Ceramic UF and Cooling Tower Blowdown Recovery
A power plant in the southwest region of the USA has been increasingly challenged with water scarcity issues reflected by higher water cost and increasing use restrictions. Cooling towers typically represent a large percentage of the water footprint in these applications, but the complex water chemistry of cooling tower blowdown (CTBD) can present a number of design challenges that often result in higher than expected CAPEX investments and excessive on-going operating expenses.
Nanostone Water was invited to pilot it’s innovative CM-151* ceramic UF monolith to explore the technology’s effectiveness and operating efficiency. Ceramic membranes have long been used in toughto- treat applications, but their historical high cost have limited their use in traditional applications. Nanostone Water is the first membrane manufacturer to deliver a high surface area ceramic monolith with all the benefits of ceramics, at cost comparable to polymeric membranes.
Figure 1: CM-151* Membrane
Table 1: CTBD Pilot Objectives
|Raw Feed||< 5 TSS||80-120||7-8||> 4|
|UF Feed||~100 TSS||80-120||10-11||> > 4|
|Sludge Blowdown||> 5% TSS||N/A||10-11||N/A|
|Dry Sludge||> 40% TSS||N/A||N/A||N/A|
|UF Filtrate For Reuse||< 0.1 NTU||< 25||10-11||< 3|
ORIGINAL SYSTEM CHALLENGES
The power plant’s original treatment system consisted of electrocoagulation (EC) followed by a polymeric UF system feeding a final RO system. Operating experience showed that the EC system metal plates had shorter than expected life, the polymeric UF cleaning frequencies were higher than expected even at a low flux of < 10 GFD, and a shortened UF membrane life cycle of only 3 years.
In an effort to reduce operating expenses, the facility decided to feed the RO system without the EC in operation. Given typical CTBD silica levels of 90- 130 mg/l, the RO recovery needed to be significantly reduced, with more frequent clean-in-place (CIP) cycles and membrane replacements. Although a reduction in operating expense was realized, the maintenance level and overall operating cost remained unacceptable.
Other CTBD Chemistry Challenges
In addition to high silica levels, CTBD contains high levels of cooling tower treatment chemicals to control scaling and biological activity. These chemicals can foul polymeric membranes. In order to effectively remove silica, chemicals used for precipitation must be dosed to overcome the dispersant and antiscalant chemistry of the CTBD feed.
With the appropriate feed chemistry applied, the use of Nanostone’s CM-151* monolith appeared viable given its inherent robustness, with high suspended solids tolerance and chemical resistance promising major benefits.
NANOSTONE'S CERAMIC UF SOLUTION
The pilot success criteria establish by the customer included:
Nanostone’s application engineers successfully developed a chemical feed dosage strategy that included precipitation with magnesium chloride and elevated pH to reduce silica. A compact mix tank combined with high rate sludge thickening will be used which offers a significant cost and foot print reduction over conventional lime softening. The high suspended solids tolerance of the ceramic membrane is 10 times greater than polymeric membranes and eliminates the need for conventional clarification and pre-filtration.
Figure 2: CTBD/CM-151* Process Design Membrane
The pilot experience was better than expected with fluxes of 100-130 GFD (170-220 LMH) (> 10 times higher than polymeric UF membranes). Silica reduction results up to 90% were consistently well below the 25 mg/l silica effluent limit. The permeate turbidity performance was also achieved consistently producing effluent water < 0.1 NTU. A steady flux was maintained with regular chemical enhanced backwashes (CEBs) to control hardness scaling.
SILICA REDUCTION PERFORMANCE/strong>
Figure 3:CTBD Silica Removal w/ Ceramic UF
During the course of developing the application, the pilot membrane was subjected to extreme conditions, including feed water in excess of 600mg/L TSS. Even at these high solids levels, the system was stable with minimum recirculation flow rates of <20% of the permeate flow. The graph below shows net driving pressure (NDP) performance during a period of very high feed turbidity levels in the direct filtration mode. This system was allowed to operate up to 5 times the initial transmembrane pressure of 5 psi (0.3 bar) NDP before chemical cleaning. Operating at a max of 25 psi (1.7 bar) NDP in this application, the pressure was well below the CM-151* module rating of 100 psi (7 bar).
Figure 4: NDP Performance w/ 400 NTU feed
Given the successful outcome in meeting both performance and cost effectiveness goals, Nanostone Water is working with its OEM partners to design and deliver a full scale solution. A detailed technical paper describing this pilot experience is underway.