Novel Surfactant-Based Adsorbent Material for Groundwater

Therefore, considerable attention has been addressed toward development of alternative technologies, such as surfactant-enhanced aquifer remediation ...
0 downloads 0 Views 182KB Size
Download PDF
Environ. Sci. Technol. 2007, 41, 6836-6840

Novel Surfactant-Based Adsorbent Material for Groundwater Remediation F. VENDITTI, R. ANGELICO, A. CEGLIE, AND L. AMBROSONE* Consorzio per lo sviluppo dei Sistemi a Grande Interfase (C.S.G.I.) c/o Department of Food Technology DISTAAM, Universita` del Molise, via De Sanctis 86100 Campobasso, Italy

Many surfactants aggregate spontaneously in aqueous media to form small spherical structures called micelles. Among the numerous technical applications it is known that micelles have the ability to dissolve in their hydrophobic part significant amounts of water-insoluble organic compounds. In this study we investigated through UV-vis spectroscopy the micellar solubilization of 2,4,5-trichlorophenol (Tcp), an intermediate product of the microbial degradation of the broad-leaf herbicide 2,4,5-trichlorophenoxyacetic (2,4,5-T). Our results show that in the presence of the anionic surfactant sodium dodecylsulfate SDS the water solubility of Tcp increases six-fold whereas with cationic CTAB and nonionic Triton-X 100 the partition of chlorinated compound is not efficient. After the excess amount of the pollutant solubilized in SDSmicelles has been precipitated with CaCl2 the remaining fraction of Tcp has been successfully reduced within the toxicological limit for drinkable water through a cocurrent multistage operation. Finally, potential use in the decontamination of wastewater or soils of the new adsorbent material has been compared with the most commonly used activated carbon and silica gel.

1. Introduction Groundwater pollution is the loss of water quality due to waste products or other substances which changes the chemical and microbiological water properties (1,2). Some of these substances are soluble in water, but many are hydrophobic liquid compounds, the so-called nonaqueous phase liquids (or NAPLs). These are divided into light (LNAPLs) and dense (DNAPLs) according to whether they are less or more dense than water (3). Most NAPLs are made up of chlorinated compounds which are largely used in the industry as solvents. They are a source of long-term contamination since they slowly dissolve, forming toxic particulate in the groundwater (4-7)· Conventional pump-and-treat techniques are very expensive and inadequate in the remediation of NAPLscontaminated aquifers. Therefore, considerable attention has been addressed toward development of alternative technologies, such as surfactant-enhanced aquifer remediation (SEAR) (8). In the last few years permeable reactive barriers (PRBs) have been used as well (9-11). As an alternative, surfactantbased remediation technologies may offer a valid and effective * Corresponding author phone: +39.0874.404715; +39.0874.404652; e-mail: [email protected]. 6836

9

fax:

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 19, 2007

method for water decontamination from chlorinated hydrocarbon compounds. However, since undefined amounts of surfactant may be released in the aquifer, one must pay attention to their environmental impact as well (12, 13). Although it has been well established that surfactants can increase the solubility of chlorinated substances such as chlorophenols (14), to the best of our knowledge, experimental data about surfactant solubilization of 2,4,5-trichlorophenol (Tcp) have never been reported. SEAR requires careful analysis of the physical chemistry of surfactant systems in the laboratory before field implementation; after that a detailed site characterization is necessary to define DNAPL zone boundaries and elucidate the hydrostratigraphy of the zones to be flooded, both to optimize remedial design and to minimize the risk of uncontrolled DNAPL migration. In this paper our attention is focused on the methodological aspects to optimize some steps of groundwater or soil treatments; moreover, the proposed suggestions can be also utilized with more environmentally friendly surfactants. For the present study, solubilization of Tcp in anionic sodium dodecylsulfate (SDS), cationic cetyltrimethyl ammonium bromide (CTAB), and nonionic TritonX-100 (TX-100) surfactants has been experimentally investigated. Tcp is used as a model of chlorinated contaminants as a chemical intermediate for herbicides, insecticides, and preservative for adhesives, textiles, rubber, wood, and paints as major applications whose genotoxic effect has been firmly assessed (15). The affinity between surfactant and contaminant is exploited to obtain, through a suitable precipitant agent, an adsorbent surfactant-based material (16). The adsorptive efficiency of the novel material is experimentally tested in batch studies and its performance is compared with those of most known adsorbents, i.e., activated carbon and silica gel. Regeneration of the adsorbent materials and surfactant recovery has been investigated as well. The experimental results are interpreted in terms of potential applications addressed to improve pump-and-treat methods. The proposed strategies may be used for both batch and continuous processes. The final goal is to identify an optimal surfactant system to increase the environmental remediation of chlorinated contaminants.

2. Experimental Procedures 2.1. Materials. Commercial-grade Tcp (MW 197.45; CAS no 95-95-4) from Fluka Chemicals was used as a model of organic chlorine compound. Three surfactants were utilized in this study, namely, cationic cetyltrimethyl ammonium bromide (CTAB) from Sigma Aldrich (purity 99.0%), nonionic Triton X-100 (TX-100) from Carlo Erba (purity 99.0%), and anionic sodium dodecylsulfate (SDS) purified through crystallization from ethanol (purity 93.0%). Calcium chloride and sodium carbonate (from Carlo Erba) were dried in a stove for 24 h before using. Activated carbon and silica gel were obtained from Carlo Erba with the specifications listed as product for chromatography and pore diameters of 0.06 ( 0.2 mm. They were used without any further treatment. Nonwastewaterrelated experiments were performed at room temperature in aqueous solutions using water from an ISCO water distiller model WTD/5. 2.2. Methods. Surfactant micellization in the presence of Tcp was monitored through UV-vis spectroscopy by varying both the type of amphiphile and the detergent concentration. Tcp is sparingly soluble in water, with a saturation concentration equal to 1.15 kg‚m-3 at 25 °C. It is also a hydrophobic weak acid with pKa ) 6.94 (17), and the mixtures for experimental tests exhibited pH values in the range 3-4 10.1021/es070643f CCC: $37.00

 2007 American Chemical Society Published on Web 09/05/2007

depending on the Tcp concentration. The samples were prepared in 15 cm3 Falcon (Sterlin) by adding small amounts of surfactant (0.1-1 wt %/wt) to a saturated aqueous solution of Tcp. The resulting mixtures were stirred at room temperature until the system reached the thermodynamic equilibrium. Then, samples were centrifuged and the absorbance of the supernatant solution was measured with a Shimadzu 1601 UV spectrophotometer at a wavelength of λmax ) 290 nm, corresponding to the maximum absorption band for Tcp dissolved in water. All experiments were carried in triplicate, and the mean values with standard deviation are presented. The maximum standard deviation was 5%. Several tests have been also performed to check either unwanted contributions due to small absorbance from surfactant solutions or possible deviations in UV absorption properties of Tcp dispersed inside the hydrophobic inner core of micelles. However, no significant artifacts were observed in the investigated systems. For the SDS-Tcp system the Lambert-Beer relationship has been tested in the range of surfactant concentration 0-0.06 wt %/wt and Tcp concentrations 0.35-0.70 mol‚m-3, the molar apparent extinction coefficient being equal to 249 m2‚mol-1in our experimental conditions.

3. Results and Discussion 3.1. Enhanced Solubilization. In very dilute aqueous solution a surfactant acts as a normal solute, but above a specific concentration it self-assembles into supramolecular small colloidal particles called micelles. This specific concentration is defined as the critical micelle concentration (cmc) (18), where monomers and micelles exist in dynamic equilibrium. Surfactants are useful in a wide variety of applications due to the ability to dissolve substantial amounts of compounds that have very low solubility in water. This process of enhanced solubility is normally referred to as solubilization (19). From an applicative point of view a quantity of interest is the maximum uptake of an apolar molecular species into a micellar solution, which occurs when the partition equilibrium is reached, e.g., between the Tcp concentration in water bulk and that dispersed in micelles. In Figure 1 the effect of the partitioning is plotted on semilogarithmic coordinates for SDS, CTAB, and TX-100. As one can see, the solubilization abilities of these three systems differ markedly from one to the other. In both CTAB/water and TX-100/water systems the measured Tcp concentration is practically equal to its maximum water solubility, while in the SDS/water system there are very large deviations for SDS concentrations above the cmc. We believe that the process of solubilization is dominated by the hydrophobic interactions between the surfactant and Tcp. This is also supported by solubility data obtained with different chlorophenols (14). It is easy to deduce that whenever SDS micelles are in solution they have a marked effect on the solubilization of chlorinated compounds. This means that SDS may be used to speed up the subsurface remediation in the surfactantenhanced aquifer remediation technique. Anionic surfactants are particularly attractive for remediation of contaminated groundwater as they potentially have low toxicity and favorable biodegradability (20) and can be more environmentally friendly than more organic-solvent-based systems. In the class of anionic surfactants, SDS is reported in several case studies and projects published in official data bases such as, e.g., the Ground Water Remediation Technologies Analysis Center (GWRTAC) (21). Moreover this surfactant also possesses U.S. FDA Direct Food Additive Status (22). Yet, researchers from Eckenfelder, Inc. and Vanderbilt University tested a pilot-scale system for recycle and reuse of spent surfactant from organic-contaminated soil washing

FIGURE 1. Semilog plot of Tcp concentration, C(Tcp), vs surfactant concentration for the micellar solubilization of Tcp in (a) SDS, (b) CTAB, and (c) TX-100 systems. Vertical lines indicate cmc values. (21). In that study a mathematical model was also used to assess the relative cleanup times as a function of location, surfactant concentration, and soil particle size. On the other hand, adsorption of pollutants from aqueous solution plays an important role in wastewater treatment since it eliminates the necessity for huge sludge-handling processes. Well-designed adsorption processes have a high efficiency resulting in a high-quality effluent after treatment which can be recycled. Furthermore, if low-cost adsorbents or adsorbent regeneration is feasible then the adsorbent material cost can be kept low. For these reasons SDS-micelles containing solubilized Tcp were turned into solid form using a precipitation technique. 3.2. Precipitation Equilibrium. In this section we shall take a brief look at some results in order to obtain a quantitative picture of how SDS micelles behave at precipitation equilibrium. A series of liquid samples containing Tcp solubilized in SDS micelles were treated with an excess of CaCl2. The monitoring of solutions indicates that, independent of the initial conditions, the final amount of chlorinated compound is always equal to its solubility in water. An explanation for this is that calcium ions bind the surfactant monomers, thus precipitating in the form of the calcium sulfate derivative of SDS (hereafter it will be referred as CDS). Due to the very low solubility of CDS in water, to restore a new equilibrium upon addition of CaCl2 micelles have to break and release the solute. The experiments were carried out in excess of calcium ions so that complete release of the solubilized Tcp was always reached. As a consequence, the amount of contaminant exceeding the saturation value phase separates and its concentration in solution equals the saturation limit. Then one can conclude that treatment of contaminated water with CaCl2 allows separation of the surfactant and part of the contaminant as a solid phase while in the liquid phase the Tcp concentration is still 1.15 kg‚m-3 (i.e., its solubility in water). Laboratory tests reveal that if the liquid phase is treated either with fresh SDS and CaCl2 or with solid CDS the contaminant concentration is equally reduced indicating that depletion of Tcp is driven by an VOL. 41, NO. 19, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6837

FIGURE 2. Ratio between the amount of Tcp remaining in the bulk after the adsorption (C1) and initial concentration (C0) vs CDS dosage; expressed as grams of solid per kilogram of liquid solution (g/kg). adsorption process. For this reason, we decided to characterize this solid material by testing, on a laboratory scale, its adsorptive ability to remove chlorinated compounds. 3.3. Cocurrent Batch Operation by Simple Multistage Contact. To achieve accurate results and know how well CDS purifies the wastewater, adsorption measurements were carried out under equilibrium conditions. Several aliquots of Tcp solution at saturation (but in the absence of precipitate) were treated with different amounts of CDS and the solutions analyzed for remaining Tcp concentration. In preliminary experiments the minimum contact time for CDS samples to reach the equilibrium was estimated to be 1 h. The amount of adsorbed Tcp was calculated as the difference between the amount originally present in the liquid, C0, and the amount remaining in the bulk after adsorption, C1. In Figure 2 C1/C0 versus the CDS dosage is directly plotted, expressed as the solid-to-liquid ratio. As it can be seen, an increase of the solid-to-liquid ratio (S/L) produces a rapid decrease of C1/C0, reaching a constant value at S/L ) 2 g/kg. A limiting ratio suggests that the material is saturated with the pollutant and no further purification occurs. In the actual remediation equipment this could be a serious limitation because it is not economically profitable to allow high ratios S/L. However, the adsorption yield is moderately high at very low S/L. This fact may be advantageously used to design a separation process with several stages in series. The required size of the operating unit, and hence its cost and operability, can be calculated from knowledge of the phase equilibrium limitations and the rate of transfer obtainable. In the case under investigation, once we have identified the minimum CDS dosage to have the highest efficiency in each stage, we can readily design a multistage system. For economic reasons, stages must accomplish the contact separation as simply possible to avoid expensive equipment. In our laboratories, cocurrent batch operation by simple multistage contact was used (23). In this equipment the total quantity of CDS used is divided into several portions. The feed undergoing decontamination is treated with each portion of fresh CDS in a series of successive stages. Precipitation of CDS leaves in solution 1.15 g of Tcp per kg of liquid, which is 440 times higher than the limit for drinking water recommended by the US EPA (24). We now consider how to purify the remaining water to satisfy State laws and municipal ordinances. The toxicological limit imposes the final conditions of the process. 6838

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 19, 2007

FIGURE 3. Semilog plot of experimental results (black circles) of relative Tcp concentrations in a series of n successive stages with constant ratio S/L ) 2 g/kg as a function of the steps n. The solid line is the graphical representation of eq 2 with k ) 0.23. Horizontal line represents the toxicological limit of Tcp. The number of stages to obtain the desired final conditions is calculated with fixing the minimum S/L in each stage. This constrains the Tcp concentration, in each stage, to remain constant, i.e., k

Cn C1 C2 ) ) ‚‚‚ ) )k C0 C1 Cn-1

(1)

The numerical value k ) 0.23 is directly evaluated from the plot of Figure 2. After n steps the overall mass fraction of Tcp remaining in the water is given by

Cn

n

)

C0

Ci

∏C i)1

) kn

(2)

i-1

which allows computing the sufficient number of stages to obtain the desired yield. For instance, letting e ) 0.0026 g/kg be the toxicological limit (24), one finds that

n)

log e ) 4.3 log k

(3)

The accuracy of eq 2 was experimentally tested by measuring the ratio Cn/C0 in each stage. Considering n a continuous variable, Cn/C0 can be easily evaluated from eq 2 (straight line of Figure 3) as a function of n. In Figure 3 the curve is drawn together with the experimental results and the toxicological limit. The experimental results, which have been discussed above, deserve some comment. A saturated aqueous solution of Tcp mimes an extreme condition of pollution. Actually, contamination of chlorinated compounds in the water is very small, so that the sufficient number of steps to make water toxicologically pure is less than five. In other words, the experimental n ) 5 has to be interpreted as the number of steps to be used in the most damaging situation possible. 3.4. Comparison with Conventional Adsorbents. The adsorbent efficiency of CDS was first compared with that of the most commonly used adsorbents in the treatment of wastewater, namely, activated carbon and silica gel. The yield, Y, defined as the ratio of the adsorbed Tcp to that present originally in the liquid phase, was preliminarily calculated for a single batch treatment with CDS, activated carbon, and

FIGURE 4. Experimental ratio between adsorbed Tcp and its initial concentration in the liquid phase for CDS, activated carbon, and silica gel systems.

FIGURE 6. Efficiency of regeneration with SDS-micelles for CDS (b), activated carbon (9), and silica gel ([). Solid lines represent best linear fits to experimental data.

FIGURE 5. Effect of solid-to-liquid ratio on adsorption yield for CDS (b) and activated carbon (9). Vertical line is the optimal CDS dosage (curves are guides for eyes). silica gel. Experiments were carried out taking S/L ) 2 g/kg for all the adsorbent materials; the results are presented in Figure 4. The lower efficiency of silica gel compared to both CDS and activated carbon is clearly due to its hydrophobicity, and therefore, it has been not further investigated. In order to understand these unexpected differences between activated carbon and CDS, we studied the adsorption isotherms, and the results are shown in Figure 5. Unlike CDS, activated carbon does not saturate and completely removes the chlorinated pollutant. However, total removal occurs to a S/L ratio near to 20 g/kg, which is about 1 order of magnitude larger than that saturates CDS. This means, for instance, that not less than 20 g of activated carbon is necessary to detoxify 1 kg of water contaminated with 1.15 g of Tcp. To obtain the same result in a five-stage adsorption operation 10-12 g of CDS are utilized. Moreover, in the actual equipment for water purification use of a large amount of activated carbon, apart from the economical problem, is an excellent support for bacterial growth (25). Regeneration of the adsorbent material and surfactant recovery are important aspects for the overall economy of the process. As discussed above, a batch adsorption on solid CDS reduces the Tcp amount originally present in solution by about 83 wt %. Interestingly, by adding 0.5 wt % of powdered SDS the Tcp quantity free in solution becomes 93 wt %. This behavior can be rationalized by considering that SDS forms micelles which resolubilize the chlorinated contaminant regenerating the adsorbent surface. When the same kind of analysis is carried through for activated carbon and silica gel, different results are observed. The comparative

FIGURE 7. (Top) Cocurrent batch by simple multistage contact proposed to detoxify wastewater. (Bottom) Conventional pumpand-treat technique used for the remediation of contaminated aquifers. results shown in Figure 6 indicate that activated carbon and silica gel are insensitive to addition of SDS; on the contrary, the Tcp amount adsorbed on CDS drastically decreases. Not only can the SDS-micelles be used to solubilize the contaminant, but also they can be used to recover the adsorbent material. Finally, the reactivated CDS may be used either exactly as it is for a new separation process or to restore the surfactant. Recovery of surfactant can be obtained by treating solid CDS with a solution of Na2CO3 which quantitatively precipitates the calcium in the form of carbonate, releasing SDS as free surfactant. 3.5. Engineering Implications. The practical value of the obtained results has to be pointed out. A convenient method for aquifer remediation is to withdraw the polluted water from the aquifer and treat it on-site. The treated water may then be returned to the aquifer, discharged to surface water, or transferred to a public water treatment plant. However, the pump-and-treat technology to achieve the desired level of cleanup requires decades of costly operations. The large VOL. 41, NO. 19, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6839

costs of cleanup make it essential to investigate alternative technologies for remediation. The results of this paper have to be considered a starting point to stimulate discussion and research on limitation of the nowadays used method and their improvement. A practical suggestion coming from the results discussed in this paper is schematically illustrated in Figure 7. A SDS solution is used to mobilize and remove large amounts of Tcp. After it has been removed from the aquifer, the contaminated water is treated above ground with CaCl2. This operation (unit 0 in Figure 7) allows separation, as a solid, of the excess of pollutant solubilized in micelles. The remaining Tcp dissolved in water is captured with a multistage operation which allows lowering its content underneath the toxicological limit for drinkable water. Although in Figure 7 a cocurrent multistage by simple contact is represented, in the actual equipment a more useful design could be used. Obviously in the engineering practice interferences of biological compounds can be very significant for surfactant/NAPL/system, and these effects must be taken into account in designing a surfactant-remediation plant. Indeed, adsorption is a function of pH and affected by competition, for adsorption sites, by other chlorophenols and other organic compounds (17). Notwithstanding, the results discussed and suggestions proposed may be particularly useful in these respects.

Acknowledgments The authors are grateful to “Consorzio Interuniversitario per lo sviluppo dei Sistemi a Grande Intefase-CSGI (Firenze)” for financial support.

Literature Cited (1) Schmitz, R. J. Introduction to water pollution biology; Elsevier: Amsterdam, 1995. (2) Acar, Y.; Zappi, M. Infrastructural needs in waste containment and environmental restoration. J. Infrastruct. Syst. 1995, 1, 8291. (3) Harwell, J. H.; Sabatini, D. A.; Knox, R. C. Surfactants for ground water remediation. Colloids Surf. A: Physicochem. Eng. Asp. 1999, 151, 255-268. (4) Miller, C. T.; Hill, E. H.; Montier, M. Remediation of DNAPLcontaminated subsurface systems using density-motivated mobilization. Environ. Sci. Technol. 2000, 34, 719-724. (5) Sabatini, D. A.; Knox, R. C.; Harwell, J. H. Surfactant Selection, Hydraulic Efficiency, and Economic Factors; U.S. EPA Environmental Research Brief number EPA/600/S-96/002, 1996. (6) Dwaraknath, W.; Jackson, R. E.; Pope, G. A. Influence of wettability on the recovery of NAPLs from alluvium. Environ. Sci. Technol. 2002, 36, 227-231. (7) Cowell, M. A.; Kibbey, T. C. G.; Zimmermann, J. B.; Hays, K. F. Partitioning of ethoxylated nonionic surfactants in water/NAPL systems: effects of surfactant and NAPL properties. Environ. Sci. Technol. 2000, 34, 1583-1588.

6840

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 19, 2007

(8) U.S. Environmental Protection Agency. Surfactant Injection for Ground Water Remediation: State Regulators’ Perspectives and Experiences; EPA/ 542/R-95/011; Technology Innovation Office: Washington, DC, 1995. (9) Gavaskar, A. R. Design and construction techniques for permeable reactive barriers. J. Hazard. Mater. 1999, 68, 41-71. (10) Baciocchi, R.; Boni, M. R.; D’Aprile, L. Characterization and performance of granular iron as reactive media for TCE degradation by permeable reactive barriers. Water, Air Soil Pollut. 2003, 149, 211-226. (11) Vogan, J. L.; Focht, R. M.; Clark, D. K.; Graham, S. L. Performance evaluation of a permeable reactive barrier for remediation of dissolved chlorinated solvents in groundwater. J. Hazard. Mater. 1999, 68, 97-108. (12) Odokuma, L. O.; Okpokwasili, G. C. Seasonal influences of the organic pollution monitoring of the new Calabar River, Nigeria. Environ. Monitoring Assessment 1997, 45, 43-57. (13) Falbe, J. In Surfactant in consumer products: theory, technology and application; Verwertungsgesellshaft Wor: Munich, 1986; Chapters 8 and 9. (14) Liu, J. C.; Chang, P. S. Solubility and adsorption behaviors of chlorophenols in the presence of surfactant. Water Sci. Technol. 1997, 35, 123-130. (15) Lampi, P.; Vartianinen, T.; Tuomisto, J. Population exposure to chlorophenols, dibenzo-p-dioxins and dibenzofurans after a prolonged ground water pollution by chlorophenols. J. Chemosphere 1990, 20, 625-634. (16) Ambrosone, L.; Ceglie, A. Italian Patent 4454PTIT, 2003. (17) Schellenberg, K.; Luenberger, C.; Schwarzenbach, R. P. Sorption of chlorinated phenols by natural sediments and aquifer materials. Environ. Sci. Technol 1984, 18, 652-657. (18) Laughlin, R. G. In The Aqueous Phase behavior of Surfactant; Academic Press: New York, 1994. (19) Mc Bain, M. E. L.; Hutchinson, E. In Solubilization and Related Phenomena; Academic Press: New York, 1955. (20) Rouse, J. D.; Sabatini, D. A.; Sulflita, J. M.; Harwell, J. H. Influence of surfactants on microbial degradation of organic compounds. Crit. Rev. Environ. Sci. Technol. 1994, 24, 325-370. (21) Roote, D. P. G. Technology Status Report in Situ Flushing; Ground-Water Remediation Technologies Analysis Center (GWRTAC), http://gwrtac.org, 1998. (22) Deshpande, S.; Shiau, B. J.; Wade, D.; Sabatini, D. A.; Harwell, J. H. Surfactant selection for enhancing ex situ soil washing. Water Res. 1999, 33, 351-360. (23) Perry, F. Chemical Engineering Handbook, 3rd ed.; McGraw Hill: New York, 1950; Section 11. (24) U.S. Environmental Protection Agency. Quality Criteria for Water; EPA 440/5-86-001; U.S. Government Printing Office: Washington, DC, 1986. (25) Camper, A. K.; LeChevallier, M. W.; Broadaway, S. C.; McFeters, G. A. Bacteria associated with granular activated carbon particles in drinking water. Appl. Environ. Microbiol., 1986, 52, 434-438.

Received for review March 14, 2007. Revised manuscript received June 28, 2007. Accepted July 30, 2007. ES070643F