Adsorption of Trichloroethene at theVapor/Water Interface

R O B E R T G . B R U A N T , J R . A N DM A R T H A H . C O N K L I N *

Department of Hydrology and Water Resources,The University of Arizona, P.O. Box 210011,Harshbarger Building (11), Tucson, Arizona 85721-0011

Aqueous solution surface tension (pressure) as a functionof vapor-phase trichloroethene (TCE) pressure isothermswere measured at atmospheric pressure at 287.2, 291.2,297.2, 303.2, and 315.2 K using a flow-through vaporadsorption method. Solute (i.e., TCE)-induced surfacetension variations were quantified using Axisymmetric DropShape Analysis-Profile (ADSA-P); vapor-phase TCEpressures were measured using automated gas chromato-graph analysis. Surface tension reductions of 5-10%from neat water at saturated TCE vapor pressure werenoted. Measured surface tension (pressure) isotherms wereused to estimate vapor/water interface adsorption as afunction of vapor-phase TCE pressure using the Gibbs relativeinterface excess (i.e., adsorption) equation and a nonidealtwo-dimensional equation of state. Complete isothermaladsorption profiles were nonlinear, with acceleratedadsorption at increasing vapor-phase TCE pressures.Comparison to other studies of adsorption at infinite dilution(i.e., linear partitioning) and corresponding to thermody-namics (i.e., ideal equilibrium standard molar free energy,enthalpy, and entropy change) indicate good agreement.Estimates of TCE planar surface area were used to calculatethe fraction of monolayer coverage at the vapor/waterinterface as a function of vapor-phase TCE pressure, whichapproached a maximum of 0.6-0.75 at saturated vaporpressure.

Introduction

It is well known that trichloroethene (TCE) is an environ-mental contaminant of significant concern. Classically,equilibrium partitioning of TCE has been quantified in termsof bulk environmental compartments (i.e., liquid, gas/vapor,organic matter) or association with mineral surfaces (e.g., 1,2). This accounting, however, neglects the influence of thevapor/water interface, which has been identified in bothsubsurface and atmospheric systems as a consequential sinkfor similar compounds (3-5). To better predict the envi-ronmental transport and fate of TCE, experiments have beenconducted to quantify equilibrium adsorption at the vapor/water interface as a function of vapor-phase TCE pressure.Aqueous solution surface tension as a function of vapor-phase TCE pressure isotherms were measured at at-mospheric pressure at five temperatures of environmentalinterest (i.e., 287.2, 291.2, 297.2, 303.2, and 315.2 K). Ap-plication of the Gibbs relative interface excess (i.e., adsorp-

tion) equation and a nonideal two-dimensional equation ofstate to surface pressure (i.e., surface tension of water lesssolution surface tension) isotherms allowed estimation ofsolute (TCE) adsorption at the vapor/water interface, includ-ing thermodynamic parameters of ideal equilibrium stand-ard molar free energy, enthalpy, and entropy change ofadsorption, and corresponding degree of monolayer cover-age.

Materials and MethodsChemicals. Trichloroethene (C2HCl3, 99.5+ % Pure) waspurchased from Fluka Chemical Corp. (Ronkonkoma, NY)and used with further processing. Water from a Milliporedeionization system (Bedford, MA) was distilled in a potas-sium permanganate/sodium hydroxide solution followed bydistillation in an all-water apparatus before use. Air for solutetransfer was standard breathing quality passed through anactivated-carbon filter to remove trace organics.

Methods. Variations in aqueous surface tension due tosolute adsorption were measured using Axisymmetric DropShape Analysis-Profile (ADSA-P) with digital image proc-essing (Applied Surface Thermodynamics, Toronto, ON).Detailed descriptions of applicable systems and protocolshave been presented previously by Cheng and co-workers(6, 7) and Bruant and Conklin (8), and only a brief sum-mary is currently given. For volatile solute adsorption analy-sis, water pendant drops were formed from a Teflon capil-lary in an inert flow-through environmental cell. The cellwas thermostated to control temperature to ( 0.2 K. Toinitiate surface tension (pressure) isotherm measurement, awell-dispersed TCE/water vapor front was delivered to aninitially solute-free drop by a slow advecting air stream.Bulk vapor-phase TCE pressures inside the cell were quanti-fied at regular time intervals using automated sample-loopinjection to a Varian Star 3600 CX gas chromatograph (Wal-nut Creek, CA) with a flame ionization detector (FID).Simultaneously, aqueous solution surface tensions (with95% confidence intervals) were measured from the dropshape using ADSA-P. Rapid TCE adsorption (i.e., instanta-neous response of aqueous surface tension to vapor-phasesolute pressure (8)) allowed correlation of vapor-phase TCEpressure and aqueous solution surface tension measurementsfor each analysis time to define equilibrium surface tensionversus vapor-phase TCE pressure pairs. Solute breakthroughprovided numerous pairs for delineation of isothermalprofiles at 287.2, 291.2, 297.2, 303.2, and 315.2 K. TCEpressures ranged from zero to saturated vapor pressure at287.2, 291.2, and 297.2 K, whereas maximum pressures at303.2 and 315.2 K were limited by ambient room temperatures(∼298.2 K).

ResultsMeasured isotherms (depicted as aqueous solution surfacepressure as a function of vapor-phase TCE pressure) arepresented in Figure 1. The fractions of total isothermmeasured (i.e., highest measured pressure value, pV

max,divided by saturated TCE vapor pressure, pV

sat (9)) are givenin Table 1. Surface pressures were small, reflecting only5-10% (dependent on temperature) decreases in aqueoussolution surface tension at saturated TCE vapor pressure. Allcomplete curves (i.e., 287.2, 291.2, and 297.2 K, where pV

max

pVsat

-1 ≈ 1) are concave to the surface pressure axis andincrease in a general monotonic fashion.

Measured surface pressure isotherms were fit with amathematical form derived from the Gibbs relative interface

* Corresponding author phone: (520) 621-5829; fax: (520) 621-1422; e-mail: martha@hwr.arizona.edu.

Environ. Sci. Technol. 2001, 35, 362-364

362 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 2, 2001 10.1021/es000994t CCC: $20.00 2001 American Chemical SocietyPublished on Web 12/08/2000

excess equation (Equation 1) (10) and a nonideal two-di-mensional equation of state (Equation 2) (11,12):

where Γ is the relative interface excess of solute moleculesper unit area of the interface (analogous to adsorption)[molecule m-2], a is the corresponding area per adsorbedsolute molecule [m2 molecule-1], pV is the vapor-phase TCEpressure (assuming ideal behavior), k is the Boltzmannconstant [N m molecule-1 K-1], T is the absolute temperature[K], π is the surface pressure [N m-1], a0 is an empirical fittingparameter [m2 molecule-1], and the subscript T indicatesevaluation at constant temperature. Substituting eq 1 intoeq 2 and integrating gives the surface pressure equation usedfor data interpolation:

where c is an integration constant [dimensionless]. Equation3 was fit to the data using a nonlinear least-squares parameterestimation routine. Derived values for each isotherm andcorresponding coefficients of determination are presentedin Table 1.

DiscussionIsothermal adsorption of TCE at the vapor/water interfacefor each measured surface tension (pressure) value was

estimated from the following modified form of eq 2 and thederived a0 values from eq 3:

Results are presented graphically as a function of vapor-phase TCE pressures in Figure 2 for the five temperaturesconsidered. Adsorption profile nonlinearity at intermediateto high vapor-phase TCE pressures (285.2, 291.2, and 297.2K experiments) is indicative of a two-dimensional liquid-expanded state with attractive solute-solute interaction (13).Additionally, because of the lack of inflection, there appearsto be no threshold value of adsorption within the vapor-phase TCE pressure ranges considered, even near or atsaturated TCE vapor pressure. Similar profile shapes andadsorption magnitudes have been defined previously foralkane and aromatic solute adsorption (8, 14, 15).

For comparison and verification, a linear partition coef-ficient (Γ pV -1) at limiting vapor-phase TCE pressure (i.e., aspV f 0) of 2.70 × 10-7 m at 297.2 K was calculated from theinterpolated parameters. Agreement with the value of 2.64× 10-7 m at 298.2 K presented by Hoff and co-workers (16)for dilute TCE adsorption is excellent (Figure 2). However,extrapolation of the linear models to moderate or high vapor-phase TCE pressures (e.g., > 0.5 pV pV

sat-1 at 297.2 or 298.2

K) leads to significant underprediction of vapor/waterinterface adsorption due to neglect of isotherm nonlinearity(Figure 2).

Ideal equilibrium standard molar free energy, enthalpy,and entropy changes of adsorption at limiting TCE pressuresalso were calculated from the presented interpolationfunctions using the mathematical construct of Kemball andRideal (11). Free energy changes ranged between -1.43 ×104 and -1.56 × 104 N m mol-1, indicating an energeticallyfavorable regime for TCE adsorption from the vapor-phaseto the interface-phase. Linear interpolations to the free energyresults as a function of temperature provided values for theenthalpy and entropy changes of -2.63 × 104 N m mol-1 and-3.85 × 101 N m mol-1 K-1, respectively. Although little dataexist regarding adsorption energetics for chlorinated hy-drocarbons at the vapor/water interface, thermodynamiccalculations presented by Hartkopf and Karger (17) at infinitedilution for chloromethane (285.7 K) of -1.5 × 104 N m mol-1,-2.3 × 104 N m mol-1, and -3 × 101 N m mol-1 K-1 (freeenergy, enthalpy, and entropy changes of adsorption,

FIGURE 1. Mean aqueous solution surface pressure (with 95%confidence intervals) as a function of vapor-phase TCE pressure at287.2 K (]); 291.2 K (0); 297.2 K (4); 303.2 K (O); and 315.2 K (3).

TABLE 1. Fraction of Total Isotherm Measured (pVmax pV

sat-1), Number of Data Points (n), and Fitted Equation 3Parameters (Including Coefficients of Determination, R2) forSurface Pressure versus Vapor-Phase TCE Pressure Profiles;Maximum Fraction of Monolayer Coverage (Γmax Γlim

-1)

Equation 3 parameters

T [K]pV

maxpV

sat-1 n

a0 × 1019

[m2 molecule-1]c × 10-1

[unitless] R2 × 101Γmax

Γlim-1

287.2 0.95 30 -6.32 1.48 9.94 0.75291.2 0.97 50 -4.90 1.48 9.95 0.61297.2 1.07 114 -4.51 1.51 9.95 0.60303.2 0.81 79 1.16 1.51 9.94 0.18315.2 0.50 122 3.44 1.58 9.95 0.09

Γ ≡ 1a

) 1kT( dπ

d ln pV)T

(1)

π(a - a0) ) kT (2)

pV ) π exp(πa0

kT+ c) (3)

FIGURE 2. TCE relative interface excess as a function of vapor-phase TCE pressure at 287.2 K (]); 291.2 K (0); 297.2 K (4); 303.2K (O); and 315.2 K (3). The dashed line represents adsorption at298.2 K based on a linear partition coefficient given by Hoff andco-workers (16).

Γ ) (kTπ

+ a0)-1(4)

VOL. 35, NO. 2, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 363

respectively) compare favorably with the presented results.Further comparison of the TCE adsorption enthalpy changewith the heat of liquefaction (-3.4 × 104 N m mol-1 at 297.2K (9)) suggests a physical adsorption mechanism. Comparisonof the entropy change with the Sackur-Tetrode/Kemballapproximation of translational entropy loss (-3.9 N m mol-1

K-1 at 297.2 K (18)) suggests little additional restriction (e.g.,vibrational or rotational) of the solute at the interface.

The maximum fraction of monolayer coverage (Γmax Γlim-1)

was calculated for each isotherm using an estimate of thelimiting two-dimensional planar surface area for TCE (alim

) Γlim-1). Assuming a spherical molecular shape, the circular

projection area was approximated from the three-dimen-sional van der Waals surface area of TCE. This approachprovides an estimate of 3.0 × 10-19 m2 molecule-1 for a three-dimensional van der Waals surface area of 1.2 × 10-18 m3

molecule-1 (9). Results for the calculation of maximumfractional coverage are presented in Table 1. At saturatedvapor pressure, current estimates indicate that only 60-75%of a completed TCE monolayer is formed at the vapor/waterinterface for the temperatures considered.

AcknowledgmentsThis manuscript was made possible by grant number P42ESO4949 from the National Institute of Environmental HealthScience, NIH, with funding provided by EPA. Its contents aresolely the responsibility of the authors and do not necessarilyrepresent the official views of the NIEHS, NIH, or EPA. Theauthors are very appreciative of the insightful commentsprovided by the reviewers.

Note Added after ASAP PostingThis paper was released ASAP on 12/08/00 with errorsintroduced during galley processing in the first line of the

Abstract and the 13th line of the Introduction. The correctversion was posted on 01/15/01.

Literature Cited(1) Grathwohl, P.; Reinhard, M. Environ. Sci. Technol. 1993, 27,

2360.(2) McCarthy, K. A.; Johnson, R. L. Water Resour. Res. 1993, 29,

1675.(3) Pennell, K. D.; Rhue, R. D.; Rao, P. S. C.; Johnston, C. T. Environ.

Sci. Technol. 1992, 26, 756.(4) Valsaraj, K. T. Water Res. 1994, 28, 819.(5) Gustafsson, O.; Gschwend, P. M. Atmos. Environ. 1999, 33, 163.(6) Cheng, P.; Li, D.; Boruvka, L.; Rotenberg, Y.; Neumann, A. W.

Colloids Surf. 1990, 43, 151.(7) Cheng, P.; Neumann, A. W. Colloids Surf. 1992, 62, 297.(8) Bruant, R. G., Jr.; Conklin, M. H. J. Phys. Chem. 2000, in press.(9) Daubert, T. E.; Danner, R. P. Physical and Thermodynamic

Properties of Pure Chemicals: Data Compilation; HemispherePublishing: New York, 1989.

(10) Adamson, A. W.; Gast, A. P. Physical Chemistry of Surfaces; JohnWiley and Sons: New York, 1997.

(11) Kemball, C.; Rideal, E. K. Proc. R. Soc. London, Ser. A 1946, A187,53.

(12) Volmer, M. Z. Phys. Chem. 1925, 115, 263.(13) MacRitchie, F. Chemistry at Interfaces; Academic: San Diego,

CA, 1990.(14) Cutting, C. L.; Jones, D. C. J. Chem. Soc. 1955, 4067.(15) Dorris, G. M.; Gray, D. G. J. Phys. Chem. 1981, 85, 3628.(16) Hoff, J. T.; Mackay, D.; Gillham, R.; Shiu, W. Y. Environ. Sci.

Technol. 1993, 27, 2174.(17) Hartkopf, A.; Karger, B. L. Acc. Chem. Res. 1973, 6, 209.(18) Kemball, C. Proc. R. Soc. London, Ser. A 1946, A187, 73.

Received for review February 11, 2000. Revised manuscriptreceived October 10, 2000. Accepted October 17, 2000.

ES000994T

364 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 2, 2001