Membrane Distillation; an attractive technology for water recovery

1.0 Introduction to Membrane Distillation

The demand for water in domestic, agricultural and industrial sectors is continuously increasing due to overconsumption, unavailability of ground/surface water and severe climatic changes. Seawater desalination and wastewater regeneration are suitable methods for generating potable water. Membrane Distillation (MD) is an emerging technology explored in seawater desalination, sewage treatment and water recovery from industrial effluents (food, pharmaceutical, healthcare, etc.). The key advantages of the MD process are relatively lower energy costs (low-pressure operation as compared to RO, lower temperature operation as compared to multiple-effect distillation), theoretically 100% rejection of non-volatile solutes (high rejection of partially volatile solutes), and a microporous membrane with significantly lower membrane fouling [1].

The MD process consists of 2 streams - hot stream (feed) and cold stream (permeate) separated by a hydrophobic membrane. The temperature difference between the 2 streams results in a vapour pressure gradient across a membrane, which acts as the driving force. The microporous membrane permits the passage of the vapour molecules (that condenses in the permeate stream) and restricts the passage of the feed liquid (water in most cases) [2].

The main challenges associated with implementing the MD process include- intricate membrane design, membrane wetting/fouling, complex process design, high energy requirements and high operational costs (Fig. 1).

A large number of laboratory research is being done to enhance permeate flux, suppress membrane fouling and form anti-wetting superhydrophobic surfaces [3]. These challenges are resolved by incorporating functionalized nanoparticles in the MD membranes. These nanoparticles are used as a surface modifier or incorporated into the membrane matrix. Silica nanoparticles, carbon nanomaterials (graphene, CNTs, quantum dots), metal/ metal-oxide, metal-organic frameworks (MOF) are some of the nanomaterials that have been investigated.

In most cases, the nanomaterials are coated with low surface energy fluoro-silanes such as Perfluoro dodecyl trichlorosilane (FTCS), Perfluoro octyl triethoxysilane (FTES/PFTS) either pre or post membrane fabrication. [4]. In recent years, the research is focused on creating new generation nanomaterials for improving the membrane performance in the MD process.

2.0 Membrane configurations and advantages of the MD process

The below table discusses various MD configurations that have been used to treat the aqueous feed solution, desalination and wastewater treatment using a microporous hydrophobic membrane.

The first patent for membrane distillation was obtained in 1963, and the first research publication was in 1967 [10]. However, there are very few large-scale MD plants that are operational worldwide. Thermodynamically, compared to an RO process, MD requires much lower amounts of electricity, whereas the thermal energy requirement is nearly 300-fold to desalinate the same amount of water.

MD is far superior as opposed to competing technologies due to the following reasons: (1) MD can achieve theoretically 100% rejection of all non-volatile components including inorganic ions, macromolecules, etc. (2) can handle feeds at higher temperatures (while operates at lower temperatures as compared to thermal desalination processes like evaporation) (3) lower operating pressures as compared to RO (Where applied pressure must be several times greater than osmotic pressure) (4) can handle high salinity feeds, (5) mechanical stability requirement is lower as compared to RO/NF membranes and (6) able to utilize waste-heat and low-grade energy sources [32].

The energy requirement for employing MD for desalination of highly-salty feeds with a latent heat recovery system lies somewhere in between RO and MSF. Further, MD requires minimum/ no pretreatment even for produced water [42], can utilize low-grade energy, is independent of feed concentrations and operates at lower temperature/pressure [43]. Low-grade energy requirement allows it to utilize renewable energy sources - solar, geothermal, tidal, wind, etc. or even co-locating the MD units close to industrial facilities and power generation systems to take advantage of the waste heat. Hybrid systems integrating MD with other separation methods are also being explored to improve water recovery [27].

For example, the cost of production of water using petroleum or electricity as the energy source in 10.8 USD/m³ using the AGMD configuration. This cost can be significantly decreased to 2.68 USD/m³ if an additional and affordable solar energy can be used, thus rendering the operating expenses equivalent to traditional desalination technologies [11]. The global interest in MD spiked in the last 10 years, as evidenced by the increase in research publications (Fig. 2).

3.0 MD for Wastewater Treatment

Polyvinylidene fluoride membranes (Millipore) could reject 99.7% of anionic and cationic dyes in DCMD experiments. The higher rejection and lowest flux decline ratio were observed for the Sodium Fluorescein (anionic dye) due to the interaction (repulsive) with the membrane surface. This avoided penetration into the pores, thus diminishing the effect of membrane fouling [12]. Polytetrafluoroethylene (PTFE) membranes were more robust and could handle industrial dyeing wastewater due to its superior hydrophobic character. The DCMD is a suitable alternative for removing COD and colour from dyeing wastewater with modest energy requirements and superior performance [13].

Excellent performance was reported using a modified stainless steel mesh/glass microfiber filter in the DCMD process to recover water from pre-treated petrochemical wastewater [14]. Omniphobic membranes prepared by spray coating silica nanoparticles and polystyrene microspheres and further making the surface hydrophobic by immersion in 2 wt. % 17-FAS. This rendered the surface omniphobic (contact angle of 176°/138.4° for water/hexadecane) [15]

The AGMD process could achieve removal of metals such as Si, Al, Cu [16], and toxic compounds such as tetramethylammonium hydroxide (TMAH) [17] from effluents of the nano-electronic industry. It was seen that pre-treatment steps were necessary to minimize membrane fouling. The achievable energy efficiency was 19%, and the estimated cost of treatment for TMAH loaded wastewater is 16 $/m³, which is significantly lower (80%) than the current disposal costs.

Most antibiotics cannot be metabolized in the human body, as a result, these organic contaminants end up in wastewater streams. Negatively charged PVDF membrane displayed a 100% rejection of negatively-charged cefotaxime (CTX) and neutral ciprofloxacin (CFX) with a water flux of ~ 19.7 LMH when tested in the DCMD mode [18]. Volatile components (trimethyl pyrazine, 2-acetyl pyrrole, phenethyl alcohol, and phenylacetic acid) and a non-volatile component (dibutyl phthalate) in the fermentation wastewater was transported to the permeate side after 12h run in the DCMD mode. This was predominantly due to the membrane wetting and fouling, which was reversible using an HCl/NaOH wash [19].

4.0 Hybrid Membrane Distillation Systems

To optimise water recovery and increase energy and cost efficiency, the scope of MD process configurations and their specific advantages provide several options for integrating and hybridising this process with other traditional and emerging membrane separation processes [21]. Various hybrid variants are available such as:

1. Reverse osmosis (RO-MD)

2. Forward osmosis (FO-MD)

3. Electrocoagulation (EC-MD)/ Electrodialysis (ED-MD)

4. Bioreactors (BR-MD)

5. Crystallisation (MDC-MD)

4.1. Reverse osmosis – MD hybrid system

The water recovery in RO is generally limited to 50% (a secondary stage at a higher operating pressure could increase the water recovery to 60%) [23]. One such way to overcome the problem was to use a hybrid RO-MD system proposed by Drioli et al. [24]. An increase in water recovery from 40% to 89% was obtained when seawater was used as the feed. Fig. 3 is the schematic of the hybrid system used to increase water recovery and reduce the environmental impact produced from the discharge [25]. Loss of vapour flux (due to the higher salinity) and membrane fouling (organic) were some of the shortcomings reported, and these could be overcome by pre-treatment of the RO wastewater using activated carbon. Moreover, the use of heat recovery from the pumps could reduce the overall thermal energy requirement and is a feasible way to achieve a near zero-liquid discharge [23].

4.2. Forward Osmosis (FO-MD)

Forward osmosis is an osmotic driven process that uses a high concentration draw solution to draw water from the feed solution across the semi-permeable membrane [26]. As this process ends with a diluted draw solution, an additional recovery such as MD, RO, or nanofiltration is required to obtain freshwater [26]. The MD system perfectly complements the FO process due to a higher factor ability to withstand high saline concentrations.

FO-MD systems are used for seawater desalination as well as effluent treatment. FO alone could recover only 95% Hg²⁺ (i.e. 1- 10 ppb gets transferred to the draw-solution) in a lab-scale study. However, coupling with the MD process could guarantee a 100% rejection of Hg²⁺. Thus, MD can also serve as an effective secondary process for eliminating organic and inorganic pollutants [28].

4.3. Electrodialysis (ED-MD)

Valuable products such as acids and metals can be recovered from industrial waste using the ED process. In ED-MD process, the ED system treats the acid content in the solution while the MD system was used to remove the cerium content and recover water [30].

4.4. Bioreactors (BR-MD)

This process is used to treat wastewater or biological wastes from food industries to remove highly volatile by-products [31]. Only volatile compounds pass through the membrane during MD separation, resulting in non-volatile feed materials like pathogens, organic matter, and salt being rejected.

The wastewater contaminants in the BR undergoes biodegradation due to heat formed in the MD [32]. Over 90-95% removal of organic compounds is achieved, leading to a reduction in chemical oxygen demand and total organic carbon [33]. From the research carried out, the BR-MD process could remove a wide variety of compounds (i.e. 30 compounds) in the pharmaceutical industry and obtained a 90-95% rejection due to the integration of MD process with BR.

4.5 Crystallisation (MDC-MD)

To eliminate wastewater discharge into the environment and to mitigate carbon, greenhouse gas emissions by treating industrial effluent are carried out using zero liquid discharge [34]. This is only achieved by evaporation of the wastewater using a heat source such as a crystalliser or large-scale concentrators. With MDC, a wide variety of water can be treated, especially brine solution, seawater and shale gas produced water at a larger scale. These techniques are also used to extract certain compounds like sodium chloride, sodium sulphate, lithium etc. Assembling a crystalliser downstream of MD decreased the fouling on the membrane [35]. MDC operating conditions need to optimise to avoid crystal deposition in the MD module and network.

5.0 Conclusion

This article explores MD as a suitable technology for desalination and wastewater treatment. Various membrane configuration, characteristics, state-of-the-art MD hybridisation techniques are focused, highlighting the advantages and potential applications. The MD technology is still at a nascent stage and extensive research and long term studies are required before the technology can be implemented on a large scale. A suitable membrane is one of the bottlenecks that limit the application of the MD process. The superhydrophobic microporous membranes resistant to organic and inorganic fouling can withstand high temperatures, and extreme conditions are required to overcome this limitation. Integrating the MD process with renewable energy sources can overcome the energy constraints, and MD technology can be developed to be sustainable stand-alone units.

Acknowledgements: The authors acknowledge the support from the SERB Start up Research Grant titled “ Development of Electrospun Membranes with graded hydrophobicity for Membrane Distillation” (Scientific Social Responsibility) to Dr. Noel Jacob Kaleekkal.

For article references kindly contact the author.