Environ. Sci. Technol. 2003, 37, 3090-3094
Identifying the Effect of Polar Constituents in Coal-Derived NAPLs on Interfacial Tension JIANZHONG ZHENG† AND SUSAN E. POWERS* Department of Civil and Environmental Engineering, Clarkson University, 8 Clarkson Avenue, Potsdam, New York 13699
Interfacial tension, which is a critical variable affecting multiphase flow of nonaqueous phase liquids (NAPLs) in the subsurface, varies greatly with the composition of complex NAPLs recovered from field sites. Much of this variability stems from the presence of acid and base molecules in the NAPL mixture. The interfacial tension and acid and base concentration in six coal-derived NAPLs were measured. Creosotes generally have lower interfacial tensions due to their higher concentrations of organic acid and base macromolecules as compared with coal tar samples. Interfacial tension is a function of pH, with lower values measured at pH greater than approximately 9. At a neutral pH, the interfacial tensions are inversely proportional to the total acid concentration. Asphaltenes in these coalderived NAPLs account for most of the acid and base constituents. It is found in this study that acid and base numbers are valuable indicators of interfacial tension and, therefore, the capillary phenomena associated with multiphase flow behavior of NAPLs in the subsurface.
Introduction There is growing evidence that the interfacial properties of dense nonaqueous phase liquids (DNAPLs) retrieved from subsurface systems are significantly different than their pure laboratory-grade counterparts that are often used in many multiphase flow experiments. These differences in wettability and interfacial tension could substantially impact our ability to accurately predict the behavior of these mixed DNAPLs in the subsurface. For example, reducing the interfacial tension would allow a DNAPL to enter smaller pores that may be less readily accessible during remediation efforts. Research on complex nonaqueous phase liquid (NAPL) mixtures indicates that the wettability can deviate significantly from water wetting conditions (1-4). Zheng and colleagues (5, 6) illustrated that these changes are correlated to electrostatic attraction associated with the presence of protonated organic base macromolecules in the NAPLs, especially for coal tars and creosotes. Values of interfacial tension (IFT) for NAPL/water interfaces are reported in DNAPL reference books to range between approximately 20 and 50 mN/m (7, 8). For complex mixtures, however, much lower values have been observed in field situations. For example, IFT values pKa would be more surface active than their neutral conjugate base and, therefore, would lead to a decrease in the IFT at high pH values. A recent study by Kanicky et al. (28) showed experimentally that the length of the hydrophobic chain has great impact on pKa for longchain carboxylic acids. For example, pKa increases from 4.83 for hexanoic acid molecules to approximately 7.5 for dodecanoic acid and about 8.7 for C16 acid. Stearic acid, with a C18 chain, is expected to have similar pKa around 9. Thus, the significant drop in IFT for octanoic acid at pH ∼5.5 (26) and for n-octadecanoic acid at pH ∼9 can be interpreted as the increase of the surfactant carboxylate ions aligning at the interface. As more and more carboxylic acid molecules are deprotonated at increasing pH values, negatively charged carboxylate ions align on the interface substantially reducing IFT. The pH dependence of the IFT of the organic base, dodecylamine (DDA, pKa ) 9) dissolved in isooctane was also tested. The experimental data are also plotted in Figure 3. As expected, starting from the high pH region, IFT decreases as pH decreases with a sharp decrease around pH 8-9. The IFT then levels off when pH is further decreased. The critical pH for IFT change agrees well with pKa of DDA. The protonation of DDA at lower pH values generates sufficient positively charged functional groups that accumulate at the oil-water interface. Because the conjugated acid is more surface-active than its neutral form, the IFT is decreased. IFT of NAPLs. Interfacial tension was measured as a function of aqueous pH for two NAPL samples to identify the impact of concentrations of acidic and basic components on this property. As shown in Figure 4, for creosote no. 2, low IFTs were observed at both low and high pH regions with an IFT plateau formed between pH 3.8 and pH 9.8. IFT decreases sharply after pH 10. A similar trend was observed for coal tar no. 2 with a much less rapid drop in IFT in the alkaline regions, no drop in IFT at low pHs, and a much higher plateau than that of the creosote sample. Similar decreases in IFT at high pH values were also observed in previous studies (3), although no substantial explanation was provided for these trends. The observed trends of IFT as a function of pH can be explained based on the acid and base reactions on the NAPLwater interface. For creosote no. 2, the sharp decrease in IFT after pH 10 can be attributed to the relatively narrow distribution of the pKa values and the higher interfacial activity of the acid constituents. As demonstrated by the acid titration experiment (Figure 1), the acid mixtures in this creosote sample have a similar strength as stearic acid and onitrophenol (pKa ∼9-10). At pH 10, substantial dissociation
FIGURE 5. Interfacial tension of ASPH-CR2 (dissolved in toluene) as a function of concentration and pH. of the acids in the NAPL and alignment of the negatively charged acid molecules on the interface cause the significant decrease in IFT. For coal tar no. 2, the shallower dip in the alkaline region could be attributed to a broader distribution of pKa values or the lower concentration of acids relative to creosote no. 2 (Figure 2). On the basis of the research presented by Hoeiland et al. (15), the differences in the slopes among the three samples can also be interpreted as differences in the molecular structure and, therefore, the interfacial activity of the acid constituents. Their analysis of three crude oil samples suggest that the steepest slope of their IFT ) f(pH) curves was associated with an oil that had a higher fraction of high molecular weight ring structures with a high degree of carboxylic acids. The decrease in IFT in the acidic region suggests that organic bases in the creosote no. 2 sample are protonated in this region. This is supported by the study of Acevedo et al. (14), who isolated a basic fraction from a crude oil that was found to decrease IFT in the acidic region. The formation of the IFT plateau in the neutral pH range is probably due to the collective effects of both acidic and basic molecules in this NAPL. Within the pH region covering the plateau, the concentrations of both protons and hydroxyl ions in the aqueous phase are insufficient to react with these weak acids and bases to generate enough charged molecules (conjugated bases and acids having higher surface activity) to substantially change the NAPL-water IFT. As compared with creosote no. 2, the high plateau of the coal tar no. 2 sample is attributed to the lower concentrations of both acidic and basic components in this NAPL (Figure 2). The IFT of coal-derived NAPL-water interfaces at pH 7 were measured for all samples to further assess the correlation between IFT and the concentration of acids and bases. As shown in Figure 2, larger acid and base numbers correspond to lower IFT at a neutral pH. The IFT of the coal tars, which have relatively low concentrations of acids and bases, is more than twice the value of creosote no. 2, which has the highest concentrations. It is expected that NAPLs with higher acid and base numbers have more polar molecules available to align on the interface resulting in the observed lower IFT. IFT of Asphaltenes. The IFT of the isolated asphaltenes dissolved in toluene was measured as a function of the solution concentration to assess the nature of interfacial activity of asphaltene molecules. Figure 5 presents the IFT of asphaltene (ASPH-CR2) as a function of asphaltene concentration at two extreme aqueous pH values (pH 3.1 and 9.8). Similar trends were observed for the other two asphaltene samples. At fixed asphaltene concentrations, the IFT is lower at pH 9.8 than at pH 3.1. For both cases, IFT decreases log linearly at lower concentration until it levels VOL. 37, NO. 14, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3093
off after a critical concentration. Lord et al. (26) observed similar trends, including the lower IFT for higher pH values, for their model system of octanoic acid in o-xylene. The observed critical concentration representing the minimum IFT is different between the two pH values (∼10 g/L at pH 9.8 and 50 g/L at pH 3.1). It was further observed that IFT increases a little for both solutions at concentrations higher than the critical concentration. The observed linear relationship between IFT and log asphaltene concentration is similar to the classic surface tension versus log concentration trend for typical surfactants in aqueous solution where a distinct critical micelle concentration is defined (29). The experimental observations strongly support that asphaltenes are naturally occurring pH-dependent surfactants. The difference in the critical concentration for the two samples at acidic and basic pH values is believed to be due to the difference in surface activity of polar molecules at these two pH values. In basic solution, the interfacial activity for asphaltenes is higher than in acidic solution due to the higher concentration of basic relative to acidic components (Figure 2). Implications and Significance. The IFTs of coal tar and creosote NAPL mixtures and, therefore, the capillary phenomena associated with multiphase flow behavior of NAPLs in the subsurface are well-correlated with the acid and base numbers of these NAPLs. On the basis of the limited number of samples tested here, we note that the acid and base numbers of the coal tar and creosote samples both contribute to the magnitude of the NAPL-water IFT over the neutral pH region. With their higher concentrations of acid and base, the creosotes have lower IFTs and are more likely to create a nonwater wetting system (5) than the coal tars. Coupled with the lower viscosity of creosote relative to coal tar, the net implication will be a higher mobility and less capillary entrapment of creosote as discrete ganglia in the subsurface. At the same capillary pressure, creosote NAPLs will migrate into smaller pore spaces than coal tar, which could reduce the efficiency of some remediation techniques.
Acknowledgments Funding for this research from the National Science Foundation (BES-9981494) is gratefully acknowledged. Jan DeWaters and Daniel Hugaboom helped with the interfacial tension measurements reported here.
Literature Cited (1) Powers, S. E.; Anckner, W. H.; Seacord, T. F. J. Environ. Eng. 1996, 122, 889-896. (2) Jackson, R. E.; Dawarakanath, V. Ground Water Monit. Rem. 1999, 19, 102-110. (3) Barranco, F. T., Jr.; Dawson, H. E. Environ. Sci. Technol. 1999, 33, 1598-1603. (4) Harrold, G.; Gooddy, D. C.; Learner, D. N.; Leharne, S. A. Environ. Sci. Technol. 2001, 35, 1504-1510. (5) Zheng, J. Z.; Powers, S. E. J. Contam. Hydrol. 1999, 39, 161-181.
3094
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 14, 2003
(6) Zheng, J. Z.; Behrens, S. H.; Borkovec, M.; Powers, S. E. Environ. Sci. Technol. 2001, 35, 2207-2213. (7) Cohen, R. M.; Mercer, J. W.; Matthews, J. DNAPL Site Evaluation; C. K. Smoley: FL, 1993. (8) Pankow, J. F.; Cherry, J. A. Dense Chlorinated Solvents and Other DNAPLs in Groundwater; Waterloo Press: Portland, OR, 1996. (9) Dwarakanath, V.; Jackson, R. E.; Pope, G. A. Environ. Sci. Technol. 2002, 36, 227-231. (10) Soe, H. S.; McCray, J. E. Environ. Sci. Technol. 2002, 36, 12921298. (11) Fu, J.; Buqiang, L.; Zihao, W. Chem. Eng. Sci. 1986, 41, 26732679. (12) Speight, J. G. The Chemistry and Technology of Petroleum, 2nd ed.; Marcel Dekker: New York, 1991. (13) Sheu, E. Y.; Storm, D. A. In AsphaltenessFundamentals and Applications; Sheu, E. Y., Mullins, O. C., Eds.; Plenum Press: New York, 1995. (14) Acevedo, S.; Ranaudo, M. A.; Gutierrez, L. B.; Escobar, G. Surfactant Sci. Ser. 1996, No. 64, 221-231. (15) Hoeiland, S.; Barth, T.; Blokhus, A. M.; Skauge, A. J. Pet. Sci. Eng. 2001, 30, 91-103. (16) Stumm, W.; Morgan, J. J. Aquatic Chemistry: An Introduction Emphasizing Chemical Equilibria in Natural Waters, 2nd ed.; John Wiley& Sons: New York, 1981. (17) ASTM D200-93, Standard Test Method for Characteristic Groups in Rubber Extender and Processing Oils and Other PetroleumDerived Oils by the Clay-Gel Absorption Chromatographic Method, ASTM International: West Conshohocken, PA, 1993. (18) Bruss, D. B.; Wyld, G. E. A. Anal. Chem. 1957, 29, 232-235. (19) ASTM D664-01, Standard Test Method for Acid Number of Petroleum Products by Potentiometric Titration. ASTM International: West Conshohocken, PA (http://www.astm.org). (20) Zheng, J. Z. Mechanisms That Affect the Wettability of Quartz Wettability of Quartz Exposed to Coal-Derived Non-Aqueous Phase Liquids. Ph.D. Dissertation, Clarkson University, Potsdam, NY, 2001. (21) ASTM D4739-95, Standard Test Method for Base Number Determination by Potentiometric Titration. ASTM International: West Conshohocken, PA (http://www.astm.org). (22) ASTM D971-91, Standard Test Method for Interfacial Tension of Oil Against Water by the Ring Method. ASTM International: West Conshohocken, PA, 1991 (http://www.astm.org). (23) Buckley, J. S. Presented at the 1994 International Symposium on Reservoir Wettability and Its Effects on Oil Recovery, Laramie, WY, September 21-23, 1994. (24) Buckley, J. S.; Liu, Y.; Monsterleet, S. Presented at the 1997 SPE International Symposium on Oil Field Chemistry, Houston, TX, February 18-21, 1997. (25) Skauge, A.; Standal, S. H.; Boe, S. O.; Skauge, T.; Blokhus, A. M. Presented at the 1999 SPE Annual Technical Conference and Exhibition, Houston, TX, 1999, SPE Paper 56673. (26) Lord, D. L.; Hayes, K. F.; Demond, A. H.; Salehzadeh, A. Environ. Sci. Technol. 1997. 31, 2045-2051. (27) Cratin, P. D. J. Dispersion Sci. Technol. 1993. 14, 559-602. (28) Kanicky, J. R.; Poniatowski, A. F.; Mehta, N. R.; Shah, D. O. Langmuir 2000, 16, 172-177. (29) Hiemenz, P. C.; Rajagopalan, R. Principles of Colloid and Surface Chemistry, 2nd ed.; Marcel Dekker: New York, 1997.
Received for review September 3, 2002. Revised manuscript received May 2, 2003. Accepted May 3, 2003. ES026118S
