Calcium Hydroxide as Precipitative Antiscalant for Nanofiltration

Abstract

Nano Filtration (NF) is an advanced treatment process that is able to remove molecules and ions from water using special, synthetic, membrane modules. Under high pressure the so– called feed water enters the membrane module. Part of the water, the permeate, passes through the membrane wall while the rejected ions and molecules are flushed out with the rest of the water, the concentrate. However, the ratio between product; the permeate, and waste; the concentrate, that can be attained with NF treatment is limited by the concentration of sparingly soluble salts in the feed water. When too much permeate is produced the concentrate stream becomes supersaturated, causing the salts to start precipitating on the membrane surface, a process called scaling. The scaling causes an increase in membrane resistance, necessitating a higher pressure, and thus more energy, to treat the same amount of water. There are several strategies to control scaling in membrane filtration installations such as feedwater alteration and antiscalant dosing, all of which aim to keep the salts dissolved for as long as possible. In this thesis a different approach to prevent scaling is proposed, aiming to promote precipitation instead, albeit in a controlled manner: precipitative antiscalants. Instead of precipitating on the membrane wall, the salts precipitate on special particles, which are transported out of the system with the waste stream. The dosing of calcium hydroxide, Ca(OH)2, was investigated as precipitative antiscalant for CaCO3 scaling. Ca(OH)2 particles exhibit a dissolve–precipitate effect, where CaCO3 precipitates on the dissolving particles, slowing down further dissolution. This effect is unwanted during application in other processes, but may prove of use as anti scaling mechanism. To better understand the mechanisms and kinetics involved with Ca(OH)2 dissolution in carbonate containing solutions a soft–sensor, capable of converting measured pH and EC to total calcium and carbonate, was developed and validated. Using this sensor it was found that, for high dosages of Ca(OH)2 , the formed layer of CaCO3 was unstable. After a certain length of time the covering layer breaks open, allowing the dissolution reaction to continue at its original rate. To investigate whether Ca(OH)2 dosing could function as antiscalant a pilot–plant installation, capable of simulating scaling on a flat–sheet polymeric NF membrane, was constructed. Dosing of Ca(OH)2 particles in combination with using a feed spacer proved problematic, as the particles got lodged between the spacers and could not be adequately removed. Without a feed spacer installed Ca(OH)2 dosing as antiscalant was a limited success. With little surface scaling taking place the runtime of the experiment could be extended from a mere 4.5 hours to over 24 hours. Excessive particulate fouling inside the pilot–plant, however, forced the experiment to be halted prematurely. In the end, although theoretically possible, the problems associated with feed spacers, the necessity of an intermittent cleaning cycle and substantial particulate fouling inside the system make the use of precipitative antiscalants for use with conventional spiral–wound polymeric membranes an unattractive option compared to more traditional anti-scaling measurements