Chemical and Process Inherent Safety Analysis of Large-Scale Suspension Poly(Vinyl Chloride) Production

Abstract

In this work, a safety assessment was carried out for the suspension polymerization method, known for the lack of studies about its sustainable performance and long history of chemical accidents. Therefore, a safety analysis was conducted using the inherent safety methodology to assess and determine the inherent risks of the poly(vinyl chloride) (PVC) suspension production process using computer-aided process engineering (CAPE). The indicators were calculated using data from safety databases and the specialized literature, considering downstream stages like vinyl chloride monomer (VCM) recovery, PVC purification and PVC drying. The obtained indicators revealed that the process has a negative performance regarding inherent safety, with a total inherent safety index of 30. The chemical inherent safety index had a value of 19, with the main chemical risk of the process being presented by the vinyl chloride monomer (with a value of 11), along with the risk of the exothermic reactions. The process safety index had a value of 15, highlighting the inventory as the primary concern of the process (with a value of 5), followed by the presence of unsafe equipment such as furnaces, burners, and dryers. The safety structure index had a score of 3, categorizing the process as probably risky, with the reaction and purification stages being more susceptible to accidents. Lastly, it is recommended to reduce the size of the process inventory and to substitute out unsafe process units.

1. Introduction

Identifying potential hazards has become imperative to develop and implement more sustainable chemical processes [1]. Since the establishment of the Sustainable Development Goals (SDGs) proposed by the UN, the analysis of risks in chemical processes has become more relevant due to the commitment of economic organizations within the industrial sector to adopting plans to achieve them [2,3]. Poly(vinyl chloride) (PVC) is the third most widely used polymer in the world [4], thanks to its low cost and versatility [5]. Its applications are mainly found in the construction, electrical, healthcare, and food sectors [6]. Eighty percent of the global PVC supply is produced using the suspension polymerization method [7], characterized by its high productivity, controllability, and reliability [8]. However, the PVC value chain exhibits serious sustainability issues, attributable to intensive energy consumption, the emission of toxic substances, and risks to human health (manufacturing and use), among others [9].

Different methods exist to quantify and evaluate the risks and hazards in chemical processes. However, these methods are related to the different design stages of industrial processes, and this can impact their effectiveness, due to the amount of information needed to carry out these methods properly [10]. Some widely used methods such as hazard and operability studies (HAZOP), failure mode and effects analysis (FMEA), health risk assessment (HRA), quantitative risk assessment (QRA) and others are able to identify risks, but normally are performed during operational processes—this condition reduces the chances of improvement by changing operative conditions, causing additional costs associated with implementing safety mechanisms (between 15% and 30% of the capital cost) [11]. Therefore, methodologies like the inherent safety analysis (ISI) of processes have become attractive tools for designing safer and more reliable chemical processing routes with the strategy of eliminating risks rather than controlling them [12,13].

Inherent safety analysis (ISI) is a methodology proposed by Heikkilä [14], and it uses indicators for risk evaluation based on the intrinsic properties associated with the operational conditions of the process (emerging or proven technologies) [15,16]. The ISI analysis includes important indexes assessing different parameters, like chemical activity between substances and their environment due to being an important factor of accidents [17]. It also includes aspects associated with the operating conditions of the units and the process as a system, such as temperature and pressure. This analysis is supported by CAPE, as it facilitates the management of hundreds of parameters, properties, and systems in processes [18]. It also considers substances of diverse composition and structures, like vaccines [19] and chitosan [20]. Normally, ISI analysis is performed in the earlier stages of design such as new and untested processes due to the fewer engineering constraints in these stages, However, in later phases, it is useful to rate processes and pinpoint sources of risks, as achieved by Tian et al. for the quantification of the effects of risk mitigation strategies in the operation of drying equipment [16].

Regarding the PVC process, some studies express safety aspects through life cycle analysis, due to the emission of high-impact substances like the VCM within PVC production plants [6,21], while the high risk of accidents has been studied using HAZOP approach [22]. On the other hand, research focuses on modelling typical risks of the PVC suspension process, such as material leakage (VCM) [23] and uncontrolled reactions using a vector machine [24]. However, these proposed models focus on the reaction stage and overlook risks in other stages of the process. Additionally, the detected risks can only be controlled through layers of safety.

Hence, in this work, a risk assessment is carried out for the PVC suspension (s-PVC) process using the inherent safety analysis method through CAPE to identify critical sources of risk in the process, emphasizing aspects such as operating conditions, substance-related risks, equipment safety and process structure. This evaluation considers the reaction stages, VCM recovery, resin purification and PVC drying.

2. Materials and Methods

2.1. Process Description

Figure 1 shows the process flow diagram of an S-PVC production process; this topology was built from information found in the scientific literature and data from operative PVC plants [25,26,27]. In the polymerization stage, the liquid VCM (fresh and recycled) is polymerized into PVC (with approximately 85% conversion) in an agitated mixture containing demineralized water, polyvinyl alcohol (PVA) as a stabilizer and 3-hydroxy-1,1-dimethylbutane-2-ethyl-2-methylheptane peroxide as an initiator. The reaction is carried out at 70 °C and 10 kg-f×cm⁻², and it is highly exothermic, meaning excess heat must be removed. At the end of the reaction, a heterogeneous mixture, known as “slurry”, is produced, containing the suspended polymer, unreacted VCM, water, initiator, and stabilizer.

In the PVC purification stage, the unreacted VCM (approximately 95%) is purged from the slurry by reducing the pressure to 1.8 kg-f×cm⁻² in a flash tank. The remaining 5% is removed via a stripping column, using steam at 225 °C and 14 kg-f×cm⁻². The overhead stream is rich in VCM, and the bottom has less than 1 ppm of VCM, following international regulations [25]. The unreacted VCM from the purification stage enters into the VCM recovery system where a series of heat exchangers (cooling) and compressors condition the unreacted VCM, removing the water and condensing the monomer for recirculation.

In the drying stage, the water in the slurry is removed from the PVC particles. First, a centrifuge (1800 rpm) removes 75% of the water. The remaining 25% is removed using hot air at 250 °C in a rotary dryer [26]. To separate the dry polymer, a cyclone separator is employed at atmospheric pressure.

2.2. Process Inherent Safety Index

A safe process can be achieved through internal (inherent) and external means. Inherent safety analysis allows the identification of risks through the real measurable parameters related to the intrinsic properties of the chemical process and aims to prevent risks rather than controlling them, following the preventing principle of eliminating risks in the first place and reducing them to a minimum if the above is not possible; the goal is to achieve a process with a minimum level of risk [27]. Inherent safety factors can be divided into two groups: (1) factors based on chemistry and (2) aspects of process engineering. Then, the total Inherent Safety Index (${\mathrm{I}}_{\mathrm{T}\mathrm{I}}$) can be calculated as the sum (Equation (1)) of the Chemical Inherent Safety Index (${\mathrm{I}}_{\mathrm{C}\mathrm{I}}$) and the Process Inherent Safety Index (${\mathrm{I}}_{\mathrm{P}\mathrm{I}}$):

$${\mathrm{I}}_{\mathrm{T}\mathrm{I}}={\mathrm{I}}_{\mathrm{C}\mathrm{I}}+{\mathrm{I}}_{\mathrm{P}\mathrm{I}}$$

(1)

2.3. Chemical Inherent Safety Index

The Chemical Inherent Safety Index includes chemical factors that affect the safety of processes and is the sum of factors such as chemical reactivity, flammability, explosiveness, toxicity, and corrosivity of the chemical substances involved in the process (Equation (2)).

$${\mathrm{I}}_{\mathrm{C}\mathrm{I}}={\mathrm{I}}_{\mathrm{R}\mathrm{M},\mathrm{m}\mathrm{a}\mathrm{x}}+{\mathrm{I}}_{\mathrm{R}\mathrm{S},\mathrm{m}\mathrm{a}\mathrm{x}}+{\mathrm{I}}_{\mathrm{I}\mathrm{N}\mathrm{T},\mathrm{m}\mathrm{a}\mathrm{x}}+{\left({\mathrm{I}}_{\mathrm{F}\mathrm{L}}+{\mathrm{I}}_{\mathrm{E}\mathrm{X}}+{\mathrm{I}}_{\mathrm{T}\mathrm{O}\mathrm{X}}\right)}_{\mathrm{m}\mathrm{a}\mathrm{x}}+{\mathrm{I}}_{\mathrm{C}\mathrm{O}\mathrm{R},\mathrm{m}\mathrm{a}\mathrm{x}}$$

(2)

where ${\mathrm{I}}_{\mathrm{R}\mathrm{M},\mathrm{m}\mathrm{a}\mathrm{x}}$ and ${\mathrm{I}}_{\mathrm{R}\mathrm{S},\mathrm{m}\mathrm{a}\mathrm{x}}$ are the index for chemical reactivity for the main and secondary reactions, ${\mathrm{I}}_{\mathrm{I}\mathrm{N}\mathrm{T},\mathrm{m}\mathrm{a}\mathrm{x}}$ is the index for chemical interactions, ${\left({\mathrm{I}}_{\mathrm{F}\mathrm{L}}+{\mathrm{I}}_{\mathrm{E}\mathrm{X}}+{\mathrm{I}}_{\mathrm{T}\mathrm{O}\mathrm{X}}\right)}_{\mathrm{m}\mathrm{a}\mathrm{x}}$ is the index of hazardous substances, and ${\mathrm{I}}_{\mathrm{C}\mathrm{O}\mathrm{R},\mathrm{m}\mathrm{a}\mathrm{x}}$ is the subindex of corrosivity.

Chemical reactivity is based on the maximum heat index values of the primary and secondary reactions, calculated based on the reaction enthalpy ($∆{\mathrm{H}}_{\mathrm{f}}$) as seen in

Table 1. Determination of the reaction heat subindices.

Heat of Reaction	Score
Thermally neutral ≤ −200 J/g	0
Thermally neutral < −600 J/g	1
Moderately exothermic< −1200 J/g	2
Strongly exothermic < −3000 J/g	3
Extremely exothermic ≥ −3000 J/g	4

, where the most exothermic reaction receives a score of 4.

The chemical interaction subindex describes unintended reactions between the chemical substances present in the process area and the materials of the plant, like undesired reactions outside the reaction section such as fires, toxic gas formation, and explosions, among others, the scores for this subindex are shown in

Table 2. Determination of the chemical interaction subindex.

Chemical Interaction	Score
Heat formation	1–3
Fire	4
Fire Formation of harmless, nonflammable gas	1
Formation of toxic gas	2–3
Formation of flammable gas	2–3
Explosion	4
Rapid polymerization	2–3
Soluble toxic chemicals	1

.

The flammability subindex describes the tendency to generate flames and is determined by the flash or boiling point (in °C).

Table 3. Determination of the flammability subindex.

Flammability	Score
Nonflammable	0
Combustible (Flash point > 55 °C)	1
flammable (Flash point < 55 °C)	2
easy flammable (Flash point < 21 °C)	3
very flammable (Flash point < 0 °C and Flash point < 35 °C)	4

shows the range of the subindex and its score.

The explosiveness subindex showcases the proclivity of a substance (gas) to form an explosive mixture with the air and is obtained according to the difference between the upper and lower explosion limits of the same substance (UEL%–LEL%). The scores for establishing this subindex are shown in

Table 4. Determination of the explosiveness subindex.

Explosivness (UEL–LEL) % v/v	Score
Non explosive	0
0–20%	1
20–45%	2
45–70%	3
70–100%	4

.

The toxicity subindex expresses the risks of a substance to human health and is determined through the threshold limit value (TLV in ppm) as seen in

Table 5. Determination of the Toxic Exposure Subindex.

Toxic Limit (ppm)	Score
TLV > 10,000	0
TLV ≤ 10,000	1
TLV ≤ 1000	2
TLV ≤ 100	3
TLV ≤ 10	4
TLV ≤ 1	5
TLV ≤ 0.1	6

.

Lastly, the corrosivity subindex indicates the risks associated with materials’ destruction by corrosive substances and is based on the materials needed for the process equipment, as shown in

Table 6. Determination of the corrosiveness subindex.

Construction Material Required	Score
Carbon Steel	0
Stainless steel	1
Special material	2

.

2.4. Process Inherent Safety Index

The Process Inherent Safety Index expresses the safety of the process in terms of a unit or system. It corresponds to the sum of the subindices of inventory, temperature, pressure, equipment safety, and safe process structure (Equation (3)).

$${\mathrm{I}}_{\mathrm{P}\mathrm{I}}={\mathrm{I}}_{\mathrm{I}}+{\mathrm{I}}_{\mathrm{T},\mathrm{m}\mathrm{a}\mathrm{x}}+{\mathrm{I}}_{\mathrm{P},\mathrm{m}\mathrm{a}\mathrm{x}}+{\mathrm{I}}_{\mathrm{E}\mathrm{Q},\mathrm{m}\mathrm{a}\mathrm{x}}+{\mathrm{I}}_{\mathrm{S}\mathrm{T},\mathrm{m}\mathrm{a}\mathrm{x}}$$

(3)

where the inventory (${\mathrm{I}}_{\mathrm{I}}$), temperature (${\mathrm{I}}_{\mathrm{T},\mathrm{m}\mathrm{a}\mathrm{x}}$), and pressure (${\mathrm{I}}_{\mathrm{T},\mathrm{m}\mathrm{a}\mathrm{x}}$) are subindices of the process, along with equipment safety (${\mathrm{I}}_{\mathrm{E}\mathrm{Q},\mathrm{m}\mathrm{a}\mathrm{x}}$), and safe process structure (${\mathrm{I}}_{\mathrm{S}\mathrm{T},\mathrm{m}\mathrm{a}\mathrm{x}}$).

The inventory subindex indicates risks associated with maintaining a certain quantity of substances, and this subindex is determined for both the internal battery limits (ISBL) and external or outside battery limits (OSBL) by considering the flows within a specific time (tons per hour), and the scores are assigned accordingly

Table 7. Determination of the inventory subindex.

Inventory		Score
ISBL	OSBL
0–1 t	0–10 t	0
1–10 t	10–100 t	1
10–50 t	100–500 t	2
50–200 t	500–2000 t	3
200–500 t	2000–5000 t	4
500–1000 t	5000–10,000 t	5

.

The temperature subindex is an indicator of the thermal energy in the system and is calculated based on the highest temperature in the process (°C), as shown in

Table 8. Determination of the process temperature subindex.

Temperature	Score
<0 °C	1
0–70 °C	0
70–150 °C	1
150–300 °C	2
300–600 °C	3
>600 °C	4

.

The pressure subindex is used to determine risks associated with leakage rates in the event of potential energy loss affecting containment (bar), as can be seen in

Table 9. Determination of the process pressure subindex.

Pressure	Score
0.5–5 bar	0
5–25 bar	1
25–50 bar	2
50–200 bar	3
200–1000 bar	4

.

The equipment safety subindex measures the risks associated with specific pieces of equipment in the ISBL and OSBL. In

Table 10. The scores of the equipment safety subindex for ISBL.

Equipment	Score
Equipment handling nonflammable, nontoxic materials	0
Heat exchangers, pumps, towers, drums	1
Air coolers, reactors, high hazard pumps	2
Compressors, high hazard reactors	3
Furnaces, fired heaters	4

and

Table 11. The scores of the equipment safety subindex for OSBL.

Equipment	Score
Equipment handling nonflammable, nontoxic materials	0
Atmospheric storage tanks, pumps	1
Cooling towers, compressors, blowdown systems, pressurized or refrigerated storage tanks	2
Flares, Boilers and furnaces	3

, the range of scores for different equipment is shown, and the ISBL entails more risks than the OSBL due to the reduced amount space used, making a specific piece of equipment riskier.

The safe process structure subindex describes the associated risks from the system engineering perspective, assessing how well components (units or equipment) work together, and it is based on accident reports, considering the scores in

Table 12. Values of the safe process structure subindex.

Safety Level of Process Structure	Score
Recommended (safety, etc. standard)	0
Sound engineering practice	1
No data or neutral	2
Probably unsafe	3
Minor accidents	4
Major accidents	5

.

It is essential to clarify that the Inherent Safety Index (${\mathrm{I}}_{\mathrm{T}\mathrm{I}}$) calculations are based on the worst-case scenario; this approach allows one to identify the most critical sources of risk in the regular PVC suspension process, enabling the planning of means or methods for the elimination or minimization of said sources of risks. For the evaluation of process safety, it is necessary to obtain information about the parameters of the different substances present in the process, such as toxicity, flammability, and explosiveness. These data can be obtained from the safety data sheets of the substances. Additionally, process simulation provides data associated with operational conditions, such as pressure, temperature, inventory, and equipment safety. The data required for determining the subindex of infrastructure safety come from the safety reports of events associated with accidents in actual chemical processes. Lastly, recommendations can be made for the improvement of the safe performance of the process, but they are tied to techno-economic considerations.

Table 13. Process inherent safety subindexes and value range, adapted from [28].

Chemical Inherent Safety Index
Symbol	IRM,max	IRS,max	INT,max	IFL,max	IEX,max	ITOX,max	ICOR,max
Score	0–4	0–4	0–4	0–4	0–4	0–6	0–2
Process Inherent Safety Index
Symbol	INV	IT,max	IP,max		IEQ,max	IEQ,max	IST,max
Score	0–5	0–4	0–4		0–4 (ISBL)	0–3 (OSBL)	0–5

shows the safety assessment indicators and their score range, where a higher number indicates more significant risks [18], while an ${\mathrm{I}}_{\mathrm{T}\mathrm{I}}$ value less than 24 represents an inherently safer process.

3. Results

Figure 2 shows the chemical inherent safety subindexes quantified for the PVC suspension process. The polymerization of VCM is the main reaction and it is highly exothermic ($∆{\mathrm{H}}_{\mathrm{f}}=-1.600 \mathrm{k}\mathrm{J}/\mathrm{k}\mathrm{g}$), requiring constant refrigeration to prevent any accidents from uncontrolled reactions. Therefore, a score of 3 is assigned to the ${\mathrm{I}}_{\mathrm{R}\mathrm{M},\mathrm{m}\mathrm{a}\mathrm{x}}$ subindex and 0 is assigned to the subindex of the side reaction ${\mathrm{I}}_{\mathrm{R}\mathrm{S},\mathrm{m}\mathrm{a}\mathrm{x}}$.

On the other hand, in the process, high-risk chemical interactions exist like the chemical and thermal degradation of the VCM and PVC by free oxygen [29] or the reaction of VCM with atmospheric air, but the processes reviewed for this work have created mechanisms that reduce the occurrence of these phenomena. Nonetheless, without proper control, unwanted interactions can provoke accidents like fire or explosions; therefore, a score of 4 is assigned to the subindex ${\mathrm{I}}_{\mathrm{I}\mathrm{N}\mathrm{T},\mathrm{m}\mathrm{a}\mathrm{x}}$, and thermal analysis is recommended such as the calorimetric analysis performed by Huang et al. [30].

When determining hazardous substances, properties related to toxicity, explosiveness, and flammability were considered. These properties were obtained from safety data sheets and information from documents using databases such as OSHA, EPA, etc., [31]. Vinyl chloride was identified as the most hazardous substance in the process, scoring 11 for the subindex ${\left({\mathrm{I}}_{\mathrm{F}\mathrm{L}}+{\mathrm{I}}_{\mathrm{E}\mathrm{X}}+{\mathrm{I}}_{\mathrm{T}\mathrm{O}\mathrm{X}}\right)}_{\mathrm{m}\mathrm{a}\mathrm{x}}$. In comparison, although toxic and flammable, other substances in the process, such as the polymer, PVA, or the initiator, are in a liquid or solid state with high flash points (above 400 °C).

Table 14. VCM safety parameters [32].

Property	Value
Threshold limit value (ppm)	1
Explosivity (LEL-UEV%)	3.6–33
Flash point (°C)	−78
Boiling point (°C)	−13.5

LEL: lower explosive limit, UEV: upper explosive limit.

shows safety data of the VCM that come from safety databases, and parameters like toxicity, explosiveness, and flammability are considered.

The selected material for the PVC process is stainless steel—the specialized literature recommended it due to safety and quality control issues. The presence of water and VCM together with high temperatures and pressure [33,34] can put heavy strain on more susceptible materials like carbon steel or aluminum alloys. The quality requirement for the PVC resin needs materials that cannot alter the conditions of the polymer, such as its color, meaning that the material needs to be highly hygienic and non-reactive, like stainless steel. Therefore, a score of 1 is assigned to the subindex ${\mathrm{I}}_{\mathrm{C}\mathrm{O}\mathrm{R},\mathrm{m}\mathrm{a}\mathrm{x}}$.

Figure 3 shows the process safety subindexes for the PVC suspension process. For the inventory subindex, the ISBL and OSBL inventories of the process were estimated with the capacity of the equipment for one hour of operation. For the ISBL, the inventory resulted in a value of 1935.8 t/h, corresponding to a score of 5. The OSBL inventory was calculated based on the storage capacities of raw materials in the process—it was 37.5 t/h, with a score of 1. Therefore, a score of 5 was assigned to the inventory subindex, as it represents the highest-risk case.

For the temperature safety subindex, ${\mathrm{I}}_{\mathrm{T},\mathrm{m}\mathrm{a}\mathrm{x}}$, the maximum temperature recorded was 250 °C in the burner in the drying section, corresponding to a score of 2. On the other hand, for the pressure subindex, the maximum pressure occurs in the stripping column at 14 kg-f×cm⁻² (13.7 bar); the score is 1. For the quantification of the equipment safety subindex, the suspension PVC production process receives a score of 4 for ${\mathrm{I}}_{\mathrm{E}\mathrm{Q},\mathrm{m}\mathrm{a}\mathrm{x}}$; this score comes from equipment categorized as insecure, with the boilers, heaters and compressor being the riskiest, all of which are present in the PVC suspension process.

Finally, the index of safe process structure can be quantified based on a local plant approach or a process approach. Accident reports serve as a basis for determining this indicator, as they provide essential safety details associated with the operation of individual equipment and their interactions. The US Chemical Safety Board (CSB) report on a PVC plant accident in Illinois describes the vulnerability to sudden discharge of hazardous materials in the reaction area [35]. Similarly, the French Ministry of Environment report on an accident at a PVC plant in Dolé highlights the release of monomer through the reactor vacuum system [33]. In addition, Ogle et al. studied the explosion of a sludge tank in a PVC plant, detailing the overpressure effects on the monomer, including in storage tanks for the slurry produced from the reactors [34]. The unexpected release of monomers into the surroundings is identified as the leading cause of accidents, along with equipment operating at high temperatures and pressures being close to the reactors. These reasons categorize the process as mostly unsafe, since using a high quantity of monomers and other highly flammable substances throughout the process can provoke explosion- or fire-related accidents. As a result, this process receives a score of 3, and it is noteworthy that most of the accidents recorded after 2000 are attributed to human-related errors. The score of the process safety index was 15, as shown in Figure 2. From this indicator perspective, inventory and equipment safety are the main process safety risks. Unlike the secure process structure, the process’s pressure and temperature conditions do not present significant risks.

4. Discussion

Figure 4 shows the PVC suspension production process has a total Inherent Safety Index value of 34, indicating a negative performance regarding process safety, pointing out that the process presents important sources of risk above the standard (24). The Chemical Safety Index had a value of 19, and the presence of the VCM in the process is the main source of chemical risk due to its properties, as the monomer, normally existing as a gas, increases the risk of accidents. The same can be observed in acetic acid production via methanol carbonylation, where CO is the most dangerous substance, with an Hazardous Substance Index (${\left({\mathrm{I}}_{\mathrm{F}\mathrm{L}}+{\mathrm{I}}_{\mathrm{E}\mathrm{X}}+{\mathrm{I}}_{\mathrm{T}\mathrm{O}\mathrm{X}}\right)}_{\mathrm{m}\mathrm{a}\mathrm{x}}$) of 10 [36]. This is unfortunate due to the VCM being the main raw material in the process; this means that the only way to control the risks associated with the VCM is through safety mechanisms associated with strict process control, planning and management.

On the other hand, the presence of unsafe equipment and a large inventory are the main problems from the process safety side. The inventory of the process is 1935.8 t/h, which is very large compared to processes like butyl acetate (70 t/h), acetone (13 t/h), and benzene (93 t/h) production [37], but is standard in the PVC industry, as seen in the VCM production process, with an inventory close to the 300 t/h [38]. The unsafe equipment in the process is mostly related to heat exchanger units such as boilers and heaters that tend to be categorized as risky, but both of these are used with energy carriers such as water and air, which makes them safer, even if the scores say otherwise.

In terms of operating conditions, the process exhibits a good score thanks to its moderate to low operating conditions (temperature and pressure) at 250 °C and 13.5 bar, while the acetone production process has a higher temperature of 507 °C and the benzene production process has a pressure of 25 bar [38]. Meanwhile, the ethylene production process involves temperatures close to 300 °C and pressures of 2000 bar [39]. The Safety Structure Index of the PVC suspension process was determined as mostly unsafe, but this could be an overestimation since Carvalho et al. estimated the ${\mathrm{I}}_{\mathrm{S}\mathrm{T},\mathrm{m}\mathrm{a}\mathrm{x}}$ for the VCM production process as neutral [38]; at the same time, Lozano assigned a score of 0 to the acetone, butyl acetate and cumene production processes, arguing that all processes are proven technologies—however, accident reports point out that there is at least a source of concern regarding how the VCM is handled throughout some process units, but the risk is lower in certain parts of the process due to the removal of the VCM, such as the drying section.

For the PVC suspension process, the risks from the chemical side cannot be removed, only contained, while the safety process indexes are more approachable in terms of elimination. For the inventory, reduction is recommended; for unsafe equipment, the replacement of the most unsafe units is also highly recommended or moving them from ISBL to OSBL and isolating the sections where the VCM has a high presence. For the operating conditions such as temperature and pressure, moderation is suggested at least in riskier equipment like boilers and heaters. Moreover, economic analysis is recommended to study the feasibility of every alternative, while at the same time extending safety analysis into the later stages of development (detailed engineering).

5. Conclusions

The inherent safety methodology was applied to assess and determine the inherent risks of the PVC suspension production process. It was shown that the process has a negative performance in terms of inherent safety, with a total index of 34. The Chemical Safety Index and Process Inherent Safety Index each reached values of 19 and 15, respectively. Many risks are associated with the chemical reactivity subindex due to the highly exothermic reaction and the hazardous substance subindex, primarily caused by the VCM. Also, several pieces of unsafe equipment, such as furnaces, burners, and dryers, are present in the process. The operating conditions also pose significant risks (high temperature and mild pressure). To improve safety, reducing the inventory of VCM and relocating the most hazardous equipment from the ISBL to the OSBL are recommended. However, these alternatives should be studied in conjunction with technical–economic analyses. Additionally, a more robust safety analysis in the future should also include resilience and sensitivity analyses.