered with very small particles. After 1 hour of hydrothermal treatment (Fig. 4c), only cubic SrF₂ particles are observed. The size and morphology of the cubes are similar to those of the particles shown in Fig. 3c. In addition, the cubic SrF₂ particles continue to grow. The average length of the cubic SrF₂ particles becomes 80 nm and the surface of the cubic SrF₂ particles becomes smooth, as shown in Fig. 4c. When the hydrothermal treatment time is prolonged to 2 hours (Fig. 4d), the cubic SrF₂ particles begin to aggregate as lengths higher than 200 nm are detected. After 4 hours (Fig. 4e), the cubic SrF₂ particles further grow. After hydrothermal treatment of 8 hours (Fig. 4f), large SrF₂ cubes are formed. However, small amounts of fine and irregular sheets are still observed in addition to the SrF₂ cubes. Furthermore, the edges of the cubic particles start to become rounded, indicating Ostwald ripening with extended hydrothermal treatment.³⁷ With further extending the treatment time to 12 hours and 24 hours, no significant change in morphology is observed except for agglomeration (Fig. 4g and h).

Fig. 4 SEM images of morphology evolution as a function of hydrothermal treatment time: (a) 30 min, (b) 45 min and the inset showing the corresponding high magnification image, (c) 1 h, (d) 2 h, (e) 4 h, (f) 8 h, (g) 12 h and (h) 24 h following hydrothermal treatment at 160 °C.

Proposed formation mechanism of SrF₂ cubes by hydrothermal synthesis. On the basis of the above analysis and discussion, the possible SrF₂ morphological evolution is proposed and depicted in Fig. 5. Following the mixing of EDTA, NH₄BF₄ and SrCl₂·6H₂O in the aqueous solution during catalyst preparation, at the early stage of the reaction, the intermediate, [H_xY]^x−4–Sr²⁺complex, was first formed through strong coordination interaction between [H_xY]^x−4 and Sr²⁺, which significantly decreased the concentration of free Sr²⁺ ions in the solution. The solution was then transferred into the hydrothermal autoclave and treated at high temperatures and pressures. Consequently, the chelation of the [H_xY]^x−4–Sr²⁺ complex was weakened by H⁺ released by the hydrolysis of NH₄BF₄ (reaction 3) and Sr²⁺ ions were released gradually from the chelating complex. At the same time, F⁻ was also released following the hydrolysis of BF₄⁻. Upon the combination of Sr²⁺ and F⁻ or an anion exchange between [H_xY]^x−4and F⁻, a fast nucleation process occurred and plenty of irregular nanoparticles of SrF₂ nuclei were formed. As shown in Fig. 4a, only with a hydrothermal treatment of 30 min, cubic-like particles are observed. With a treatment of 45 min, growth and cube formation are observed as shown in Fig. 4b. As shown in the inset of Fig. 4b, the surface of the cubes is covered by significant amounts of small particles. Clearly, Ostwald ripening plays a major role in the formation and growth of SrF₂ cubes. After repeated deposition in the surface of SrF₂ particles, SrF₂ grew into large cubes. Finally, complete cube-like SrF₂ particles were achieved following the hydrothermal process for about an hour. In addition, as demonstrated in Fig. 3d, with a high concentration of EDTA, more [H_xY]^x−4–Sr²⁺ are available, which facilitates the adsorption on the SrF₂ surface. Therefore, during preparation, the surface of SrF₂ particles was rapidly covered by large amounts of small particles. As a result, a sheet-like structure is observed in Fig. 3d. In this process, Ostwald ripening and surface reconstruction take place to form the cube-like SrF₂ particles with larger sizes and smoother surfaces. However, the growth of cubes does not mean the growth of crystals. As indicated in Fig. S3,† the intensity of SrF₂ increases with hydrothermal time. However, only slight growth of SrF₂ crystals is noted (Table S2†). This is consistent with the results of TEM in Fig. 2 wherein the cubes are polycrystalline in nature. Therefore, the growth of cubes represents the assembly of SrF₂ rather than its crystallization.

Fig. 5 Schematic illustration of the morphological evolution process of the SrF₂ cube architecture.

The leaving [H_xY]^x−4 ions still surrounded the particle and some of the [H_xY]^x−4 ions were adsorbed over the particle surface function as the surfactant.^38,39 The adsorption of [H_xY]^x−4 is not favourable for the aggregation of SrF₂ cubes with several cube-like SrF₂ particles.^40,41 The surface elements of SrF₂ obtained with hydrothermal treatment for 24 hours at a pH of 9, a temperature of 160 °C and an EDTA to Sr²⁺ ratio of 1 : 1 include C, N and O, in addition to Sr and F determined by XPS (Table 1). This indicates that the particles adsorbed significant amounts of [H_xY]^x−4 during preparation. In the formation process, similar to the surfactant, the chelating agent EDTA also plays a role in preventing the particles from aggregating and the evolution of the SrF₂ shape. In addition, the above discussion is supported by the following results. Firstly, in Fig. 3a, it is observed that the SrF₂ particles are irregular and larger in the absence of EDTA. Secondly, with low pH of the solution, the cubic SrF₂ particles are not observed (Fig. 2a). Clearly, EDTA plays an important role in obtaining cubic SrF₂ particles. Thirdly, as indicated in Fig. 2, cubic SrF₂ particles possess clear multi-layer structures. Fig. 3d reinforces this structure. Hence, as illustrated in Fig. 5, we suggest that the initial grains of SrF₂ particles (nuclei) with high surface energy adsorb the unsaturated [H_xY]^x−4–Sr²⁺. Then [H_xY]^x−4–Sr²⁺ bonds to fluorine ions deposited on the surface of the initial grains. In this way, numerous initial grains obtained through deposition of [H_xY]^x−4–Sr²⁺ form cubic SrF₂ particles after several hours.

Table 1 Surface element content of SrF₂ with hydrothermal treatment at a pH of 9, a temperature of 160 °C and an EDTA to Sr²⁺ ratio of 1 : 1 (determined by XPS)

Samples	Atomic contents (atom%)
	C	N	O	F	Sr
SrF2	10.6	2.6	13.9	51.1	21.8

---

Although the adsorption of [H_xY]^x−4 prevents the cubes from agglomerating, a combination of cubes is still observed with extension of the hydrothermal treatment to longer than 2 h (Fig. 4e–h). As demonstrated in Fig. 4c, with a hydrothermal treatment of 1 h, the particle size of the cubes is around 80 nm. However, sizes of 150 nm to 300 nm are detected in Fig. 4e–h. We suggest that 2 to 4 cubes agglomerate into large cubes (Fig. 5).⁴²

Catalytic activity and characterization of SrF₂ catalysts prepared by precipitation and hydrothermal synthesis

The SrF₂ catalysts prepared by precipitation and hydrothermal synthesis were first calcined in air at 500 °C for 4 h with a flow rate of 50 mL min⁻¹. The derived catalysts were denoted as SrF₂-P for precipitation and SrF₂-H for hydrothermal synthesis, respectively. Pyrolysis of HCFC-142b (CH₃CClF₂) was adopted as a model reaction for the evaluation of the catalytic activity.

Fig. 6 displays the catalytic activities of the SrF₂ catalysts for the pyrolysis of HCFC-142b to vinylidene fluoride as a function of time on stream (TOS). The reactions were carried out at 350 °C, a pressure of 1 bar and a GHSV (HCFC-142b) of 600 h⁻¹. In addition to the major product VDF, the by-products obtained during the catalyst activity evaluation include CH₂=CClF (VCF, dehydrofluorination product of 1,1-chlorofluoroethane) and trace amounts of 1,1,1-trifluoroethane (HFC-143a, CH₃CF₃). As expected, the activity increases with reaction temperature (Fig. 6a). SrF₂ starts to catalyse the decomposition of HCFC-142b at temperatures below 250 °C. It is worth noting that the SrF₂ catalyst exhibits high activity as the conversion levels of HCFC-142b are about 70% and close to 100% at reaction temperatures of 350 °C and 450 °C, respectively. These temperatures are significantly lower than the temperatures for industrial manufacture of VDF of 650–700 °C. As demonstrated in Fig. 6a, the activity of SrF₂-P is slightly higher than that of the SrF₂-H catalyst within the temperature range investigated. However, at temperatures above 300 °C, the activities of both catalysts are close. The selectivity to the target product, VDF, over these two catalysts differs dramatically (Fig. 6b). A selectivity of 90–97% is achieved over the SrF₂-H catalyst and it only changes with reaction temperature slightly. By contrast, the selectivity to VDF over the SrF₂-P catalyst is significantly lower than that over SrF₂-H. Although the SrF₂ catalyst (SrF₂-P) prepared by the conventional precipitation method shows relatively high conversion, its selectivity to VDF is only around 50–70%. As demonstrated in Fig. 6c and d, both catalysts exhibit stable activity at time on stream of 30 h at 350 °C.

Fig. 6 The performance of SrF₂ catalysts prepared by precipitation (P) and the hydrothermal (H) method for the pyrolysis of HCFC-142b to vinylidene fluoride. (a) The conversion of HCFC-142b and (b) the selectivity to VDF as a function of reaction temperature. (c) The conversion of HCFC-142b and (d) the selectivity to VDF as a function of time on stream at 350 °C. Reaction conditions: 1 bar, N₂ : HCFC-142b of 1 : 1, GHSV (HCFC-142b) of 600 h⁻¹.

To elucidate the difference in catalytic activity between SrF₂-P and SrF₂-H catalysts, both catalysts were characterized by XRD and N₂ adsorption–desorption isotherms. As shown in Fig. 7a, it is clearly observed that all the diffraction peaks of both catalysts are perfectly consistent with those of the standard SrF₂ pattern (PDF# 88-2294) following calcination in air at 500 °C for 4 h. SrF₂-P and SrF₂-H are composed of the same crystalline grains. The strong, sharp diffraction peaks reveal that both catalysts are highly crystallized. However, the intensity of SrF₂-P is a little higher than that of SrF₂-H, indicating that the SrF₂-P catalyst has higher crystallinity and possesses larger crystal particles. This agrees well with the SEM images which show that the particles of the SrF₂-P catalyst are larger than those of SrF₂-H, as shown in Fig. 8a and b. It is reasonable to infer that SrF₂-P should possess a lower specific surface area than SrF₂-H.

Fig. 7 (a) X-ray diffraction patterns and (b) N₂ adsorption–desorption isotherms of SrF₂-H and SrF₂-P catalysts.

Fig. 8 SEM images of (a) calcined SrF₂-P and (b) calcined SrF₂-H catalysts. Low and high magnification TEM images of the calcined SrF₂-H catalyst (c₁, c₂ and c₃). HRTEM image of the calcined SrF₂-H catalyst (c₄) and the inset showing the corresponding SAED pattern.

The specific surface area was determined by N₂ adsorption–desorption. The isotherms of both catalysts are illustrated in Fig. 7b. Clearly, both catalysts reveal typical type III isotherms and H3 hysteresis loops which are usually observed with aggregates of particles giving rise to slit-shape pores.⁴³ The results of N₂ adsorption indicate that the porous structures of both catalysts result from the aggregation of particles which agrees well with the observations of SEM and TEM. The surface area of SrF₂-H and SrF₂-P was determined to be 15.0 m² g⁻¹ and 5.5 m² g⁻¹, respectively (Table S1†). During the catalytic reaction of HCFC-142b pyrolysis, the metal fluoride serves as the active site for the dehydrochlorination and dehydrofluorination reactions of HCFC-142b.⁴⁴ This suggests that the catalyst with a higher surface area usually has higher catalytic activity. However, the SrF₂-H catalyst with a higher surface area and smaller crystal size does not lead to higher conversion of HCFC-142 compared with the SrF₂-P catalyst. Hence, the crystal size and the surface area are not the key factors affecting the catalytic activity although they play important roles in the conversion of HCFC-142b.

Following calcination in air at 500 °C for 4 h, aggregation of fine particles is noted for the SrF₂-P catalyst (Fig. 8a). By contrast, no obvious change in morphology is found over the SrF₂-H catalyst (Fig. 8b). As shown in Fig. 8b and c₁, c₂, c₃, the SrF₂-H catalyst still maintains the cube-like particles after treatment at 500 °C for 4 h. It is observed that slight aggregation takes place. As indicated by the arrow in Fig. 8c₄, it is observed that the surface of SrF₂-H was capped by the carbon-containing compound derived from the calcination of adsorbed EDTA. As mentioned previously, the particles of SrF₂ prepared by the hydrothermal method adsorb small amounts of EDTA. The adsorption sites prefer to locate at sites with large surface energy, more specifically the Lewis acid sites. Hence, the carbon-containing compound reduces the adsorption of the feed gas over the SrF₂-H catalyst during the catalytic reaction.

The surface element content of the calcined SrF₂-H catalyst determined by XPS is listed in Table 2, and the whole XPS spectrum of SrF₂-H is shown in Fig. S1.† The results indicate that the surface of the SrF₂-H catalyst after treatment at 500 °C still contains certain amounts of carbonaceous species. This further confirms that part of the SrF₂-H surface is covered by the carbonaceous compound.

Table 2 Surface element content of the SrF₂-H catalyst

Samples	Atomic contents (atom%)
	C	N	O	F	Sr
SrF2	6.2	2.5	11.2	56.8	23.3

---

Generally, the acidity, more specifically, the Lewis acidic site of the catalyst plays a key role in dehydrochlorination and dehydrofluorination reactions.⁴⁵ Teinz⁴⁴ demonstrated that catalysts possessing strong Lewis acidity exhibit enhanced catalytic activity. However, strong Lewis acidity usually facilitates selective dehydrofluorination rather than dehydrochlorination leading to the poor selectivity to the dehydrochlorination product.^46,47

As exhibited in Fig. 9a, the surface acidity of SrF₂-P is significantly higher than that of SrF₂-H. In addition, the SrF₂-P catalyst displays relatively uniform acidic sites with an ammonia desorption temperature of around 520 °C. By contrast, NH₃-TPD of the SrF₂-H catalyst shows a weak and broad peak with desorption temperature between 100 and 600 °C. This indicates that acidic sites over SrF₂-H are much fewer and weaker than those over SrF₂-P. Consequently, it leads to the low conversion of HCFC-142b over SrF₂-H. Although parts of the surface, especially the strong Lewis acidic sites, are covered by carbonaceous species, SrF₂-H possesses a high specific surface area. Consequently, conversion of HCFC-142b close to SrF₂-P is approached. Furthermore, as the strong acidic sites over SrF₂-H are covered by the carbonaceous species, the SrF₂-H catalyst exhibits a much higher selectivity to VDF (by dehydrochlorination) than VCF (dehydrofluorination).

Fig. 9 (a) The profiles of temperature-programmed desorption of ammonia on SrF₂-H and SrF₂-P catalysts and (b) DTG spectra of SrF₂-H and SrF₂-P catalysts.

It is generally accepted that the Lewis acidic sites are coke formation centres.^48–50 As a result, the catalyst with large amounts of Lewis acidic sites is favourable for carbon deposition. The coke formation over the catalysts was characterized by DTG as shown in Fig. 9b. It is observed that the weight loss rate of both catalysts is very small, indicating that there are very small amounts of coke after catalytic reaction for 30 h. Furthermore, SrF₂-P has a higher weight loss than the SrF₂-H catalyst, indicating that SrF₂-P suffers from greater coke formation. In addition, a weight loss over 700 °C is noted for the spent SrF₂-H catalyst. It is reasonable to infer that the weight loss is ascribed to the carbonaceous species covering the surface of the SrF₂-H catalyst.

Conclusions

In summary, cube-like architectures of SrF₂ particles were synthesized through a convenient and facile hydrothermal synthesis route with ethylenediaminetetraacetic acid as the chelating agent. During the hydrothermal synthesis of SrF₂, the pH of the solution plays an important role in the morphology and size of SrF₂ particles. The suitable synthesis conditions for cube-like SrF₂ were determined to be a pH of 9 and an EDTA : Sr²⁺ of 1 : 1 at 160 °C. The formation process of the free-standing and uniform cube-like sub-micro particles was proposed based on the above results.

The obtained SrF₂ samples were adopted as efficient catalysts for the pyrolysis of 1-chloro-1,1-difluoroethane to vinylidene fluoride. The SrF₂-P catalyst prepared by the precipitation method is included as a reference. The results indicate that both SrF₂ catalysts exhibit high catalytic activity with high HCFC-142b conversion and VDF selectivity at temperatures between 250 °C and 450 °C. No deactivation was observed following time on stream of 30 h at 350 °C. The selectivity to the target product, VDF, over the SrF₂-H catalyst is much higher than that over SrF₂-P. The surface acidity, the size of particles and the specific surface area of SrF₂ play important roles in VDF selectivity and catalytic performance. The carbonaceous species capping the surface of the SrF₂-H catalyst are responsible for the high selectivity to VDF.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The financial support from the Zhejiang Provincial Natural Science Foundation of China (LY19B060009), the Qianjiang Talent Project B in Zhejiang Province (2013R10056), and Special Programs for Research Institutes in Zhejiang (2015F50031) is acknowledged.

References

1. P. P. Fedorov, A. A. Luginina, S. V. Kuznetsov and V. V. Osiko, J. Fluorine Chem., 2011, 132, 1012–1039.
2. P. P. Fedorov, S. V. Kuznetsov and V. V. Osiko, in Photonic and Electronic Properties of Fluoride Materials, ed. A. Tressaud and K. Poeppelmeier, Elsevier, Boston, 2016, pp. 7–31.
3. C. M. Zhang, Z. Y. Hou, R. T. Chai, Z. Y. Cheng, Z. H. Xu, C. X. Li, L. Huang and J. Lin, J. Phys. Chem. C, 2010, 114, 6928–6936.
4. P. P. Fedorov, A. A. Luginina, J. A. Ermakova, S. V. Kuznetsov, V. V. Voronov, O. V. Uvarov, A. A. Pynenkov and K. N. Nishchev, J. Fluorine Chem., 2017, 194, 8–15.
5. S. Kuznetsov, Y. Ermakova, V. Voronov, P. Fedorov, D. Busko, I. A. Howard, B. S. Richards and A. Turshatov, J. Mater. Chem. C, 2018, 6, 598–604.
6. A. A. Lyapin, S. V. Gushchin, A. S. Ermakov, S. V. Kuznetsov, P. A. Ryabochkina, V. Y. Proydakova, V. V. Voronov, P. P. Fedorov and M. V. Chernov, Chin. Opt. Lett., 2018, 16, 091901.
7. Y. A. Rozhnova, S. V. Kuznetsov, A. A. Luginina, V. V. Voronov, A. V. Ryabova, D. V. Pominova, R. P. Ermakov, V. A. Usachev, N. E. Kononenko and A. E. Baranchikov, Mater. Chem. Phys., 2016, 172, 150–157.
8. X. Zhang, Z. Quan, J. Yang, P. Yang, H. Lian and J. Lin, Nanotechnology, 2008, 19, 075603.
9. Y. P. Du, X. Sun, Y. W. Zhang, Z. G. Yan, L. D. Sun and C. H. Yan, Cryst. Growth Des., 2009, 9, 2013–2019.
10. Y. P. Sun and P. Y. Jia, J. Nanosci. Nanotechnol., 2014, 14, 3957–3960.
11. J. Y. Sun, J. B. Xian, X. Y. Zhang and H. Y. Du, J. Rare Earths, 2011, 29, 32–38.
12. R. Q. Long and H. L. Wan, J. Chem. Soc., Faraday Trans., 1998, 94, 1129–1135.
13. L. Wang, X. Yi, W. Weng, C. Zhang, X. Xu and H. Wan, Catal. Lett., 2007, 118, 238–243.
14. L. H. Wang, X. D. Yi, W. Z. Weng and H. L. Wan, Catal. Today, 2008, 131, 135–139.
15. C. L. Yaws, McGraw-Hill handbooks, 1999, vol. 4, pp. 48–52.
16. B. Ameduri, Chem. Rev., 2009, 109, 6632–6686.
17. B. Hu, C. Lei, R. Xu, W. Shi, Q. Cai, H. Mo and C. Chen, J. Plast. Film Sheeting, 2014, 30, 300–313.
18. T. H. Han, R. Nirmala, T. W. Kim, R. Navamathavan, H. Y. Kim and S. J. Park, J. Nanosci. Nanotechnol., 2016, 16, 595–600.
19. C. G. Zhang, W. W. Liu, C. Cao, F. Y. Zhang, Q. M. Tang, S. Q. Ma, J. J. Zhao, L. Hu, Y. Shen and L. L. Chen, Adv. Healthcare Mater., 2018, 7, e1701466.
20. J. X. Zhang, X. Y. Du, C. C. Wang and K. L. Ren, J. Phys. D: Appl. Phys., 2018, 51.
21. J. Chen, S. B. Tan, G. X. Gao, H. Y. Li and Z. C. Zhang, Polym. Chem., 2014, 5, 2130–2141.
22. S. B. Tan, J. J. Li, G. X. Gao, H. Y. Li and Z. C. Zhang, J. Mater. Chem., 2012, 22, 18496–18504.
23. Y. Kano, S. Akiyama, T. Kasemura and S. Takahashi, Polym. Networks Blends, 1996, 6, 153–159.
24. T. Soulestin, V. Ladmiral, F. D. Dos Santos and B. Améduri, Prog. Polym. Sci., 2017, 72, 16–60.
25. Y. M. Cheng, J. C. Wang, W. F. Han, Y. Y. Song, W. C. Liu, L. T. Yang, S. C. Wang, Z. X. Wu, H. D. Tang, J. J. Zhang, M. Stockenhuber and E. M. Kennedy, Greenhouse Gases: Sci. Technol., 2018, 8, 587–602.
26. W. F. Han, Y. Y. Song, W. C. Liu, L. T. Yang, H. D. Tang, S. C. Wang, Z. X. Wu and J. J. Zhang, Greenhouse Gases: Sci. Technol., 2017, 7, 891–902.
27. W. F. Han, Y. Li, H. D. Tang and H. Z. Liu, J. Fluorine Chem., 2012, 140, 7–16.
28. G. Wang, H. F. Zheng, H. Yin, S. F. Yuan and Z. R. Chen, J. Zhejiang Univ., Sci., A, 2015, 49, 1812–1816.
29. M. Herbert, E. Gerhard and H. Hanns, Chem. Ing. Tech., 1984, 56, 626–628.
30. R. Geetha, Pop. Plast. Packag., 1998, 43(9), 67–70.
31. G. Wang, H. Zheng, H. Yin, S. Yuan and Z. Chen, Chem. React. Eng. Technol., 2016, 32, 73–77.
32. Z. Wang, W. Han, C. Zhang, S. Zhou, H. Wang, H. Tang and H. Liu, ChemistrySelect, 2018, 3.
33. H. Guo, Y. Guo, H. M. Noh, B. K. Moon, S. H. Park, J. H. Jeong and K. H. Kim, J. Nanosci. Nanotechnol., 2016, 16, 1146–1150.
34. B. L. Gersten, M. M. Lencka and R. E. Riman, J. Am. Ceram. Soc., 2004, 87, 2025–2032.
35. R. S. Juang, Y. J. Chen and I. P. Huang, Sep. Sci. Technol., 1999, 34, 3099–3112.
36. Y. Zhang, Q. Zhao, B. Shao, W. Lu, X. Dong and H. You, RSC Adv., 2014, 4, 35750–35756.
37. Y. K. Bai, R. T. Zheng, Q. Gu, J. J. Wang, B. S. Wang, G. A. Cheng and G. Chen, J. Mater. Chem. A, 2014, 2, 12770–12775.
38. M. Chatti, S. Sarkar and V. Mahalingam, Microchim. Acta, 2016, 183, 133–140.
39. K. Ariga, J. P. Hill and Q. M. Ji, Phys. Chem. Chem. Phys., 2007, 9, 2319–2340.
40. R. L. Penn and J. F. Banfield, Science, 1998, 281, 969–971.
41. J. J. De Yoreo, P. Gilbert, N. Sommerdijk, R. L. Penn, S. Whitelam, D. Joester, H. Z. Zhang, J. D. Rimer, A. Navrotsky, J. F. Banfield, A. F. Wallace, F. M. Michel, F. C. Meldrum, H. Colfen and P. M. Dove, Science, 2015, 349, aaa6760.
42. H. Suo, Z. Yang, C. F. Guo, X. Q. Zhao and L. L. Zhang, CrystEngComm, 2016, 18, 3566–3573.
43. G. C. Bond, K. S. W. Sing, D. H. Everett, R. A. W. Haul, L. Moscou, R. A. Pierotti, J. Rouquerol and T. Siemieniewska, Annexes, Wiley-VCH Verlag GmbH, 2008.
44. K. Teinz, S. Wuttke, F. Börno, J. Eicher and E. Kemnitz, J. Catal., 2011, 282, 175–182.
45. L. E. Manzer and V. N. M. Rao, Adv. Catal., 1993, 39, 329–350.
46. W. Mao, Y. Bai, W. Wang, B. Wang, Q. Xu, L. Shi, C. Li and J. Lu, ChemCatChem, 2017, 9, 824–832.
47. K. Teinz, S. R. Manuel, B. Bin Chen, A. Pigamo, N. Doucet and E. Kemnitz, Appl. Catal., B, 2015, 165, 200–208.
48. R. M. Navarro, M. C. Alvarez-Galvan, M. C. Sanchez-Sanchez, F. Rosa and J. L. G. Fierro, Appl. Catal., B, 2005, 55, 229–241.
49. J. Ni, L. Chen, J. Lin and S. Kawi, Nano Energy, 2012, 1, 674–686.
50. W. F. Han, C. P. Zhang, H. L. Wang, S. L. Zhou, H. D. Tang, L. T. Yang and Z. K. Wang, Catal. Sci. Technol., 2017, 7, 6000–6012.

---

Footnote

† Electronic supplementary information (ESI) available: Textural parameters of SrF₂ samples, XPS spectra, and SEM images obtained with different solution concentrations. See DOI: 10.1039/c8ce01546e

---