Thermal Stability Determination of Propylene Glycol Sodium Alginate and Ammonium Sulfate with Calorimetry Technology

Abstract

Propylene Glycol Alginate Sodium Sulfate (PSS) is widely produced and used in medicine as a marine drug for treating hyperlipidemia. During the sulfonation synthesis of PSS, the sulfonation of chlorosulfonic acid is exothermic. At high temperatures, the process can easily produce a large amount of ammonium sulfate. Ammonium sulfate adheres to PSS in crystal and participates in the sulfonation reaction. In this study, the sulfonation process of commercial PSS was reproduced in the laboratory using chlorosulfonic acid and formamide. We used differential scanning calorimetry and thermogravimetric analyzer to examine the thermal stability of PSS, and we used both differential and integral conversional methods to determine the appropriate thermokinetic models for this substance. We also established an autocatalytic model to study the conversion limit time and the maximum rate time of this substance. After calculation, the activation energy of this substance is no more than 60 kJ/mol, and it has other exothermic performances at different heating rates. The results help to optimize the sulfonation process of PSS and analyze the thermal risk of PSS with ammonium sulfate.

1. Introduction

For the past 15 years, propylene glycol alginate sodium sulfate (PSS) has been a marine-sulfated polysaccharide drug produced in large quantities in China [1]. The heparin-like activity of the drug can reduce the viscosity of blood, has antithrombotic effects, and can reduce peripheral vascular dilation. It has a significant impact on treating cerebrovascular diseases [2,3]. The sulfonyl and propylene glycol groups are added by degradation, esterification, and sulfonation into the hydroxyl and carboxyl groups in Figure 1. The sodium alginate extracted from seaweed synthesizes the final product PSS.

The typical PSS production process in commercial pharmaceutical companies is to emulsify and acidify the raw materials, then add propylene oxide and sodium hydroxide to the reactor for esterification under pressure. The propylene glycol alginate (PGA) formed in the esterification reaction is fused with formamide and added to chlorosulfonic acid at low temperatures. Then the temperature is increased to promote the sulfonation reaction and obtain the crude PSS sample by washing and drying. Formamide is used as a reaction solvent because of its good solubility and high boiling point [4,5]. However, formamide is easy to decompose into ammonium and carbon monoxide at high temperatures. The ammonium is also easy to form a large amount of ammonium salt (ammonium sulfate ((NH₄)₂SO₄)) in an acidic environment [5]. Although the products will be dissolved and precipitated with different concentrations of ethanol to remove sulfate salts at the end of the process, the products are still mixed with a large amount of (NH₄)₂SO₄ in the process. This is prone to causing a thermal runaway reaction [6].

The process solves the problems of high raw material viscosity, incomplete reaction, and low yield. However, the actual sulfonation process is a high-risk process involving many dangerous substances, including highly corrosive chlorosulfonic acid and (NH₄)₂SO₄. They are prone to producing harmful gas at high temperatures, which has a substantial risk [7,8,9]. In May 2012, an explosion occurred in Jiangxi Haichen Honghua Chemical Co., Ltd. (Jiangxi, China), where the cooling water of the condenser entered the sulfonation kettle and reacted violently with chlorosulfonic acid to cause an explosion. In May 2005, a chemical burn accident occurred in the production process of (NH₄)₂SO₄ workshop of a pharmaceutical chemical enterprise with PSS production process (Henan, China), in which (NH₄)₂SO₄ was heated and decomposed into sulfur dioxide and water, causing a chemical burn accident.

In this study, the sulfonation reaction process in the above process is performed in the laboratory to explore the potential risks in the reaction. In addition, the thermal stability of PSS + (NH₄)₂SO₄ produced in the process is examined. Differential scanning calorimetry (DSC) and thermogravimetric analysis (TG) were adopted to conduct thermal tests. The thermokinetic models were adopted to simulate the exothermic situation of PSS + (NH₄)₂SO₄ under different heat-flow environments. Through the calculation results of model fitting, the activation energy (E_a) of the PSS + (NH₄)₂SO₄ was obtained. The findings imply an optimization of the sulfonation process of PSS and establishing thermal safety parameters for PSS + (NH₄)₂SO₄.

2. Experimental Materials and Methods

2.1. Materials

In the laboratory reproduction of the sulfonation production process of commercial PSS, the used raw materials and reagents include propylene glycol alginate (PGA), formamide, and chlorosulfonic acid. The specific information of the reagents used is listed in

Table 1. Specific information of materials used.

Reagent	CAS	Reagent Purity	Usage	Manufacturers
Propylene glycol alginate(PGA)	9005-37-2	CP	10 g	Zhejiang Yinuo Biotechnology Co., Ltd. (Hangzhou, China)
Formamide	75-12-7	AR	100 mL	Yatai Chemical Co., Ltd. (Wuxi, China)
Chlorosulfonic acid	7790-94-5	AR	30 mL	Qianyan Chemical Technology Co., Ltd. (Wuhan, China)

.

2.2. Sulfonation Synthesis of PSS

At present, the commercial production methods of PSS have quite a few risk factors. This is mainly through the hydrolysis of sodium alginate and esterification with propylene oxide to form the propylene diester alginate. Then, the propylene diester alginate is sulfonated with formamide and chlorosulfonic acid to form sodium diester alginate [2]. In the second reaction, because of the immense heat released by the sulfonation reaction, the formamide involved in the reaction can easily decompose into ammonia and CO when heated. Ammonia can easily form (NH₄)₂SO₄ under the environment of solid acids, such as chlorosulfonic acid.

In the experiment of reproducing sulfonation synthesis, we stirred and fused 100 mL formamide with 10 g propylene alginate as solvent. The reaction temperature was controlled at about 5 °C, and the chlorosulfonic acid was added dropwise. After dropping, we raised the total temperature to 110 °C. A large amount of (NH₄)₂SO₄ will be formed in the reaction. To further explore PSS + (NH₄)₂SO₄ thermal safety performance, we measured it with an advanced calorimeter and fitted its E_a with the thermokinetic model [10].

2.3. Differential Scanning Calorimetry Experiments

The DSC developed by Mettler Toledo (Mettler Toledo Co., Zurich, Switzerland) was used to measure the heat release of PSS + (NH₄)₂SO₄ produced in the sulfonation synthesis experiment under different β [11,12,13]. According to the physical and chemical properties of PSS + (NH₄)₂SO₄, the high-density alumina crucible was selected to conduct the experience, evenly spread 5.75 ± 0.06 mg of PSS + (NH₄)₂SO₄ in an identical alumina crucible, and calorimetric experiments with different β were carried out [10]. Combined with the suggestions of the International Federation Of Thermal Analysis And Calorimetric Algorithms and the actual situation of DSC, we calculated five groups of experiments, set to 1.0, 2.0, 3.0, 5.0, and 8.0 °C/min, respectively [14,15,16]. The characteristic thermal parameters of each series of experiments (including peak temperature (T_p), initial temperature (T₀), and conversion.) were obtained through multiple experiments. These important kinetic parameters are also utilized for subsequent thermokinetic calculation.

2.4. Thermogravimetric Analysis Experiments

The TG analyzer (Mettler Toledo Co., Zurich, Switzerland) was utilized to measure the overall mass loss of PSS + (NH₄)₂SO₄ under different β [17,18,19]. The acidity and alkalinity of the PSS + (NH₄)₂SO₄ was tested, the pH value was 5.8. The alumina was chosen as the experimental vessel in TG experiments because of its weakly acidic [20]. According to the overall mass loss of PSS + (NH₄)₂SO₄ in the experimental process. The five groups of experiments about 2.0, 3.0, 5.0, 8.0, and 10.0 °C/min was set under the same standard air atmosphere, respectively. The characteristic thermal parameters of each group (including mass loss rate, mass loss, T_p) were obtained through multiple experiments. These are combined with the parameters to draw the derivative thermogravimetric curve (DTG), to observe the thermal mass loss of PSS at different β.

2.5. Thermokinetic Analysis

For the study of the apparent E_a, it is a characteristic value related to temperature. When the E_a of the substance is lower, the energy required for the reaction of the substance is lower, which illustrates that reaction can occur easier [21,22]. Through the following thermokinetic calculation methods, the kinetic parameters were obtained in the measurement experiment to determine the value of E_a. This reliable calculation method is widely used in the thermal analysis experiments of various reactions or materials [23,24,25].

The Kissinger model was used to simulate the linear relationship of the data by the T_p of the reaction and the corresponding maximum heat flow. E_a of the substance and the determination coefficient (R²) of the linear relationship can be obtained through the slope. The method is illustrated in Equation (1) [26,27].

$$\mathit{ln}\left(\frac{\beta }{T^2}\right)=\mathit{ln}\left(\frac{AR}{E_{a}g\left(\alpha \right)}\right)-\frac{E_a}{R}\frac{1}{T}$$

(1)

where A is the pre-exponential factor, T is the reaction temperature, R is the ideal gas constant (8.314 J/(mol·K)).

Further, the FWO model was utilized to analyze the conversion, time, and corresponding temperature and deduces their linear relationship, The E_a and R² of the linear relationship was obtained according to the slope. The method can be expressed as Equation (2) [28,29].

$$\mathit{lg}\beta =\mathit{lg}\left(\frac{AE}{RG\left(\alpha \right)}\right)-2.315-0.4567\frac{E}{RT}$$

(2)

The Kissinger–Akahira–Sunose (KAS) method is derived by taking temperature as integral and using Coats–Redfern approximation. In this method, the conversion is substituted by the relationship between the conversion rate (α) and the temperature integral to improve E_a’s calculation accuracy and study the risk of the product [30,31]. The KAS method can be expressed as Equation (3).

$$\mathit{ln}\frac{\beta }{{T_{\alpha }}^2}=\mathit{ln}\left[\frac{R{k}_0}{E_{a}G\left(\alpha \right)}\right]-\frac{E_a}{R}\frac{1}{T_{\alpha }}$$

(3)

3. Results and discussion

3.1. Process Safety of Sulfonation Synthesis of PSS

The sulfonation synthesis process of commercial PSS was performed in the laboratory. This involved mixing propylene diester alginate with solvent formamide. A colloid with high viscosity was formed, which is not easy to stir and has poor heat transfer. This means uneven heating or high local heat can occur easily, as shown in Figure 2. After the temperature was reduced to 5 °C, chlorosulfonic acid was added dropwise. In the dropping process, chlorosulfonic acid participates in the reaction to form a sulfonation reaction, leading to exothermic and viscosity reduction of the colloid. After dropping, we raised the temperature to 110 °C for 3 h. During this period, the formamide involved in the reaction can quickly decompose into ammonia and CO when heated. Ammonia can easily form (NH₄)₂SO₄ under the environment of solid acids [32]. In the meantime, a large amount of CO was released, and the colloid could not discharge the gas quickly. This led to many bubbles, as expressed in Figure 2, which can easily cause an increase in reactant volume, a leak of CO, and personnel poisoning.

Figure 3 shows the products after experiment (PSS + (NH₄)₂SO₄) were obtained. Then we dissolved and precipitated the products with different concentrations of ethanol 3 times to separate the sulfate salts and dry them in a drying oven for 24 h. Finally, it is characterized by Fourier transform infrared spectrometer. Figure 4 displays the similarity of the spectral curve between the dried sulfate salts and (NH₄)₂SO₄ is 94.86%. Therefore, the PSS is mixed with a large amount of (NH₄)₂SO₄ in the production process.

3.2. Thermal Analysis Technology

(1)

Thermal decomposition analysis by DSC

Figure 5 shows PSS + (NH₄)₂SO₄ at different β in DSC curves.

Table 2. Characteristic temperature of PSS + (NH₄)₂SO₄ in the DSC at different β.

β (°C /min)	T0 (°C)	Tp (°C)	Te (°C)
1	132.7	148.5	161.7
2	146.9	165.3	182.9
3	120.5	168.5	190.7
5	174.4	183.7	213.7
8	189.5	209.3	288.7

details the thermokinetic parameters (including starting temperatures T₀, T_p and T_e) of the substance during DSC experiment. PSS + (NH₄)₂SO₄ at each heating rate is stable below 130 °C. As the temperature rises, PSS + (NH₄)₂SO₄ begins to release heat independently. Above 290 °C, the heat release ends, and the temperature change tends to be stable.

The DSC curves at different β were compared. It can be seen that the exothermic situation of PSS + (NH₄)₂SO₄ at different β is also different. T₀, T_e, and T_p of PSS + (NH₄)₂SO₄ increased with the higher β of PSS + (NH₄)₂SO₄. When the β is 1.0, 2.0, 3.0, and 5.0, the peak value of the curve increases slowly. The largest heat flow value occurred at the β of 8.0 °C/min, the heat release is obvious, and the critical temperature parameters increase significantly. So, the β of PSS + (NH₄)₂SO₄ has obvious effects on the initial temperature, duration and effect of heating release [33,34,35].

(2)

Thermal decomposition analysis by TG.

Figure 6 describes PSS + (NH₄)₂SO₄ produced in the sulfonation synthesis process at different β in DTG curve. DTG curve is the first-order partial derivative of TG measured data, indicating the DTG of the substance. In mass reduction, the DTG increased twice and formed two mass loss peaks. Furthermore, the β varied also affect the DTG. When the β value increases, the peak value of DTG curve becomes sharper. The DTG of the substance at the two peaks and its corresponding temperatures are T₁ and T₂. The PSS + (NH₄)₂SO₄ was stable before 180 °C in the standard air atmosphere environment, as listed in

Table 3. Characteristic temperature of PSS + (NH₄)₂SO₄ in the DTG at different β.

	Stage I		Stage II
β (°C/min)	T1 (°C)	DTG (mg/s)	T2 (°C)	DTG (mg/s)
2	260.5	−0.5511	323.9	−0.3587
3	268.8	−0.5252	326.9	−0.3499
5	276.6	−0.4999	337.4	−0.3385
8	289.2	−0.4787	345.2	−0.3203
10	292.8	−0.4825	349.3	−0.4821

. After 180 °C, PSS + (NH₄)₂SO₄ shows evident exothermic decomposition with increasing temperature. This is because it was generated into CO and water, and the mass decreased. So at the same temperature, when the β is higher, the mass loss rate of PSS + (NH₄)₂SO₄ is also higher. The β value affected the temperature at which mass loss begins, the peak temperature in the DTG curve took place at the lower temperature with the higher β. It illustrated that at the same mass loss rate, the β value prevents an inversely proportional trend with mass loss temperature of PSS + (NH₄)₂SO₄.

(3)

Analysis of thermokinetic results

The first exothermic peak was studied, and some thermokinetic parameters of the substance were obtained. According to the relevant research of the International Conference on Thermal Analysis and Calorimetry, we can combine several different β, establish the thermokinetic model, and the E_a of the material is solved by various model calculation methods [36,37,38]. The Kissinger model was used to establish a linear equation and solve the E_a according to the temperature parameters. In the meantime, according to the α recorded in the experimental process, FWO, Vyazovkin, and KAS models were used for calculation. This was done to achieve the effect of mutual verification, reduce errors, and improve the accuracy of the E_a [39,40,41].

Figure 7 diagrams the linear relationship between ln(β/T_p²) and 1000/T(K) by substituting β, the exothermic peak and the system temperature corresponding to the exothermic peak into the Kissinger model. According to the relevant results, the E_a of PSS + (NH₄)₂SO₄ is 50.8418 kJ/mol, and the R² is 0.9341.

By substituting the α, the β, and corresponding system temperature into FWO model calculation, we select the α of 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 0.95, and 0.99. the E_a for all the samples was calculated from the slope of the lines within the conversion range of 0.05–0.99 are shown in Figure 8. It can be seen from the figure that the fitting degree is not directly proportional to the E_a value. The average value E_a is 51.4959 kJ/mol, and R² value is 0.9562.

The fitting results through the linear relationship between ln(α) and 1/T in Vyazovkin model is shown in Figure 9. Among them, the calculated value of E_a is higher with the decrease of the fitting degree. The average value of E_a is 46.9087 kJ/mol, and the R² is 0.9006. The KAS model was used to calculate the average E_a of PSS + (NH₄)₂SO₄—results are shown in

Table 4. E_a and R² under different α based on the KAS model.

α	Ea (kJ/mol)	R2
0.05	66.2145	0.9920
0.10	60.9222	0.9789
0.20	55.0361	0.9591
0.30	51.5291	0.9485
0.40	48.7408	0.9407
0.50	46.2739	0.9348
0.60	44.0605	0.9294
0.70	42.1805	0.9247
0.80	40.3987	0.9235
0.90	38.5649	0.9242
0.95	37.3125	0.9261
0.99	33.8099	0.9176

. As the value of α is less than 0.3, the value of E_a is greater than 50 kJ/mol. On the contrary, the value of E_a decreases continuously and is less than 50 kJ/mol when the value of α is greater than 0.3. The average value of E_a is 47.0870 kJ/mol, and the R² is 0.9417.

The E_a and R² of PSS + (NH₄)₂SO₄ can be obtained through the fitting calculation of four thermokinetic models. The results show the R² is close to 1.0, and the difference of each fitted E_a is slight, which illustrates the fitting results obtained by these four methods are relatively reasonable and scientific. Finally,

Table 5. E_a and R² values were obtained by Kissinger, FWO, Vyazovkin, and KAS methods.

	Ea (kJ/mol)	R2
Kissinger	50.8418	0.9341
FWO	51.4959	0.9562
Vyazovkin	46.9087	0.9006
KAS	47.0870	0.9417
average value	49.0834	0.9332

presents the fitting results of the four thermokinetic models. The E_a of PSS + (NH₄)₂SO₄ is 49.0833 kJ/mol.

3.3. Thermokinetic Parameters Determined by Autocatalytic Model

According to the DSC curve of PSS + (NH₄)₂SO₄ in the exothermic process (Figure 4). The curves of the initial stage of the exothermic process did not overlap, and the whole curve is biased to the side with higher temperatures. According to the empirical judgment method of spectrum and a multitude of simulation experiments. It is preliminarily concluded that the exothermic process of PSS + (NH₄)₂SO₄ in the sulfonation process is a two-stage autocatalytic process. The following reaction formats are considered in Equations (4)–(6) [42,43].

$$A+nB⟷\left(n+1\right)B$$

(4)

$$A⟷B$$

(5)

$$B\stackrel{}{⟷}C$$

(6)

In this autocatalytic reaction process, the reaction is often accelerated with the rapid consumption of reaction substances, and autocatalytic substances are produced at the same time. The autocatalytic model Equation (7) is as follows:

$$\frac{d\alpha }{dt}=K_0{e}^{-Ea/RT}{\left(1-\alpha \right)}^{n1}\left(z+{\alpha }^{n2}\right)$$

(7)

where n1 and n2, respectively, represent the first and second stages of the reaction and z is the autocatalytic factor.

As listed in

Table 6. Thermokinetic evaluation of the multistage reaction models of PSS + (NH₄)₂SO₄.

	PSS + (NH4)2SO4 5 °C/min		PSS + (NH4)2SO4 8 °C/min
	Autocatalysis A to B1	Autocatalysis B1 to B2	Autocatalysis A to B1	Autocatalysis B1 to B2
ln A (ln 1/s)	15.2094	30.9340	18.3841	20.1172
Ea (kJ/mol)	70.9749	132.3971	87.4477	100.1652
Reaction order n1	1.3955	2.4947	1.2565	1.1530
Reaction order n2	0.9987	0.5087	0.9257	0.3684

, In different conditions β at (1, 2, 3, 5, and 8 °C/min), the relationship between the β and time and the relationship between heat release and time are shown in Figure 10 and Figure 11, respectively, where sim and exp represent simulation and experimental data, respectively, comparing the model simulation with the actual DSC curve. The fitting results of the autocatalytic model are completely scattered on the same line as the DSC experimental data. This illustrates that the simulation results are consistent with the experimental results. The fitting results of the autocatalytic model showed the E_a fitted by autocatalytic model was 68.43 kJ/mol. Therefore, the kinetic parameters simulated by autocatalytic model are not different from those calculated by Kissinger, FWO, and KAS model.

Figure 12 shows the curve of TCL and TMR_ad, where TCL depends on the temperature of PSS + (NH₄)₂SO₄ calculated according to the kinetic model. The dependence of time instant is the conversion limit time when the reaction conversion reaches a predetermined value. Therefore, estimating TCL studied the safe conversion time of the substance. Furthermore, the parameters studied can evaluate the thermal stability of PSS + (NH₄)₂SO₄ to establish the thermal safety parameters of the substance in the production process. The results showed that in some hot areas, when the temperature reaches 42 °C, the conversion of the substance could also reach a limit value of 200 days.

The estimation of dynamic model simulation (TMR_ad) shows the correlation of the maximum rate time according to the kinetic model. At present, the probability evaluation standard of thermal runaway accidents in this regard is mainly based on the suggestions of Stossel [44]. When TMR < 480 min, it is considered that the probability of an accident is high. According to the simulated TMR curve, the temperature of PSS + (NH₄)₂SO₄ in daily production and use should not be higher than 63 °C.

4. Conclusions

We reproduced the sulfonation reaction in PSS production in the laboratory and studied the thermal stability of PSS + (NH₄)₂SO₄ produced by increasing the temperature. The experimental results of DSC show that PSS + (NH₄)₂SO₄ caused an exothermic reaction under continuous heating. There is a positive proportional relationship between T₀, T_p, and β. The PSS + (NH₄)₂SO₄ began decomposition after reaching 160 °C. The starting temperature and rate of decomposition were consistent with β positive correlation. They were reflected in the TG experiment. Because the heat released by the substance is relatively low during the experiment, the thermal risk of this substance in daily production is not too great. Through the fitting calculation of four reliable kinetic models (Kissinger, FWO, Vyazovkin, and KAS models), the E_a of PSS + (NH₄)₂SO₄ is 49.0834 kJ/mol, and the R² is 0.9417. Since the value of E_a is relatively low, the substance is easy to react to when heated. In order to prevent the energy released after PSS + (NH₄)₂SO₄ reaction. It is recommended that the substances be stored separately and avoid the temperature of the storage environment not exceeding 61 °C. The autocatalytic methods were used to study the conversion limit time and the maximum rate of PSS + (NH₄)₂SO₄. It provides help for optimizing PSS production process and establishing thermal safety parameters of PSS + (NH₄)₂SO₄ in the future. It also provides a basis for studying the influence of (NH₄)₂SO₄ on the thermal stability of the system under the thermal runaway.