3-Step approach towards evaluation and elimination of acid use in pre-treatment for a brackish water reverse osmosis process Jayapregasham Tharamapalan, Christopher C. Boyd, Steven J. Duranceau * University of Central Florida, Department of Civil, Environmental and Construction Engineering, 4000 Central Florida Blvd., POB 162450, Engineering Building 2, Suite 211, Orlando, FL 32816-2450, USA article info Article history: Received 20 December 2012 Received in revised form 5 March 2013 Accepted 19 March 2013 Available online 3 May 2013 Keywords: Acid elimination Canary unit Scale inhibitor Reverse osmosis abstract To control carbonate scale formation on reverse osmosis (RO) membrane surfaces, it is common practice to add acid as a pre-treatment chemical. However, advancements in the formulation of proprietary scale inhibitors have resulted in a re-evaluation of the need to suppress the pH of RO feedwater. In this study, a 3-step approach was taken to eliminate sulfuric acid pre-treatment at a 4.5 MGD (17,000 m 3 /day) brackish water RO plant operating for over 7 years without previous membrane replacement. The 3-Step approach adopted in this study to evaluate and eliminate use of acid in pre-treatment process involved first pilot testing the plan to reduce the dependence on acid. Secondly, implementing the plan on the full-scale system with conservative pH increments and thirdly continuously screening for scale formation potential using a “canary” monitoring device. This 3-step approach resulted in the successful elimination of sulfuric acid pre-treatment at the brackish water RO plant, with an estimated $105,000 minimum annual cost savings. Ó 2013 Elsevier Ltd. All rights reserved. 1. Introduction Acid pre-treatment of reverse osmosis (RO) feed water reduces the pH to control calcium carbonate scaling by increasing the sol- ubility of calcium carbonate, while also suppressing organic and biological fouling of the membranes (Ghafour, 2003). Hydrochloric [HCl] or sulfuric [H 2 SO 4 ] acids are typically used for acid pre- treatment to reduce the feed water pH to between 5 and 7 pH units. Sulfuric acid is sometimes preferred over hydrochloric acid because of the expected higher membrane rejection of the divalent sulfate ion compared to the monovalent chloride ion. A drawback of sulfuric acid addition, however, is that the scaling potential for sulfate-based salts is also increased with increasing sulfate ion concentration. Scaling occurs as sparingly soluble salts are concentrated in the reject (concentrate) stream of membrane processes beyond their solubility limit, which is a function of temperature, pressure and pH (Singh, 2006). Three major perfor- mance issues are related to membrane scaling: increasing oper- ating pressure (or alternatively a reduced flux at constant pressure), increasing pressure drop across membrane elements, and decreasing salt rejection rates (Kucera, 2010; Pontie et al., 2005). In order to improve RO plant performance, chemicals are added prior to the feed water entering the RO modules to inhibit scale formation. There are generally three types of scale inhibitors: so- dium hexametaphosphate, organophosphates and polyacrylates (Prihasto et al., 2009). These scale inhibitors limit the precipitation of sparingly soluble salts by increasing the ion clustering concen- tration threshold and disrupt nuclei ordering and crystal structure formation. Some scale inhibitors repel other ions in solution by adsorbing onto crystal surfaces, or they fully chelate with dissolved ions (Greenlee et al., 2009). However, scale inhibitors do not completely eliminate precipitation because precipitation will eventually occur at sufficient salt concentration and given sufficient time. The increasing availability of proprietary scale inhibitors that can inhibit both alkaline and sulfate-based scales has resulted in a re-evaluation of the need to suppress the pH of feedwater to RO membranes (Ning and Netwig, 2002). When RO feed water is adjusted to approximately pH 6, car- bonate is present as either bicarbonate or carbonic acid (i.e. soluble form of carbon dioxide (CO 2 )). These species of carbonate do not form scales with calcium. However, the soluble form of CO 2 will freely pass through the membrane into the RO permeate, thereby requiring downstream degasification for corrosion control. If the feedwater also contains sulfide (S 2 , and HS  ) ions, acidification will generate unionized hydrogen sulfide (H 2 S) gas (Gare, 2002). Since hydrogen sulfide is also a corrosive gas that passes through RO membranes, post-treatment is normally required. Untreated * Corresponding author. Tel.: þ1 407 8231440; fax: þ1 407 8233315. E-mail addresses: tjaya@knights.ucf.edu (J. Tharamapalan), chrisboyd@ knights.ucf.edu (C.C. Boyd), steven.duranceau@ucf.edu, sduranceau@aol.com (S.J. Duranceau). Contents lists available at SciVerse ScienceDirect Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman 0301-4797/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jenvman.2013.03.038 Journal of Environmental Management 124 (2013) 115e120