Experimental analysis of a novel helical air gap membrane distillation system

Membrane distillation presents one of the feasible solutions to fresh water problems. The present study aims to develop an innovative helical air gap membrane distillation (HAGMD) system and to analyze its behavior under different operating conditions. In this design the condenser is made up of a cylindrical copper tube with continuous helical fins over it, that increases the total available condensation area by almost 45% and enhances the overall heat transfer throughout the module. The presence of fins in the gap also reduces the total air gap width by almost 64% and therefore improves the flux production. A detailed experimental analysis is carried out for a better understanding of the underlying phenomenon. The effect of feed water temperature, feed flow rate, cold flow rate, coolant temperature and feed salinity on the performance of HAGMD is investigated experimentally. The analysis shows that the finned condenser results in very high flux. The maximum flux obtained from the system was 20 kg/m hr with feed of 5 g/liter salinity and a diving force temperature difference of 45 C.


GRAPHICAL ABSTRACT INTRODUCTION
The world is moving towards achieving more and more energy efficiency and reducing improper usage of resources.
With an increase in population growth, industrialization and urbanization it is evident that the availability of natural resources will be diminished which poses a severe problem to humanity. Water is one of the essential elements of life.
Due to the reasons mentioned above, the availability of potable water to many communities is reduced. Membrane distillation (MD) is a rate governed separation process that produces freshwater from a brackish or saline solution.
The number of studies on different membrane modules, membrane properties, membrane modification and overall membrane systems are already available in the literature (Eykens et al. ; Mahmoudi et al. a). Four basic con- The difference between them appears in the permeate side arrangement. Most studied and explored configuration out of all above is the direct contact MD (Wu et al. ) still, it has been shown that with renewable energy sources, AGMD is the most preferred configuration with which to go (El Amali et al. ). Some other new configurations are also being derived from these basic configurations and are studied nowadays. Permeate or water gap MD, material gap MD, conductive gap MD, vacuum-enhanced DCMD (VEDCMD), multi-effect MD and vacuum-multi-effect MD are some of the arrangements. Air gap MD offers an advantage in terms of lower internal conduction losses from the membrane to cold fluid, higher driving temperature difference, and ability to separate expensive solutes without dilution (Aryapratama et al. ). This process has excellent thermal efficiency when used with an internal heat recovery option and employing solar energy to provide thermal assistance would definitely help in establishing an energy efficient and cost-effective solution for potable water needs (Koschikowski et al. ). Several modifications are performed in the basic design of air gap MD to explore new configurations that can be more productive than the conventional AGMD system (Mahmoudi et  (WGMD) worked best and it could produce 800% more distillate flux than AGMD. They also tested the effect of gap width and showed that with an increase in gap width from 9 to 13 mm the flux increases by a minimum of 340%. The reason attributed to this is the availability of more surface area for the dissipation of latent heat of condensation (Francis et al. ). A novel conductive gap MD was developed by Swaminathan et al. (b). In this configuration the gap was filled with a conductive metal mesh along with the permeate. In this way the overall conductivity of the gap was increased, which resulted in higher thermal efficiency. The effect of permeate flow direction in the gap was also analyzed. It was asserted that the counter-flow direction of the permeate is the most suitable for higher energy efficiency (Swaminathan et al. b). Khalifa and colleagues analyzed the performance of multi-stage AGMD and WGMD under series, parallel and mixed stage connections; for the feed solution with a salinity of 61,400 μs/cm it was found that multistage WGMD was superior to that of AGMD. Out of all the three arrangements, the parallel flow was the best in terms of flux production (Khalifa & Alawad ). Another permeate gap MD (also known as WGMD) module in a plate and frame configuration was studied and analyzed theoretically and experimentally by Mahmoudi et al. (a). Most of the studies have preferred plate and frame configurations, as they are easy for fabrication, operation and maintenance. But recently a design flexibility in terms of cylindrical module AGMD was studied statistically and optimized for minimum energy consumption and maximum flux conditions by Shahu and colleages (Shahu & Thombre ). Another finned tubular module with vertical grooves over the metallic condenser was studied by Cheng and co-workers in small-scale and scaled-up configurations with or without the use of solar energy. They were able to achieve as high as 50 kg/m 2 per hour flux with 11 vertical grooves on a condenser with 5 lpm flow rates (Cheng et al. ). Therefore, it is clear from the literature that air gap and the condenser design can be one of the critical areas to improve the performance, however not many research studies are available concerning these issues. This motivated us to explore a different air gap and condenser design and to identify the improvements over the basic AGMD design.
In the present study a cylindrical helical AGMD (HAGMD) system was developed that resembles a shell and tube heat exchanger in its design. It has a hollow copper condenser with helical fins over it. Copper was used as the condenser material here because copper has very high thermal conductivity that can help to achieve higher thermal efficiency as well as higher flux. Its higher corrosion resistance properties make it a perfect option for using with brines. Besides this, copper is known for imparting beneficial antioxidant and antimicrobial properties and is an essential micronutrient for human health (Dietrich hour was achieved with this HAGMD system for a feed with 5 gm/liter salinity and for a diving force temperature difference of 45 C. The present experimental facility is not equipped with an energy recovery method, and therefore the present study is not intended for energy studies.

Design and principle of HAGMD system
The design of HAGMD module is similar to a shell and tube heat exchanger as shown in Figure 2. In this system, the shell forms a passage for hot feed flow and, condenser tube forms a passage for cold-water flow. The shell houses the condenser tube. Helical fins are provided on the cylindrical copper condenser tube and a flat sheet PTFE membrane is wrapped over the finned condenser tube. The height of the fins forms the air gap, the shape of the fins is trapezoidal and fin tips act as the support for the membrane. On one side of the membrane, there is a hot foods and on other side there is an air gap followed by a helical condenser.
The fins create a continuous helix on the condenser surface, therefore the total condenser length L is now divided into 'n' number of smaller length regions with number of smaller condensate films. As the length of condensate film L b is reduced, it ultimately increases the overall condensation heat transfer coefficient as per the Nusselt theory of condensation for vertical plate (Swaminathan et al. b).
It is assumed that all the 'n' sections will behave in the same manner as far as the heat and mass transfer is con- diffusion through the membrane is accelerated by this suction effect on the incoming vapors.

HAGMD module preparation and membrane arrangement
The dimensions of the HAGMD module are given in Table 1.
The membrane used in the present study is PTFE hydrophobic membrane in the form of a flat sheet, which is wrapped over the finned condenser tube. The length of the condenser tube is 500 mm. The length of condenser over which the membrane is wrapped is termed as the active length of the module and is 270 mm. The extra length of the condenser is given before and after the membrane to facilitate a fully developed flow at the membrane surface. The properties of the membrane used in this study are given in the Table 2.   Table 3.
After passing through the membrane, some vapors get condensed over the fin surface and remaining travel through the air gap and condense over the vertical wall of the condenser surface. Condensate formed by both ways gets merged and travels downwards through the continuous helical fins and leaves through the permeate tapping at the condenser's bottom.
The driving temperature difference (ΔT) in HAGMD is that between the membrane surface temperature at hot feed side (T fm ) and the permeate film temperature over the condenser surface (T film ), as shown in Figure 4. The hot feed and cold water, both travel upwards in parallel flows.    The space between the helical pitch provides enough space to accommodate the higher permeate coming from the membrane as well as permeate traveling from the fins above.
As the flux production rate changes according to the operating conditions, the resulting permeate accumulation inside the air gap also varies. This process results in different MD configurations to get established in the HAGMD system as discussed here.

HAGMD module configuration at lower permeate production
When the flux production rate is lower, such as that with lower driving force temperature difference and low flow rates, there may be some air always available in the upper portions of the air gap, but the lower most gaps will be mostly filled with permeate only. It is because, at the bottom, vapors from two regions become merged, that is the travels condensed vapors traveling downwards, as well as vapors diffusing through the membrane. However, the air gap region will be randomly occupied with air and water droplets along the length of helical fins. Therefore, it will be considered as a mixed HAGMD configuration. In this case the heat transfer through the gap may be higher than AGMD configuration but less than DCMD configuration and therefore the thermal efficiency may be lower than AGMD.
HAGMD module configuration at higher permeate production rate When the permeate production rate is high, such as with higher driving force temperature difference and higher flow rates, the air gap may be completely filled with the permeate and results in a conductive gap configuration (Swaminathan et al. b). The velocity of the condensate traveling downwards will be higher than that in the previous case. This process will further increase the heat and mass transfer coefficient in the air gap that facilitates more permeate production. The conductive heat transfer will also be higher due to the flowing permeate in the gap, which may increase energy consumption and reduce the system's thermal efficiency, if heat recovery is not used. Therefore, the thermal efficiency may be lower than AGMD. The presence of permeate in the air gap further facilitate increased area for the latent heat of condensation to dissipated and resulting in increased flux production. All these factors result in higher permeate production in the conductive HAGMD configuration.

Experimental methods
Numbers of experimental runs were performed to determine the performance of the HAGMD module under different operating conditions. The experimental setup is shown schematically in Figure 5. The whole setup is divided into three circuits or channels for better control and understanding: a feed channel, a coolant channel and a permeate channel.
The temperatures at the inlet and outlet of the HAGMD module were measured using PT-100 thermocouples having ±0.01 C accuracy. The flow rate in feed and coolant  Some make-up feed is added to the heater tank at regular intervals to maintain the feed supply's desired salinity. The saline feed water is prepared in the laboratory by adding the required amount of NaCl to tap water. The outer shell of the module is insulated using rock wool insulation to reduce the heat loss from the module to the environment.

Coolant channel
Due to easy availability, tap water is used as the coolant, and for this reason, the temperature of the coolant was not controllable. The temperature of the coolant was always maintained in the range of 28-30 C. The cold tap water is stored in a tank, and from there, it flows through a small centrifugal pump to the module. After the module, the cold water is sent to a small cooling tower where extra heat is rejected, and the cold water returns to its initial temperature. A small laboratory scale cooling tower was used for maintaining the coolant temperature within the considered range.
This cooled water is again sent to the cold-water tank, and this completes the coolant channel. Any loss of cold water is compensated by adding make-up water from the tap.

Permeate channel
The permeate formed over the condenser surface (finned surface of the tube) flows to the bottom of the tube, from where it is taken out and collected in a small container.
The copper condenser tube with fins and a hole for condensate removal is shown in Figure 6. The experiment is performed for a fixed time; the quantity of condensate collected in the container is weighed. The total flux is calculated using the following formula: where J is the total flux for that particular parameter set (kg/ m 2 .hr), Wis the weight of the permeate (kg), A m is the membrane area (m 2 ) and t is the total time (hr). The quality of the permeate is measured by checking its salinity with the help of a conductivity check apparatus (Systronics-306).

RESULTS AND DISCUSSION
A HAGMD module is developed and analyzed in the present study. The performance of the HAGMD system can be indicated by the flux as the end product. The current experimental setup is not designed with an energy recovery option and energy efficiency studies is not the scope of current study. The effect of different operating parameters on the flux production will now be discussed in detail.  ficients. When the FT was increased by 15% from 65 C to 75 C, the flux for 2 lpm FFR was increased around 30% from 9.80 kg/m 2 hr to 13.1 kg/m 2 hr, and when the FT was increased 1.7 times from 45 C to 75 C, the distillate flux was increased 3.3 times from 4.5 kg/m 2 hr to 14.85 kg/ m 2 hr. In both cases the temperature of cooling water is constant to 25 C. This shows that higher will be temperature difference higher will be the flux, and also that the flux

Salinity
The effect of salinity (S) on the distillate production rate for the different feed flow rates is shown in Figure 10.   Figure 12 shows the effect of coolant flow rate on the flux for different feed temperatures.
The coolant temperature was maintained constant at 29 C.
This figure shows that flux increases as the cold flow rate increases. Increasing the cold flow rate helps in faster heat removal because of increase in cold side heat transfer coefficient and thus maintains a higher driving force at any given inlet cold fluid temperature that increases the flux. From the figure it is clear that at higher feed temperatures the variation of flux with CFR is more than that at the lower values. The reason for this is that at lower feed temperatures the latent heat of condensation is higher than that at higher feed temperatures. For a given temperature difference the latent heat will be removed more effectively at higher cold flow rates. Additionally, provision of fins at the condenser surface presents more area for condensation and therefore more latent heat needs to be removed. Therefore, at higher cold flow rates the flux is higher as the heat is more effectively removed. Also, at higher feed temperatures the latent heat of vaporization required is less and therefore  The same comparison is shown in a bar graph using

CONCLUSIONS AND FUTURE SCOPE
A novel design of the HAGMD system is developed and ana-  and salinity of feed are analyzed on the system's performance.
Following conclusions can be drawn from the analysis: • The fins on the copper condenser tube are helical in a design that starts from the top and continues to the bottom. The membrane is flat sheet and is wrapped and pasted over the condenser and forms a cylindrical shape that rests over the fins. The condensing vapors travel through the helically finned path from the top toward the exit at the bottom. The provision of fins facilitates the smaller air gap width with a very large condensation area, that is always favorable for the higher flux hence improving performance of HAGMD system.
• The maximum flux obtained from the system is 20 kg/ m 2 .hr with feed of 5gm/liter salinity and with a driving force temperature difference of 45˚C, which is very high.
• The flux increases with an increase in FT feed flow rate, and cold flow rate, and reduces with an increase in salinity. The heat transfer resistance in air gap is reduced by employment of conductive fins and therefore the cold side flow rate helps to increase the heat and mass transfer coefficient that helps to increase the flux. It is suggested that to reduce the overall cost of the system, costly chilling units can be eliminated. The idea behind this is to result in higher average driving force by employing higher cold flow rates with normal room temperature cold water instead of that with lower cold flow rates and very low temperature cold water, as the advantage in flux overrides the cost of pumping in HAGMD systems.
• At lower permeate production rate, the module configuration is AGMD at the top half while conductive gap at the bottom half. This configuration is thus referred as a mixed HAGMD configuration. While at a higher distillate production rate, the HAGMD module behaves like a conductive HAGMD configuration.
The overall analysis reveals that the helical air gap MD system is very suitable for producing high distillate flux. An economic analysis and energy efficiency studies can be performed in future to further analyze the system's better feasibility.

DATA AVAILABILITY STATEMENT
Data cannot be made publicly available; readers should contact the corresponding author for details.