Introduction

The drastic increase in energy-related carbon dioxide emissions is alarming. Therefore, the development of renewable energy sources as alternatives to fossil fuels is of great concern. More usage of renewable energy is expected by 2050 (Capuano 2020). As a result, biodiesel is used extensively to replace fossil fuel. However, crude glycerol is generated as the primary by-product during biodiesel processing, which is approximately 10% of the biodiesel produced (Alves et al. 2020). Before crude glycerol can be marketed, it needs to undergo further purification. Treatment is required before the crude glycerol waste can be released into the environment. This purification and treatment processes can be expensive (Chilakamarry et al. 2021a). Generally, the waste stream of crude glycerol from biodiesel synthesis is disposed of by burning (Sun et al. 2016). Research reveals that crude glycerol waste can be utilized as feedstock for value-added products. One of the products of crude glycerol is dihydroxyacetone (DHA) (Chilakamarry et al. 2021a, b; Dennis et al. 2011; Ciriminna et al. 2018; Zheng et al. 2012; Liu et al. 2019, 2022; Black and Nair 2013; Ripoll et al. 2021; Carbajal et al. 2020). Repurpose crude glycerol waste into DHA offers environmental benefits, and it is a sustainable alternative to crude glycerol disposal. This is aligned with Sustainable Development Goal no. 12 on ensuring sustainable consumption and production patterns.

DHA is a 3-carbon ketose. Its major applications are for cosmetic, chemical, and nutraceutical purposes (Kimura et al. 1993; Fu et al. 2004) whereby the biggest market is in the cosmetic industry. In cosmetic applications, DHA is used as a color additive in sunless tanning products in the form of lotions, creams, mousses, sprays, and wipes (Ciriminna et al. 2018; Braunberger et al. 2018; Lidia et al. 2018). DHA reacts with keratin in the stratum corneum, which allows the user to achieve a tanned appearance for between two and six hours without exposure to the sun (Braunberger et al. 2018; Monfrecola and Prizio 2001). DHA was first commercialized in sunless tanner lotions in 1945 in California before debuting in the US cosmetic market in 1959 (Ciriminna et al. 2018; Nguyen and Kochevar 2003). DHA is also the only color additive approved by the Food and Drug Administration (FDA) as a color additive in sunless tanning products for external application (Fu et al. 2004). In chemical applications, DHA is used as a raw material for the synthesis of chemicals and pharmaceutical products such as methotrexate and 1, 2-propylene glycerol (Black and Nair 2013; Liebminger et al. 2014). Nonetheless, despite various uses of DHA, the primary and major application of DHA is as a color additive in sunless tanning products (Ciriminna et al. 2018; Braunberger et al. 2018; Lidia et al. 2018).

The market breakdown of DHA can be described according to the industry, geographical locations, and current key manufacturers (Industry_Research 2021). The industrial market share of DHA is currently governed by the cosmetic and medical industries. The key global market of DHA is dominated by five regions, namely North America, Europe, Asia–Pacific, South America, and the Middle East and Africa. In addition, the major global manufacturers of DHA are Merck KGaA originated from Darmstadt, Germany, Adina Cosmetic Ingredients originated from Kent, UK, and Hubei Marvel-Bio Medicine Co. Ltd. originated from Hubei, China (Adroit Market Research 2023).

The annual production rate of DHA is at least 2000 tons (Habe et al. 2010). DHA is estimated to have a selling price ranging from USD 20 000 to USD 100 000 per ton (Alibaba.com 2024; Made-in-China 2024). The global market size of DHA has been growing since 2011 from USD 135.1 million and projected to reach USD 219.8 million with an approximate compound annual growth rate of 5.59% by 2028 (Industry_Research 2021; Adroit Market Research 2023). With the cosmetic industry being the biggest market for DHA, the key driver for the market growth of DHA is the production of self-tanning products. Therefore, it is worth studying the production of DHA from waste crude glycerol from biodiesel production.

Some laboratory-scale experimental works related to the DHA synthesis from glycerol were available. Different types of DHA production techniques were scrutinized, including microbial fermentation (Black and Nair 2013; Ripoll et al. 2021; Lidia et al. 2018; Liebminger et al. 2014; Tan et al. 2019; Hu et al. 2010; Sudarshan and Sanjay 2018), direct oxidation of glycerol (Ulgheri and Spanu 2018; Choi et al. 2018), photocatalytic conversion of glycerol (Liu et al. 2019; Imbault et al. 2020), and indirect oxidation of glycerol (Zheng et al. 2012; Wang et al. 2013). These researches (Zheng et al. 2012; Liu et al. 2019; Black and Nair 2013; Ripoll et al. 2021; Lidia et al. 2018; Liebminger et al. 2014; Tan et al. 2019; Hu et al. 2010; Sudarshan and Sanjay 2018; Ulgheri and Spanu 2018; Choi et al. 2018; Imbault et al. 2020; Wang et al. 2013) focused on using different means to obtain higher glycerol conversion. Simulation-related study on DHA production was relatively lacking, and one of them was contributed by Carbajal et al. (Carbajal et al. 2020). Carbajal et al. (Carbajal et al. 2020) experimentally and numerically evaluated the economic viability of a biodiesel refinery with the introduction of a glycerol bioconversion for DHA generation based on the Mexican context.

To the best of knowledge, (1) process simulation of DHA production from glycerol is scarce, (2) there are limited studies on reviewing the feasibility of a DHA production plant, and (3) no in-depth economic analysis has been conducted on an industrial-scale DHA production from crude glycerol in Malaysia. Therefore, it was the aim of this project to design and perform a feasibility study on a DHA manufacturing plant with a production rate of more than 2000 tons per annum. Technical, safety, and economic assessment was carried out to scrutinize the viability of the production of DHA from crude glycerol. The results from this study will be beneficial for the optimization of the parameters in the DHA production process via simulation. In addition, the findings of this work can provide insights into how to promote a circular economy through glycerol waste valorization into a value-added product–DHA.

Methodology

Process of technology selection

The well-known technologies for the production of DHA from glycerol include direct and indirect oxidation of glycerol and the reaction of glycerol with acetaldehyde. In direct oxidation, DHA is either produced by the fermentation of glycerol with Gluconobacter oxydans or by oxidation of glycerol in the presence of oxygen and a metal catalyst such as iron zeolite. In indirect oxidation, glycerol undergoes a series of reactions such as acetalization, oxidation, and hydrolysis to produce DHA. Besides these conventional methods, there is also a recent patented technique that involves the reaction of glycerol and acetaldehyde to produce DHA. The evaluation of different DHA production techniques is presented in Table 1.

Table 1 Evaluation of alternative process technologies
Full size table

In terms of technical feasibility, microbial fermentation can produce DHA with a purity of > 97%. Although the purity is considerably high, the lower bound of DHA product purity using microbial fermentation is the lowest among the techniques. Nonetheless, the purity achieved is still high and there is a potential for the DHA produced to achieve higher purity. Furthermore, as the fermentation is conducted under a batch environment, it has the advantage to achieve increased and high flexibility in the process. The plant will be able to ferment batches of different sizes at a time and alter its supply according to the customer’s demand. It is also one of the most common techniques used for the commercial production of DHA from glycerol. The current status of the application gives high confidence in the reliability and maturity of this given technique. Lastly, the key technical hurdle associated with microbial fermentation is substrate inhibition. When the concentration of glycerol exceeds more than 20 g/L, substrate inhibition occurs which causes the productivity of the process to be reduced. Nevertheless, this challenge can be overcome by the implementation of a fed-batch reactor and enhanced process control. Glycerol concentration can be ensured to never exceed the inhibitory levels by periodically injecting the feed and extracting the product. Moreover, the operating temperature and pressure of microbial fermentation are all at ambient conditions which can significantly reduce any risk associated with fire and explosion (Morales et al. 2015).

The main concern of direct oxidation of glycerol over iron zeolite catalyst is that this technology is still at a preliminary designed level. Hence, complex interactions and scaling-up issues may occur during pilot plant production (Lari et al. 2016). Not only that, but its operating conditions are also the most extreme compared to the other techniques which resulted in its relatively poorer safety performance.

For indirect oxidation, the novelty of this technique reduces its reliability. It is still a laboratory-scale study that requires more research and development before moving into large-scale production. Moreover, there is a lack of information and assessment in terms of environmental sustainability, energy efficiency, and process data. Its reaction scheme is also complicated. Glycerol is first acetalized with benzaldehyde to produce the desired product HPD, 5-hydroxy-2-phenyl-1,3-diozane, and the undesired product HMPD, 4-hydroxymethyl-2-phenyl-1,3-diozane. The desired product will undergo oxidation and hydrolysis to produce DHA, whereas the HPMD has to be isomerized in an acid-catalyzed equilibrium to transform its ring to the structure of the desired compound (Zheng et al. 2012).

Production of DHA through the reaction of glycerol and acetaldehyde is a patented technique. Its technology is mature and reliable. However, the reaction involved is very complex compared to the other process technologies. Undesired side reactions may occur during the reaction of glycerol and acetaldehyde. There may be difficulties in selecting the most optimal catalyst for G-tone as there are a variety of catalysts which are available (Dennis et al. 2011). To successfully implement this technology in the plant, there will be additional expenses to hire a specialist and pay patent royalties to utilize this technology in the production plant. Hence, considering the complexities and higher costs, this technique is not the most favorable. After comparison and evaluation, microbial fermentation of glycerol is the most desirable DHA production method.

Process simulation

In this study, DHA was produced from glycerol via microbial fermentation using Gluconobacter oxydans. The block flow diagram of the synthesis process is illustrated in Fig. 1. The whole production was divided into two parts, consisting of upstream and downstream processes. The upstream processes included glycerol pre-treatment and DHA production while the downstream processes were crystallization and purification of DHA as well as the recovery of n-butanol.

Fig. 1
figure 1

Block flow diagram of DHA production from crude glycerol

Full size image

Fluid package selection

The DHA synthesis process from the crude glycerol was simulated using Aspen Plus version 12.1. The fluid package selected followed the Non-Random Two Liquid (NRTL) fluid package which was used to estimate the thermodynamic and transport properties of the fluid present in the system. NRTL is one of the most commonly used fluid packages to model the properties of polar mixtures (Morales et al. 2015). The components involved in this present system which include water, glycerol, acetone, methanol, n-butanol, and DHA are all polar components; hence, the selection of NRTL was suitable. Furthermore, the Aspen Property Package Selection Assistant was also used and NRTL was suggested to be adopted as well.

Chemical reaction equation

There were only two reactions occurring in the plant for the production of DHA from glycerol, which is summarized in Table 2.

Table 2 Reactions occurring in DHA synthesis process
Full size table

Basis and assumptions of simulation

The basis was sourced from the literature and was used for the simulation. Table 3 summarizes the basis of the simulation. Several assumptions were made for the simulation, and they are listed in Table 4.

Table 3 Basis for simulation
Full size table
Table 4 Assumptions made for Aspen Plus simulation
Full size table

Preliminary process hazards and safety review

The process hazards were assessed using the Hazard and Operability study (HAZOP). The HAZOP analysis is a systematic examination of planned or existing operations to identify and assess issues such as potential human error and external influences that may pose risks to the workers or the equipment. It is useful as a preliminary study to analyze risk-based decision-making while enhancing operational safety and availability.

The scope of the HAZOP for this study is limited to major equipments like DHA synthesis reactor (R-201), azeotropic distillation column (D-301), water distillation column (D-302), methanol evaporator (E-101), water evaporator (E-201), and acetone evaporator (E-301). For the equipment considered, the process parameters which are essential to the system safety like temperature, pressure, flow rate were identified. A set of keywords were adopted in order to find out the potential deviations from the intended operations of those equipment which could lead to hazards and undesirable conditions. Some guide words were applied to define each potential deviation. The causes and risks or consequences associated with the deviation in the operation of the considered equipment were then assessed, and potential risk mitigation techniques or safeguards were proposed. In addition, the risks of chemicals used in the production process were assessed using process material risk assessment.

Economic and sensitivity analysis

In the study, the economic analysis was carried out to determine the profitability and the economic sensitivity of the DHA production plant in Malaysia context. The economic analysis conducted was a preliminary estimate classified as Class IV, accounting for accuracy of ± 30%, and generally is used for the study of the feasibility of the project. The total cost of the proposed plant was estimated using the initial investment of resources and its annual operating cost which can be categorized into capital expenses (CAPEX) and operating expenses (OPEX).

The estimation of the capital expenses was conducted by the percentage of delivery equipment method based on the total purchased equipment cost of the overall project. This estimation method gives an accuracy of ± 30% and is sufficient for a preliminary estimation. The total purchased equipment cost followed the Chemical Engineering Plant Cost Index (CEPCI) in 2022 which was 816.0 (Maxwell 2024).

After obtaining the total purchased equipment cost, the capital expenses (CAPEX) were calculated. The estimation of CAPEX includes the consideration of the total fixed capital investment (TFCI), working capital investment (WCI), and contingency costs. TFCI covers the direct and indirect costs. On the other hand, the WCI covers any unexpected expenses or expansion of business and was assumed to be 15% of the TFCI. The contingency cost was taken to be 15% of the sum of the Inside Battery Limit (ISBL) and Outside Battery Limit (OSBL). The ISBL plant cost was estimated using the total fixed capital investment (TFCI), and OSBL was assumed to be 40% of the ISBL cost (Walas 1990; Towler and Sinnott 2013). Moreover, the total fixed operating cost (FOC), total variable operating cost (VOC), and total plant overhead cost (POC) were estimated to determine the total OPEX needed for the operation of the plant. The FOC comprised insurance, residual wages and salaries, taxation, labor and supervision, pre-operating cost, interest on utilized or drawn load, maintenance, site remediation, and depreciation. The VOC included costs that vary depending on the operating rate of the project such as raw material, utilities, effluent disposal, and transportation. The purchase price for the crude glycerol, Gluconobacter oxydans, n-butanol, and acetone was RM 1.47/kg (Wamucil 2024), RM 433.64/kg (Leibniz Institute 2020), RM 4.40/kg (Mike 2024), and RM 6.94/L (Warehouse 2024), respectively. The exchange rate of 1 US Dollar is taken to be approximately RM 4.73 (Xe Currency Converter 2024).

Finally, the POC involved the expenses for general and administration (G&A), selling and marketing (S&M), as well as research and development (R&D).

The plant service year was proposed to be 20 years. According to Inland Revenue Board of Malaysia, the corporate tax rate considered was 24% (Lembaga Hasil Dalam Negeri Malaysia 2024). Property tax was estimated to be 1–2% of the Inside Battery Limit (ISBL) fixed capital. Besides, the depreciation of the plant was estimated by the straight-line depreciation method with a rate of 3% annually (Walas 1990).

The economic sensitivity analysis of the project was also performed to determine and evaluate the profitability and economic feasibility of the plant. In this analysis, the DHA selling price was varied at ± 25% to represent a dynamic economic condition (Don and Marylee 2019). The return on investment (ROI) and payback period served as indicators for the analysis.

Results and discussion

Process simulation results

The process flow diagram of DHA production from glycerol via microbial fermentation is depicted in Fig. 2. The upstream process can be separated into two sections which are the pre-treatment of the feed and the DHA production section. First, the feed was fed from its storage tank (V-102) and was sent to the filter, F-101. The feed was crude glycerol and consisted of impurities such as ash, methanol, and water. Hence, it was necessary to perform pre-treatment before its introduction into the reaction. When crude glycerol from V-102 had been sent to F-101, ash present in the stream is removed and sent to the waste treatment through Stream 2. Stream 3A exited as the filtrate of F-101 was now glycerol without the presence of ash, and it was sent to HX-101. Stream 3A entering HX-101 was then heated from a temperature of 30–160 °C. The increase in temperature was to reduce the load of the stream on its next unit operation. Hence, Stream 3B which was the exiting stream from HX-101 was then fed into the evaporator E-101. E-101 assisted in the removal of most of the methanol and a part of the water in Stream 3B by operating at an operating temperature of 180 °C and a pressure of 1 bar. The methanol and water exited from E-101 through Stream 4 as vapor and were sent to the waste treatment. On the other hand, the purified glycerol with a minimal amount of impurities came out from E-101 as Stream 5A as a liquid. The pre-treatment of the feed ended at this unit, and the next section was the DHA production section.

Fig. 2
figure 2

Process flow diagram for a upstream and b downstream DHA production processes

Full size image

Stream 5A was sent to heat exchanger HX-102 to cool its temperature from 180 to 40 °C. Stream 5A was cooled before entering the reactor as the reaction worked in the presence of Gluconobacter oxydans which tend to experience thermal degradation at higher temperatures. Stream 5B then exited from HX-102 and was sent to a mixing point, M-101, whereby it was mixed with Streams 6A and 6B and fed to the reactor R-201. Stream 6A was the fresh nutrients and Gluconobacter oxydans from V-101 and Stream 6B was the recycled Gluconobacter oxydans. For the DHA production, Stream 8 from M-101 was fed to the reactor, R-201, along with oxygen in Stream 7 to allow the microbial fermentation process to occur. Once the reaction had achieved 85% conversion of the glycerol, DHA was formed along with glyceric acid and water. They left R-201 as Stream 9. Stream 9 was fed to the filter F-201. As Gluconobacter oxydans was not consumed in the reaction, it was removed by F-201 from Stream 9 and recycled back to R-201 in Stream 6B. The filtrate, which consisted of the products, by-products, and un-reacted reactant, was then exited F-201 as Stream 10 and entered HX-201. The temperature of Stream 10 was increased from 30 to101 °C. The increase in temperature was to pre-heat the stream before it entered the evaporator E-201 for water separation. The E-201 operates at 102 °C and 100 kPa. It removed the water, which was left as vapor in Stream 13. The DHA and its remaining components then exited E-201 as liquid in Stream 12. Stream 12 was then sent to a mixing point M-201 to the downstream process. In theory, the upstream process should be a batch process whereby M-201 is an equalizing tank ensuring that the downstream process is continuous. However, in the Aspen Plus simulation, it was assumed that the operating hours of 7920 h per year were sufficient to account for any shutdowns and batch scheduling (Walas 1990). Therefore, the upstream process was modeled as a continuous process with a flow rate that can meet the minimum production rate of 2000 tons DHA per year with the operating hours set.

In the downstream process, there are two sections which are the crystallization and purification of DHA and the recovery of n-butanol. N-butanol from V-103, Stream 12, and Stream 37 all entered M-201. Stream 12 was the product stream from the upstream process, whereas Stream 37 was the recovered n-butanol from the downstream processes. Stream 15 left M-201 and was sent to the D-301 which was an azeotropic distillation column. As n-butanol has a higher solubility in water than DHA, it formed a heterogenous azeotrope with water and removed water from DHA in D-301.

Crystallization and purification of DHA started from the bottom product of D-301. As DHA and glyceric acid were less volatile than the heterogenous azeotrope of n-butanol and water, they came out from the bottoms of D-301 as Stream 16. Stream 16 was then sent to the crystallizer C-301 whereby DHA solids were formed while glyceric acid and n-butanol remained in the mother liquor. The slurry was sent to filter F-301 whereby glyceric acid & n-butanol were recycled back to the system as Stream 20. DHA solids exited F-301 as Stream 19 and entered washer TE-301. Washer TE-301 was fed with acetone from Stream 22. Stream 22 was the combination of fresh acetone from V-301 and recycled acetone from Stream 31. TE-301 used acetone to remove any impurities from the DHA solids to produce extra pure DHA. Purified DHA left TE-301 as Stream 24, and it entered the dryer DE-301. DE-301 was fed with air from Stream 25. The air removed any moisture present in the DHA, effectively drying it into a powder form. The DHA powder was then stored in its respective storage tank. On the other hand, the acetone which was used in TE-301 came out as Stream 23. Stream 23 and Stream 26 (air) were fed to a mixing point, M-301, whereby Stream 27 was its exiting stream. Stream 27 was then fed to the evaporator E-301 which separated water and air from acetone. Methanol was recovered at the liquid outlet of E-301 as Stream 28, whereas water & air was removed from the vapor outlet as Stream 29. Stream 28 was recycled back into the system. The crystallization and purification of DHA ended at this unit.

The recovery of the n-butanol section started from the top product of D-301. The azeotropic mixture of n-butanol and water was sent to a decanter FL-301. At the decanter, the n-butanol-rich phase was recycled back to S-301 whereby it was fed back to D-301. The water-rich phase was sent to the distillation column D-302. At D-302, water was separated as the top product of D-302 and was sent to the waste treatment process. N-butanol was recovered at the bottom of D-302, and it would be recycled back to S-301 again to be fed back to D-301.

Process optimization

The objective of the optimization was to increase the profit of the plant by maximizing the production rate of DHA. The constraints for optimization were to ensure that the production rate of DHA was at least 2000 tons/year with 7920 operation hours/year. To maximize the production rate of DHA, sensitivity studies were performed on the operating conditions of the reactor, distillation column, and crystallizer. Based on the results obtained from the studies, the production rate of DHA is found to be not sensitive to these variables and has insignificant changes. However, when the flow rate of glycerol is varied from 400 to 800 kg/h, a peak in the profit is observed as shown in Fig. 3. The optimal flow rate is 480 kg/h of treated glycerol, which gives a total profit of RM 6728/h. To further optimize the profit, Steam 20 is recycled back into the system and this achieved an overall 17.62% increment in the DHA production rate. The optimized flow rate was used in the final simulation. After optimization, the production rate of DHA with a purity of 99.9% is found to be 3871 tons/year. The steam table obtained from the optimized simulation is presented in Table S1 of supplementary material.

Fig. 3
figure 3

Glycerol mass flow rate (kg/h) versus profit (RM/h)

Full size image

Utility usage summary

The utility used for heat transfer is shown in Table 5. The major utilities include high-pressure (HP) steam, low-pressure (LP) steam, and cooling water (CW). The total utility usage could be found by summing up the utility usage of each unit which uses the same type of utility. A total of 204.73 kg/h is used for HP steam, 3255.53 kg/h for LP steam, and 5498.26 kg/h for cooling water.

Table 5 Summary of utilities used
Full size table

Preliminary process hazards and safety review

Preliminary process hazards and safety reviews in terms of HAZOP are presented in Table S2 of supplementary material. The overall risk associated with the plant operations is tolerable as residual risks are assessed to be within the as low as reasonably practicable (ALARP) region. The highest risk among all the equipment evaluated is the risk of explosion or fire hazards which might destroy the sites and lead to injuries or fatalities. However, mild operating conditions and safeguards were placed to reduce the risk to a moderate rating.

Besides, equipment such as distillation column, reactor, and evaporator may cause overheating, underheating, or malfunctioning temperature controls. Hence, the thermocouple indicator and temperature alarm could be installed for the process control for prevention purposes. Moreover, the temperature detected must be regularly controlled by the output signal when the actual temperature deviates from the set point to activate other temperature-regulating devices to bring the equipment temperature back to the desired set point.

Diaphragm pressure indicators, pressure alarms, and vents could be installed to prevent the equipment from facing problems such as Gluconobacter oxydans being killed under high pressure above 220 MPa. When this happens, R-201 would most likely deteriorate with a high-pressure difference, and the safety valve would then be destroyed. Hence, operators must calibrate the alarm and indicators regularly and monitor the pressure detected to avoid pipeline blockage, pipeline leakage, and pressure controller.

Lastly, a fail-closed valve, one-way valve, and shock arrestor could also be installed in the process control system to avoid the potential occurrence of a clogged pipeline and leakage. The operators should also ensure that the emergency evacuation pathway is practical and that the process safety valve should have regular inspection and maintenance. They should also check whether the turbulent sources could be minimized and that the emergency shut-off valve would be activated to solve the potential problems.

The chemicals used were also reviewed. Process material risk assessment is presented in Table S3 of supplementary material. The chemicals involved in the process mostly have fire and explosion risks. Nonetheless, accidental release measures, handling, and storage measures are proposed to ensure that proper handling is performed to reduce the risk associated with the chemicals.

It is to be noted that the process hazards and safety review carried out for this work are merely a preliminary study with limitations. Not all the equipment or process involved in the DHA synthesis from glycerol is taken into account. The process should be divided into smaller nodes or units for further detailed analysis, and experts from multiple discipline (including commissioning, operations, engineering, maintenance, etc.) need to be engaged in the HAZOP team to ensure complete safety.

Results of economic and sensitivity analysis

The calculated total purchased equipment cost and the corresponding assumptions are stated in Table S4 of supplementary material. The total purchased equipment cost is approximately RM 33.65 million. From Table 6, the total fixed capital investment (TFCI) of the project is approximately RM 158.14 million which includes the total direct cost and total indirect cost. The working capital investment (WCI) is RM 23.72 million while the contingency is RM 33.21 million. The total CAPEX is estimated to be RM 215.07 million. The total operating cost (OPEX) is approximately RM 446.14 million which includes the total fixed operating cost (FOC), total variable operating cost (VOC), and the total plant overhead cost (POC) as shown in Table 7. The FOC, VOC, and POC are RM 85.66 million, RM 334.54 million, and RM 25.95 million, respectively.

Table 6 Total capital expenses (CAPEX)
Full size table
Table 7 Total operating expenses (OPEX)
Full size table

From Fig. 4, it is clearly noticed that the major expenses of the DHA production plant are on the OPEX. The CAPEX only contributes to 32% of the cost. The majority of the OPEX comes from the total variable operating cost, which is 51% out of the 68% of the total OPEX. Most of the VOC is spent on raw materials and utilities. Therefore, the type of feedstock for DHA production plays an important role in determining whether the plant is generating profit.

Fig. 4
figure 4

Percentage distribution of the total CAPEX and OPEX of DHA production plant

Full size image

As a result, the sensitivity analysis was carried out to examine the economic feasibility of the production by varying the DHA selling price. The DHA selling price is RM 124.16/kg (or USD 26.25/kg) for the worst case, RM 165.55/kg (or USD 35/kg) for the base case and RM 206.94/kg (or USD 43.75/kg) for the best case as shown in Table 8. After paying a corporate tax of 24%, the net earnings of the plant are estimated to be approximately RM 72.94 million in the worst case, RM 194.70 million in the base case, and RM 316.47 million in the best case. The cumulative cash flow diagram is plotted in Fig. 5. The negative value in the plant service year indicates the construction period. The payback periods are 2.95, 1.10, and 0.68 year for the worst case, base case, and best case, respectively, if the plant is operating at full capacity after the construction period. However, the plant is usually expected to operate at an initial capacity of 50% before gradually ramping up to 90.4% in 1.5 years (Braunberger et al. 2018); then, the payback periods for all three cases will be one to two years more than the one estimated in the sensitivity study. When the selling price for DHA is in the best condition, the return on investment is 147.04%, which is promising. Even if the DHA selling price is at the worst scenario, the return on investment is 33.80%. Therefore, the economic performance of the plant is highly viable. It is feasible to synthesize DHA from crude glycerol.

Table 8 Market analysis when DHA selling price is varied
Full size table
Fig. 5
figure 5

Cumulative cash flow diagram

Full size image

Conclusion

In this study, a feasibility study on the production of DHA from crude glycerol via microbial fermentation using Gluconobacter oxydans was performed. Optimization was carried out to maximize profit. For this, the n-butanol stream from the crystallizer was recycled back to the azeotropic distillation column and this promoted an overall 17.62% increment in the DHA production rate. The total production rate of DHA was 3871 tons/year from 6266 tons of crude glycerol. From the feasibility study carried out, the preliminary process hazards and safety were reviewed. The overall operation of the plant was relatively safe with risks ranging from moderate to low. With appropriate mitigation in place for process hazards as well as proper accidental release measure, storage, and handling of chemicals, the risks involved could be kept as low as reasonably practicable. From the economic study, it was revealed that the total capital investment and total operating expenses were RM 215.07 million and RM 446.14 million, respectively. As a whole, the economic performance of the plant was viable even when the DHA selling price was low (RM 124.16/kg). When the selling price of DHA was at the best condition (RM 206.94/kg), RM 801.06 million of annual revenue could be attained with a return on investment of 147.04%. Even though the DHA selling price might fluctuate, the payback periods for the investment were less than five years. Overall, the DHA production from microbial fermentation of glycerol is techno-economically feasible. DHA synthesis from crude glycerol offers an alternative to reuse the waste from biodiesel production, hence promoting responsible consumption and production.

It is recommended to carry out a detailed environmental sustainability study such as a life cycle assessment of the DHA production plant for future research. Moreover, the development of a forecasting economical model is another area of potential future work to provide more insights for potential investors.