water filter research paper

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Research Highlight
• Published: 11 September 2019

CLEAN WATER

Algae-based water filters

• Aimee Guha Roy 1  

Nature Sustainability volume  2 ,  page 788 ( 2019 ) Cite this article

4057 Accesses

1 Altmetric

Metrics details

• Biopolymers
• Developing world
• Water microbiology
• Water resources

ACS Sustain. Chem. Eng . http://doi.org/c9nj (2019)

Access to clean drinking water remains a problem in many developing countries, and affordable point-of-use water treatment is needed to reduce the spread of water-borne diseases. In Bangladesh, piped water is often contaminated and only available for a few hours a day. As such, it must be boiled prior to consumption, but this requires either fuel that may be unavailable to the poorest families or wood that may cause deforestation. One solution is water filtration, but abundant low-cost materials are required.

Tim Gainey / Alamy Stock Photo

Olof Gustafsson and colleagues from Uppsala University, Sweden and the University of Dhaka, Bangladesh made filter papers using nanocellulose from a locally available macroalgae called Pithophora . This robust algae thrives in nitrogen- and phosphorus-contaminated water and could be cultivated and harvested in large quantities in Bangladesh. The filter paper’s performance was measured using simulated wastewater containing latex nanobeads and in vitro viruses. The filters also showed complete removal of all infectious pathogens when tested on water samples collected from the Turag River and Dhanmondi Lake in Bangladesh. This study provides a way for locally grown algae to be converted into advanced filters for point-of-use water filtration.

Author information

Authors and affiliations.

Nature Sustainability https://www.nature.com/natsustain

Aimee Guha Roy

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Aimee Guha Roy .

Rights and permissions

Reprints and permissions

About this article

Cite this article.

Guha Roy, A. Algae-based water filters. Nat Sustain 2 , 788 (2019). https://doi.org/10.1038/s41893-019-0382-3

Download citation

Published : 11 September 2019

Issue Date : September 2019

DOI : https://doi.org/10.1038/s41893-019-0382-3

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing: Anthropocene newsletter — what matters in anthropocene research, free to your inbox weekly.

Microbiology and Drinking Water Filtration

Cite this chapter.

• Gary S. Logsdon 3  

Part of the book series: Brock/Springer Series in Contemporary Bioscience ((BROCK/SPRINGER))

886 Accesses

5 Citations

Water filtration research has been undertaken for a variety of reasons. Studies have been performed to develop information for filtration theories and for design of filtration plants to remove suspended matter such as clays, algae, suspended matter in general, and asbestos fibers from water. Filtration studies related to removal of microorganisms have generally been motivated by the need to learn about the removal of pathogens or indicator organisms, or both. Reducing the risk of waterborne disease has been a goal of microbiologically related filtration research for nearly 100 years. This chapter briefly reviews that research and then discusses the results of recent investigations.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

• Available as PDF
• Read on any device
• Instant download
• Own it forever
• Compact, lightweight edition
• Dispatched in 3 to 5 business days
• Free shipping worldwide - see info
• Durable hardcover edition

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Unable to display preview.  Download preview PDF.

Al-Ani, M.Y., Hendricks, D.W., Logsdon, G.S., and Hibler, C.P. 1986. Removing Giardia cysts from low turbidity water by rapid rate filtration. Journal American Water Works Association 78 (5): 66–73.

CAS   Google Scholar  

Al-Ani, M.Y., McElroy, J.M., Hibler, C.P., and Hendricks. D.W. 1985. Filtration of Giardia Cysts and Other Substances: Volume 3. Rapid Rate Filtration . EPA/600/2-85/027. U.S. Environmental Protection Agency, Cincinnati, Ohio.

Google Scholar  

U.S. Army and U.S. Public Health Service. 1944. Report 834. Efficiency of Standard Army Water Purification Equipment and of Diatomite Filters in Removing Cysts of Endamoeba histolytica from Water.

Arozarena, MM . 1979. Removal of Ciardia muris Cysts by Granular Media Filtration. Unpublished Master2019s Thesis, University of Cincinnati, Cincinnati, Ohio.

Baylis, J.R., Gullans, O., and Spector, B.K. 1936. The efficiency of rapid sand filters in removing the cysts of the amoebic dysentery organisms from water. Public Health Reports 51: 1567–1575.

Article   CAS   Google Scholar  

Bellamy, W.D., Silverman, G.P., Hendricks, D.W., and Logsdon, G.S. 1985a. Removing Giardia cysts with slow sand filtration. Journal American Water Works Association 77 (2): 52–60.

Bellamy, W.D., Hendricks, D.W., and Logsdon, G.S. 1985b. Slow sand filtration: influences of selected process variables. Journal American Water Works Association 77 (12): 62–66.

Brazos, B.J., O’Connor, J.T., and Lenau, C.W. 1987. Seasonal effects on total bacterial removals in a rapid sand filtration plant, pp 795–833 in Proceedings of the 1986 Water Quality Technology Conference . American Water Works Association, Denver, Colorado.

Brown, T.S., Malina, J.F., Jr. and Moore, B.D. 1974a. Virus removal by diatomaceous earth filtration-Part 1. Journal American Water Works Association 66 (2): 98–102.

Brown, T.S., Malina, J.F., Jr., and Moore, B.D. 1974b. Virus removal by diatomaecous earth filtration-Part 2. Journal American Water Works Association 66 (12): 735–738.

Bryck, J., Walker, B., and Mills, G. 1987. Giardia removal by slow sand filtration-pilot to full scale, pp. 49–58 in Proceedings AWWA Seminar on Coagulation and Filtration: Pilot to Full Scale , American Water Works Association, Denver, Colorado.

Bucklin, K., and Amirtharajah, A. 1988. The Characteristics of Initial Effluent Quality and Its Implications for the Filter to Waste Procedure . American Water Works Association Research Report. American Water Works Association, Denver, Colorado.

Chang, S.L. and Kabler, P.W. 1958. Removal of coxsackie and bacterial viruses in water by flocculation. Part 1. American Journal of Public Health 48: 51–61.

Cleasby, J.L., Hilmoe, D.J., and Dimitracopoulos, C.J. 1984. Slow sand and direct inline filtration of a surface water. Journal American Water Works Association 76 (12): 44–55.

Cullen, T.R. and Letterman, R.D. 1985. The effect of slow sand filter maintenance on water quality. Journal American Water Works Association 77 (12): 48–55.

DeWalle, F.B., Engeset, J., and Lawrence, W. 1984. Removal of Giardia lamblia Cysts by Drinking Water Treatment Plants. EPA-600/2-84-069. U.S. Environmental Protection Agency, Cincinnati, Ohio.

Edberg, S.C., Allen, M.J., and Smith, D.B. 1988. National field evaluation of a defined substrate method for simultaneous enumeration of total coliforms and Escherichia coli from drinking water: comparison with the standard multiple tube fermentation method. Applied and Environmental Microbiology 54: 1595–1601.

Fayer, R. and Ungar, B.L.P. 1986. Cryptosporidium spp. and Cryptosporidiosis. Microbiological Reviews 50: 458–483.

Fox, K.R., Miltner, R.J., Logsdon, G.S., Dicks, D.L., and Drolet, L.F 1984. Pilot plant studies of slow rate filtration. Journal American Water Works Association 76 (12): 62–68.

Guy, M.D., Mclver, J.D., and Lewis, M.J. 1977. The removal of virus by a pilot treatment plant. Water Research 11: 421–428.

Article   Google Scholar  

Hazen, A. 1913. The Filtration of Public Water Supplies, 3rd Ed . John Wiley & Sons, New York.

Heer, J.M. 1987. Turbidity Standards, Calibration, and Practice. Water World News March/ April 18–22.

Hendricks, D.W., Seelaus, T. Janonis, B., Mosher, R.R., Gertig, K.R., Williamson-Jones, G.L., Jones, F.E., Alexander, B.D., Saterdal, R., and Blair, J. 1988. Filtration of Giardia Cysts and Other Particles Under Treatment Plant Conditions . American Water Works Association Research Foundation Research Report, American Water Works Association, Denver, Colorado.

Hilmoe, D.J. and Cleasby, J.L. 1986. Comparing constant-rate and declining-rate filtration. Journal American Water Works Association 78 (12): 26–34.

Howell, D.G. 1987. The Evaluation of Montana Surface Water Treatment Plants for Their Ability to Remove Giardia . pp. 437–448. In Proceedings of the 1986 Water Quality Technology Conference , American Water Works Association, Denver, Colorado.

Huisman, L. and Wood, WE. 1974. Slow Sand Filtration . World Health Organization, Geneva, Switzerland.

Johnson, G. 1916. The typhoid toll. Journal American Water Works Association 3:249–326 and 791–868.

Jordan, H.E. 1937. Epidemic amebic dysentery-The Chicago outbreak of 1933. Journal American Water Works Association 29: 133–137.

Lance, J.C., Gerba, C.P., and Melnick, J.L. 1976. Virus movement in soil columns flooded with secondary sewage effluent. Applied and Environmental Microbiology 32: 520–526.

Lange, K.P., Bellamy, W.D., Hendricks, D.W., and Logsdon, G.S. 1986. Diatomaceous earth filtration of Giardia cysts and other substances. Journal American Water Works Association 78 (1): 76–84.

Logsdon, G.S. 1987. Evaluating treatment plants for particulate contaminant removal. Journal American Water Works Association , 79 (9): 82–92.

Logsdon, G.S. and Rice, E.W. 1985. Evaluation of sedimentation and filtration for microorganism removal, pp. 1177–1197 In: Proceedings of the 1985 American Water Works Association Annual Conference , American Water Works Association, Denver, Colorado.

Logsdon, G.S., Symons, J.M., Hoye, R.L. Jr., and Arozarena, M.M. 1981. Alternative filtration methods for removal of cysts and cyst models. Journal American Water Works Association 73: 111–118.

McConnell, L.K. 1984. Evaluation of the Slow Rate Sand Filtration Process for Treatment of Drinking Water Containing Viruses and Bacteria . Unpublished Master of Science Thesis. Utah State University, Logan, Utah.

McCormick, R.F. and King, P.H. 1982. Factors that affect the use of direct filtration in treating surface waters. Journal American Water Works Association 74: 234–242.

Mosher, R.R. and Hendricks. D.W. 1986. Rapid rate filtration of low turbidity water using field-scale pilot filters. Journal American Water Works Association . 78 (12): 42–51.

O’Connor, J.T., Brazos, B.J., Ford, W.C., Plaskett, J.L., and Dusenberg, L.L. 1985. Chemical and Microbiological Evaluations of Drinking Water Systems in Missouri: Summer Conditions, pp. 1199–1225 in: Proceedings of the 1985 American Water Works Association Annual Conference , American Water Works Association, Denver, Colorado.

Ongerth, J. 1988. A study of water treatment practices for the removal of Giardia lamblia cysts, pp. 171–175 in: Proceedings of Conference on Current Research in Drinking Water Treatment, EPA/600/9-88/004, U.S. Environmental Protection Agency, Cincinnati, Ohio.

Poynter, S.F.B. and Slade, J.S. 1977. The removal of viruses by slow sand filtration. Progress in Water Technology 9: 75–88.

Pyper, G.R. 1985. Slow Sand Filter and Package Treatment Plant Evaluation: Operating Costs and Removal of Bacteria, Giardia and Trihalomethanes . EPA-600/2-85-052, U.S. Environmental Protection Agency, Cincinnati, Ohio.

Rao, V.C., Symons, J.M., Ling, A., Wang, P., Metcalf, T.G., Hoff, J.C., and Melnick, J.L. 1988. Removal of hepatitis A virus and rotavirus by drinking water treatment. Journal American Water Worfes Association 80 (2): 59–67.

Robeck, G.G., Clarke, N.A., and Dostal, K.A. 1962. Effectiveness of water treatment processes for virus removal. Journal American Water Works Association 54: 1275–1290.

Sayre, I.M. 1984. Filtration. Journal American Water Works Association 76 (12): 33.

Seelaus, T.J., Hendricks, D.W., and Janonis, B.A. 1986. Design and operation of a slow sand filter. Journal American Water Works Association 78 (12): 35–41.

Slade, J.S. 1978. Enteroviruses in slow sand filtered water. Journal of the Institution of Water Engineers and Scientists 32: 530–536.

Slezak, LA. and Sims, R.C. 1984. The application and effectiveness of slow sand filtration in the United States. Journal American Water Works Association 76 (12): 38–43.

Spector, B.K., Baylis, J.R., and Gullams, O. 1934. Effectiveness of filtration in removing from water, and of chlorine in killing, the causative organism of amoebic dysentery. Public Health Reports 49: 786–800.

Stetler, R.E., Ward, R.L., and Waltrip, S.C. 1984. Enteric virus and indicator bacteria levels in a water treatment system modified to reduce trihalomethane production. Applied and Environmental Microbiology 47: 319–324.

Streeter, H.W. 1927. Studies of the Efficiency of Water Purification Processes. Public Health Bulletin No . 172 , U.S. Public Health Service, Government Printing Office.

Streeter, H.W. 1929. Studies of the Efficiency of Water Purification Processes. Public Health Bulletin 193 . U.S. Public Health Service, Washington, D.C.

Tanner, S.A. 1987. Evaluation of Sand Filters in Northern Idaho . Unpublished Master’s Thesis, University of Washington, Seattle, Washington.

Taylor, E.W 1969–1970. Forty-Fourth Report on the Results of the Bacteriological. Chemical and Biological Examination of the London Waters for the Years 1969–1970 . Metropolitan Water Board, London, England.

Walton, H.W. 1988. Diatomite filtration: how it removes Giardia from water, pp. 113–116 In: Wallis, P.M. and Hammond, B.R. (editors), Advances in Giardia Research , University of Calgary Press, Calgary, Alberta.

Download references

Author information

Authors and affiliations.

Environmental Protection Agency Cincinnati, Ohio, USA

Gary S. Logsdon

You can also search for this author in PubMed   Google Scholar

Editor information

Editors and affiliations.

Department of Microbiology, Montana State University, 59717, Bozeman, Montana, USA

Gordon A. McFeters

Rights and permissions

Reprints and permissions

Copyright information

© 1990 Springer-Verlag New York Inc.

About this chapter

Logsdon, G.S. (1990). Microbiology and Drinking Water Filtration. In: McFeters, G.A. (eds) Drinking Water Microbiology. Brock/Springer Series in Contemporary Bioscience. Springer, New York, NY. https://doi.org/10.1007/978-1-4612-4464-6_6

Download citation

DOI : https://doi.org/10.1007/978-1-4612-4464-6_6

Publisher Name : Springer, New York, NY

Print ISBN : 978-1-4612-8786-5

Online ISBN : 978-1-4612-4464-6

eBook Packages : Springer Book Archive

Share this chapter

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Publish with us

Policies and ethics

• Find a journal
• Track your research

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here .

Loading metrics

Open Access

Peer-reviewed

Research Article

Research on drinking water purification technologies for household use by reducing total dissolved solids (TDS)

Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

* E-mail: [email protected]

Affiliation Redlands East Valley High School, Redlands, California, United States of America

• Bill B. Wang

• Published: September 28, 2021
• https://doi.org/10.1371/journal.pone.0257865
• Reader Comments

This study, based in San Bernardino County, Southern California, collected and examined tap water samples within the area to explore the feasibility of adopting non-industrial equipment and methods to reduce water hardness and total dissolved solids(TDS). We investigated how water quality could be improved by utilizing water boiling, activated carbon and sodium bicarbonate additives, as well as electrolysis methods. The results show that heating is effective at lower temperatures rather than long boils, as none of the boiling tests were lower than the original value. Activated carbon is unable to lower TDS, because it is unable to bind to any impurities present in the water. This resulted in an overall TDS increase of 3.5%. However, adding small amounts of sodium bicarbonate(NaHCO 3 ) will further eliminate water hardness by reacting with magnesium ions and improve taste, while increasing the pH. When added to room temperature tap water, there is a continuous increase in TDS of 24.8% at the 30 mg/L mark. The new findings presented in this study showed that electrolysis was the most successful method in eliminating TDS, showing an inverse proportion where an increasing electrical current and duration of electrical lowers more amounts of solids. This method created a maximum decrease in TDS by a maximum of 22.7%, with 3 tests resulting in 15.3–16.6% decreases. Furthermore, when water is heated to a temperature around 50°C (122°F), a decrease in TDS of around 16% was also shown. The reduction of these solids will help lower water hardness and improve the taste of tap water. These results will help direct residents to drink more tap water rather than bottled water with similar taste and health benefits for a cheaper price as well as a reduction on plastic usage.

Citation: Wang BB (2021) Research on drinking water purification technologies for household use by reducing total dissolved solids (TDS). PLoS ONE 16(9): e0257865. https://doi.org/10.1371/journal.pone.0257865

Editor: Mahendra Singh Dhaka, Mohanlal Sukhadia University, INDIA

Received: June 22, 2021; Accepted: September 14, 2021; Published: September 28, 2021

Copyright: © 2021 Bill B. Wang. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the manuscript and its Supporting Information files.

Funding: The author received no specific funding for this work.

Competing interests: The authors have declared that no competing interests exist.

Introduction

The concentration of total dissolved solids(TDS) present in water is one of the most significant factors in giving water taste and also provides important ions such as calcium, magnesium, potassium, and sodium [ 1 – 3 ]. However, water with high TDS measurements usually indicates contamination by human activities, such as soil and agricultural runoff caused by irrigation, unregulated animal grazing and wildlife impacts, environmentally damaging farming methods such as slash and burn agriculture, and the overuse of nitrate-based fertilizer [ 4 , 5 ], etc. Around tourist areas as well as state parks, these factors will slowly add up over time and influence the water sources nearby [ 5 ]. Water that flows through natural springs and waterways with high concentrations of organic salts within minerals and rocks, or groundwater that originates from wells with high salt concentration will also result in higher particle measurements [ 6 ].

Water sources can be contaminated by substances and ions such as nitrate, lead, arsenic, and copper [ 7 , 8 ] and may cause many health problems related to heavy metal consumption and poisoning. Water reservoirs and treatments plants that do not consider water contamination by motor vehicles, as well as locations that struggle to provide the necessary components required for water treatment will be more prone to indirect contamination [ 9 – 11 ]. Many plants are effective in ensuring the quality and reduction of these contaminants, but often leave out the secondary considerations, The United States Environmental Protection Agency(US EPA)’s secondary regulations recommend that TDS should be below 500 mg/L [ 2 ], which is also supported by the World Health Organization(WHO) recommendation of below 600 mg/L and an absolute maximum of less than 1,000 mg/L [ 3 ]. These substances also form calcium or magnesium scales within water boilers, heaters, and pipes, causing excess buildup and drain problems, and nitrate ions may pose a risk to human health by risking the formation of N -nitroso compounds(NOC) and less public knowledge about such substances [ 12 – 15 ]. Nitrates can pose a non-carcinogenic threat to different communities, but continue to slip past water treatment standards [ 15 ]. Furthermore, most people do not tolerate or prefer water with high hardness or chlorine additives [ 16 ], as the taste changes tremendously and becomes unpreferable. Even so, TDS levels are not accounted for in mandatory water regulations, because the essential removal of harmful toxins and heavy metals is what matters the most in water safety. Some companies indicate risks in certain ions and alkali metals, showing how water hardness is mostly disregarded and is not as well treated as commercial water bottling companies [ 17 , 18 ].

In Southern California, water quality is not as well maintained than the northern counties as most treatment plants in violation of a regulation or standard are located in Central-Southern California [ 19 ], with southern counties having the largest number of people affected [ 20 ]. This study is focused on the Redlands area, which has had no state code violations within the last decade [ 21 ]. A previous study has analyzed TDS concentrations throughout the Santa Ana Basin, and found concentrations ranging from 190–600 ppm as treated wastewater and samples obtained from mountain sites, taking into account the urban runoff and untreated groundwater as reasons for elevated levels of TDS but providing no solution in helping reduce TDS [ 22 ]. Also, samples have not been taken directly through home water supplies, where the consumer is most affected. Other water quality studies in this region have been focused on the elimination of perchlorates in soil and groundwater and distribution of nitrates, but such research on chemicals have ceased for the last decade, demonstrated by safe levels of perchlorates and nitrates in water reports [ 23 , 24 ]. In addition to these studies, despite the improving quality of the local water treatment process, people prefer bottled water instead of tap water because of the taste and hardness of tap water [ 25 ]. Although water quality tests are taken and documented regularly, the taste of the water is not a factor to be accounted for in city water supplies, and neither is the residue left behind after boiling water. The residue can build up over time and cause appliance damage or clogs in drainage pipes.

This study will build upon previous analyses of TDS studies and attempt to raise new solutions to help develop a more efficient method in reducing local TDS levels, as well as compare current measurements to previous analyses to determine the magnitude to which local treatment plants have improved and regulated its treatment processes.

Several methods that lower TDS are reviewed: boiling and heating tap water with and without NaHCO₃, absorption by food-grade activated carbon [ 26 , 27 ], and battery-powered electrolysis [ 28 – 30 ]. By obtaining water samples and determining the difference in TDS before and after the listed experiments, we can determine the effectiveness of lowering TDS. The results of this study will provide options for residents and water treatment plants to find ways to maintain the general taste of the tap water, but also preserve the lifespan of accessories and pipelines. By determining a better way to lower TDS and treat water hardness, water standards can be updated to include TDS levels as a mandatory measurement.

Materials and methods

All experiments utilized tap water sourced from Redlands homes. This water is partially supplied from the Mill Creek (Henry Tate) and Santa Ana (Hinckley) Water Sheds/Treatment Plants, as well as local groundwater pumps. Water sampling and sourcing were done at relatively stable temperatures of 26.9°C (80.42°F) through tap water supplies. The average TDS was measured at 159 ppm, which is slightly lower than the reported 175 ppm by the City of Redlands. Permission is obtained by the author from the San Bernardino Municipal Water Department website to permit the testing procedures and the usage of private water treatment devices for the purpose of lowering water hardness and improving taste and odor. The turbidity was reported as 0.03 Nephelometric Turbidity Units (NTU) post-treatment. Residual nitrate measured at 2.3mg/L in groundwater before treatment and 0.2 mg/L after treatment and perchlorate measured at 0.9 μg/L before treatment, barely staying below the standard of 1 μg/L; it was not detected within post-treatment water. Lead content was not detected at all, while copper was detected at 0.15 mg/L.

For each test, all procedures were done indoors under controlled temperatures, and 20 L of fresh water was retrieved before each test. Water samples were taken before each experimental set and measured for TDS and temperature, and all equipment were cleaned thoroughly with purified water before and after each measurement. TDS consists of inorganic salts and organic material present in solution, and consists mostly of calcium, magnesium, sodium, potassium, carbonate, chloride, nitrate, and sulfate ions. These ions can be drawn out by leaving the water to settle, or binding to added ions and purified by directly separating the water and ions. Equipment include a 50 L container, 1 L beakers for water, a graduated cylinder, a stir rod, a measuring spoon, tweezers, a scale, purified water, and a TDS meter. A standard TDS meter is used, operated by measuring the conductivity of the total amount of ionized solids in the water, and is also cleaned in the same manner as aforementioned equipment. The instrument is also calibrated by 3 pH solutions prior to testing.All results were recorded for and then compiled for graphing and analysis.

Heating/Boiling water for various lengths of time

The heating method was selected because heat is able to break down calcium bicarbonate into calcium carbonate ions that are able to settle to the bottom of the sample. Four flasks of 1 L of tap water were each heated to 40°C, 50°C, 60°C, and 80°C (104–176°F) and observed using a laser thermometer. The heated water was then left to cool and measurements were made using a TDS meter at the 5, 10, 20, 30, and 60-minute marks.

For the boiling experiments, five flasks of 1 L of tap water were heated to boil at 100°C (212°F). Each flask, which was labeled corresponding to its boiling duration, was marked with 2, 4, 6, 10, and 20 minutes. Each flask was boiled for its designated time, left to cool under open air, and measurements were made using a TDS meter at the 5, 10, 20, 30, 60, and 120-minute marks. The reason that the boiling experiment was extended to 120 minutes was to allow the water to cool down to room temperature.

Activated carbon as a water purification additive

This test was performed to see if food-grade, powdered activated carbon had any possibility of binding with and settling out residual particles. Activated carbon was measured using a milligram scale and separated into batches of 1, 2, 4, 5, 10, 30, and 50 mg. Each batch of the activated carbon were added to a separate flask of water and stirred for five minutes, and finally left to settle for another five minutes. TDS measurements were recorded after the water settled.

Baking soda as a water purification additive

To lower scale error and increase experimental accuracy, a concentration of 200 mg/L NaHCO₃ solution was made with purified water and pure NaHCO₃. For each part, an initial TDS measurement was taken before each experiment.

In separate flasks of 1 L tap water, each labeled 1, 2, 4, 5, 10, and 30 mg of NaHCO 3 , a batch was added to each flask appropriately and stirred for 5 minutes to ensure that everything dissolved. Measurements were taken after the water was left to settle for another 5 minutes for any TDS to settle.

Next, 6 flasks of 1 L tap water were labeled, with 5 mg (25 mL solution) of NaHCO₃ added to three flasks and 10 mg (50 mL solution) of NaHCO₃ added to the remaining three. One flask from each concentration of NaHCO₃ was boiled for 2 mins., 4 mins., or 6 mins., and then left to cool. A TDS measurement was taken at the 5, 10, 20, 30, 60, and 120-minute marks after removal from heat.

Electrolysis under low voltages

This test was performed because the ionization of the TDS could be manipulated with electricity to isolate an area of water with lower TDS. For this test, two 10cm long graphite pieces were connected via copper wiring to a group of batteries, with each end of the graphite pieces submerged in a beaker of tap water, ~3 cm apart.

Using groups of 1.5 V double-A batteries, 4 beakers with 40mL of tap water were each treated with either 7.5, 9.0, 10.5, and 12.0 V of current. Electrolysis was observed to be present by the bubbling of the water each test, and measurements were taken at the 3, 5, 7, and 10 minute marks.

Results/Discussion

Heating water to various temperatures until the boiling point.

The goal for this test was to use heat to reduce the amount of dissolved oxygen and carbon dioxide within the water, as shown by this chemical equation: Heat: Ca(HCO 3 ) 2 → CaCO 3 ↓ + H 2 O + CO 2 ↑.

This would decompose ions of calcium bicarbonate down into calcium carbonate and water and carbon dioxide byproducts.

Patterns and trends in decreasing temperatures.

The following trend lines are based on a dataset of changes in temperature obtained from the test results and graphed as Fig 1 .

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

https://doi.org/10.1371/journal.pone.0257865.g001

To predict the precise temperature measurements of the tap water at 26.9°C, calculations were made based on Fig 1 . The fitting equations are in the format, y = a.e bx . The values for the fitting coefficients a and b, and correlation coefficient R 2 are listed in Table 1 as column a, b and R 2 . The calculated values and the target temperature are listed in Table 1 .

https://doi.org/10.1371/journal.pone.0257865.t001

Fig 2 was obtained by compiling TDS results with different temperatures and times.

https://doi.org/10.1371/journal.pone.0257865.g002

The fitting equations for Fig 2 are also in the format, y = a.e bx . The fitting coefficients a and b, and correlation coefficient R 2 values are listed in Table 2 . Based on the fitting curves in Fig 2 and the duration to the target temperature in Table 1 , We calculated the TDS at 26.9°C as listed in column calculated TDS in Table 2 based on the values we reported on Fig 2 .

https://doi.org/10.1371/journal.pone.0257865.t002

Based on the heating temperature and the calculated TDS with the same target water temperature, we obtained the following heating temperature vs TDS removal trend line and its corresponding fitting curve in Table 2 .

In Fig 1 , a trend in the rate of cooling is seen, where a higher heating temperature creates a steeper curve. During the first five minutes of cooling, the water cools quicker as the absorbed heat is quickly released into the surrounding environment. By the 10-minute mark, the water begins to cool in a linear rate of change. One detail to note is that the 100°C water cools quicker than the 80°C and eventually cools even faster than the 60°C graph. Table 1 supports this observation as the duration to target temperature begins to decrease from a maximum point of 94.8 mins to 80.95 mins after the 80°C mark.

As shown in Fig 2 , all TDS values decrease as the temperature starts to cool to room temperature, demonstrating a proportional relationship where a lower temperature shows lower TDS. This can partially be explained by the ions settling in the flasks. Visible particles can also be observed during experimentation as small white masses on the bottom, as well as a thin ring that forms where the edge of the water contacts the flask. When the water is heated to 40°C and cooled, a 3.8% decrease in TDS is observed. When 50°C is reached, the TDS drops at its fastest rate from an initial value of 202 ppm to 160 ppm after 60 minutes of settling and cooling. The TDS measurements in these experiements reach a maximum of 204 ppm at the 60°C mark. However, an interesting phenomenon to point out is that the water does not hit a new maximum at 100°C. meaning that TDS reaches a plateau at 60°C. Also, the rate of decrease begins to slow down after 20 minutes, showing that an unknown factor is affecting the rate of decrease. It is also hypothesized that the slight increase in TDS between the 5–20 minute range is caused by a disturbance in the settling of the water, where the temperature starts to decrease at a more gradual and constant rate. The unstable and easy formation of CaCO 3 scaling has also been the subject of a study of antiscaling methods, which also supports the result that temperature is a significant influence for scale formation [ 12 ].

In Table 2 , calculations for TDS and the time it takes for each test to cool were made. Using the data, it is determined that the test with 50°C water decreased the most by 16% from the initial measurement of 159ppm. This means that it is most effective when water is heated between temperatures of 40–60°C when it comes to lowering TDS, with a difference of ~7–16%. When water is heated to temperatures greater than 80°C, the water begins to evaporate, increasing the concentration of the ions, causing the TDS to increase substantially when cooled to room temperature.

Finally, in Fig 3 , a line of best fit of function f(x) = -0.0007x 3 + 0.1641x 2 –10.962x + 369.36 is used with R 2 = 0.9341. Using this function, the local minimum of the graph would be reached at 48.4°C.

https://doi.org/10.1371/journal.pone.0257865.g003

This data shows that heating water at low temperatures (i.e. 40–50°C) may be more beneficial than heating water to higher temperatures. This study segment has not been presented in any section within the United States EPA Report on water management for different residual particles/substances. However, warmer water temperatures are more prone to microorganism growth and algal blooms, requiring more intensive treatment in other areas such as chlorine, ozone, and ultraviolet disinfection.

Using the specific heat capacity equation, we can also determine the amount of energy and voltage needed to heat 1 L of water up to 50°C: Q = mcΔT, where c, the specific heat capacity of water, is 4.186 J/g°C, ΔT, the change in temperature from the experimental maximum to room temperature, is 30°C, and m, the mass of the water, is 1000 g. This means that the amount of energy required will be 125580 J, which is 0.035 kWh or 2.1 kW.

After taking all of the different measurements obtained during TDS testing, and compiling the data onto this plot, Fig 4 is created with a corresponding line of best fit:

https://doi.org/10.1371/journal.pone.0257865.g004

In Fig 4 , it can be observed that the relationship between the temperature of the water and its relative TDS value is a downwards facing parabolic graph. As the temperature increases, the TDS begins to decrease after the steep incline at 50–60°C. The line of best fit is represented by the function f(x) = -0.0142x 2 + 2.258x + 105.84. R 2 = 0.6781. Because the R 2 value is less than expected, factors such as the time spent settling and the reaction rate of the ions should be considered. To determine the specifics within this experiment, deeper research and prolonged studies with more highly accurate analyses must be utilized to solve this problem.

Boiling water for various amounts of time

Trend of boiling duration and rate of cooling..

Using the same methods to create the figures and tables for the previous section, Fig 5 depicts how the duration of time spent boiling water affects how fast the water cools.

https://doi.org/10.1371/journal.pone.0257865.g005

As seen in Fig 5 , within the first 10 minutes of the cooling time, the five different graphs are entwined with each other, with all lines following a similar pattern. However, the graph showing 20 minutes of boiling is much steeper than the other graphs, showing a faster rate of cooling. This data continues to support a previous claim in Fig 2 , as this is most likely represented by a relationship a longer the boil creates a faster cooling curve. This also shows that the first 5 minutes of cooling have the largest deviance compared to any other time frame.

The cooling pattern is hypothesized by possible changes in the orderly structure of the hydrogen bonds in the water molecules, or the decreased heat capacity of water due to the increasing concentration of TDS.

Effect on TDS as boiling duration increases.

In Fig 6 , all lines except for the 20-minute line are clustered in the bottom area of the graph. By excluding the last measurement temporarily due to it being an outlier, we have observed that the difference between the initial and final TDS value of each test decreases.

https://doi.org/10.1371/journal.pone.0257865.g006

Despite following a similar trend of an increase in TDS at the start of the tests and a slow decrease overtime, this experiment had an interesting result, with the final test measuring nearly twice the amount of particles compared to any previous tests at 310 ppm, as shown in Fig 6 . It is confirmed that the long boiling time caused a significant amount of water to evaporate, causing the minerals to be more concentrated, thus resulting in a 300 ppm reading. Fig 6 follows the same trend as Fig 2 , except the TDS reading veers away when the boiling duration reaches 20 minutes. Also, with the long duration of heating, the water has developed an unfavorable taste from intense concentrations of CaCO₃. This also causes a buildup of a thin crust of CaCO₃ and other impurities around the container that is difficult to remove entirely. This finding is in accordance with the introductory statement of hot boiling water causing mineral buildups within pipes and appliances [ 9 ]. A TDS reading of 300ppm is still well below federal secondary standards of TDS, and can still even be compared to bottled water, in which companies may fluctuate and contain 335ppm within their water [ 1 , 2 ].

This experiment continues to stupport that the cooling rate of the water increases as the time spent boiling increases. Based on this test, a prediction can be made in which an increased concentration of dissolved solids lowers the total specific heat capacity of the sample, as the total volume of water decreases. This means that a method can be derived to measure TDS using the heat capacity of a tap water mixture and volume, in addition to current methods of using the electrical conductivity of aqueous ions.

Adding food-grade activated carbon to untreated tap water

Fig 7 presents a line graph with little to no change in TDS, with an initial spike from 157 to 163 ppm. The insoluble carbon remains in the water and shows no benefit.

https://doi.org/10.1371/journal.pone.0257865.g007

The food-grade activated carbon proved no benefit to removing TDS from tap water, and instead added around 5–7 ppm extra, which settled down to around +4 ppm at 120 minutes. The carbon, which is not 100% pure from inorganic compounds and materials present in the carbon, can dissolve into the water, adding to the existing concentration of TDS. Furthermore, household tap water has already been treated in processing facilities using a variety of filters, including carbon, so household charcoal filters are not effective in further reducing dissolved solids [ 18 ].

Adding sodium bicarbonate solution to boiled tap water

As seen in Fig 8 , after adding 1 mg of NaHCO 3 in, the TDS rises to 161 ppm, showing a minuscule increase. When 4 mg was added, the TDS drops down to 158 ppm. Then, when 5 mg was added, a sudden spike to 172 ppm was observed. This means that NaHCO 3 is able to ionize some Ca 2+ and Mg 2+ ions, but also adds Na + back into the water. This also means that adding NaHCO 3 has little to no effect on TDS, with 4mg being the upper limit of effectiveness.

https://doi.org/10.1371/journal.pone.0257865.g008

To examine whether or not the temperature plays a role in the effectiveness in adding NaHCO 3 , a boiling experiment was performed, and the data is graphed in Fig 9 .

https://doi.org/10.1371/journal.pone.0257865.g009

Fig 9 presents the relationship between the amount of common baking soda(NaHCO₃) added, the boiling time involved, and the resulting TDS measurements. After boiling each flask for designated amounts of time, the results showed a downward trend line from a spike but does not reach a TDS value significantly lower than the initial sample. It is apparent that the NaHCO₃ has not lowered the TDS of the boiling water, but instead adds smaller quantities of ions, raising the final value. This additive does not contribute to the lowering of the hardness of the tap water. However, tests boiled with 5 mg/L of baking soda maintained a downward pattern as the water was boiled for an increasing amount of time, compared to the seemingly random graphs of boiling with 10 mg/L.

In some households, however, people often add NaHCO₃ to increase the pH for taste and health benefits. However, as shown in the test results, it is not an effective way of reducing TDS levels in the water [ 10 , 16 ], but instead raises the pH, determined by the concentration added. Even under boiling conditions, the water continues to follow the trend of high growth in TDS, of +25–43 ppm right after boiling and the slow drop in TDS (but maintaining a high concentration) as the particles settle to the bottom.

Utilizing the experimental results, we can summarize that after adding small batches of NaHCO3 and waiting up to 5 minutes will reduce water hardness making it less prone to crystallizing within household appliances such as water brewers. Also, this process raises the pH, which is used more within commercial water companies. However, the cost comes at increasing TDS.

Using electrolysis to treat TDS in tap water

Different voltages were passed through the water to observe the change in TDS overtime, with the data being compiled as Fig 10 .

https://doi.org/10.1371/journal.pone.0257865.g010

The process of electrolysis in this experiment was not to and directly remove the existing TDS, but to separate the water sample into three different areas: the anode, cathode, and an area of clean water between the two nodes [ 19 ]. The anions in the water such as OH - , SO 4 2- , HCO 3 - move to the anode, while the cations such as H + , Ca 2+ 、Mg 2+ 、Na + move to the cathode. The middle area would then be left as an area that is more deprived of such ions, with Fig 10 proving this.

As shown in Fig 10 , electrolysis is effective in lower the TDS within tap water. Despite the lines being extremely tangled and unpredictable, the general trend was a larger decrease with a longer duration of time. At 10 minutes, all lines except 10.5 V are approaching the same value, meaning that the deviation was most likely caused by disturbances to the water during measurement from the low volume of water. With each different voltage test, a decrease of 12.7% for 6.0 V, 14.9% for 9.0 V, 22.7% for 10.5 V. and 19.5% for 12.0 V respectfully were observed. In the treatment of wastewate leachate, a study has shown that with 90 minutes of electrical treatment, 34.58% of TDS content were removed, supporting the effectiveness of electricity and its usage in wastewater treatment [ 29 ].

This experiment concludes that electrolysis is effective in lowering TDS, with the possibility to improve this process by further experimentation, development of a water cleaning system utilizing this cathode-anode setup to process water. This system would be a more specific and limited version of a reverse osmosis system by taking away ions through attraction, rather than a filter.

The Southern Californian tap water supply maintains TDS values below the federal regulations. However, crystalline scale buildup in household appliances is a major issue as it is hard to clean and eliminate. To easily improve the taste and quality of tap water at home as well as eliminating the formation of scales, the following methods were demonstrated as viable:

• By heating water to around 50°C (122°F), TDS and water hardness will decrease the most. Also, the boiling process is effective in killing microorganisms and removing contaminants. This process cannot surpass 10 minutes, as the concentration of the ions in the water is too high, which poses human health risks if consumed. These, along with activated carbon and NaHCO₃ additives, are inefficient methods that have minimal effects for lowering TDS.
• Electrolysis is one of the most effective methods of eliminating TDS. Experiments have proven that increased current and duration of time helps lower TDS. However, this method has yet to be implemented into conventional commercial water filtration systems.

Also, some observations made in these experiments could not be explained, and require further research and experimentation to resolve these problems. The first observation is that TDS and increasing water temperature maintain a parabolic relationship, with a maximum being reached at 80°C, followed by a gradual decrease. The second observation is that when water is boiled for an increased duration of time, the rate of cooling also increases.

This experiment utilized non-professional scientific equipment which are prone to mistakes and less precise. These results may deviate from professionally derived data, and will require further study using more advanced equipment to support these findings.

Acknowledgments

The author thanks Tsinghua University Professor and PLOS ONE editor Dr. Huan Li for assisting in experimental setups as well as data processing and treatment. The author also thanks Redlands East Valley High School’s Dr. Melissa Cartagena for her experimental guidance, and Tsinghua University Professor Dr. Cheng Yang for proofreading the manuscript.

• View Article
• Google Scholar
• 2. United States Environmental Protection Agency. 2018 Edition of the Drinking Water Standards and Health Advisories Tables (EPA 822-F-18-001). US EPA, Washington D.C., USA, 2018; pp. 9–19.
• 3. World Health Organization. Guidelines for Drinking-Water Quality: Fourth Edition Incorporating the First Addendum. WHO, Geneva, Switzerland, 2017; pp. 7, 219–230, 423.
• PubMed/NCBI
• 6. Chloride, Salinity, and Dissolved Solids. 2019 Mar 1, [cited on 20 September 2020] Available from: https://www.usgs.gov/mission-areas/water-resources/science/chloride-salinity-and-dissolved-solids?qt-science_center_objects=0#qt-science_center_objects .
• 13. Shoukat, Ammara, Hussain, M., Shoukat, Asra. Effects of Temperature on Total dissolved Solid in water. Water Quality Study Conference, Mehran University Sindh, Pakistan, February 2020.
• 17. United States Environmental Protection Agency. Drinking Water Treatment Plant Residuals Management—Technical Report: Summary of Residuals Generation, Treatment, and Disposal at Large Community Water Systems. US EPA, Washington D.C., USA, 2011; pp. 177–182.
• 19. Exceedance and Compliance Status of Public Water Systems. 2012 Feb 1, [cited 22 September 2020] Available from: https://www.arcgis.com/apps/MapJournal/index.html?appid=143794cd74e344a29eb8b96190f4658b# .
• 20. 2019 Water Quality Status Report California. California Water Boards, 2019 July 1, [cited 22 September 2020] Available from: https://gispublic.waterboards.ca.gov/portal/apps/MapJournal/index.html?appid=6cde29ac0afc4d55b0fdaaae6bfc1aa4 .
• 21. City of Redlands—Water Quality Consumer Confidence Reports. City of Redlands, 2010–2020, [cited 23 September 2020] Available from: https://www.cityofredlands.org/post/water-quality .
• 25. Consumers’ Preference For Bottled Water Is Growing And They Want It Available Wherever Drinks Are Sold. 2019 Jan 8, [cited 23 September 2020] Available from: https://www.bottledwater.org/consumers’-preference-bottled-water-growing-and-they-want-it-available-wherever-drinks-are-sold .

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Membranes (Basel)
• PMC10051433

Existing Filtration Treatment on Drinking Water Process and Concerns Issues

Mashitah che razali.

1 Faculty of Electrical Engineering, Universiti Teknikal Malaysia Melaka, Hang Tuah Jaya, Durian Tunggal, Melaka 76100, Malaysia

2 Faculty of Electrical Engineering, Universiti Teknologi Malaysia, Johor Bahru 81310, Malaysia

Norhaliza Abdul Wahab

Noorhazirah sunar, nur hazahsha shamsudin, associated data.

Data sharing not applicable.

Water is one of the main sources of life’s survival. It is mandatory to have good-quality water, especially for drinking. Many types of available filtration treatment can produce high-quality drinking water. As a result, it is intriguing to determine which treatment is the best. This paper provides a review of available filtration technology specifically for drinking water treatment, including both conventional and advanced treatments, while focusing on membrane filtration treatment. This review covers the concerns that usually exist in membrane filtration treatment, namely membrane fouling. Here, the parameters that influence fouling are identified. This paper also discusses the different ways to handle fouling, either based on prevention, prediction, or control automation. According to the findings, the most common treatment for fouling was prevention. However, this treatment required the use of chemical agents, which will eventually affect human health. The prediction process was usually used to circumvent the process of fouling development. Based on our reviews up to now, there are a limited number of researchers who study membrane fouling control based on automation. Frequently, the treatment method and control strategy are determined individually.

1. Introduction

The quality of drinking water resources is being enthusiastically addressed around the world since it is essential to health and development issues. Due to uncontrolled industrial waste and low public awareness, water pollutants can be discharged either directly or indirectly to water resources such as lakes, ponds, rivers, seawater, and groundwater, which later become contaminated. The contaminated or poor quality of drinking water can cause various infectious diseases and negatively impact our overall health [ 1 ]. According to the World Health Organization (WHO), contaminated drinking water can cause serious diseases such as diarrhea, cholera, dysentery, hepatitis A, typhoid, and polio [ 2 ]. It is estimated that around 502,000 people die each year from diarrhea due to unsafe drinking water. The quality of water resources has been gradually depreciating due to industrialization and urbanization [ 3 ]. It has become a crucial problem due to the difficulty of meeting effluent quality standards with conventional treatment processes [ 4 , 5 , 6 ]. Good-quality drinking water helps people achieve maximum body health and well-being.

To obtain high-quality drinking water, a good and reliable water treatment process is desirable. Traditional drinking water treatment includes five common units such as coagulation, flocculation, sedimentation, filtration, and disinfection [ 7 , 8 , 9 ]. More than ten decades ago, the only treatment processes used in municipal and industrial water treatment were conventional filtration, such as clarification and granular media filtration, and chlorination methods. However, in the past twenty years, industrial water has shown high interest in the implementation of advanced water treatment technologies, particularly for water purification technologies such as membrane filtration, ultraviolet irradiation, the advanced oxidation process (AOP), ion exchange, and biological filtration for the removal of water contaminants in drinking water [ 10 ]. As of today, the latest water purification technologies are nanotechnology, acoustic nanotube, photocatalytic water purification, aquaporin inside TM , and automatic variable filtration. The use of technologies in water treatment is mainly due to three main reasons: a new standard for water quality, an increase in water contamination, and cost. Certainly, the new technology to be introduced should provide more advantages over the conventional treatment processes, such as lower operation and maintenance costs, being more efficient and simple to operate, having higher effluent quality and a high degree of reliability, having lower waste production, and most importantly, meeting regulatory requirements.

This paper identifies and reviews some of the available technologies ‘often’ used in drinking water treatment. Many of them are certainly not new to the water industry, but their application has been limited due to many circumstances, which are highlighted in this paper. This review focuses on membrane filtration technology and its application to municipal and industrial water systems. This review was motivated to establish an understanding of the related issues that come up in the drinking water treatment process.

We gather information about the available filtration systems, with a focus on the differences between conventional and membrane filtration, from these studies. Membrane filtration problems such as fouling phenomena, membrane cleaning, fouling prediction, and fouling prevention are discussed thoroughly. We also discussed the consequences of this review for the selection of a control strategy to overcome the problem in drinking water treatment, particularly due to the fouling phenomena. The goal is to organize and summarize most of the work and to identify the research focus and the trends in the literature on filtration treatment methods for drinking water processes.

2. Available Drinking Water Treatment Technologies

In general, the treatment technologies for treating water depend on the type of raw intake water that comes from various water sources, such as surface water and groundwater. The existing filtration treatments that are covered in this section are divided into conventional and advanced methods. Some of the available drinking water filtration treatment technologies, both conventional and advanced, as well as their concerns, are described.

2.1. Conventional Treatment

Conventional treatment is one of the popular approaches that has been used for water and wastewater treatment systems, where it involves several processes, including bar screening, grit removal, pre-oxidation, coagulation, flocculation, sedimentation, rapid/slow sand, granular active carbon filtration, and/or disinfection [ 11 ]. These processes can remove various solid sizes and organic matter from the liquid phase. It is also able to contribute to the reduction of microorganisms that cause concern for public health. There are several types of conventional filtration treatments, such as simple screen filters, slow and fast sand filters, diatom filters, and charcoal filters. The effect of filter media on the filtration process needs to be considered when designing the filtration unit. Additionally, the design of the backwash filter needs to be taken into account when high turbidity in effluent water increases head losses and requires long filtration operations [ 12 , 13 ].

Many studies have been performed to investigate the effectiveness of conventional filtration in treating drinking water. The previous study of the removal of diclofenac from drinking water is reported by Rigobello et al. [ 14 ], where the conventional sand filter is compared with granular activated carbon (GAC) filtration. The results showed that a sand filter could not effectively remove diclofenac, whereas a combination of a sand filter and GAC filtration could remove diclofenac with ≥99.7% efficiency. A slow sand filter and charcoal filter have been used in the study by Murugan and Ram [ 15 ]. The application of a slow sand filter can help in the reduction of water turbidity and prevent fouling at the reactor tubes. The charcoal filter is used to help in the absorption of heavy metals that are present in the water. In this work, slow sand filters require periodic removal of the microbial layer, while charcoal must be replaced in the filter every month as there are no indications that the charcoal has reached its breakthrough.

Zheng et al. [ 16 ] investigate the use of a slow sand filter as a pre-treatment for the removal of organic foulants in secondary effluent. The investigation was conducted with different filtration rates and showed that the proposed pre-treatment can effectively control the fouling rate at low filtration rates with respect to biopolymer removal and cycle time. Another study on the effect of a flow configuration based on a slow sand filter was performed by Sabogal-Paz et al. [ 17 ], where a comparison study was performed for the household system between intermittent and continuous flows. The authors observe that the flow configuration of a slow sand filter cannot be applied as a single treatment because it is not able to remove the organic foulants effectively. The work proposed by Ahammed and Darva [ 18 ] investigates the effect of a modified slow sand filter by introducing a thin layer of iron oxide-coated sand. The performance of the proposed method is measured based on its capability to remove bacteria and turbidity. Results showed that the modified slow sand filter was able to increase the removal rate of bacteria, but there was no significant reduction in turbidity. Work by Mizuta et al. [ 19 ] presents bamboo powder charcoal and activated carbon filtration in the removal of nitrate and nitrogen from drinking water. The results showed that bamboo powder charcoal filtration was able to provide higher adsorption and less influence on temperature compared to activated carbon filtration. Bamboo charcoal filtration was studied by Zhang et al. [ 20 ] to remove microcystin-LR from drinking water. In this study, bamboo charcoal filtration was modified with chitosan, and the results indicate that the applied treatment was able to effectively remove the microcystin-LR, especially when the amount of bamboo charcoal was increased.

Based on previous studies of conventional treatment methods, it is clear that the method is incapable of producing satisfactory effluent quality. Most of the treatments require either modification or combination with other methods, which is costly due to frequent maintenance. Moreover, this treatment is considered economically unbeneficial for developing countries [ 21 ], where the treatments require a long operating period and a large footprint [ 22 ]. Due to the importance of having safe and healthy water, water utilities have started to consider alternative treatment technologies to traditional drinking water treatment.

2.2. Advanced Treatment

Here, several advanced treatments of water technologies, particularly for water purification technologies such as membrane filtration, ultraviolet irradiation, the advanced oxidation process, ion exchange, and biological filtration, are discussed. Recently, membrane filtration is increasingly being accepted and implemented in drinking water treatment plants [ 23 ]. Membrane technology is widely used in filtration systems, particularly for the removal of particulate matter in solid-liquid separation processes [ 24 , 25 ]. Moreover, the combination of membrane technology with a bioreactor is called a membrane bioreactor, and this technology has proven its high capacity for the removal of pollutants in water and wastewater treatment processes [ 26 , 27 ]. The main issue in membrane filtration is the fouling phenomenon, which, if not prevented, will affect the overall filtration performance in the long run.

Another advanced technology that is primarily used in drinking water is ultraviolet (UV) irradiation technology [ 28 ]. UV irradiation is used as a disinfection process and is commonly designed with a series of UV lamps so that the microorganisms in the water will be inactivated when exposed to UV light [ 29 ]. Although UV irradiation is a promising disinfection technology due to its compactness and low cost, it faces a challenge due to its reliance on electrical component sensitivity [ 30 ], which can result in high failure rates.

The advanced oxidation process (AOP) is another technology generally applied in water treatment. The AOP includes several processes that produce hydroxyl radicals for the oxidation of organic and inorganic water impurities [ 31 ]. Among the three main AOP processes are ozone, ozone with hydrogen peroxide addition, and UV irradiation with hydrogen peroxide addition. Each of the processes has its challenges and will not be discussed in detail here. To summarize, AOP can provide multiple uses in water treatment, such as color, oxidation of synthetic organic chemicals, taste and odor, and many more. However, the complexity of AOPs in terms of chemical reactions between processes makes it hard to achieve an optimum treatment system design [ 32 ]. The next advanced water treatment is ion exchange (IX) technology. This technology was previously limited to only softening water for use in water treatment plants. However, the limits are now also being set on several inorganic chemicals, making the IX a more interesting technology to explore in water treatment applications. Lastly, biological filtration is another type of advanced treatment in water technology. The filtration is based on biological processes, which are different from the previously mentioned technologies that are based on physical and/or chemical processes. Works by Wang et al. [ 33 ] claim this biological filtration is the most effective process to produce biologically stable water. However, there are still unanswered issues regarding the proper design and implementation of biological filtration, particularly in terms of the size and type of filter media to be used. Figure 1 summarizes the conventional and advanced filtration methods for drinking water treatment.

Available treatment for drinking water [ 5 , 11 , 15 , 21 , 34 , 35 , 36 ].

2.3. Hybrid Treatment

In general, most industrial drinking water treatments still involve conventional and advanced treatment processes [ 8 ]. Figure 2 shows an example of industry-standard potable reuse water plants that involve conventional and advanced treatment processes [ 8 ]. In the primary treatment, the sedimentation of solid waste is performed. Water from secondary and tertiary treatment can be used for potable and non-potable reuse applications. The secondary treatment involves biological processes (e.g., the activated sludge process), and the tertiary treatment involves physical and/or chemical processes. For the disinfection process, chlorine is used to disinfect water to kill bacteria, parasites, and viruses in drinking water [ 37 ]. Alternatively, disinfectants such as chlorine dioxide, ozone, and ultraviolet radiation are also used. In advanced treatment, the integrated membrane system (IMS) and full advanced treatment (FAT) are implemented. The IMS uses a low-pressure membrane filtration process either microfiltration (MF) or ultrafiltration (UF). Meanwhile, FAT applies called either nanofiltration (NF) or reverse osmosis (RO), which are high-pressure membrane filtration processes. The application of IMS can provide high efficacy in the removal rate of particulate matter, microbial pathogens, and natural organic matter, whereas FAT is capable of removing magnificently organic–inorganic dissolved constituents such as salts and organic chemicals that are impossible to be removed by IMS. Ultraviolet and advanced oxidation processes act as post-treatment disinfection. In this stage, it will break down small neutral organic compounds that pass-through FAT. The final stage is known as degassing and lime dosing, which act as a water stabilizer and increase the pH and alkalinity of the water. The industry standard potable reuse water plant shown in Figure 2 can meet the specification for drinking water quality, but there are several drawbacks, including a large footprint, high capital cost, and high energy consumption, which make it essential to discover another technology that can overcome the drawbacks [ 38 ].

Industry-standard potable reuse plant.

The conventional design of the drinking water treatment process includes five common units, and four of them (coagulation, flocculation, sedimentation, and filtration) are the lines that remove suspended particles from surface water treatment plants. Filtration is the final step in the removal of suspended particles, and without it, the plants are considered untreatable. Therefore, proper control, design, and implementation of the filtration operation unit are crucial to improving the effluent quality and reducing the risk of waterborne diseases. The next section then focuses on a review of numerous types of membrane filtration technologies. The advantages and disadvantages of each type of filtration are also discussed, and this will provide some hints for researchers on how to choose the most suitable membrane filtration for their applications.

3. Membrane Filtration Technology

Membrane filtration is an advanced drinking water treatment that is widely used nowadays in water treatment processes, mainly for drinking water. Examples of types of membranes include microfiltration (MF), ultrafiltration (UF), nanofiltration (NF), reverse osmosis (RO), electrodialysis (ED), forward osmosis (FO), and membrane distillation (MD). Each method has its own specific range of membrane pore sizes, surface charge, and hydrophobicity that is produced from different materials [ 39 ]. Table 1 shows the pore size ranges of various membrane filtration systems as compared to the size of common water contaminants.

Contaminant with respective membrane filtration type.

The application of membrane filtration technology to drinking water treatment on a large-scale [ 40 ] has received attention due to its advantages, including excellent effluent quality [ 41 ], simple process management [ 42 ], and strict solid-liquid separation with a small footprint requirement [ 43 , 44 ]. The technology is also easy to adapt to the existing treatment facilities [ 45 ], provides low energy consumption [ 11 ], and removes various contaminants [ 46 ]. The removal rate of contaminants depends on the characteristics of the membrane and the properties of the contaminant [ 36 ]. Aside from these benefits, the main disadvantage of this technology is the cost of the membrane itself, which can be reduced or eliminated if the membrane filtration process is handled properly. Figure 3 shows the advantages and disadvantages of each membrane filtration treatment applied to drinking water treatment.

In general, membrane filtration can be classified into two categories: low-pressure membrane (10 to 30 psi) and high-pressure membrane (75 to 250 psi). The low-pressure membrane system includes MF and UF, while NF and RO are categorized as high-pressure membrane systems.

The low-pressure MF and UF membranes for the application of municipal surface water treatment have been studied and implemented since the 1980s. In these studies, the MF (nominal pore size of 0.2 mm) and UF (nominal pore size of 0.01 mm) have proven their high capabilities for the removal of particulate matter (turbidity) and microorganisms [ 47 , 48 ]. MF and UF membranes were proven to provide a barrier to microorganisms such as Giardia cysts and Cryptosporidium oocysts, while the UF was proven to be an absolute barrier to viruses due to its smaller pore size of 0.01 mm [ 49 , 50 ]. Previous studies [ 51 , 52 ] also demonstrated that low-pressure membranes were able to treat turbidity efficiently using pilot and full-scale plants. The low-pressure MF and UF membrane systems provide high performance for the removal of contaminants from surface water, and other advantages include a smaller footprint, low chemical usage, and more automation. However, the limitation of membrane technology, including MF and UF, is the high cost of membrane replacement and the lower effectiveness in removing dissolved organic matter in the treated water. The study of modified MF membrane technology is reported by Sinclair et al. [ 53 ], and it showed an improvement in reducing cost as they do not require any external driving force. Unfortunately, the modification resulted in an approximately 22% loss of membrane permeability.

Meanwhile, He et al. [ 54 ] published a study on improved UF technology in which they combined heterogeneous catalytic ozonation and a UF membrane filtration technique for the long-term degradation of bisphenol A (BPA) and humid acid (HA). Results have shown improvements in removal efficiency, reduction of membrane resistance, and mitigation of membrane fouling. Another study concerning UF was reported by Chew et al. [ 55 ], which compared and evaluated industrial-scale UF with conventional drinking water treatment systems. The study showed that UF systems can provide reliable filtrate quality even with the existence of fluctuation in the raw water quality. In addition, the UF system offers promising sustainability, with no coagulant required for high-quality filtrate and non-toxic sludge discharge.

High-pressure NF and RO membranes can provide an alternative method for removing organic and inorganic matter. The NF process is already known for its capabilities in the removal of total organic carbon (TOC) in surface water treatment [ 56 ]. This process has been implemented in several drinking water industries [ 57 , 58 , 59 ]. In an experiment conducted using pre-ozonation as a pre-treatment process for NF membranes proposed by Vatankhah et al. [ 60 ], it was found that pre-ozonation with a low specific ozone dose could effectively mitigate a significant portion of fouling. However, the removal performance of dissolved organic carbon (DOC) of the NF membrane did not show a substantial change, which may be due to the relatively low applied ozone dose. The RO process is applied for drinking water treatment, whether the source water comes from seawater, brackish water, or groundwater [ 21 ]. However, RO has a problem with the ability of suspended solids, colloidal material, and dissolved ions in raw water to foul the system [ 61 ]. A study conducted by Touati et al. [ 62 ] combined UF, NF, and RO processes for isotonic and drinking water treatment. Results showed that the UF process used as pre-treatment was able to eliminate natural organic matter (NOM), while the NF process was able to characterize the fouling mechanism. The overall performance’s energy consumption is determined by salt rejection during the NF process.

Apart from RO, ED is another process that can be used to treat brackish water with high performance and energy efficiency [ 63 , 64 ]. The process involved the transfer of electrolytes or ions through a solution and membranes based on an applied electric field as the driving force [ 65 ]. Walha et al. [ 66 ] investigated the use of the NF, RO, and ED processes in producing drinking water from a brackish water source. The results showed the treatment based on RO and ED processes is more efficient, as shown by the high rejection of inorganic matters present in the feed waters. The concentration of ions in the permeate flux can achieve World Health Organization (WHO) standards, and it is more economical than the NF process.

Forward osmosis (FO) and membrane distillation (MD) processes are driven by heat, which is different from the pressure-driven process usually used for potable water reuse [ 67 ]. FO processing operates at low or no hydraulic pressure, which may reduce irreversible fouling and achieve high rejection of contaminants [ 68 ]. However, Li et al. [ 69 ] reported that the water flux produced by the FO process was still inadequate compared to the RO process under a similar applied pressure. FO processes involve a permeable membrane and two solutions, known as feed and draw solutions. The feed and draw solution consists of different concentrations that produce the osmotic pressure gradient that acts as the driving force for water permeation across a semi-permeable membrane [ 70 ]. An experiment conducted by Tow et al. [ 71 ] studied the fouling propensity between RO, FO, and MD. The experiment was conducted using a single membrane module and showed that both FO and MD exhibit a significant advantage in fouling resistance but neither of them performed well with both organic and inorganic foulants.

Available membrane filtration treatment for drinking water [ 34 , 61 , 68 , 72 , 73 , 74 , 75 , 76 , 77 ].

Membrane Filtration and Fouling Issue

In membrane filtration, fouling is still the main reason for flux decline, and it needs to be reduced appropriately. Fouling is formed during the membrane filtration process. It is a very complex phenomenon that developed based on a combination of physical, chemical, and biological aspects. Membrane fouling can cause a reduction in permeate flux [ 78 ], an increment of trans-membrane pressure (TMP) [ 79 ], a shorter membrane life span [ 80 ], and consequently, cause a reduction of water quality [ 81 ]. Another work in [ 82 ] also claimed that membrane filtration is fraught with disadvantages regarding the amount of permeate flux and fouling tendency. Membrane filtration treatment is also struggling with a downside where it requires high operation and maintenance costs including labor, chemicals, membrane replacement, energy, and sludge disposal [ 83 , 84 ], when irreversible fouling on the membrane surface is not properly controlled.

Fouling takes place in the membrane filtration process based on four types of foulants: particulates/colloidal, organics, inorganics, and micro-biological organisms [ 85 ]. Table 2 illustrates each foulant type and its associated membrane fouling mode. Particulates or colloids with a similar or close diameter to the membrane pores can cause the membrane pores to become clogged, whereas larger particulates that are unable to pass through the membrane pores can cause the formation of a cake layer on the upstream face of a membrane. Organic and inorganic foulants tend to adsorb and precipitate in the membrane pores and consequently cause blockage of the membrane pores, whereas the accumulation of microorganisms on the membrane surface will cause the development of biofouling. Pore blocking, cake layering, adsorption and precipitation of organic-inorganic fouling, and biofouling occurrences cause a reduction in the rate of permeate production and escalate the complexity of the filtration process.

Types of foulant.

In general, there are two categories of fouling: reversible and irreversible fouling. The reversible fouling which is back washable and non-back washable occurs when organic or inorganic materials accumulate on either side of the membrane surface as operating time increases [ 86 , 87 ]. Back washable reversible fouling can be restored based on physical and hydrodynamic methods, while non-back washable reversible fouling can only be removed based on chemical cleaning. The irreversible fouling is usually occurring after quite a long run of filtration process where the particles formed a matrix that strongly attached to the membrane surface like pore blocking, clogging, biofilm, and cake gel [ 43 , 88 ]. Arise of irreversible fouling caused a loss in transmembrane flux. In this case, the membranes can only be fixed by extensive chemical cleaning. In the worst case, the membrane needs to be replaced.

The occurrence of membrane fouling in membrane filtration processes is due to many factors [ 89 ]. It is due to the characteristics of the feed water, membrane properties, and configuration of the filtration system itself [ 90 ]. Several studies on the factors that influence fouling have been conducted to control and mitigate its development. For example, Mozia et al. [ 91 ] showed the effect of process parameters which are feed cross-flow velocity and TMP, on the fouling behavior of the MF/UF system, whereas Kola et al. [ 92 ] discovered the fouling behavior for different feed water types and different membrane pore sizes. Results indicated that the parameters involved in both studies closely influence the fouling growth rate. In the work of Zhao et al. [ 93 ], the fouling was mitigated by controlling the membrane surface shear rate. The authors observed that by providing a high shear rate, the filtration process was able to achieve high critical flux. High shear rates cause algae to foul the membrane. This claim can be supported by similar research done by Jaffrin [ 94 ]. Table 3 tabularizes the parameters that influence membrane fouling during the filtration process. From the table, systematic approaches can be strategized to provide high-quality drinking water.

Parameter that influences the fouling growth rate.

4. Current Solutions and Way Forward

Numerous fouling reduction techniques have been studied by many researchers to ensure the successful application of membrane filtration systems. In this review, the fouling reduction methods proposed by the previous researchers can be classified into three main categories: chemical cleaning, physical cleaning, and hydrodynamic cleaning [ 111 ], as summarized in Table 4 . Chemical cleaning is a process that is usually used as a pre-treatment method. The process is recognized as a prevention method. It involved chemical agents as a tool to reduce or eliminate the deposition of fouling. Reversible fouling based on the natural organic matter can be partially or fully restored by chemical cleaning. Reversible fouling can also be removed physically. Irreversible fouling can only be removed by chemical cleaning. In general, chemical cleaning is executed when physical cleaning no longer provides effective cleaning performance and the flux cannot restore the environment sufficiently. However, the cleaning method for each filtration process is dependent on many factors. Still, trial-and-error practice is the most suitable method to get the best strategy for any process.

In this review, different views and perspectives on fouling reduction methods are discussed as a way forward to solving the issue, which are prevention, prediction, or control automation. Figure 4 summarizes the main strategies that were used for membrane fouling control. The prevention method is usually related to chemical cleaning, while the prediction and control automation methods are related to physical cleaning. The hydrodynamic technique involves modification of module design and arrangement of flow such as for feed and permeate. The hydrodynamic technique has been studied by Lee et al. [ 112 ] to control the fouling during the forward osmosis-reverse osmosis (FO-RO) hybrid process. The study evaluated the influence of feed flow rate, draw flow rate, and hydraulic pressure difference. The results showed that the high feed flow rate was able to effectively mitigate the fouling. The high draw flow rate, on the other hand, causes an increase in the fouling growth rate. In addition, increasing hydraulic pressure does not affect reducing the fouling growth rate.

Membrane fouling control.

Fouling reduction techniques.

There are also efforts to improve methods by optimizing operational conditions [ 123 , 124 ], both in chemical and physical cleaning operations. To optimize operating conditions, it is important to understand the characteristics of accumulated irreversible fouling. The irreversible fouling is a very complicated phenomenon in which membrane characteristics (membrane materials, pore size, configuration, hydrophobicity, charge), process operating conditions (TMP, temperature, permeate flux), and influent physicochemical properties (particle size distribution, inorganic or organic mattes) closely influence each other. Soft computing optimization is one of the best solutions to handle complex and nonlinear processes.

4.1. Fouling Prevention

Fouling prevention is important in order to prevent fouling (reversible or irreversible) from occurring or arising. In the prevention step, the foulants that can cause fouling are eliminated before the feed water enters and passes through the membrane [ 125 ]. In practice, an irreversible type of fouling is removed using chemical cleaning methods [ 126 ]. For safer operation, acids, bases, and oxidants are usually used in chemical cleaning [ 127 , 128 ]. The quantity of chemical cleaning is monitored to avoid excessive chemical use that can damage the membrane surface and increase the cost of operations [ 129 ]. Therefore, it is important to optimize the operating conditions of chemical cleaning, which involve cleaning intervals, cleaning duration time, chemical type, and chemical concentration. Yoo et al. [ 3 ] showed that proper optimization of operating conditions for chemical cleaning was able to reduce energy consumption, chemical use, and sludge production. However, the process causes an increase in the membrane replacement cycle.

Previous researchers applied numerous techniques as a pre-treatment method of preventing fouling from occurring, such as coagulation [ 130 ], oxidation, ozonation, and adsorption methods [ 73 ]. The coagulation process disperses and suspends contaminants and is suitable for natural organic matter (NOM) with a high molecular weight [ 131 , 132 ]. The process commonly uses pre-hydrolyzed salts such as polyaluminium and sulfates as coagulant agents. The process required low cost due to the simple operation, conversely it produces high sludge formation. Unlike coagulation, the oxidation process produces a lesser amount of sludge formation. The process is useful for the removal of dissolved organic contaminants such as arsenic and humic acid [ 133 ], the same as the ozonation process [ 60 , 134 ]. The main advantage of ozonation treatment is that it does not produce any sludge, but it can cause the degradation of biopolymers and high energy consumption. The adsorption process is frequently used to remove organic and inorganic micro-contaminants from pharmaceuticals and personal care products, such as pesticides, antibiotics, detergents, soaps, and oils [ 135 ]. The process is drive by the electrostatic interaction of negative and positive charged that produce between influent and adsorbent [ 136 ]. The influent, which is a liquid/water or gaseous contaminant, will change into solid formation [ 137 , 138 ]. The absorbent can be restored and reused. The process is capable of removing micro-contaminants even when they are present in trace amounts in water. Adsorption process is more efficient when ozonation is used as a pre-treatment.

Oloibiri et al. [ 139 ] show that discovered that combining pre-treatment methods yields better results in reducing fouling tendency than a single pre-treatment method [ 140 ]. Yu et al. [ 125 ] studied the effect of the conventional coagulation technique and hydrogen peroxide (H 2 O 2 ) addition at three doses level during the backwash process. The performance of the technique was measured based on the rate of TMP. The authors found that the addition of H 2 O 2 at all doses was able to prevent any measurable increase in TMP, which represents the success of the proposed technique to prevent the development of membrane fouling. Wang et al. [ 141 ] used H 2 O 2 in the pre-oxidation process before executing the coagulation process. In this study, the TMP value, microorganism development, and cake layer rate were monitored to observe the effect of biofouling in the presence of H 2 O 2 . Results showed that the proposed technique was able to decelerate the microorganisms’ growth rate and reduce the cake layer, hence decreasing the TMP value, which indicated a reduced membrane fouling tendency.

Park et al. [ 142 ], investigated a pre-ozonation technique based on two doses for the NF process’s surface water brine. The technique was mainly applied to reduce or control membrane fouling, where the doses are determined based on the residual ozone dose. A precise ozone dose is required to avoid membrane damage and an increase in operating costs. The authors observe that the pre-ozonation technique was able to reduce a significant amount of organic fouling potential with relatively low ozone doses. Results also indicated that the applied technique was able to act as a barrier for the removal of trace organic compounds which are important for water treatment.

Another study on the pre-ozonation technique as a membrane fouling prevention method was reported by Wang et al. [ 143 ]. In this work, the effects of pre-ozonation as a pre-treatment for the UF process on secondary wastewater effluents are investigated. The research is based on two types of UF membranes: hydrophilic regenerated cellulose membranes and hydrophobic polyethersulfone membranes. The result showed that high fouling reduction was attained for the hydrophobic membrane at high ozone doses. Table 5 presents the settings for membrane fouling prevention from previous researchers. The results from the previous studies cannot be generalized because the result and consequence of each pre-treatment method are expected to diverge according to the feed water, filtration technology, and pre-treatment material, such as the types of absorbent and oxidation agent.

Setting of membrane fouling prevention in drinking water treatment.

4.2. Fouling Prediction

Many researchers are interested in foul prediction. The prediction will be able to help the researchers forecast the best operating conditions for a particular process and determine the parameter that can trigger the fouling. It is also useful to circumvent or slow down the process of fouling developments. The prediction process is a part of modeling and controlling development. Some researchers used prediction terms to describe the modeling process, which can be categorized as mathematical and empirical processes.

As mentioned previously, the prediction or modeling of physical cleaning operations has been widely utilized by researchers to understand fouling behavior. Physical cleaning operations include air scouring, backwashing, and relaxation operations. Air scouring, also known as aeration or air bubble control, is a widely used method of membrane cleaning. The method boosts the saturation of oxygen by applying air bubbles that exhibit cross-flow velocity and can eliminate reversible fouling [ 150 ]. The backwashing process involves the pumping of permeate or water backward through the filtration module (membrane) in order to remove the particles attached to the membrane surface. Other than permeate, the backwash can be implemented using either chemicals, clean water, or air. Finally, relaxation is a process where the permeating or filtration process is temporarily idle, but with the air bubbles scouring continuously working to relieve the membrane from the generated pressure [ 151 , 152 ]. A comparative study of physical cleaning involving air scouring, backwashing, and relaxation techniques to control the fouling in drinking water treatment was conducted by De Souza and Basu [ 153 ]. In this study, it was shown that in some cases, backwashing and relaxation durations have integrated results for the reduction of fouling, while air scouring can reduce fouling at the highest level with the highest air scour rate. Overall, the result indicated that the combination of the three techniques outperformed air-assisted backwashing alone in terms of fouling reduction. It is crucial to understand the effectiveness of each operation (air scouring, backwashing, and relaxation) when controlling membrane fouling in order to properly strategize the coordination of the operations. Fouling may also be controlled by operating UF under its critical flux [ 26 ]. When UF is operated under its critical flux, foulant deposition on the membrane surface can be avoided. Thus, membranes can be operated with a stable flux. Vigneswaran et al. [ 154 ] also mentioned that the performance of the membrane combined with the adsorption process is influenced by the reactor configuration, mode of operation, carbon dosage, adsorption, and influent characteristics.

Work by Kovacs et al. [ 155 ] proposes a mathematical framework for batch and semi-batch modeling techniques for membrane filtration processes. The proposed method uses feed concentrations as the basis for calculation and can be applied to all pressure-driven membrane filtration processes. The main advantage of the proposed method is that it can capture the dynamic behavior of all types of batches and semi-batch configurations without changing the general mathematical framework. However, it required challenging mathematical problem-solving to obtain the general framework, whereas Ghandehari et al. [ 156 ] proposed a semi-empirical and artificial neural network (ANN) modeling technique to predict the characteristics of microfiltration systems based on permeate flux decline and membrane rejection. Results showed that the semi-empirical method was able to predict the flux only for a specific time, unlike the ANN method. The ANN method can model the membrane filtration system over the entire filtration time for all tested operating conditions.

Ling et al. [ 157 ] proposed a tent sparrow search algorithm back propagation network (Tent-SSA-BP) technique for predicting membrane flux in a membrane bioreactor (MBR) fouling model. They utilized the principal component analysis (PCA) algorithm to reduce the initial auxiliary variables. A study was conducted to compare the genetic algorithm back propagation (GA-BP), particle swarm optimization back propagation (PSO-BP), sparrow search algorithm extreme learning machine (SSA-ELM), sparrow search algorithm back propagation (SSA-BP), and tent particle swarm optimization back propagation (Tent-PSO-BP) networks. The results indicated that the Tent-SSA-BP technique provided the best performance in terms of training speed and prediction accuracy. The Tent-SSA-BP technique predicts with 97.4% accuracy, whereas BP predicts with only 48.52% accuracy. A model for MBR prediction has also been studied by Kovacs et al. [ 158 ], where it predicts transmembrane pressure (TMP) at various stages of the MBR production cycle. The prediction was performed based on a data-driven machine learning technique involving a random forest (RF), artificial neural network (ANN), and long-short-term memory (LSTM) network. Among the proposed methods, RF models provide the best statistical measures. The obtained prediction models produce promising results, but their ability to predict the data is limited at this time. Yao et al. [ 159 ] predict the variation of the TMP in the constant flux mode by proposing a novel method based on the loss of effective filtration area. The result showed a high correlation coefficient, which indicates a good model prediction.

Another study on fouling prediction was conducted by Chew et al. [ 160 ], where the first principle equation of Darcy’s law on cake filtration and ANN were combined to predict the models that represent the dead-end ultrafiltration process. In this study, the turbidity of the feed water, filtration time, and TMP were used as the input parameters. The sensitivity analysis showed that there was a strong linear correlation between specific cake resistance and turbidity. The proposed models can predict the specific cake resistance and total suspended solids (TSS) of feed water with high accuracy, which provide an early indication of fouling development.

Another study on fouling prediction was reported by Lie et al. [ 161 ]. In this work, experiments were conducted on a constant flow microfiltration membrane system at critical flux and supra-critical flux conditions with various permeate fluxes and feed water qualities. In this study, five input variables of the ANN model, including permeate flux, turbidity, UV 254 , time, and backwash frequency were used for the prediction of TMP. The results show that the ANN model with five input parameters can predict TMP behaviors, where the TMP value is used to indicate fouling propensity. A similar study using ANN models was done by Hazrati et al. [ 162 ], where back propagation algorithms were used to predict the effluent chemical oxygen demand (COD) and TMP. The research indicated that the ANN model can easily be used to predict the concentration of COD and TMP in effluent. The study also investigated the specifications of the cake layer at different hydraulic retention times (HRTs) in order to control membrane fouling. Results indicated a linear relationship between the reduction in HRT and the particle size of the cake layer.

An ANN technique was also found by Abbas and Al-Bastaki [ 163 ], where an experiment was conducted using a spiral wound reverse osmosis membrane system with three operating conditions of inputs were studied. The first one was trained using a total of sixty-three data points from different operating temperatures for training purposes. The second one is trained only using forty-two data points corresponding to the operating temperatures of 10 °C and 30 °C; another twenty-one data points corresponding to the operating temperature of 20 °C were employed for testing purposes. The third condition was trained using the data corresponding to the operating temperatures of 10 °C and 20 °C, whereas another 21 data points corresponding to the operating temperature of 30 °C were employed for testing purposes. It was found that ANN was able to interpolate the data with good accuracy but was unable to produce acceptable results for data extrapolation, whereby works by Chen and Kim [ 164 ] used 17% of experimental data for training purposes and 83% for verification. The authors studied the capability of a radial basis function neural network (RBFNN) and a multilayer feed-forward backpropagation neural network (BPNN) to predict the permeate flux in cross-flow membrane filtration. The predictions are based on five input parameters, which are particle size, ionic strength, pH, TMP, and elapsed time. The result shows that a single RBFNN is able to predict the permeate flux and provide better predictability than a BPNN.

Based on the review conducted, it was found that fouling prediction has been broadly applied for the mitigation of membrane fouling. Various techniques have been implemented, but most of them apply ANN as a primary strategy. Due to the difficulty of solving tricky mathematical problems, only a few studies use mathematical frameworks in prediction. Many of them also combine the ANN technique with other methods. The ability of ANN to solve highly complex and nonlinear problems makes it extremely useful in the treatment of drinking water. ANN is capable of providing good predictions even without detailed information about the physical parameters of the system, relying solely on input-output data. Even so, the process of determining appropriate input-output parameters is crucial and plays a significant role. Without a good relationship between the selected input and output parameters, acceptable prediction cannot be achieved. Therefore, it is important to decide the respective input-output parameter before proceeding with the ANN architecture. Every process comes with differences and complex characteristics that are likely due to the system itself. Certainly, changing the concentration of the feed water will change the entire process. As a result, understanding the process in terms of which parameter caused an effect on which parameter is critical in the ANN technique. Table 6 shows the various ANN settings for the prediction of membrane fouling in drinking water treatment. Based on Table 6 , it is clearly shown that each system with different feed water characteristics involves different input and output parameters. Hence, this will affect the ANN architecture.

ANN setting for the prediction of membrane fouling in drinking water treatment.

4.3. Fouling Control and Automation

To overwhelm the problem that fouling causes, it is essential to equip the membrane filtration process with an effective controller. The effective design of the controller will be able to improve the overall efficiency, increase the membrane’s lifespan, and reduce the total operating costs. However, the design of the controller is not an easy task due to the many impediments, such as the dynamic processes of the system itself, the difficulty of modeling the system, variations in feed water quality, system faults, membrane fouling, and the requirement of continuous monitoring for membrane cleaning.

As the system becomes more complex, the control strategies make it easier to handle the membrane filtration process by estimating the uncertainties and making control systems that are robust and reliable. However, based on the literature, there is still a lack of research that applies control automation to the membrane fouling problem [ 171 ]. Most of the previous research focused on open-loop control, membrane modifications, physical cleaning, and pre-treatment methods.

In the previous study on control automation, Azman et al. [ 172 ] applied a proportional-integral-derivative (PID) controller with the Ziegler Nichols (ZN) and Cohen-Coon (CC) tuning methods for the coagulation and flocculation filtration processes. The robustness of the controller’s performance was measured based on the step test, set point change, and load disturbance test. At the end of the study, it was shown that the PID controller with the ZN tuning method exhibits better performance than the PID controller with the CC.

The design of model predictive control (MPC) based on a support vector machine (SVM) model for the ozone dosing process is reported by Dongsheng et al. [ 173 ]. The results have shown an improvement in maintaining a constant ozone exposure compared to the use of the proportional-integral (PI) controller. However, the controller design was only tested for a plant-scale experiment. The design of MPC was also found in the works by Bartman et al. [ 174 ], where the purpose was to determine and control the optimal switching path of flow operating conditions, thereby reducing the fouling problem for a RO desalination process. Results showed that the proposed controller was able to reduce the variation of system pressure, and hence, provide smaller pressure fluctuations with a shorter transition time. The designed MPC can control and prevail over the disturbance that comes through the system and reduce the percentage error between the actual and the desired final steady-state value.

Multiple model predictive control (MMPC) was used in the simulation works of Bello et al. [ 175 ] to control and optimize the amount of chemicals used in the coagulation process of water treatment plants. They applied switching mechanisms to deal with the control input constraints explicitly. Simulation results show that the proposed MMPC provides better performance than conventional control. However, the work is only conducted based on the linear model; future work may use the nonlinear model, which represents the real system. Rivas-Perez et al. [ 176 ] designed an expert model of predictive control (EMPC) to control the critical variables of the pilot scale RO desalination plant. Based on known information, an expert system was created that can lead to decision-making strategies. The robustness of the proposed controller was evaluated based on two real-time cases. In the first case, the performance of EMPC and the ability to ignore disturbances were tested. In the second case, the performance of the proposed EMPC was compared to the performance of the standard MPC. The results showed that the control plant with EMPC provided higher accuracy and robustness than standard MPC, especially for time-varying parameter rejection. Table 7 shows the modeling and control strategy that has been reported based on several techniques by the previous researcher to maximize and control the quality of drinking water treatment.

Setting of control strategy in drinking water treatment.

5. Conclusions

This paper summarizes the available filtration treatments for treating drinking water. Filtration treatment can be categorized into two main types, i.e., conventional and advanced treatment. As discussed in the relevant section, conventional treatment entailed additional costs due to the need for additional treatment and a large footprint, whereas advanced treatment, specifically membrane filtration treatment, is now well-established in the industry because it is capable of overcoming the disadvantages caused by conventional treatment. Membrane filtration treatment achieves satisfactory results in the elimination of different kinds of contaminants from effluent. As a result, the rate of permeate flux (effluent) production will increase. However, membrane filtration is facing problems with membrane fouling as the operating time increases.

Until now, many researchers’ studies on the parameter that causes fouling have resulted in the development of a model for prediction and prevention. Membrane fouling is affected by many factors, including feed water type, feed water and membrane properties, membrane material, filtration strategy, and process operating conditions such as transmembrane pressure and sludge retention time. Previous studies showed that membrane fouling in processes can be very diverse, and it is mainly due to the feed water type and the process treatment itself. In this case, understanding the composition of the feed water and the characteristics of the process treatment are crucial. Fouling mitigation is typically based on prevention, prediction, and control automation process. The prevention method has been utilized broadly and presents promising results for water treatment. The procedure involving the use of chemicals as an agent to mitigate fouling is the method’s main shortcoming. Since the process discussed drinking water, which is closely related to human health, it is remarkable to prevent any approach that could cause undesirable consequences. For the prediction method, a former researcher mostly applied ANN as a tool to predict the development of fouling. Conversely, the study did not discuss in detail the technique to reduce fouling but instead focused only on prediction purposes. Nevertheless, there is not much information presented on control automation strategies. The majority of researchers control membrane fouling through pre-treatment or modification of membrane characteristics, both of which required the use of chemicals. It is critical for ecologically mitigating membrane fouling. Future research is needed to add value to the control automation method via the application of control strategies such as controllers (proportional-integral-derivative controllers, model predictive controllers, etc.). A study on membrane automation is necessary to control the occurrence of fouling without the use of chemical agents. It is thought that this will lead to more exciting discoveries, directly encounter fouling, and produce high-quality drinking water. In the years ahead, it might be switched to fresh strategies and technologies.

Acknowledgments

This work was supported financially by the Faculty of Electrical Engineering and the Centre for Research and Innovation Management (CRIM) from the Universiti Teknikal Malaysia Melaka (UTeM) and the Universiti Teknologi Malaysia High Impact University Grant (UTMHI) vote Q.J130000.2451.08G74. The first author wants to thank the UTeM and the Ministry of Higher Education (MOHE) for the ‘Skim Latihan Akademik Bumiputera’ (SLAB) scholarship.

Funding Statement

This research was funded by the Faculty of Electrical Engineering and the Centre for Research and Innovation Management (CRIM) from the Universiti Teknikal Malaysia Melaka (UTeM) and the Universiti Teknologi Malaysia High Impact University Grant (UTMHI) vote Q.J130000.2451.08G74.

Author Contributions

Conceptualization, M.C.R. and N.A.W.; formal analysis, M.C.R.; investigation, M.C.R.; resources, M.C.R.; data curation, M.C.R. and N.A.W.; writing—original draft preparation, M.C.R.; writing—review and editing, N.A.W. and N.S.; supervision, N.A.W. and N.S.; project administration, N.H.S.; funding acquisition, M.C.R., N.A.W., N.S. and N.H.S. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Data availability statement, conflicts of interest.

The authors declare no conflict of interest.

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

An Investigatory Study of Water Filtration

• M. L. Ferrer
• C. Florentino
• D. L. Gabato
• E. F. Lacson
• D. N. Ventura
• Reynold R. Bangalisan, LPT, MRIEdr

Water filtration project is a process that can remove unwanted substances in water using materials such as pebbles, sand, and charcoal and turn it into water that can be used on everyday habits and can eventually be drunk. Charcoal is activated to remove chlorine. Pebble is used to trap and strain particles in the water. Sand is naturally occurring granular materials composed of finely divided rock and mineral particles. It is defined by size, being finer than gravel and coarser than silt. This study utilized a quantitative analysis method of research. It also utilized a questionnaire to test the level of acceptability of the water filtration process in terms of color, suspended particles, and usefulness and assessed by 5 Science Experts and 5 Science Teachers of Bestlink College of the Philippines. To determine the level of acceptability of the water filtration process, the researchers used a t-test to test the significant differences between the assessment of the Science Experts and Science Teachers at the level of significance of α= 0.01 and the degree of freedom of 8. The level of acceptability of the water filtration process in terms of color, suspended particles, and usefulness is moderately acceptable as assessed by the Science Experts and Science Teachers. There is no significant difference between the assessment of the Science Experts and Science Teachers on the level of acceptability of the water filtration process in terms of color, suspended particles, and usefulness. The Science Teachers said that the color and suspended particles were acceptable but it is not advisable for drinking. The Science Experts said that there are many tests that we need to conduct to know if it is potable and safe. The level of acceptability of the water filtration process is rejected. This study was connected to another study that used the same ingredients as pebbles, sand, and charcoal. This study was not as effective as we used normal charcoal to filtrate the water. Another study that used activated charcoal was reliable. The flaws of the study were acknowledged and the future researchers were advised to research more effective ingredients to filter the water.

How to Cite

• Endnote/Zotero/Mendeley (RIS)

Most read articles by the same author(s)

• J. C. Licayan, M. S. Lopez, F. Manasan, J. Ramirez, K. R. Raquid, Reynold R. Bangalisan, LPT, MRIEdr, Generating Electricity through Running Water Turbine with the Use of Non-Potable Water , Ascendens Asia Singapore – Bestlink College of the Philippines Journal of Multidisciplinary Research: Vol. 1 No. 1 (2019): Ascendens Asia Singapore – Bestlink College of the Philippines Journal of Multidisciplinary Research Abstracts, Vol.1, No.1, March 2019
• M. J. Acob, J. E. Camarillo, MJ. P. Catipay, MJ. P. Catipay, A. T. Miole Jr., J. M. Tapacion, Reynold R. Bangalisan, LPT, MRIEdr, Dual-Purpose Bus for Water and Land Transportation , Ascendens Asia Singapore – Bestlink College of the Philippines Journal of Multidisciplinary Research: Vol. 1 No. 1 (2019): Ascendens Asia Singapore – Bestlink College of the Philippines Journal of Multidisciplinary Research Abstracts, Vol.1, No.1, March 2019