June 13, 2024 Ι In the gwf Wasser/Abwasser issue 5/24, Dr. Jean-Jacques Lagref and Mr. Robert Reisewitz contributed an article on silicic acid incrustation in reverse osmosis systems. This article is now also available in English.

In the field of water purification, Reverse Osmosis (RO) systems are fundamental, especially for producing drinking water or ultrapure water essential in several industries, including municipal, pharmaceutical and technological sectors. Although they are efficient and widely used, operating RO systems at high recovery rates presents challenges. Scaling, a phenomenon that results in significant operational difficulties and efficiency losses, is a particular issue. Some scaling events remain manageable, like the calcium carbonate scale, if some cleanings in place are done on a cost-efficient time frequency. Among the various types of scaling, silicate scaling poses however a unique and complex challenge due to the nature of silica and silicate compounds in water [1]. Natural water contains a certain level of silica, and its concentration is region and location specific. The silicon element is by weight the second most abundant element on earth, thus explaining why natural water sources usually contain a level of silica from 1 to 40 mg/L, and in some less common situation from 40 to 100 mg/L. Volcanic islands for example are known to have natural waters highly rich in silicate salts.

Silicates, when present in RO feed water, can lead to reduce rapidly the entire system performance and consequently raise its maintenance and operational costs. Reverse osmosis membrane autopsies revealed that silica is the major inorganic mineral fouling problems accounting for nearly 50% of all scaling cases (Chart 1). Silica is known to be difficult to deal with because it is multi-form. Below pH 7, monomeric silica tends to polymerize to form oligomeric or colloidal silica. At higher pH particularly above about pH 9.5, silica can form monomeric silicate ions. All these forms may exist at any time, depending on the history of the system. Due to these factors, it’s typical to encounter a variety of silica forms, including monomeric, oligomeric, colloidal, or particulate, as well as silicates salts of magnesium, calcium, iron, and aluminum, among other unique silicate salts. Throughout this study, we will simply refer to this mixture as either silica or use the term silicate.

The problem intensifies with the combined presence of some additional cations even in low concentration. The aluminum cation is one example [2], Iron (II/III) is another common one. The probability of such scale incident is often increased when certain coagulants are used in pre-treatment, which, despite their effectiveness in removing particulates, inadvertently might foster an environment conducive to a rapid alumino-silicate scale formation. Alternative coagulant will be described here. For example, in some severe cases, when a concentration of aluminum is superior to 100 ug/L, it takes only few hours to see the membrane flux declining by 60%. The RO membranes being then hardly recoverable in terms of flux and pressure performances. Although dangerous cleaning chemicals like ammonium bifluoride (NH4HF2) or hydrofluoric acid (HF) can be used to remove silica fouling efficiently, the use of such harsh cleaning chemicals brings the risk of equipment damage, operators’ safety and environmental concerns.

Understanding, as much as possible, the parameters intricacies, interactions, their dynamics, the binding forces in place and their final impact on the silica crystalline phase formation is crucial for the development of an effective management strategy and new efficient antiscalant ingredients. Despite tried, the thousands of possible interaction’s combinations with cations, short silicates dimers/trimers, long carboxylate polymers chains, polymerized silicate chains partially protonated or totally deprotonated makes any reliable computer simulation impossible. Would an intensive learning AI algorithm, able to learn from thousands of plant data, be the future option is a realistic scenario. In parallel, a side revolution is also taking place with the introduction of „environmental friendlier“ technologies and novel treatment methodologies for addressing these challenges, emphasizing even more the need for stable, sustainable, and performing solutions in highly pure water production.

In the present work, we review the current state of silica treatment technology in RO desalination. We need to delve into the latest understandings, discoveries and review the promising innovative approaches. From the chemistry behind silicate interactions to advanced mitigation strategies, we explore the latest research findings and practical applications that set new guidance in the management of silica and silicate scaling, ensuring the longevity and efficiency of RO systems in the quest for sustainable and reliable high-quality water production.

Chart 1: Representing the proportion of minerals typically found in RO membrane scaling based on common occurrences reported in literature. The silica (SiO₂) is a major component, making up 50 % of the scaling, followed by calcium carbonate (CaCO₃) and calcium sulfate (CaSO₄). Barium sulfate (BaSO₄), strontium sulfate (SrSO₄), and magnesium hydroxide (Mg(OH)₂) are less common but still significant contributors to scaling

The Scaling Mechanisms and the critical Chemical Interactions

The process of silicate scaling in reverse osmosis (RO) systems is intricately tied to the chemical interactions between silica, cations, anions, water molecules, “free” protons and other elements present in the feed water. Silica, primarily found as monomeric silicic acid in natural waters, can undergo several routes like the polymerization, transforming into polymeric and colloidal forms that progressively lose affinity with the surrounding water molecules. Interesting is that this polymerization is significantly influenced by the presence of multivalent cations such as calcium (Ca²⁺), magnesium (Mg²⁺), and iron (Fe²⁺/Fe³⁺). These ions can interact with silica to form complex nanomolecular entities which will be pressed on the RO membrane surfaces, starting then a layer-by-layer growing process. This growth will translate into a progressive reduction of the system efficiency and operability, leading to increasing maintenance and repair costs.

One explanation for the mechanism influencing the polymerization is that the silica in feedwater is existing under the predominant monomeric silicic acid (H4SiO4) form. During the RO process of water passing the polyamide membrane, the concentration phenomenon of silica leads to an unstable supersaturated state, which instability will materialize by the initiation of the silica polymerization. At this precise moment, a tremendous number of near-by hydrogen bonds are created, shifting the polymer chain in or the other way, while a dynamical flow of cations will initiate random electrostatic forces. The multivalent cations such as Ca²⁺, Mg²⁺, Al3+ and/or Fe²⁺/Fe³⁺ will act as bridges between negatively charged silica particles and atoms, bending chains in a closed loop, reducing the direct contacts with water molecules. These cations will also neutralize the surface polymer’s charges and facilitate the agglomeration of silica particles into soft/sticky colloid particles. The process is known as „destabilization“ or „coagulation“. Observed is a different behavior based on the relative presence of this metallic cations. Why a reactivity difference between Calcium/Magnesium and Aluminum/Iron? The Pearson concept of Hard and Soft acid/base to explain the thermodynamic of reaction is the one which could explain the situation. Person reported that hard acids prefer to associate with hard bases (forming ionic bonds) and that soft acids prefer to associate with soft bases (making covalent bonds). The “hardness” concept being defined as
Hardness n = (Ionization energy – Electron Affinity)/2

The Aluminum (III) or Iron (III) do have a high ionization energy and given their tiny radius size do have strong charge density, scouting for strong electron donors like a Si-O(-) site.

The Calcium mainly and Magnesium in a lower extent, in contrary have a lower ionization energy and do have a larger radius, giving them the ability to make bonds with a higher covalent degree. The resulting mineral silica scale being more in the form of a Calcium and Magnesium complexes with silicate anions. This explains the appearance of that scale as a fine white powder when the water contains Ca²⁺ and Mg²⁺.

For Iron, the interaction with silica is way more complex [3]. Indeed, both species are influencing themselves: iron triggers the scale of aqueous silica, and aqueous silica significantly inhibit the Iron polymerization and solid-phase formation [4]. Iron can trigger the precipitation of Silica when under the iron oxide form forming a mixed scale: the tiny iron oxide particles (ferrihydrite) will then adsorbate silica under monomeric or oligomeric species and both species will scale. The binding energy of the adsorbate and the oligomeric structure of silica will even depend on the rate of ferrihydrite precipitation, further detailing how complex the mechanism is in reality [5]. Understanding these chemical interactions is crucial to anticipate the difficulties during the RO design phase. The careful management of these parameters makes possible to reduce the likelihood of silica polymerization and its subsequent scaling, while choosing the most appropriate antiscalant formulation.

Parameters: Impact of pH, Concentration step and Temperature

It is commonly understood that the behavior of silica in water treatment systems is profoundly affected by both pH and temperature parameters. These ones directly influence the solubility and the polymerization rate of silica, thereby affecting its potential scaling behavior. At lower pH levels, the hydrolysis of silicate ions to form silicic acid is favored, keeping silica in a more soluble and less reactive form. However, while the pH increases, the condensation reactions become more prevalent, leading to the formation of siloxane bonds and the subsequent polymerization of silica. The polymerized Silica will later adopt a new solubility behavior, making more complex the management of the silicate scaling probability. In parallel, side reactions take place as the pH moves towards neutral and then to alkaline (higher pH) and the solubility of silica increases significantly. This is due to the deprotonation of silicic acid (H₄SiO₄) into various silicate species (such as H₃SiO₄⁻, H₂SiO₄²⁻, etc.) that are more soluble in water a priori. Indeed, the deprotonation partial or total is not a guaranty of a full solubility situation, because it will now depend on the nature of the cations associated with the negatively charged silicate molecules. As an example, if the water also contains calcium carbonates (CaCO₃), the situation becomes trickier for 3 reasons:

• We have seen that calcium ions (Ca²⁺) can interact with silicate ions in the water. This leads to the formation of calcium silicate, which is much less soluble than many other silicate forms. As a result, in water systems containing both silicates and calcium carbonate, a reaction can take place where the calcium ion is shared or exchanged between the present salts and calcium silicate is forming because of the increased pH. In this case, both calcium silicate and calcium carbonate salts would enter a low solubility momentum.
• The presence of calcium carbonate can act as a soft alkaline salt and affect the pH of the water. The buffer capacity of the calcium carbonate salt will be sufficient to stabilizing the pH in a near neutral to slightly alkaline pH range, which represents a low silicate solubility zone. This multi-dimensional dynamic cocktail explains why such brackish waters, rich in silicates and carbonate salts, are difficult to operate in a RO unit. Additionally, despite not clearly demonstrated yet [6], it is thought that a supersaturation and precipitation of metal carbonates could trap the dissolved silica particles smaller than 220 nm and trigger the scale of silicate. X-ray diffraction analysis did not confirm the presence of common calcium carbonates or sulphates but instead showed the presence of a suite of complex minerals, to which amorphous silica and/or silica rich compounds could have adhered.
• In reverse osmosis operation, the salt concentration phenomenon usually leads to a higher concentration of salts, and thus finding a solubility limit remains easy. However, in that case, the higher concentration of species will lead to a higher reactivity and cross-reaction frequency between silicates, carbonates and cations, emphasizing the ideal condition for Silicate to form larger particles like colloids, silts or clays. Recent studies revealed that the higher the recovery rate, the higher silicate will adopt the colloidal particle form, with a spherical nanoparticle shape and diameters ranging from 20-50 nm [7]. Christopher Turner in 2020 reported [8] in its publication that dissolution and polymerization reaction of silica fouling was dependent on the colloids’ radius of curvature. The experiments were done with colloids of 100 and 300 nm diameter and at pH 8.5 to 9.5. The study concluded that the particles’ surface energy had a major impact on the silica dissolution rate constant. The discovery let the team to conclude that dissolution and redeposition is responsible for the problematic silica fouling behavior during RO treatment.

Temperature plays a critical role in how well silicate dissolves in water, on top of other unpredictable factors. Generally, warmer water dissolves salts better, and this holds true for silicate due to the endothermic nature of dissolution. As the temperature goes up, water molecules move faster, their vibrations frequencies increase, making it easier to break the ionic and Van der Waals bonds in silicate structures. This effect of temperature is crucial to consider during the design of reverse osmosis (RO) systems, especially in relation to feedwater temperature and seasonal weather changes. It’s also important during the plant’s design phase to remember that metallic pipes exposed to the external environment can cool down, potentially becoming sites where scaling, like silicate starts to crystallize.

Additionally, Ken Sorbie and Lorraine Boak, observed in 2021 that silicon tends to react with magnesium instead of calcium when both ions are present at the same time [9]. However, at higher concentrations of calcium ions, these ions may bridge with the silica, thus forming a calcium silicate 3D network. The tendency to form calcium silicate scale increased with the increased initial calcium concentration and at higher temperatures. It was also demonstrated that the morphology of the scale precipitates was highly dependent on the relative ion concentration as well as temperature.

The interplay of numerous parameters like the pH, temperature, ratio of Ca²⁺/Mg²⁺, concentration process and the concentration of sodium cations means that the optimal condition for minimizing a silicate scaling involves the absolutely necessity to maintain the feed water in a constant physicochemical status in which the silicate solubility is maximal.

These parameters form a fundamental basis of the strategy aimed at mitigating silica scaling in a reverse osmosis unit. By understanding and manipulating these parameters, water treatment operators can significantly improve the stability of a RO plant even at high recovery rate, thereby improving the efficiency and longevity of RO membranes.

Mechanical collection of scaling/fouling on the UO membrane surface

Operational Strategy, Pretreatment, Antiscalan

Real-world applications provide valuable insights into effective strategies for managing silicate scaling. One striking example is related to a coal seam gas-associated water treatment plant in Australia where the silica values in the feedwater are particularly elevated. A set of research studies have shown that managing colloidal silica through regular cleaning cycles and regularly monitoring silica concentrations could significantly mitigate the scaling issues, enhancing RO membrane operational efficiency, and also reducing the treatment costs [10]. The research uncovered that colloidal silica predominantly forms in the later stages of the RO process, particularly under high recovery conditions (>80%) and when silica concentrations surpass its solubility limits. This colloids formation could be managed through regular cleaning cycles, underlying the importance of monitoring and managing colloidal silica to optimize RO operations, particularly in high-recovery applications.

Operational strategies and system design play crucial roles as well in mitigating silicate scaling in RO systems. One approach involves the careful management of chemical additives during the pre-treatment phase [11]. The study of Park and Yeo (Desalination 2024) highlighted the detrimental effects of using polyaluminum chloride as a coagulant, which increased aluminum concentration, consequently exacerbating aluminum silicate fouling on RO membranes. An effective mitigation strategy is the use of polydiallyldimethylammonium chloride as an alternative coagulant, significantly reducing both organic and silicate fouling. This emphasizes the importance of coagulant selection in controlling residual aluminum ions and preventing fouling layers‘ accumulation on RO membrane surfaces, ensuring the longevity and efficiency of the RO process in producing ultrapure water.
From experience, a solution was to remove the silica present in the RO through a pretreatment involving the use of a lime-soda ash process. The calcium carbonate added in conjunction to sodium aluminate and ferric chloride triggered the rapid precipitation of a complex of silicic acid, aluminum oxide and iron. Operating it at 60°C the process would allow a RO feedwater to bear a remaining of <5ppm of silicate, solving the inherent silica scaling risk.

The type of RO membrane will also influence the severity and the rapidity of the silica scale formation. Silica scale deposition occurs by monomeric silica species at low silica saturation levels. At high saturation level, however, silica deposition involves the colloidal particles, followed by some monomeric deposition. There the surface property of the RO membrane will influence the silica deposition. According to the study by Tong et al. [12], silica scaling on positive RO surface was more severe compared to negatively charged surfaces. This implies that silica deposition is affected by the charge interaction between silica species and RO filter layer surface. Once the silica deposition occurred under a neutral pH, the polymerization of dissolved silica occurs, building up then a set of amorphous layers transforming into a glass-like film.

From a prevention point of view, the best actual solution is focusing on the use of effective antiscalant chemical products, with ingredients being able to act on the dispersion mainly. We saw that the particulate character of silica is emerging when the concentration phenomenon is applied and where the monomolecular species starts to condense in oligomers, then to colloid particles. Acting early on this particle formation step is key to keep silica soluble or at least non filtrable. Here the efficiency of the antiscalant solution will directly come from the suited components and their purity. It is critical that the raw materials are well chosen and free from unknown side products. Two possible mechanisms for controlling silica or silicate salts from fouling or depositing on a surface during a process are: 1) inhibiting precipitation of the material from the process water and 2) dispersing the material once it has formed in the bulk water to prevent it from attaching to surfaces. The exact mechanism by which a specific scale inhibitor functions, however, is not well understood. The inventions made in the 1990s are concerned with the use of certain low molecular weight water-soluble (meth)acrylic or maleic acid-based polymers considering a mixture of poly (meth)acrylic acid or polymaleic acid (molecular weight from 1,000 to 25,000) and a water-soluble copolymer or terpolymer (molecular weight from 1,000 to 25,000) [13]. This is still the state of art effective technology in place in 2020-2024. Pragmatically, we could expect that the amount of dispersing antiscalant ingredient would be proportional to the concentration of silica, but by experience a limit of this relation is achieved, above which an excess of antiscalant starts to behave negatively promoting the bio-fouling reaction.

Future: Emerging Research and Technologies

As the demand for water produced through reverse osmosis (RO) technology increasingly becomes a standard expectation, there’s a growing push for more affordable solutions. This shift means any approach to managing silica scaling needs to be both effective and cost-efficient to gain acceptance in the market. However, tackling and preventing silica scale effectively requires a deep understanding of the complex chemistry involved. Over time, this knowledge fosters the development of innovative solutions.

In 2007, despite potential risks to RO membranes, a Greek research team pursued an inventive strategy by utilizing cationic polymers to increase silicate solubility. Their approach relied on creating an electrostatic complex between two polymeric structures [14]. They tested the effectiveness of polymers like polyethyleneimine (PEI), polyallylamine hydrochloride (PALAM), and poly(acrylamide-co-diallyl-dimethylammonium chloride) (PAMALAM), finding that the right balance between the cationic charge and the polymer’s structure is crucial for enhancing silica solubility and advancing eco-friendly water treatment methods. However, it’s important to proceed with caution when applying this technology to RO systems, as cationic polymers can permanently attach to the polyamide layer of RO membranes, leading to reduced performance. A decade ago, a groundbreaking approach by an Indian research team showcased the innovative use of algae biomass. This method kept silica dissolved in water for 12 hours at concentrations exceeding 450ppm. The technique involved attaching polyacrylamide and polyacrylic acid branches to a core, creating a dendrimer with a star-shaped structure derived from algae biomass. These algae-based core dendrimers work by slowing down silica’s polymerization process and acting as nucleation sites, which prevents the formation of large silica particles and enhances silica’s solubility in water.
These cutting-edge solutions represent the latest successful polymers in combating silicate scaling, especially important in today’s global push towards high recovery rates and a zero liquid discharge approach in water treatment.

Conclusion and Guidance for safe RO operations

Considering the intricate chemistry involved, the increased occurrence of silicate scaling, and the widespread adoption of reverse osmosis technology, effectively managing silica scaling remains a significant challenge into 2024. This situation highlights the need for deeper understanding to fully address the issue. This situation also highlights the importance for using high quality, high performance specialized antiscalant formulations. Those requires specific polymers able to interact with the silica oligomers, preventing their further growth. Future approaches will likely require adjustments to a variety of physical, chemical, and design factors. The way silica interacts with other ions, such as calcium, magnesium, iron, and aluminum, along with variables like pH level, temperature, and system recovery rates, adds layers of complexity to this challenge. Recent research and case studies have underscored the importance of considering these issues early in the system design process, suggesting a proactive approach is necessary for effective management. The following provides basic guidance based on the knowledge accumulated up to 2024.

1. System Design and Operation: Consider incorporating early system design strategies that minimize the likelihood of any scaling. This includes finely setting reasonable recovery rates and maintaining consistent cleaning cycles, especially in applications with high recovery (>70%). Fine-tuning operational conditions, such as pH levels, temperature, and smooth 24/7 constant operation, will reduce the chances of scaling.
2. Regular Maintenance: To maintain peak performance of reverse osmosis (RO) systems, it’s crucial to regularly clean the system and replace membranes. Crafting a maintenance plan tailored to the unique scaling challenges faced by your RO system can prolong membrane longevity and enhance water recovery efficiency.
3. Monitoring and Management: Consistently track silica levels in both the feed water and the reverse osmosis system to detect potential scaling issues early on. Monitor the bank operating pressures: few grams of silicate scale per RO element is enough to increase the operation pressures by 2 bars. Establish a thorough water analysis program that examines the unique chemical makeup of your water, allowing you to customize your treatment approaches effectively.
4. Chemical Treatment: At the pretreatment stage, choose coagulants that are gentle on the system and do not trigger a silica scaling. Polydiallyldimethylammonium chloride is a promising option that minimizes both organic and silicate fouling, especially in the semiconductor UPW application. In non-membranes application, consider using cationic polymers like polyethyleneimine and polyallylamine hydrochloride. These have been effective in increasing silicate solubility and inhibiting scale buildup.
5. Future Technologies: Keep up to date with the latest studies and advancements, like the introduction of poly(amido amine) dendrimers as a way to prevent silica scaling. These new materials are not only environmentally friendly but also effective. They work by changing the shape, compactness, and stickiness of scale deposits, which greatly reduces the formation of both silicate and calcium carbonate scales.
6. Research and Collaboration: Case studies revealed the benefits of staying updated on the latest in scaling prevention and water treatment by continuously working with industry professionals and universities.

By employing a tailored and high-performance antiscalant, you can safely manage most risks associated with silicate scaling. The six recommendations provided offer a comprehensive strategy for operating a reverse osmosis plant efficiently and cost-effectively.

References

1. Kucera.J., 2015, Reverse Osmosis: Industrial Processes and Applications, May 2015, Online ISBN:9781119145776, DOI:10.1002/9781119145776.
2. D. Park, I.-H. Yeo, J. Lee, et al., Combined impacts of aluminum and silica ions on RO membrane fouling in full-scale ultrapure water production facilities, Desalination (2024), https://doi.org/10.1016/j.desal.2024.117380
3. Sahachaiyunta, P & Koo, T & Sheikholeslami, R. (2002). Effect of Several Inorganic Species on Silica Fouling in RO Membranes. Desalination. 144. 373-378. 10.1016/S0011-9164(02)00346-6.
4. Pokrovski, Gleb & Farges, Francois & Hazemann, Jean-Louis. (2003). Iron (III)-silica interactions in aqueous solution: Insights from X-ray absorption fine structure spectroscopy. Geochimica et Cosmochimica Acta. 67. 3559-3573. 10.1016/S0016-7037(03)00160-1.
5. Swedlund, Peter & Sivaloganathan, Sivashanthy & Miskelly, Gordon & Waterhouse, Geoffrey. (2011). Assessing the role of silicate polymerization on metal oxyhydroxide surfaces using X-ray photoelectron spectroscopy. Chemical Geology. 285. 62-69. 10.1016/j.chemgeo.2011.02.022.
6. Zaman M, Birkett G, Pratt C, Stuart B, Pratt S. Downstream processing of reverse osmosis brine: Characterisation of potential scaling compounds. Water Res. 2015 Sep 1;80:227-34. doi: 10.1016/j.watres.2015.05.004. Epub 2015 May 9. PMID: 26001825.
7. Sarabian E, Birkett G, Pratt S. Occurrence and behaviour of colloidal silica and silica-rich nanoparticles through stages of reverse osmosis treating coal seam gas associated water. Water Res. 2024 Feb 1;249:120866. doi: 10.1016/j.watres.2023.120866. Epub 2023 Nov 13. PMID: 38101050.
8. Turner C, Donose BC, Kezia, Birkett G, Pratt S. Silica fouling during groundwater RO treatment: The effect of colloids‘ radius of curvature on dissolution and polymerisation. Water Res. 2020 Jan 1;168:115135. doi: 10.1016/j.watres.2019.115135. Epub 2019 Sep 30. PMID: 31622911.
9. Sazali, R.A., Ramli, N., Sorbie, K.S., and Boak, L.S. (2021). Impacts of Temperature on the Silicate Scale Morphologies Studies and Severity. International Transaction Journal of Engineering, Management, & Applied Sciences & Technologies, 12(9), 12A9R, 1-16. http://TUENGR.COM/V12/12A9R.pdf DOI: 10.14456/ITJEMAST.2021.186
10. Esmaeil Sarabian, Greg Birkett, Steven Pratt, Occurrence and behaviour of colloidal silica and silica-rich nanoparticles through stages of reverse osmosis treating coal seam gas associated water, Water Research, Volume 249, 2024, 120866, ISSN 0043-1354.
11. Christopher J. Gabelich, Wei R. Chen, Tae I. Yun, Bradley M. Coffey, I.H. “Mel” Suffet, The role of dissolved aluminum in silica chemistry for membrane processes, Desalination, Volume 180, Issues 1–3, 2005, Pages 307-319, ISSN 0011-9164.
12. Tong T, Zhao S, Boo C, Hashmi SM, Elimelech M. Relating silica scaling in reverse osmosis to membrane surface properties. Environ Sci Technol. 2017;51:4396–4406.
13. Judy Hughes Bardsley, William Mathis Hann, Susan Tabb Robertson, Jan Edward Schulman, 1991, patent EP0459661B1.
14. Kalpana Chauhan, Priyanka Patiyal, Ghanshyam S. Chauhan, Praveen Sharma, Star-shaped polymers of bio-inspired algae core and poly(acrylamide) and poly(acrylic acid) as arms in dissolution of silica/silicate, Water Research, Volume 56, 2014, Pages 225-233, ISSN 0043-1354.

Authors:

Dr. Jean-Jacques Lagref and Mr. Robert Reisewitz

Toray Membrane Europe AG