Energy aspects and cost analysis in direct contact membrane distillation using commercial PTFE membranes

Energy aspects and cost analysis in DCMD using different commercial PTFE membranes are experimentally examined under the impact of operating conditions and membrane pore sizes. Regarding energy efficiency (EE), gain output ratio (GOR), and specific energy consumption (SEC), feed inlet temperature is the most influential factor in comparison to feed concentration and volume flow rate. Regarding the membrane pore size, the DCMD system using a larger pore of membrane had a lower cost per cubic meter of freshwater production than the system using a smaller pore of membrane.

Therefore, the optimization of process parameters such as feed inlet temperature, feed concentration, volume flow rate, and thermal sources is involved.Solar photovoltaic thermal collectors are becoming increasingly common in desalination technologies (Maqbool et al., 2024;Al-Hrari et al., 2020;Anand et al., 2021;Choi et al., 2022).Thermal desalination techniques typically use between 40 to 80 kilowatt-hours per cubic meter for heating, along with 2.5 to 5 kilowatt-hours per cubic meter for auxiliary equipment.The average energy consumption of the widely used reverse osmosis (RO) desalination method is approximately 100 terawatt-hours per year, leading to emissions of 60 to 100 million metric tons of CO2 annually (Ahmed et al., 2019;Anand et al., 2021).Therefore, a robust solution is required to reduce both its energy consumption and the resulting emissions.
In addition, membrane configuration or membrane properties can contribute to a significant improvement in permeate flux (Al-Obaidani et al., 2008;Alkhudhiri et al., 2012).Furthermore, the membrane properties such as membrane thickness, membrane pore size, and membrane porosity also affected the permeate flux.The membrane thickness ranged from 50 to 180 µm (Phattaranawik et al., 2003;Termpiyakul et al., 2005).Additionally, a wider range of membrane thickness up to 1550 µm was also examined in other studies (Laganà et al., 2000).There are two opposite effects of membrane thickness relating to permeate flux issues.First, the membrane thickness should be minimized to reduce the resistance to the mass transfer; hence the permeate flux gets higher.However, the conductive heat loss across membranes rises in the case of thinner membranes.Therefore, the permeate flux drops because of the lower transmembrane temperature difference (Li et al., 2014).
In conclusion, the membrane thickness impacted permeate flux directly or indirectly.Another important factor influencing the permeate flux and energy efficiency is membrane materials and membrane pore size.The membrane materials and pore size affected the liquid entry pressure (LEP), and as a result, it influenced membrane penetration when the LEP was lower than the applied pressure (Sayed et al., 2024;Zare et al., 2024).Zare et al., (2024) create an affordable and highly hydrophobic ceramic membrane for use in DCMD applications.The contact angle and the LEP of the membrane were 1600 and 1.5 bar, respectively.According to the obtained findings, an average permeate flux of 3.15 kilograms per square meter per hour (kg/(m²•h)) and a salt rejection rate (R%) of 99.62% were observed for the 3.5% sodium chloride (NaCl) solution (Zare et al., 2024).Furthermore, the membrane pore size is the crucial factor to determine the primary mass transfer mechanism through membrane pores.
According to Wang and Chung, (2015), the ideal MD membrane pore size should be in the range of (0.1 µm -0.3 µm) to maximize freshwater production and energy efficiency because it reduces the temperature polarization effect and mass transfer resistance.If the surface tension of the solution is low, the role of membrane pore size is more important (Kimura et al., 1987).Generally, the previous studies' usual membrane pore sizes were from 0.2 µm to 0.45 µm to investigate the improvement in permeate flux and thermal efficiency (Courel et al., 2000;Durham and Nguyen, 1994;Kim et al., 2013).According to Li et al., (2014), both mass flux and thermal efficiency rose when the mean pore radius of the membrane increased, and the ideal values should be from 0.1 µm to 0.2 µm.
If the mean radius of membrane pores exceeds 0.2 µm, the growth of permeate flux and thermal efficiency is slow (Li et al., 2014).Furthermore, according to Ve et al., (2024a), the mass transfer mechanism should include the Poiseuille flow for the membrane having a pore diameter of 1 µm.This study aims to investigate the influence of operating conditions and pore size of commercial PTFE membranes on gain output ratio (GOR), energy efficiency (EE), and specific energy consumption (SEC).The cost analysis in terms of the unit cost for 1 m3 freshwater production is also calculated.

Mass transfer in DCMD
The relationship between permeate flux and vapor pressure difference on the membrane surfaces in DCMD can be expressed in ( ) Based on the experimental works from (Ve et al., 2024a;Ve et al., 2024b), the membrane permeability Cm should be calculated based on Ding's model in the case of (0.22 µm -1 µm) PTFE membrane (Ding et al., 2003), as shown in Equation (2):

Energy efficiency in DCMD configuration
In desalination plants, energy costs contribute to 30% -50% of the cost of freshwater production (Al-Karaghouli and Kazmerski, 2013).
In most studies, the energy aspects regarding EE, GOR, and SEC are commonly considered in the DCMD system (Khayet, 2013).As shown in Equation (3), the EE represents the proportion of heat utilized for evaporation compared to the overall heat transmission across the membrane: in the DCMD process at steady state conditions, as shown in Equation (4) (Gryta and Tomaszewska, 1998): ( ) Generally, SEC represents the total required energy for producing 1 m3 of freshwater from saline water (Khayet et al., 2011).The SEC values are much higher for laboratory-scale DCMD than for larger pilot plants.The specific energy consumption can be evaluated (Elmarghany et al., 2019): The GOR indicates how effectively the input energy generates freshwater within the DCMD module, as shown in Equation ( 6) (Khayet, 2013).For optimal performance in a DCMD configuration, the GOR value should be maximized. )

Experiment protocols
The DCMD system and two types of commercial PTFE membranes in this study were similar to previous studies (Ve et al., 2024a;Ve et al., 2024b).The feed inlet temperature, the feed and permeate volume flow rate, and feed concentration were in the range of (400C -500C), (0.017L.s-1 -0.03L.s-1), and (20000ppm -40000ppm), respectively.The temperature on the permeate side remains constant at 200C.The feed and permeate solution were pumped counter-currently.The commercially available PTFE membranes come with pore dimensions of 0.22 µm and 1 µm and a membrane porosity of 75%. Figure 1     Figure 2 and Figure 3 showed that the difference in EE and GOR under the effect of different membrane pore sizes was insignificant.The DCMD configuration using the PTFE1 membrane had nearly 6% higher EE and GOR than the DCMD configuration using PTFE022.Furthermore, the DCMD system using the PTFE1 membrane consumed more energy by almost 20% and 11% compared to the DCMD system using the PTFE022 membrane when the feed solution was 20000 ppm and 40000 ppm, respectively.

Effect of feed inlet temperature
In the DCMD configuration, temperature was one of the critical factors contributing to the significant fluctuation of energy issues.
Figure 4 shows that the GOR values rose significantly by up to 40% when the feed inlet temperature ranged between 40°C and 50°C for both DCMD configurations.Additionally, the improvement of EE was insignificant, with nearly 12.6%.According to the previous study (Deshmukh and Elimelech, 2017), the relationship between the partial vapor pressure difference and the temperature difference across the membrane increased as the average temperature at the membrane surface rose.Consequently, there was a considerable improvement in freshwater production.Moreover, latent heat accounted for (50% -80%) of the total energy for vapor evaporation through the membrane, and the remaining heat was conductive heat loss.At higher feed inlet temperatures, the conductive heat loss was less significant, leading to higher energy efficiency (Khayet et al., 2011;Phattaranawik and Jiraratananon, 2001).
In contrast, the SEC decreased considerably when the feed inlet temperature rose.Figure 5 shows that the SEC decreased by nearly 47% as the feed inlet temperature rose from 400C to 500C.The significant rise in freshwater production was the main reason for that massive drop in SEC.In conclusion, the feed inlet temperature had a more significant impact on EE, GOR, and SEC compared to the feed concentration.To significantly enhance permeate flux and energy efficiency, the DCMD system should be operated at the highest feasible temperature.
Regarding the influence of membrane pore sizes, there was an insignificant discrepancy in EE and GOR, whereas the fluctuation of SEC was considerable.Figure 4 indicates that the DCMD configuration with the PTFE1 membrane had slightly higher EE and GOR compared to the configuration with the PTFE022 membrane.However, the DCMD configuration with the PTFE1 membrane consumed about 20.1% more energy than the configuration with the PTFE022 membrane to produce freshwater at a feed inlet temperature of 50°C, as shown in (Figure 5).

Effect of volume flow rate
As shown in Figure 6b, there was a slight rise in EE when the volume flow rate increased from 0.017 L.s-1 to 0.03 L.s-1.Additionally, Figure 6 revealed the same pattern happening in SEC with substantial growth.The SEC increased up to under 30% for both DCMD configurations using different PTFE membranes.Conversely, the GOR dropped by nearly 14% in both DCMD configurations as the volume flow rate rose, as illustrated in (Figure 6a).The fluctuation of EE, GOR, and SEC depended on the insignificant improvement of permeate flux, the considerable increase in heat transfer rate, and the pressure drop when the volume flow rate increased, as shown in Eqs. ( 4), (7). Figure 6 demonstrates that the difference in EE and GOR for DCMD configurations using different PTFE membranes was insignificant, with a variation of less than 7%.However, the SEC for the DCMD system using the PTFE1 membrane was 20.1% higher than that for the DCMD system using PTFE022, as shown in (Figure 7).

Cost analysis
To evaluate the costs associated with the DCMD system using various commercial PTFE membranes, this study implemented the method mentioned in previous studies (Ali and Hassan-Ali, 2023; Qasem and Zubair, 2019).Based on other studies noted in Table 1, certain assumptions were proposed to estimate the freshwater production cost.In addition, Table 2 summarizes the total capital investment cost for the entire DCMD system.All capital expenditures were initially calculated in Vietnamese currency (VND) and subsequently converted into US dollars ($) using the exchange rate as of May 22, 2024.To assess the economic efficiency of the DCMD system, it is essential to analyse the cost per cubic meter of freshwater production (Cfw) over the assumed lifespan of the plant and the permeate flow rate (Jp).The process for calculating Cfw was detailed in Table 3, and Jp was determined using Eq. ( 16).

Factor Equation
Capital recovery ratio Annual capital cost ($/Year) Annual power cost ($/Year) Labor cost ($/Year) Maintenance cost ($/Year) Management cost ($/Year) Total freshwater cost ($/Year) Cost of freshwater ($/m3) Figure 8 shows that the Cfw dropped when the feed inlet temperature and volume flow rate increased.The cost per cubic meter of freshwater production decreased remarkably by nearly 50% with increasing feed inlet temperature.However, there was a notable rise in Cfw when the volume flow rate increased.In contrast, as shown in Figure 9, when the feed concentration increased from 20000 ppm to 40000 ppm, the Cfw rose insignificantly by nearly 2%.The cost per cubic meter of desalinated water for the DCMD configuration using the PTFE1 membrane was lower than that for the DCMD configuration using the PTFE022 membrane.At the same operating condition (Tfi=500C; Vf=Vp=0,03L.s-1;Sf=20000 ppm), the Cfw for the PTFE1 membrane in the DCMD system was nearly 14% lower than that for PTFE022 membrane, as shown in (Figure 8a).

CONCLUSIONS
Energy-related aspects including EE, GOR, SEC, and the cost per cubic meter of freshwater production were experimentally studied under various operating conditions and membrane pore sizes.The feed inlet temperature had the most significant influence on the EE, GOR, SEC, and Cfw, compared the volume flow rate and feed concentration.Although the specific energy consumption of the DCMD system using the PTFE1 membrane was higher than that of the DCMD system using the PTFE022 membrane, the Cfw for the former system was lower than that for the latter system.Therefore, incorporating renewable energy such as solar energy into the DCMD system should be considered to freduce the SEC and the cost per cubic meter of freshwater production.
All data associated with this study are present in the paper.

Nomenclature
distillation (MD) is a potential technology because it can integrate lowgrade heat sources or renewable energy into operation (Khayet, 2013; Baghbanzadeh et al., 2017).MD is more competitive than reverse osmosis (RO) which needs high energy requirements.Direct contact membrane distillation (DCMD) is a more attractive technology in terms of energy efficiency (Singh and Sirkar, 2012).The effect of experimental conditions such as feed inlet temperature, volume flowrates, and solution concentration on energy aspects is critically investigated (Singh and Sirkar, 2014; Dahiru and Atia, 2014; Levy and Earle, 1994; Manawi et al., 2014; Phattaranawik et al., 2001; Schofield et al., 1990; Taamneh and Bataineh, 2017).The energy requirements affected mainly the implementation feasibility of DCMD.It is the critical factor that impacts the freshwater production cost (Ali and Hassan-Ali, 2023).In comparison to RO, the permeate flux of DCMD is lower.

Figure 1
Figure 1 Experimental setup in DCMD configuration

Figure 8
Figure 8 Effect of (a) feed inlet temperature; (b) volume flow rate on Jp and Cfw in DCMD configuration

Figure 9
Figure 9 Effect of feed concentration on Jp and Cfw in DCMD configuration

Table 1
Cost calculation assumptions

Table 2
Total capital investment cost for DCMD