Craig Bartels

Optimization of seawater RO systems design

The paper describes the configuration and operating parameters of current large seawater desalination systems. Major advances of RO seawater desalination technology that lead to a remarkable decrease of desalted water costs are evaluated. Process improvements that enable compliance with more stringent requirements of permeate water quality are discussed. Results of field tests conducted to demonstrate a new process approach are described. Some examples of process optimization resulting in lower power consumption and more efficient system operation are presented.

A surface science investigation of composite membranes

Abstract The chemical composition, physical structure and morphology of interfacially formed composite membranes were studied by a variety of surface science techniques, including attenuated total reflectance infrared spectrometry (ATR-IR), Rutherford backscattering spectroscopy (RBS), Xray photoelectron spectroscopy (XPS), scanning electron microscopy (SEM) and transmission electron microscopy (TEM). These techniques allow the characterization of ultrathin membrane barrier layers, as opposed to many techniques which are used only to study bulk samples. Also, the membrane performance was determined in reverse osmosis using a 5% aqueous isopropanol feed stream. p]The thickness of the barrier layer could be readily determined by microscopic techniques; however, this measurement may be in error due to penetration of the barrier layer into the porous support. By using a heavy metal dye which preferentially stained the barrier layer, TEM studies clearly showed the extent of this interpenetration. The chemical species present in the barrier layer were best characterized by ATR-IR and RBS techniques, since these are not as sensitive as XPS to surface contamination. RBS was particularly effective because it also provides a nondestructive determination of the element gradients within the top few microns of the sample. p]This type of study with complementary surface science techniques is especially needed for the study of composite materials which are formed in situ , because the chemical structure can be very complex and different from that of bulk reactions. Extensive characterization by a variety of methods can provide greater certainty for drawing conclusions about structure-performance relationships.

Structure-performance relationships of composite membranes: Porous support densification

Abstract Changes in the structure and chemical composition of a composite membrane surface were correlated with changes in the procedure used to prepare the membrane and with resultant changes in performance. Membranes were formed by the interfacial polymerization of a water-soluble multifunctional amine with an organic-soluble crosslinking agent on top of a porous polysulfone support. Scanning and transmission electron microscopy results indicate that with some reactions a dense layer (0.30 μm thick or less) was formed on top of a nodular substructure, and that the dense layer penetrates at least 0.1 μm into the porous support. In some other reactions little or no dense layer was formed on top of the porous support; for these membranes the nodules comprising the surface of the porous support compacted during the heat-curing process. fp]The membranes were used to remove water from a dilute aqueous solution of isopropyl alcohol by reverse osmosis. Performance results indicate that the lack of a generated dense layer decreases the permeate flux through the membrane. It is believed that the dense layer may act as a spacer to prevent densification of the support during heat curing; alternatively, it may favor water retention, which would prevent surface pore densification during the heat curing step.

Consideration of energy savings in SWRO

Abstract Seawater reverse osmosis (SWRO) processes have become the dominant desalination technology in the industry due to the low energy costs of this process. Thanks in large part to improvements in membrane and energy recovery devices, the SWRO process has become the accepted technology for desalination. In one report, current energy costs of SWRO processes have reduced by nearly 90% compared to SWRO in the 1970s and by 75% compared to SWRO in the 1980s. The best current SWRO processes require 2–2.5 kWh/m3 of electrical energy. The question is how much further reduction can be made with a conventional SWRO process. Recent theoretical analysis has shown that there is a diminishing potential for further energy savings in any desalination process. One such report has stated that the ideal SWRO energy consumption is 1.06 kWh/m3 for a 50% recovery plant treating 35,000 mg/l total dissolved solids, while the more realistic practical minimum energy consumption is thought to be 1.56 kWh/m3. With current plants...

A nanofiltration membrane for the removal of color from surface water to meet Norwegian standards

The color removal from a lake on the border between Engerdal municipality in Norway and Älvdalen municipality in Sweden by a Norwegian water treatment plant using a new nanofiltration (NF) membrane is demonstrated in this study. This water source has low turbidity but a high concentration of natural organic matter (NOM), which gives the water undesired color, smell and taste. In order to fulfill the Norwegian Drinking Water Regulations, this plant installed new NF membranes, the sulfonated polyethersulfone HYDRACoRe 50. The typical Norwegian water treatment design was used for the water treatment, which consisted of a screen filter of 50 μm, the NF membrane rig, a UV unit and an alkaline filter treatment. The initial flux, 14 L/m2∙h, was recovered to around 90% after all the main cleanings were applied during 4 years of service. This effort helped to maintain a good performance of the plant. Each year, the normalized flux declined 7%, suggesting the formation of the common fouling layer over the membrane surface. This performance loss is normal after 4 years of continuous operation with NF membranes. In fact, there have been no membrane replacements during the time under study. The color and infectious microorganisms removal with this selected process was higher than the 90% that shows its adequacy to treat the typical high colored Norwegian water sources. This paper presents the successful application of the NF membranes in the typical Norwegian water treatment design.

Operational experience from hybrid RO system at SAFI wastewater re-use RO plant

AbstractOperation of wastewater treatment plants can be very challenging due to the high fouling tendency of wastewater. Much knowledge has been gained over the past few years to guide design engineers on the optimum design for wastewater reverse osmosis (RO) plant. The implementation of this knowledge helps to improve the operation and performance of many wastewater plants. One such example is the SAFI (BESIX Group Company) plant in Ajman, UAE. Here, an microfiltration (MF) and RO plant was installed in 2010 to treat sewage effluent in order to supply high quality water for industrial and domestic reuse. The 6,800 m3/d plant treats secondary effluent with MF and RO. There are two RO trains designed to operate at 18.6 l/m2 h and 75% recovery, for treating municipal treated effluent with 2,000–4,000 mg/l of total dissolved solids (TDS). When operation started in 2010, the rate of fouling on RO membranes was extremely high, with up to 66% flow decline in 3 months of operation. This was eventually attributed...

Turbidity and pH dissipations in a dosage of high concentration NaOH solution to seawater

Abstract It is sometimes necessary to dose into a desalination feed stream a caustic solution to achieve a controlled pH increase, as, for example, in the application of boron suppression measures. To restrict the size of the caustic feed system, the NaOH is dosed from a high concentration solution. The solution is injected at the center of the raw water feed pipe and its concentration dissipates as it mixes with the main flowing solution. The problem is that desalination feed waters usually contain ions of a sparingly soluble salt such as CaCO3 and Mg(OH)2, which tend to precipitate at high pH levels. The very high local concentrations of the dosed NaOH at the injection region can lead to very high local supersaturations and induce the nucleation and precipitation of scale particles. The main goal of the present work was to delineate flow and concentration conditions at which dosage of a concentrated NaOH solution to seawater concentrates will not induce significant precipitation of CaCO3 and Mg(OH)2.

Improving Performance and Reducing Costs in Reclaiming Secondary Municipal Waste

AbstractThe Orange County Water District of Southern California conducted side-by-side testing of the latest reverse osmosis (RO) membrane technology and membranes currently operating in the districts Groundwater Replenishment System. The new RO elements were installed in a satellite vessel attached to the first stage of an existing three-stage train. New membrane performance was monitored for seven months and compared with the performance of the existing membranes, which had been operating for two years. Construction of the new elements included a thicker (34-mil) spacer, leading to less differential pressure loss and less fouling potential. The new elements had the same membrane area as the existing elements and, therefore, could be operated at the same flux and flow. This article compares operating data to demonstrate the new membranes improved performance, RO membrane chemistry, and element design. The data demonstrate the new membranes potential for reducing operating costs in a full-scale system.

Pilot filtration study to reduce fouling on Marbella seawater RO plant

Abstract The seawater RO plant at Marbella had operated intermittently, and at low flowrates, since its construction 10 years ago. This was because the plant had been designed to operate at times of water shortage, to provide water when the conventional supplies were not available. During the early part of the 21st century, the mode of the plant changed, and it was expected to operate at design capacity for most of the time. This brought two facts to light — that during high abstraction rates the intake water quality deteriorated significantly, and that the filtration system that had been installed was not capable of adequately filtering the feed water to make it suitable for feeding to a reverse osmosis plant. The plant was originally built using DuPont B10 permeators. Due to the unavailabilty of these permeators, the plant is currently being converted, stream by stream, to spiral wound membranes supplied by Hydranautics. At design flowrates, the DuPont system fouled rapidly, and required cleaning every two weeks. The spiral trains fouled more slowly, but still required cleaning at a higher frequency than would be considered normal for this type of plant. A pilot filtration plant was installed on site to attempt to find the following; • An optimum coagulant for the water. • An approximate dose rate for that coagulant. • The effect of different media on the quality of filtered water. • The length of run between backwashes using different media. • The quality of water that can be achieved using this coagulant and media. By installing pressure tapping points along the length of the filter, the area of differential pressure could be measured. This was used to ensure depth filtration was taking place, and the foulants were being removed through the length of the bed rather than surface filtration. The trials lasted a total of three months and achieved all of the targets set. The SDI typically achieved by the main plant was approximately 5. The pilot filter showed that the SDI could be reduced to below 2 by modifying the filters and applying a coagulant. Filter runs achieved by the pilot filters were in excess of 48 h. Following the trials, the plant commenced replacing the media in the filtration system, and is expected to install a coagulant dosing system once this was complete. This paper describes the pilot plant built, the selection of the media, and the coagulants used, and presents the operating data produced from the trials.