Remediation of saturated soil contaminated with petroleum · PDF file Remediation of saturated...

Click here to load reader

  • date post

  • Category


  • view

  • download


Embed Size (px)

Transcript of Remediation of saturated soil contaminated with petroleum · PDF file Remediation of saturated...


    0301-4797/$ - se


    �Correspond E-mail addr

    Journal of Environmental Management 83 (2007) 339–350

    Remediation of saturated soil contaminated with petroleum products using air sparging with thermal enhancement

    A.M.I. Mohamed�, Nabil El-menshawy, Amany M. Saif

    Mechanical Power Department, Faculty of Engineering, Suez Canal University, Egypt

    Received 18 April 2005; received in revised form 3 April 2006; accepted 4 April 2006

    Available online 17 July 2006


    Pollutants in the form of non-aqueous phase liquids (NAPLs), such as petroleum products, pose a serious threat to the soil and

    groundwater. A mathematical model was derived to study the unsteady pollutant concentrations through water saturated contaminated

    soil under air sparging conditions for different NAPLs and soil properties. The comparison between the numerical model results and the

    published experimental results showed acceptable agreement. Furthermore, an experimental study was conducted to remove NAPLs

    from the contaminated soil using the sparging air technique, considering the sparging air velocity, air temperature, soil grain size and

    different contaminant properties. This study showed that sparging air at ambient temperature through the contaminated soil can remove

    NAPLs, however, employing hot air sparging can provide higher contaminant removal efficiency, by about 9%. An empirical correlation

    for the volatilization mass transfer coefficient was developed from the experimental results. The dimensionless numbers used were

    Sherwood number (Sh), Peclet number (Pe), Schmidt number (Sc) and several physical-chemical properties of VOCs and porous media.

    Finally, the estimated volatilization mass transfer coefficient was used for calculation of the influence of heated sparging air on the

    spreading of the NAPL plume through the contaminated soil.

    r 2006 Elsevier Ltd. All rights reserved.

    Keywords: NAPLs; Air sparging; Volatilization mass transfer coefficient; Soil

    1. Introduction

    Non-aqueous phase liquids (NAPLs) pose a significant threat to ground water resources. When NAPLs infiltrate to the subsurface they descend as an immiscible phase. In cases where of the spilled quantity exceeds the retention capacity of the ground water saturated zone; the NAPLs will reach the capillary fringe. A NAPL less dense than water (LNAPL) will form pools at the water table, while a more dense one (DNAPL) will move through the saturated zone, spreading along the less permeable layers and leaving behind a portion of its volume as immobilized pockets of liquid called residual saturation (Okeson et al., 1997).

    In the case of both DNAPL and LNAPL, pumping to remove free product within a highly NAPL saturated lens cannot completely recover the NAPL. With the rising of the water table or DNAPL lens migration, NAPL will

    e front matter r 2006 Elsevier Ltd. All rights reserved.


    ing author. Tel.: +20 0 10 1597112; fax: +20 0 66 3400936.

    ess: (A.M.I. Mohamed).

    become trapped during pumping as a discontinuous residual. Entrapped NAPLs act as long-term sources of groundwater contamination, (Fisher et al., 1999; Sprague and Delahaye, 1996; Held and Celia, 2001). The present research work aims at the enhancement of

    the air sparging remedial technology. Air sparging is a cost- effective, time-efficient system for the remediation of volatile and/or biodegradable contaminants. This techni- que involves introducing forced air into the saturated zone of an aquifer to encourage volatilization of contaminants into the unsaturated zone where the contaminants can then be removed with another complementary technology such as soil vapor extraction (SVE), bioventing, horizontal wells, or heating (Sprague and Delahaye, 1996; Reddy et al., 1999). The airflow behavior induced by air sparging is typically

    characterized by a conical air plume, often known as the radius of influence (Johnson, 1998; Mohtar et al., 1996). Ji et al. (1993) visualized the steady state air distribution patterns using a thin Plexiglas tank with uniform lighting



    A surface area, m2

    a specific interfacial area, m2/m3

    C concentration, mg/cm3

    Cs aqueous saturation concentration, mg/cm 3

    D molecular diffusion coefficient, cm2/s dp particle diameter of the soil, mm d0 normalized mean particle size d1 pipe diameter, ‘mm H dimensionless Henry’s constant h Henry’s constant, atm.m3/mol K permeability, mm2

    kL liquid phase volatilization mass transfer coeffi- cient, cm/min

    kG gas phase volatilization mass transfer coeffi- cient, cm/min

    ms initial mass of VOC injected into the employed soil

    _m rate of mass transferred, mg/min Pe Peclet number Q aeration rate, L/min Sc Schmidt number Sh Sherwood number T temperature, K t time, s UC uniformity coefficient of the porous media u air velocity, cm/s Va air phase volume, cm


    x distance in x-dir., cm z depth in z-dir., cm X soil reactor width, cm Z soil reactor depth, cm

    Greek letters

    Zrem The contaminant removal efficiency l weight factor for the finite difference technique t tortuosity factor of the porous media e porosity of the porous media n kinematic viscosity, cm2/s ym normalized mean temperature o uncertainty of measured or estimated values


    a air phase diff diffusion diss dissolution i interfacial in inlet G gas phase L liquid phase NAPL non-aqueous phase liquid out outlet ref reference w water phase

    A.M.I. Mohamed et al. / Journal of Environmental Management 83 (2007) 339–350340

    behind the tank and glass beads as porous media. They observed two patterns of airflow in soils. The first pattern was observed for medium to coarse grained media where the airflow is characterized by a plume of discrete air bubbles. The second pattern describes the airflow regime for coarse to fine grained media that resemble the textures of sand, silts and clays of natural aquifer material. This flow regime was dominated by channel flow where air plumes are formed from discrete and continuous air channels. As the injection rate increases air channels increase in number and grow into a condensed continuous cone-shape cavity.

    Braida and Ong (2000) conducted experiments using a single air channel setup to study the influence of porous media properties and air velocity on the fate of NAPLs under air sparging conditions. Their study showed that the presence of advective airflow in air channels controlled the spreading of the dissolved phase but the overall removal efficiency was independent of the air flowrate. In addition, they noted that the removal efficiencies and dissolution rates of the NAPL were strongly affected by the mean particle size of the porous media during air sparging. This agrees with the investigations made by Reddy and Adams (1998). They performed a series of one dimensional column experiments to study the effects of soil type, air injection mode, and the synergistic effects of co-

    contaminants on air sparging removal efficiency. They found that there is a threshold value for the effective particle size (dp10), which is equal to 0.2mm; above this threshold value, the rate of removal is linearly proportional to the (dp10) value; while below this value, there is a drastic increase in the time required for contaminant removal. Additionally, they found that the pulsed air injection mode has no advantage over continuous injection for coarse sand; however, pulsed air injection led to substantial reductions in system operating time for fine sand. Also they observed a slight increase in removal rate when benzene and toluene coexisted in the test soil compared to when they existed alone. Chao et al. (1998) developed non-equilibrium water-to-

    air mass transfer experiments for six volatile organic compounds during air sparging in soil columns packed with coarse, medium, or fine sand or glass beads. They performed a numerical study and assumed that the concentration in the bulk phase remained constant due to slow diffusion of VOCs in the aqueous phase to the air–water interface as compared to the rapid volatilization of VOCs at the air–water interface. Therefore, they modeled the interface mass transfer alone. At an in-situ air sparging remediation site contaminated

    through the spillage of a LNAPL waste, the influence of the system design parameters in terms of contaminant


    Fig. 1. Schematic diagram of the approach model.

    A.M.I. Mohamed et al. / Journal of Environmental Management 83 (2007) 339–350 341

    removal time was reported by Benner et al. (2000). They suggested that only the type of sparging operation (i.e. pulsed or continuous) was significant in terms of total contaminant removal time, while both the sparging operation and air injection rate were significant in terms of removal of critical xylene species.

    However, the contaminants within the radius of influ- ence will not be removed with equal efficiencies due to the natural inherited heterogeneity of the porous medium that yields non-uniform and asymmetrical air plumes (Ahlfeld et al., 1994). Rahbeh and Mohtar (2001) studied the influence of porous media heterogeneity and air channeli- zation on contaminant removal by air sparging. Their results showed that the contaminant removal is propor- tional to channel spacing, and controlled by th