Adsorption of Methyl Tertiary-Butyl Ether on High-Silica Zeolites: Effects of Adsorbent Characteristics and Natural Organic Matter on Adsorption Isotherms

Show full item record

Title: Adsorption of Methyl Tertiary-Butyl Ether on High-Silica Zeolites: Effects of Adsorbent Characteristics and Natural Organic Matter on Adsorption Isotherms
Author: Rossner Campos, Alfred Armin
Advisors: Detlef R.U. Knappe, Committee Chair
Joel J. Ducoste, Committee Member
Morton A. Barlaz, Committee Member
Abstract: Methyl tertiary-butyl ether (MTBE) is frequently detected in surface and ground water. Because of its hydrophilicity, MTBE is difficult to remove from aqueous solution by activated carbon adsorption processes. In drinking water treatment applications, natural organic matter (NOM) also adsorbs on activated carbons, which further decreases the MTBE adsorption capacity. Unlike activated carbons, high-silica zeolites are adsorbents with well-defined pore sizes. From a drinking water treatment perspective, it may be possible to select high-silica zeolites with pore sizes that are suitable for the adsorption of smaller organic contaminants while preventing the adsorption of competing NOM components of larger molecular size. Therefore, the objectives of this research were to evaluate the effects of zeolite pore structure and hydrophobicity on the adsorption of MTBE in the presence of NOM. MTBE adsorption isotherm data were collected for a matrix of high-silica zeolites with different pore sizes (ZSM-5/silicalite, Mordenite, Beta, Y), exchangeable cations (H+, Na+, NH4+), and hydrophobicities (SiO2/Al2O3 ratios). MTBE adsorption capacities of high-silica zeolites were compared to those of three GACs (one coconut-shell-based, two coal-based) and a carbonaceous resin (Ambersorb 563). Single-solute isotherm tests were conducted in ultrapure water buffered at pH 7.2. Additional isotherm studies were conducted to determine the effects of co-adsorbing and preloaded NOM on MTBE adsorption from Tar River water (Greenville, NC). Single-solute MTBE adsorption isotherm data showed that high-silica zeolites with smaller pores (ZSM-5/silicalite, Mordenite) were more effective adsorbents for MTBE than zeolites with somewhat larger pores (Beta, Y). Over a range of 90-700, the SiO2/Al2O3 ratio of the tested ZSM-5 zeolites had no effect on MTBE adsorption capacity. Similarly, the exchangeable cation (H+, Na+, NH4+) of high-silica ZSM-5 zeolites had little effect on MTBE adsorption at the tested conditions, and no zeolite-catalyzed MTBE hydrolysis was apparent for ZSM-5 zeolites in the H+ form. For high-silica zeolites, co-adsorbing and preloaded NOM lowered the single-solute MTBE adsorption capacity at a liquid-phase MTBE concentration of 10 μg/L (q10) by 0-23%. Similar decreases in MTBE adsorption capacity as a result of NOM adsorption were measured for the coconut-shell-based activated carbon CC-602; however, its MTBE adsorption capacity (q10) was only about 25% of that obtained for ZSM-5 zeolites. In the presence of preloaded NOM, the MTBE adsorption capacity of the carbonaceous resin (q10) decreased by about 47% relative to the single-solute value while that of one coal-based activated carbon decreased by almost 60%. Overall, the coal-based activated carbons exhibited the smallest MTBE adsorption capacities (q10), which were approximately one order of magnitude lower than those of the ZSM-5 zeolites. Using an equilibrium model, adsorbent usage rates (AURs) and costs associated with adsorbent usage were calculated to evaluate the feasibility of zeolite-based adsorption systems. The latter analysis showed that the lowest treatment cost ($0.7/1000 gal) was associated with the usage of HiSiv3000 zeolite (cost: ~$7/lb) and CC-602 GAC (cost: ~1.50/lb). Despite the similar cost for these two adsorbents, the zeolite-based adsorption system may be more advantageous because (1) the AUR calculated for the HiSiv3000 zeolite was less than 25% of that calculated for the CC-602 GAC and (2) the calculated bed life for a packed bed adsorber containing HiSiv3000 zeolite was more than 6 times longer than that for an equally sized packed bed adsorber containing CC-602 GAC, a result that was affected by the lower packed bed density of the latter adsorbent. Thus, adsorbent replacement/ regeneration would have to occur on a less frequent basis when zeolites are used. Finally, it may be possible to regenerate spent high-silica zeolite with steam or microwave methods rather than with more energy-intensive thermal methods. This opportunity could further lower the cost of zeolite-based adsorption systems for MTBE removal from water.
Date: 2004-08-06
Degree: MS
Discipline: Civil Engineering

Files in this item

Files Size Format View
etd.pdf 517.4Kb PDF View/Open

This item appears in the following Collection(s)

Show full item record