The Long Read: Do we have the technology for potable water reuse?

Water reuse for drinking purposes is racing up the agenda in the US and elsewhere. Work to draw up new rules for potable reuse is advancing in states including California and Arizona, as well as in Canada and the European Union. Technologies are being piloted, and plants permitted.

Water reuse: Do we have the technology?

As the market for water reuse grows, how confident are water technologists and engineers that they can produce consistently safe sources of drinking water from wastewater? Very confident, is the industry consensus. Reverse osmosis (RO) systems of the type used in seawater desalination are emerging as the dominant technology for water reuse, and confidence is high that they are robust, although work continues on a handful of challenges in reuse applications.

“There is a very high level of confidence surrounding RO. It’s not a new technology. It has been around for a long time, it’s proven, and has been studied many, many times over. There is a high degree of reliability and confidence that when RO is coupled with other processes you can achieve the desired results of removing pathogens, compounds of emerging concern, salts and other solids,” says US Water Environment & Reuse Foundation research manager Justin Mattingly.

This strength of confidence echoes across the industry. “We know exactly what the process train is going to look like. It’s going to be reverse osmosis with commodity RO membranes. There doesn’t need to be a lot of innovation in terms of inventing new treatments steps to make it work,” says water industry consultant Rick Stover.

“We know exactly what the process train is going to look like.
It’s going to be reverse osmosis with commodity RO membranes.”
Rick Stover, water industry consultant

And as well as the shared confidence, there is broad consensus about what technologies will comprise a typical water reuse treatment train. “In water reuse we apply our ultrafiltration (UF) technology upfront, which acts as a pre-filter to the RO downstream. The effluent that comes out of a UF membrane is ideal to feed an RO because it has the silt density index (SDI) and the turbidity and the total suspended solids (TSS), and all the characteristics to keep the RO performing optimally,” says Taylour Johnson, product manager, water and wastewater at Koch Membrane Systems.

“So a lot of times we will treat surface water or wastewater with a membrane bioreactor with our UF technology, either pressurised or submerged, depending on the application and hydraulic and things like that, and then we’ll directly feed an RO,” he adds.

In one example of a non-potable reuse plant, a fertiliser producer wanted to take effluent from a municipal wastewater plant and recycle it. “Historically they were buying water from the city and using it for in-process stuff, then they wanted to expand their plant and the city said, ‘We can’t give you more water.’ They took effluent from the wastewater plant, polished it with ultrafiltration membranes, our PURON MP membranes, and then sent it through one of our eight inch RO skids, and took that to feed their boilers and cooling towers. It was really the only the option for this industrial customer if they wanted to expand their capacity,” says Johnson.

In the case of potable reuse, this type of multi-stage treatment process acts like a belt-and-braces approach to safety. “In California, after the RO they use a combination of UV light, hydrogen peroxide, and advanced oxidants,” says Mattingly. “Let’s say one of the contaminants that needs removing in this process is a pathogen. If you have a treatment train that runs microfiltration then reverse osmosis then UV then advanced oxidation: pathogens can be removed by microfiltration, they can be removed by RO, and they can be removed by UV light, so it is not just relying on a single barrier to remove the contaminant, but multiple barriers. If for any reason there is an upset in one barrier, you have other barriers in place to prevent contamination.”

Lessons learned through previous desalination plant projects are now being applied to the reuse market. One is to always study the specifics of each project. “When you’re talking about regular water, the composition is somewhat similar, but in wastewater, you might have characteristics in the wastewater because of industrial discharge or potential natural characteristics. Looking at projects individually and understanding their details can help a lot. Piloting in advance of going full scale is one way to address this,” says Felipe Pinto, Americas marketing manager, reverse osmosis, at Dow Water & Process Solutions. The lesson is that every wastewater is different and every reuse project is different.

The good news is that RO is substantially cheaper when applied to reuse than to seawater desalination, because reuse uses less energy. “Reuse plants typically run at about 10 bar as opposed to 100 bar for seawater RO, the pumps are a lot smaller, there’s no need for energy recovery devices, and you don’t have to use exotic metals — seawater is highly corrosive, so the metals you use are super-expensive,” says Greg Wetterau, vice president at CDM Smith, an engineering and construction company that advised the US Environmental Protection Agency (EPA) on its reuse guidelines.

However, though confidence is high and the technology is considered proven, five key challenges remain to moving ahead with water reuse.

Challenge 1: NDMA

One specific piece of work that’s underway in the US is on the contaminant Nitrosodiemethylamine (NDMA), a disinfection byproduct that passes through RO systems and can survive in groundwater. In Singapore, another region where water reuse is growing fast, NDMA is less of a problem because it degrades in sunlight, and the Singaporeans store water above ground in a reservoir. In Australia, the approach is to control NDMA’s formation, rather than to remove it.

“In California, where we inject water into the ground, there’s no sunlight, and so it persists all the way to the drinking water well. The target for NDMA in California is 10 nanograms a litre, so it’s really, really low. In Australia it’s 100,” says Wetterau. There is currently no World Health Organisation guideline, or US Environmental Protection Agency (EPA) regulation, on NDMA.

“It’s an unregulated contaminant, but it’s something that’s in wastewater and it’s not in drinking water, so we want to make sure that we’re addressing it. That kind of stuff is the driver for why we use membrane technologies in the first place. There are things in wastewater that we haven’t really thought about that much in drinking water and we want to make absolutely sure that we address all of them. If you’re going to be turning this into a drinking water source, you have to get everything that might hurt people, out of it,” says Wetterau.

Challenge 2: Log removal credits

How to demonstrate the reliability of potable reuse systems is the next big challenge. The discussion between engineers and regulators is about log removal credits – a regulatory mechanism that provides a way to compare the effectiveness of technologies at removing contaminants. In challenge validation testing, contaminants are put into water that is then run through the treatment step to be tested, and the quality of what comes out is measured. “It has to have been reduced by somewhere between eight and 10 removal, this log is a factor of 10, so it really is at the parts-per-billion level of purity. They’re looking for a couple of specific organisms and the extent to which they get reduced by the treatment steps. These are dangerous bacteria that they have earmarked as representative, molecules that have properties that make them good indicators for families of contaminants,” says Stover.

There are already log removal credits for ultrafiltration membranes for drinking water, and the discussion now is about log removal credits for membrane bioreactor (MBR) systems. “In a lot of cases the same type of membrane that’s being used for drinking water is being used for MBR. So there’s really no reason why they shouldn’t get a credit for log removal for pathogens. A lot of discussions are happening, it’s slow, it’s bureaucratic, but there are a lot of areas in the US that don’t have enough water,” says Pinto.

Challenge 3: Integrity testing

Another area up for further investigation is how to conduct integrity testing of reverse osmosis systems while they’re online. “The challenge with RO there is no convenient way to do integrity testing to make sure that there are no leaks or bypasses in the filter. Whereas for ultrafiltration, microfiltration, and UV disinfection, these other steps in the treatment chain can be verified online,” says Stover.

“Integrity testing for reverse osmosis is another big thing that
the industry is developing.” Greg Wetterau, CDM Smith

Particularly for direct potable reuse, it’s important to have controls in place to ensure that the treatment steps are working correctly. “You have to have extra measures so that if anything goes wrong, you know immediately and you can address it. Part of it is monitoring water quality, and part of it is making sure that the processes are doing their job. So you have to do integrity testing of the membranes, so you need something that’s continuous, more sensitive than once-a-day testing. We do that for surface water membrane plants for specific pathogens, cryptosporidium and giardia, but they haven’t done it for RO before, and so that’s another big thing that the industry is developing,” says Wetterau.

Mattingly adds: “It can be, let’s say, a continuous monitor for total organic carbon, that’s a way of confirming reliability that the systems are operating properly. It’s not just having confidence in the system, but if you have proper monitoring procedures you’ll have data confirming that. Over time, you can see the system performance. Reporting requirements all feed into the concept of reliability.”

Challenge 4: Brine disposal

Potable reuse poses a different kind of challenge at inland locations, where disposal of the brine stream created by RO systems can be impractical. Two responses to this problem are mooted. The first is to use an alternative treatment process to RO, and the second seeks to minimise the amount of brine that RO produces.

“Probably the main challenge with reverse osmosis is the disposal of brine. If you’re in a coastal area, in California or in Israel, or the Middle East – if you’re close to the ocean – brine disposal isn’t all that difficult because you can simply run a line into the ocean,” says Mattingly. “If you’re an inland community, say in central California or in many places nearer the US East Coast, disposal of brine can be a lot more difficult.”

In Gwinnett County, in the southeastern US state of Georgia, a long-established indirect potable reuse facility outside of the capital city, Atlanta, is not RO. “They don’t use RO in part because if they wanted to dispose of their brine, it’s really not possible because the ocean is a couple of hundred miles away,” says Mattingly. Instead, the plant uses a combination of ozone and biologically activated carbon.

Meanwhile in Arizona, in the southwest, water authorities are investigating soil aquifer treatment for water reuse. Here again, one of the main drivers for not using RO is because disposing of brine is impractical. Soil aquifer treatment comprises spreading grounds where treated wastewater flows through soil, using the natural processes of the soil to remove contaminants.

While such alternatives continue to be of interest, particularly at inland locations, the reality is that the “majority of the plants operating today are membrane based,” says Wetterau. Therefore the other way to address the challenge of an RO brine stream is to limit or eradicate it.

The industry’s holy grail in recent years has been to achieve so-called zero liquid discharge (ZLD), which has happened with qualified success. Dow, conscious particularly of the cost of ZLD, has begun developing the related concept of minimal liquid discharge (MLD), which gets close to ZLD but stops short of it, avoiding some of the potentially high costs involved.

“If you look at what the industry has been trying to do with ZLD, it sounds great in practice but it means a very high cost, and in many cases, the high cost prohibits people from taking the first step. We want to promote MLD, meaning that you go very, very close to zero, but you don’t necessarily get there,” says Pinto. “That pretty much cuts the cost in half because that last five per cent of the curve in the liquid discharge treatment is where your cost rises exponentially.”

MLD is particularly relevant in water reuse because the processes and technologies involved can become costly if using a combination of different processes, membranes, and crystallisers. “With MLD, you can potentially have a small crystalliser system by adding pre-treatment stacks with membranes that are more active and that, overall, reduces your footprint, your capex, and makes your application more efficient from an economic perspective,” says Pinto.

“Minimal liquid discharge pretty much cuts the cost in half, because that last five
per cent of the curve in liquid discharge treatment is where cost rises exponentially.”
Felipe Pinto, Dow Water & Process Technologies

In 2016, Dow launched its new range of Filmtec Fortilife membranes specifically for MLD and wastewater reuse applications. “In the next five to 10 years you’re really going to see a growth in these membrane applications. The drivers are water scarcity as well as environmental regulations,” says Pinto.

Challenge 5: Public perceptions

The fifth and final challenge to the potential future success of water reuse is one that water industry insiders arguably struggle with the most: Public perception.

“From a technology standpoint, those of us working in filtration techniques, and particularly membranes techniques, have absolute confidence that we have the technology, and that the issue is more hung up on the policy side and on the public acceptance side,” says Stover. He argues that from a safety perspective, controlled direct potable reuse is preferable to some of the ad hoc indirect potable reuse processes that have grown up in the US over the past decades.

“In the US, most drinking water comes out of rivers, and most wastewater treatment plants discharge to those same rivers. You literally can have a wastewater treatment plant discharging effluent into the river a mile upstream from where you’re pulling it out and disinfecting it and putting it into the public water supply. The prospect for DPR is, in my view, a lot safer and smarter, because you’re controlling and moderating the process, rather than just tossing it into the river and hoping that nature is going to do something to it,” he says.

Attitudes vary globally, with acceptance in Australia, Singapore, and the US developing faster than in Europe, where the market is still in the early stages. “I get the sense that Europe is a lot more nervous about potable reuse than Australia, Singapore, or the West Coast of the US. It rains more there, and they are not in as dire a need as we are. The places that have the least water are the ones that are looking at this first,” says Wettereau.

What next for water reuse?

Two plants are doing DPR in the world right now: Windhoek Goreangab Water Reclamation Plant (WGWRP), near Windhoek, Namibia, and Big Spring Water Treatment Plant in Texas, US.

The Namibian plant, designed by Wabag, started up in 2001 and has capacity to treat 21,000 m3/d of municipal wastewater. The water treatment begins with a nutrient removal process, and it is then polished in maturation ponds before going through an advanced, multi-barrier system at WGWRP. The system uses 10 process steps in total (see infographic, above), from oxidation and pre-ozonation, to coagulation and flocculation, to ultrafiltration and disinfection.

Meanwhile the Big Spring Water Treatment Plant in Texas, which started up in 2013, uses the treatment train that is emerging as the leader for water reuse, namely microfiltration, reverse osmosis, and ultraviolet disinfection.

In California, a string of indirect potable reuse projects, that is, groundwater replenishment schemes, are planned: There are eight approved projects, mostly in the south, including a couple at advanced pilot testing stage and which could potentially move to direct potable reuse; and a further 12 at various stages of development.

“Potable reuse is gaining traction very quickly. It’s moving faster than a lot of people realise.”
Justin Mattingly, Water Environment & Reuse Foundation

The sheer scale of some of California’s planned plants is pioneering for reuse. “They’re talking about 50 million gallons a day (190,000 m3/d) plants in California – they are big plants. On indirect potable reuse there are already two plants that are 100 million gallons a day or larger. So we already have large plants, but not for direct potable reuse. Orange County Groundwater Replenishment Scheme is 100 million gallons a day. The Changi indirect reuse plant in Singapore is 120 million gallons a day,” says Wettereau.

“California is really driving attention, and a lot of municipalities are copying that. California is a really influential state, and we’ll see that happening in other states. Once you have the US adopting these things, it gives a case for other countries to follow suit as well,” says Pinto.
Water reuse is evolving rapidly across the US, with Arizona moving toward approving reuse standards, Florida poised to build on its experience of non-potable reuse, and states like Colorado, Oklahoma, and Virginia weighing the options.

“Potable reuse is something that’s happening faster than most people realise. It’s no longer just a Southern California and Texas topic of interest. It’s a concept that’s gaining traction very quickly and it’s moving faster than a lot of people realise,” says Mattingly. “Based on the amount of requests for information that we get, I believe that it is a market that will be growing.”

Feed spacers can address the top three challenges in every RO system

Ivan Soltero, senior strategic marketing leader, Conwed

Conwed manufactures plastic netting for hundreds of industrial and consumer products in diverse industries. In filtration, we specialise in reverse osmosis (RO) feed spacers for RO wound elements, extruded cylinder tubes, sleeves, diamond and square extruded netting for filter pleat support, and core protection with a wide range of customisation options.

In 2013, we began a huge research and development effort to understand how feed spacers could impact overall RO filtration performance. Our journey has helped us to identify key challenges in the RO industry, how they are addressed and how our products can add value. We believe that feed spacers can address the top three challenges in every RO system: membrane damage, pressure drop and biofouling.

In addition to the full portfolio of standard feed spacers that we have commercialised for decades, for industrial water, brackish water reverse osmosis (BWRO), and seawater reverse osmosis (SWRO), we developed next-generation configurations that address energy consumption. Our Alternating Strand Design (ASD) for SWRO has been tested in reverse osmosis systems, providing improved pressure drop results and consequently impacting energy consumption. More recently, we launched our newest Ultra High Porosity (UHP) feed spacers, which have a higher porosity ratio to improve water pressure drop in BWRO systems, while maintaining equivalent flux and biofouling performance when compared to standard feed spacer configurations.

For decades, feed spacers have been perceived as an adjacent material in RO wound elements’ configurations that perform a specific function. We set ourselves on a path of discovery to push the limits of our feed spacer technology and understand what we could do to help improve the water filtration system. We developed our Reverse Osmosis Series of technical papers to share our insights, and have collaborated with industry researchers through Conwed Joint Efforts to better understand the role of feed spacers. We want to drive feed spacer innovation and find the magic that may have been missed until now.

* In December 2016, Conwed was acquired by Schweitzer-Mauduit International, a global provider of highly engineered solutions and advanced materials for a variety of industries. In addition to Conwed, SWM’s Advanced Materials and Structures (AMS) division includes DelStar Technologies, acquired in 2013, and Argotec, acquired in 2015. DelStar and Conwed, competitors for decades, have now joined talents and technologies under SWM to lead feed spacer development globally.

The Long Read on water reuse is in association with Conwed.

The Long Read: Where’s the innovation in brine treatment?