Overview of ultrafiltration in municipal drinking water markets

By Christophe Cassant

• In/out or out/in and custom engineered or rack systems, the CEO of Aquasource, a Degrémont/SUEZ Environnement company, takes a 'bipolar' look at the UF market approach in Europe vs. North America.

Aquasource UF membranesTOULOUSE, France, Sept. 29, 2009 -- There has been a lot of activity in the ultrafiltration (UF) membrane market in recent years. The in/out PES module of manufacturers Norit and inge AG has shown fast growth over these last two years. Dow and GE launched last June 2009 their new out/in PVDF modules. Why today two different technologies and strategies? How can they both be successful? This article gives an understanding of the bipolarisation of the UF markets, technologies and applications sustained by these two different approaches in North America and Europe. But further than that, it points out cultural and economic factors that are at the origin of the two different philosophies. At the end, Aquasource presents its UF product range strategy to competitively match with the all markets, taking into account their respective applications and philosophies.

1 - Introduction to the water treatment line and its UF applications

A water treatment plant designed to produce drinking water generally comprises several treatment stages. Combining these stages enables the required water quality to be achieved. This is then referred to as the treatment line.

The simplest treatment line comprises a single step: UF applied, for example to karst water -- or water from aquifers influenced hydrogeologically by limestone and/or dolomite and extremely susceptible to contamination. For most raw waters, however, satisfactory treatment can only be achieved through use of "multi-barrier" treatment lines, composed of several treatment stages.

These treatment lines comprise specific technologies for each step in the treatment process. Equipment manufacturers and engineers talk of applications for their technologies. These applications are defined either with reference to position of the technology in the system, or to reflect the nature of the water to be treated -- and even sometimes by mentioning the destination of the water treated using this application. For example, UF "technology" can be found in the following "applications": polishing or pre-treatment of reverse osmosis (RO), treatment of coagulated clarified water and wastewater recycling.

The choice of the sequence of treatment steps in the treatment line and design of each one requires a high level of competence, which is only in the possession of companies specialized in water chemistry and water treatment processes.

2 - Two ways of optimizing the costs of a water treatment plant

The drinking water plant is a costly infrastructure burden for a local authority and a large number of studies have to be carried out in order to optimize the cost. Close observation enables us to identify two major trends, each associated with different histories and cultures, different economic models, or administration models that are centralized to varying degrees.

The integrated approach (or Design & Build), which is predominant in France for example, is characterized by the fact that an "expert" stakeholder -- the water "treater" as named by the French big companies -- is given plant input and output water quality data and is then responsible for optimizing the overall design of the treatment line. In this arrangement, the scenario is that the best plant capital investment and operating costs (CAPEX and OPEX) will be the result of a tailor-made treatment line. Under the terms of "design and build" contracts, the French water "treaters" have developed a very real capability in choosing and designing the best possible sequence of treatment steps in order to achieve the required performance. This know-how is based on expert knowledge of the optimum efficiency and effectiveness of each technology according to the variety and concentration of pollutants.

Figure 1. Capital expenditures (CAPEX) design approach

For the water "treaters", this integrated solution has high engineering costs, which are necessary for implementing tailor-made designs for the various treatment steps, in order to produce a treatment line that's optimized overall. But the plant overall design-to-performance may also lead to inadequate safety coefficients. And meeting the plant construction deadline is often compromised by the complexity of the engineering studies. Perfect project management is required if an approach such as this is to be a success.

The segmented approach (part of the Design-Bid-Build or DBB process), which is historically predominant in North America, for example, is underpinned by industrial prefabrication dictated by marketing analyses. In an arrangement such as this, the consultant engineers approximately design the treatment steps in the system and then choose standard off-the-shelf structures and equipment. The standard range of equipment is industrially manufactured, with the initial investment necessary for manufacturing being amortized by sales. Each plant project thus avoids incurring significant dedicated engineering design expenditure, thereby bringing down its cost. Adopting a standard equipment item in a system, however, generally entails the plant design being larger than what's theoretically strictly necessary. Therefore, savings gained from the standard equipments can be partly offset by the mismatch between requirements and performance offered by the product ranges.

Another observation is that the industrial approach becomes less applicable, the greater the plant capacity, because industrialization of a standard range of technologies is only profitable above a minimum volume of sales which rules out rare or unique plants. For each technology/application pairing, there's thus a certain capacity threshold below which the technology must be standardized and above which it's preferable to opt for engineering dedicated to each project.

3 - Time factor vs. sustainable development

Still, all of the above considerations fail to take account of the time factor. Demographic and sociological changes, expanding health legislation and the increasing share of financial services in the economy mean that this criterion is today a priority in the evaluation and success of a project, sometimes ahead of the plant investment cost.

Two observations must be made concerning the municipal sector:

• It takes too long to adopt new technologies and to build new water treatment plants. It's easy to see that the time to build a plant will be shorter if standard technologies are adopted, rather than engineering a tailor-made structure. The industrial approach described above is therefore perfectly logical in that it helps shorten these lead-times.

• Technological innovations already proven at the industrial customers, where development benefited from the stimulus of competitiveness, dynamic management techniques, better risk management by pilot tests, small plant sizes and generally short business cycles, are already virtually standardized when they hit the municipal scene. The risk that then arises is above all in the changes of scale demanded by the larger municipal plants. In the model which gives priority to engineering, however, the water "treaters" must first of all take on-board what's new, then evaluate it (often through pilot tests) before integrating it into optimized systems or creating new ones that are competitive.

Figure 2. Operating expenditures (OPEX) design approach

In the light of the above, it would seem that the trend is towards the industrial model because standard equipment offers economical and proven solutions that are quicker to implement. Based on this observation, the technology manufacturers will tend to constantly raise the above-mentioned threshold, in order to produce ever-larger standard equipment ranges.

The limit of this strategy is determined by the level of risk taken by the equipment manufacturer. The innovation cycle is constantly becoming shorter and standardized products have to pay for themselves before the next technological breakthrough. It's therefore often preferable for an equipment manufacturer to make a strategic decision to abandon expanding the range upwards and focus its research and development (R&D) efforts on a future generation of products.

It would seem therefore that the die is cast: The "marketed" industrial approach will be gradually eroding the overall engineering model owing to the faster innovation and project implementation times.

The present crisis could indirectly affect this situation. Reflecting conclusions of the Environment Summit in France, governments and populations are today increasingly concerned about the environment. Energy and water have to be preserved, cars must be fuel-efficient and public transport is back in vogue, etc... Water treatment stakeholders are gradually taking these social changes on board and the first positive energy water treatment plants are making their appearance.

One of the main features of sustainable development is that it calls for long-term thinking, the main aim being to preserve the human race, its environment, the climate and biodiversity, in particular by limiting our consumption of non-renewable energies. With regard to water treatment, this change in attitudes obliges us to conduct analyses in which plant operating expenditure (OPEX) is accorded at least as much importance as capital expenditure (CAPEX) investment costs.

So one might say the current crisis could have the effect of promoting the overall approach and water treatment engineering which is more OPEX focused. Maybe, but boosting government level infrastructure investment could help the world out of its economic troubles before the long-term intentions can be realized. Will there then be anything left?

4 - The example of the French municipal water market

Historically, France has adopted development master plans consisting of extensive networks, and high-capacity plants as much as possible -- 2% of the total number of drinking water plants produce 50% of the total cubic meter (m³) capacity.

Figure 3. Aquasource moduleArrangements such as treatment at the point of consumption or point of discharge were not chosen for reasons of infrastructure sharing and expenditure rationalization. In France, there are no private residences or compounds, leisure centers, hospitals or schools supplied by private drinking water plants, as in North America for example.

Traditionally, the cities (and urban communities) have been well-equipped and their plants regularly renovated. Still, France is lagging behind in ensuring compliance of its small and medium municipal installations with the standards (98% of the total number of DW plants has capacity below 10.000 m³/day). It's in these municipalities -- which often have limited resources to finance their projects -- that efforts will be needed in the next few years.

Although the French local authorities' municipal market is financially vulnerable, it's also subject to water quality health constraints and quality demands of consumers (taste, color, odour, etc.). These rising demands, combined with deterioration of resources and the appearance of new pollutants, is driving adoption of innovative technologies and systems in the drinking water sector. Failing which, sales of bottled water will rise and the environment will continue to deteriorate.

The government recognised the problem when it issued a decree in February 2009 implementing a proactive policy in favor of innovative small and medium companies and technologies. This essential step forward will provide the ministries and local authorities with a decisive tool in opening the door to innovation in public contracts.

Sales of UF technology, which is still innovative when compared with the number of installations in operation in France (approximately 160 UF plants compare to 15.300 existing drinking water plants, all capacities included), are currently on the rise. Equipment design has continuously improved in recent years, in order to meet the strict requirements of the municipal customers and their operators (both public and private), while the current upward trend in the drinking water market will only confirm that UF is here to stay.

The issue that the various water treatment stakeholders face today is the following one: how to equip small or medium sized local authorities with drinking treatment meeting current or future quality requirements, both rapidly and at least cost?

5 - Ultrafiltration approach in Europe and North America

Either alone, or as part of a larger treatment line, UF provides the solution to a good number of tomorrow's challenges. It eliminates bacteria and above all the risk of contamination by Cryptosporidium and Giarda, while maintaining the natural mineral balance of the water. It improves the colour and taste of the water. It guarantees the hygienic quality of reused wastewater (preservation of resources). When combined with activated carbon it makes it easier to treat pesticides, and so on.

At the end of the 1980s, the SUEZ Environnement group pioneered UF in the French municipal market, using a cellulose acetate material (naturally hydrophilic) to produce hollow fibers initially used for treating karst waters for drinking water applications. These fibers filter the water from the inside to the outside (in/out technology). On the European market, most membrane promoters have since opted for pressurized in/out technologies (Norit, Aquasource, inge, Hydranautics, etc.). In the steps of SUEZ, when it came later on to the surface water application, the European promoters mainly opted for multi-barrier treatment lines involving clarification, filtration and then UF stages. Also, and to increase the fiber tolerance to the pH range of chemical cleanings required for the surface water application, European membrane manufacturers developed polysulfone (PS) and polyethersulfone (PES) fibers.

Figure 4. Aquasource membraneIn the drinking water sector, the multi-barrier system is characterized by systematic use of UF as the final treatment phase (polishing), for example, in Angers, France, producing daily 120.000 m³ of drinking water or Orleans, France, producing 40.000 m³ of water daily. This choice decreases the actual costly UF surface area through use of filters upstream. It also brings longer membrane lifetime, improves UF operating costs and increases global plant safety.

In North America, now that GE (Zeeweed 1500) and DOW launched in June 2009, their new modules, one can say that the trendy technology is clearly based on pressurized hollow fibers filtering from the outside to the inside. This out/in technology is promoted by Zenon, Pall, Westech, Toray… mainly with PVDF fiber material. As for PS and PES, PVDF offers a great tolerance to the pH range of chemical cleaning agents and, therefore, historically enabled North American promoters to expand UF applications in particular by opening up the settled water or direct treatment market.

The North American companies were thus able to boost their equipment sales by claiming that drinking water plants no longer needed the traditional filtration stage. But this argument often ignored the total operating cost of the plant. With this type of short treatment line, the UF flows are lower (unfiltered water) and cleaning requirements with reagents are frequent (25 times more than in polishing, for example). The North American solution demonstrates the market dominance of standard equipment to the detriment of engineering that supports the European multi-barrier approach. In France, the two world's water treatment leaders SUEZ and Veolia have managed to impose themselves as the primary contacts for customers, in particular ahead of civil engineering firms. Their powerful engineering resources have given them a clear view of the total cost of plants, enabling them to determine the optimum balance within the treatment line. Still, it would seem that the highly logical multi-barrier systems approach in UF isn't one that appeals to North America, where the influence of standardized equipment manufacturers has encouraged consumption of constantly rising volumes of membranes.

It should, however, be noted that differences in regulations also have impacted the treatment approach. For example, the regulation level for atrazine in Europe is 30 times lower than in the USA, requiring most of the time an adsorption treatment stage and therefore promoting a multi-barrier approach not only for disinfections but also for contaminant removals.

6 - Aquasource strategy

Aquasource is a Degrémont (SUEZ Environnement) affiliated company that has been specializing in UF for 20 years. The first worldwide municipal UF drinking water treatment plant was built by Aquasource in France in 1988 (Amoncourt). Since then, the company has installed more than 130 UF municipal drinking water treatment plants only in France.

As the company's market is global and it's permanently confronted by the two economic models described above, Aquasource therefore had to make strategic choices to be competitive in all market configurations.

Aquasource skid-mounted UF systemThe company today sells two main categories of standard products:

• Stand-alone, standardized, skid-mounted UF units, equipped with all the peripherals: pumps, pre-filters, valves, etc., necessary for treatment by UF. These units are delivered on a turnkey basis, ready for connection.

• UF modules, accompanied by the associated support system: the rack. The rack facilitates implementation of the UF technology by water "treaters", who can then easily install the modules in its plant layout and focus on the process to be used and the installation of all peripheral equipments.

All these products are designed and manufactured using an industrial approach giving competitive costs and minimum lead-times. Consequently, as mentioned above, Aquasource was faced with the challenge of defining the limit threshold for the size of its standard equipment.

For the stand-alone UF units (skids), the threshold adopted corresponds to units outputting a unit flow of 200 m³/hour, currently the highest in the world. These standard units are designed for drinking water treatment plants with a capacity of less than 10.000 m³/day, because once you have more than three stand-alone units installed side by side, it becomes more economical to share some of the peripheral equipment (pumps, valves, pre-filters), and to switch to the rack range. This does entail certain engineering costs for the water treater, but they remain lower than would result from having redundant peripheral equipment on each stand-alone unit.

For drinking water plants with a capacity of more than 10.000 m³/d, Aquasource can offer racks to support UF modules. Its rack was designed to be the common denominator for all UF plants, regardless of the applications and configurations. This choice makes the rack the largest possible "standardisable" equipment item in a high-capacity UF plant. The main criteria driving the design of its rack were its price, its floor footprint -- in order to minimize civil engineering work -- and its user-friendliness.

Finally, the Aquasource industrial site in Toulouse can produce fibers made from both materials, hydrophilic polysulfone and cellulose acetate. Each of these materials has its own advantages, as mentioned earlier. The racks can be equipped with one or other of these materials, depending on characteristics of the water to be treated and the system chosen by the water "treater". For example, for treatment of surface waters feeding the new UF drinking water plant (10.000 m³/day) of the city Nantong, China, Aquasource decided to supply hydrophilic polysulfone modules. Its ability to produce two membrane materials enables it to place its technology in the short, more North American treatment lines (after clarification) and in the longer, more European treatment lines (polishing). Both CAPEX and OPEX oriented UF plant designs are taken into consideration with the Aquasource product range.

The strategic choices made by the Aquasource company today guarantee access to all ultrafiltration markets, regardless the customer, the application, the geographical situation of the plant, the system or the size of the projects.

Author's Note: Christophe Cassant is vice president of the SIEP -- the French National Potable, Process & Leisure Water Association -- and CEO of Aquasource SAS, a business unit of Degrémont Technologies, based in Toulouse, France. Contact: [email protected] orwww.aquasource-membrane.com

###