FR2784911A1 - Triple phase extraction procedure for polluted soils pumps out gases in vadose zone, floating pollutants and water - Google Patents

Abstract

The procedure consists of using triple phase extraction to pump out gases contained in the vadose zone (1), floating liquid pollutants (2) and dissolved substances contained in the upper zone (3) of the aquifer which have to be extracted, leaving the lower unpolluted zone (4). The extraction is carried out using small-diameter needles for an optimum area of influence and to adapt to the nature of the terrain, and can go down to a depth of tens or hundreds of metres.

Description

PROCESS FOR REHABILITATION OF POLLUTED SOILS BY

TRIPLE PHASE EXTRACTION

  The technical sector of the invention The present invention has developed an original method of simultaneously pumping gases contained in the vadose zone, pollutants

  supernatants and groundwater.

  PRIOR ART Pollution of the soil by liquid products requires the implementation of various extraction and treatment techniques: pumping of groundwater and supernatants, extraction of air and evaporated products, treatment by action organic in the tablecloth and

in the vadose zone.

  In such a situation, it is essential to manage flows as well as possible in all phases of the process. The extraction is carried out from small diameter boreholes, typically of the order of 50 mm, similar to those used in the drawdown processes by needles. The number of boreholes can reach thirty, or even fifty for very large sites. In some cases, like the treatment of petrol stations, a few needles may suffice. This three-phase extraction technique (Water, Oil and Gas) was applied

  successfully at several sites in France and abroad.

  The advantages of the method are numerous, both technically and economically: - drilling is inexpensive, all the more so as the permeability of the

soil is weak.

  - It is not necessary to install a pump at the bottom of each well

to extract liquids.

  - We can pump three phases simultaneously, the water which can contain

  dissolved pollutants, gas and supernatants.

  - It is possible to pump to great depth.

  - The gas and liquid flow rates can be adjusted independently.

  The optimization of the installations requires a precise design methodology taking into account the interactions between the different components. It is, in particular, fundamental to precisely calculate the flows in the various zones of the circuit, including the soil, the extraction needles, the transport network (s), the pumping machines and the effluent treatment devices. Investments and operating costs over the processing period can vary considerably depending on the choices

  technological and machines implemented.

  Detailed description of the invention

  * Principle of triple phase extraction Figure 1 gives a schematic representation of polluted soil, with its four characteristic zones. In the upper vadose zone 1 the gas circulates. The flows in this zone are particularly complex, since it generally contains liquid inclusions and adsorbed components. It can be modified under the effect of precipitation or variation in groundwater levels. Zone 2 consists of supernatant liquids which must be eliminated. Zone 3 is the upper part of the aquifer. It contains dissolved components which must be extracted. Zone 4 is made up

unpolluted water.

  Figure 1 Soil can be treated by pumping the liquid it contains. This pumping creates a drawdown in the water table and at the level of the well (Figure 2.a). It is also possible to extract the gases alone. It is the classic operation of "Venting" or gas ventilation which leads to an elevation of the water table at the level of the well (Figure 2.b). If the two methods are applied simultaneously, the two fluids can be extracted in any proportions by varying the vacuum applied by the air flow and the drawdown imposed on the well. It's extraction

three-phase (Figure 2.c).

  Figure 2 * Specifications, preliminary tests In any treatment of polluted soils, we must first determine the characteristics of the pollution, the water table and the ground. To do this, test drillings are carried out and piezometers are used according to the usual hydrogeology methods. The characteristic quantities of the terrain, transmissivity and storage coefficient in particular are thus determined. This first phase is based on calculation models in porous media and must make it possible to determine the number of needles to be used, the flow rates to be extracted from each of them and the processing time. This calculation will be done first on the water and the supernatants and will take into account the slope of the aquifer. Equivalent procedures are applied to the vadose area. These are conventional dimensioning methods for

  <, Venting ", with possible consideration of biological aspects.

  Treatment tests in the laboratory or in situ will generally be

desirable.

  * Simulation of flows in the soil during the treatment time One of the particularly interesting points of the three-phase treatment is that it is possible to significantly increase the flow rates of water and extracted supernatants by applying a sufficiently large depression in the needle. The flow of water extracted from the water table is in fact proportional to the sum of the effective drawdown of the water table and the pressure difference between atmospheric pressure and pressure

in the needle.

  The flow rate of water extracted per meter of aquifer is given in FIG. 3 for different values of the total drawdown, sum of the effective drawdown and the pressure difference applied. This highlights the interest

depression.

  Optimizing the number of needles and the treatment time requires a good simulation of these parameters by means of hydrology calculation codes. The basic formulas used are the formulas of TheïsJacob for flows in the aquifer and those of Theim for flows

in the Vadose area.

  At the end of this analysis, the following information is available - the pressure in each needle, - the air flow in each needle,

  - the flow of liquid to be extracted from each needle.

  We perform a simulation of these parameters as a function of time and we thus have the input data to size the network.

  best suited for the intended duration of treatment.

  Figure 3 It goes without saying that this phase of study is particularly important and that the success of a treatment is often linked to a good knowledge of the soil, the water table, the vadose zone and the behavior of the pollutants subjected to such or such such treatment. When these technical choices are made, the processing time is completely defined. * The extraction network Although a three-phase extraction system does not require the use of machines on each needle, two basic systems are possible, extraction with separate networks, one for liquids and

  the other for gas, and single-tube extraction in three-phase flow.

  In both cases, the networks are branched.

  The choice between these two methods involves an estimate of the investment costs, operating costs and conditions

operating.

  1. In separate networks: The needles The two-phase extraction needles are made up of two main parts: - the needle proper which is formed by a screened piping over all or part of its height, and which is introduced into the borehole - the liquid extraction rod, the end of which is located in part

  lower part of the needle and which is therefore completely submerged.

  The two conduits are connected, as shown in Figure 4 to their respective extraction networks. The adjustment of the flow rates at each needle is ensured by a fixed or adjustable pressure drop, the

  the former being generally more reliable.

  Figure 4 Transport Transport in separate networks requires the implementation of codes

  of classical calculation of fluid mechanics.

  For the liquid network, essentially consisting of water, the flow rate is given by needle. We therefore know by summation the flow at any point in the network. <"YLec Consultants>, (TM) has developed for the Applicant an easy to use software which gives the pressures at any point of the network, and, in particular at the inlet of the extraction pump which can possibly be coupled to a We can determine the characteristics of the pump to be used, flow rate, height generated, NPSH and power. The flow rates at each needle are adjusted by means of thin diaphragms

  wall which are calculated automatically.

  For the gas network, account must be taken of the flow in the soil, which gives the pressures and mass flow rates at each needle. With optimized three-phase extraction, the compressibility of the gas cannot be overlooked. The calculation programs therefore take account of this behavior of the gas and of a thermodynamic scheme which will be

  function of the geometry and the thermal environment of the network.

  As with the liquid network, this calculation makes it possible to automatically determine the characteristics of the vacuum pump or, more generally, of the pumping group to be used as well as the

  various flow adjustment diaphragms to be placed at the needle head.

  2 In a three-phase network: The needles A three-phase network makes it possible to simultaneously transport liquids and gases which enter the needle through the strainer. Unlike pumping with separate networks, the end of the cane

  extraction fixes the point of physical drawdown in the well.

  Essentially, the aspirated flow consists of air and vapors.

  However, when liquid enters the needle, it is automatically driven by the predominant air flow. The water level in the well is therefore automatically positioned at the end of

the walking stick.

  Figure 5 illustrates the operation of a two-phase extraction needle. Figure 5 We can easily calculate the density of the two-phase mixture sucked into a rod: PL being the density of the liquid (or of the mixture of liquids), PG being that of the air under the suction pressure pp in the well

considered.

  The PD density of the two-phase mixture is calculated by: P = cxP + (1- a) P PD pc Ldfn) p ac being the rate of presence of local liquid defined by: (x = VL vo

VG + VL

  VG and VL being the volumes of air and liquid respectively in a given total volume. When the flow velocities are sufficiently high in the pipes, the ratio of the volumes becomes almost equal to the ratio of the volume flow rates qG and qL. The void rate can then be written: qL qL - qG _ qL qG + qL] + qL qG qG The flow calculations in the porous medium show that the order of magnitude of the ratio of gas and liquid flow rates by meter of land height is 50 when the vacuum applied to the well is greater than the physical drawdown. If we assume that the height of the vadose zone is equal to the height of the water table, we will find

  a liquid presence rate of approximately 2% in the mixture.

  Therefore, the density of the fluid is immediately calculated

  two-phase taking into account the density of water equal to 1000.

  This is therefore typically around 20 kg / m3.

  In conventional pumping of water by suction, we are theoretically limited to a depth such that Pa, being atmospheric pressure, Pa = pgh

  We find, classic result, a maximum height of 10 m.

  With a three-phase medium, the theoretical maximum pumping height is given by the same expression: Pa pa pgh With a density of 20, the theoretical pumping limit is therefore 500 meters! The accessible practical values are rather of the order of 0.7 to 0.8 times the theoretical values in both cases. It can be seen that it is therefore possible to treat land using the Applicant's three-phase process regardless of the depth of the water table. It is therefore a decisive advantage of

  the present invention over traditional methods.

  The choice between the two methods, separate networks or common network is made on economic criteria, knowing that pumping with only one

  network is feasible in all cases.

  Automatic priming One of the difficulties encountered during the operation of a two-phase network is that it can be difficult to prime the last needles

  when a large number of them are already in operation.

  This delicate point was the subject of specific studies and several solutions were proposed as part of the collaboration between the Applicant and YLec Consultants. They are subject to specific sizing procedures and no longer require the intervention of an operator, either at the first start-up or during untimely stops. The flow regime It is important that the flow velocities in the sampling rod are large enough to avoid instabilities

  and low enough to limit pressure drops.

  Figure 6 gives an example of the flow configurations that

  may meet in vertical piping.

  The flow in the vertical rods was therefore modeled so that we find an optimum according to the flow rates and pressures imposed

  by treatment with each needle.

  To do this, we apply the characteristic equations of two-phase flows, based on the determination of the pressure gradient resulting from the sum of the contributions of gravity, friction and momentum transfers.

  1 D. CHISHOLM. Two Pause-Flows in Pipelines and Heat Exchangers.

  op _ dpF + dPM + dPK dx dx dx dx We deduce the ideal diameters of the rods as well as their operating range in water and air flow rates and the corresponding pressures. Figure 6 Flow in the network Flow in the network is treated analogously to

compressible flow.

  It is, of course, important to take into account the compressibility of the medium as well as the thermodynamic transformations including the

  heat and mass transfers between phases.

  The pressure drops are calculated using the concepts of

  two-phase multipliers and Lockhart and Martinelli correlations.

  The two-phase multiplier ç G is the ratio of the pressure gradient PG caused by the two-phase flow to the pressure gradient due to only one of the constituents, for example gas under local operating conditions 2 = Dpp

PG DPPG

  We show that this two-phase multiplier only depends on the Lockhart and Martinelli parameter, defined by:

X = DPP

  X DpPG Figure 7 gives one of the correlation curves obtained by these

  authors on a large number of test points.

  Figure 7 Given the relative complexity of the problem, an original specific calculation program has been developed. The results are, as in the

previous case.

  - The characteristics of the pipes (diameter and length).

  - Pressure losses and pressures at any point on the network.

  - The dimensions of the pressure drop adjustment diaphragms in

head of needles.

  From this data, it is possible to size the extraction network, vacuum tank, vacuum pump and extraction pumps

oil and water accessories.

  We deduct the corresponding investment and operating costs.

  The calculation procedure makes it possible to carry out a multi-criteria optimization from which we will decide on the type of network and its most suitable operating conditions. We must obviously take into account

  of the evolution of the behavior parameters of the water table.

  Figure 8 * Effluent treatment Effluents are treated with the aim of recovering pollutants and

  meet the standards for discharging water and air into the natural environment.

  The following components are generally found on the Applicant's installations. In the case of an extraction with two networks, the air is automatically separated at the needle. It is therefore not necessary to plan

  this function in the vacuum tank.

  In the case of extraction with a single three-phase network, the vacuum tank is systematically used as a gas-liquid separator. The gas

is evacuated by the vacuum pump.

  The liquid can be extracted from the vacuum tank by different pumping or pneumatic emptying processes. We can also use this

  vacuum tank as decanter, the tank operating at constant level.

  By using a compact coalescer with filling or honeycomb, we can thus make the best use of the volume of the tank and avoid

  implementation of an external decanter at atmospheric pressure.

  In the case of separate networks, another solution may consist in pumping the two liquids through a low emulsifying pump and in

  use a compact conventional gravity separator.

  There is therefore a range of solutions for the function of gravity extraction and separation of species. The most suitable choice must take into account all of the pumping extraction and separation functions. Very important differences on investments and especially on operating costs may appear depending on whether one

  choose one or the other of the solutions.

  It will be emphasized, however, that there is no universal optimal solution.

  The Applicant systematically conducts technical-

  cost-effective to compare solutions using software

  specific developed with the collaboration of YLec Consultants.

  The other treatments are more conventional since it involves implementing a "stripping" (or entrainment) with air of water in order to remove the dissolved gaseous products. The Applicant also worked on the optimization of "stripping" devices to reduce energy costs. Finally, the air is treated by combustion in catalysis ovens or by passage over activated carbon. In practice, the rejection standards impose the type of treatment and the gains on this part of the process

are quite limited.

Claims (9)

  1. - Process for the rehabilitation of polluted soils of medium or low permeability, characterized in that it comprises a treatment of the soil by a triple phase extraction process of simultaneous pumping of the gases contained in the vadose zone, supernatant pollutants and water from the tablecloth.   2. - Process for the rehabilitation of polluted soils of medium or low permeability, according to claim 1, characterized in that in the upper vadose zone 1 the gas circulates, the zone 2 consists of supernatants which it is necessary to eliminate, zone 3 is the top of the aquifer, and contains dissolved components which   must be extracted, and zone 4 consists of unpolluted water.   3. - Method for the rehabilitation of polluted soils of medium or low permeability, according to claim 1 or 2, characterized in that needles of small diameter are used which optimize their   influence radius and adapt it to the characteristics of the terrain.   4. - Process for the rehabilitation of polluted permeability soils   medium or low, according to any one of claims 1 to 3,   characterized in that vacuum pumping is carried out.   5. - Process for the rehabilitation of polluted permeability soils   medium or low, according to any one of claims 1 to 4,   characterized in that vacuum pumping is carried out and in that the flow rate obtained with vacuum pumping is from 10 to 20   times higher than that of a conventional pumping.   6. - Process for the rehabilitation of polluted permeability soils   medium or low, according to any one of claims 1 to 5,   characterized in that the method is applied with an almost horizontal sheet, without drawdown, around the well, 7. - Method for the rehabilitation of polluted soils of medium or low permeability, according to claim 6, characterized in that the tablecloth is thin.   8. - Process for the rehabilitation of polluted permeability soils   medium or low, according to any one of claims 1 to 7,   characterized in that the pumping is carried out at any depth, which can reach several tens, even hundreds of meters. 9. Process for the rehabilitation of polluted permeability soils   medium or low, according to any one of claims 1 to 8,   characterized in that the relative flow rates of liquid and gas.   10. - Process for the rehabilitation of polluted permeability soils   medium or low, according to any one of claims 1 to 9,   characterized in that the two-phase sampling rods are primed automatically, without any intervention operators.   11. - Process for the rehabilitation of polluted permeability soils   medium or low, according to any one of claims 1 to 10,   characterized in that the installations are dimensioned by a calculation procedure allowing the best possible choices to be made   depending on the characteristics of the site and the processing objectives.