WO2009144390A1

WO2009144390A1 - Method for checking the integrity of geological storage of co<sb>2</sb>

Info

Publication number: WO2009144390A1
Application number: PCT/FR2009/000278
Authority: WO
Inventors: François Kalaydjian; Bernard Bourbiaux; Brigitte Bazin
Original assignee: Ifp
Priority date: 2008-03-19
Filing date: 2009-03-17
Publication date: 2009-12-03
Also published as: FR2929007A1; FR2929007B1

Abstract

The present invention relates to a method for checking the integrity of geological storage containing CO2, wherein the following stages are carried out: CO2 is injected into a geological formation (1) closed by a caprock (4) that prevents the CO2 from resurfacing; and a reducing agent solution is injected, reducing CO2 mobility in the vicinity (A) of a vulnerable or endangered area of the covering where CO2 has escaped or may escape.

Description

METHOD OF CONTROLLING THE INTEGRITY OF GEOLOGICAL STORAGE CONTAINING CO2

The present invention relates to the field of CO2 storage in the subsoil, for example in a geological reservoir whose structure and the nature of the sediment cover naturally ensure the containment of the fluid stored in the reservoir, in this case CO2.

Maintaining, controlling and controlling the CO2 containment qualities during the storage life require:

(a) to provide for devices and / or methods of monitoring the unexpected occurrence of any migration of CO2 from the storage tank,

(b) to take steps to ensure the possible risks that such leakage would be likely to cause to the environment.

The present invention relates to a method for stopping or greatly slowing down the CO2 migration process in case of loss of storage integrity, in particular by fracturing or microcracking of the rock cover. For this, we put in place a reducing agent for CO2 mobility in storage leaks.

Thus, the present invention relates to a method for controlling the integrity of a geological storage containing CO2, in which the following steps are carried out:

- CO2 is injected into a geological formation closed by a rock cover preventing the rise of CO2 to the surface,

a solution of CO2 mobility reducing agent is injected in the vicinity of a fragile or weakened zone of the blanket where a CO2 leak has occurred, or may take place.

The mobility reducing agent may be a foaming agent selected for foaming the leaking CO2 through said blanket. The foaming agent may be a surfactant or a mixture of surfactants, in aqueous solution.

The injection point of the mobility reducing agent may be in the vicinity of the roof of the tank. Sensors can detect the embrittlement of the cracking cover to trigger the injection of a determined volume of CO2 mobility reducing agent.

One can inject said agent through a well opening to the tank roof, through a tube down into said well.

The present invention will be better understood and its advantages will appear more clearly on reading the following embodiment, in no way limiting, and illustrated by the appended figures, of which: FIG. 1 schematically shows a fractured reservoir according to FIG. In the prior art, FIG. 2 shows the implementation of the present invention in this same reservoir.

FIG. 1 schematically shows an aquifer reservoir 1 whose top 2 is located at a depth Z of the surface 3, in which CO2 has been stored on a height H.

As a first approximation, and for purposes of order of magnitude estimation, it is assumed that within the aquifer and the surrounding terrain there is a pressure close to the hydrostatic pressure, but this hypothesis restrictive condition.

While the cover 4 normally acts as a barrier to the fluid contained in the reservoir, it is assumed that an unforeseen event or evolution has led to a fracturing of the cover, in particular located at the roof of the tank 2. In the presence of the CO2 bubble of height H, the fluid overpressure exerted under the cover at the top of the tank is:

(p _w - P _g ) gH p _w and pg being the average densities of water and CO2 fluid (under background conditions) over the height H of the stored bubble.

For the example, we consider the practical case of storage at a moderate depth, but nevertheless sufficient to concern a saline aquifer (and not drinking water) and store CO2 in a sufficiently dense fluid form. The storage is at 1000m depth, corresponding to a temperature of the order of 50 ⁰ C (given the average geothermal gradient). Under normal hydrostatic pressure conditions (P = 100 bar), the densities of CO2 and water, p _w and p _g , are then of the order of 1.00 g / cm 3 and 0.43 g respectively. / cm3. The value of p _w considered here corresponds to a salinity of approximately 10 to 15 g / l of NaCl, which is likely to vary by only a few percent for most common salinities.

The overpressure exerted by the CO2 on the roof of the storage tank 100 m thick H, is then close to 6 bar in said hydrostatic conditions. The situation considered corresponds to the situation where an unforeseen and / or accidental event (overpressure in CO2 storage, tectonic movement, or other cause) leads to a fracturing of the cover. An example is where this fracturing occurs at the highest point of the reservoir, where the curvature of the terrain (in the case of a classical anticlinal structure) and the pressure is usually the highest. The occurrence of this event is supposed to be brought to the attention of the operators in charge of the conduct and monitoring of the storage, either implicitly (if the latter are at the origin of it by an accidental operation) or by means of monitoring of the tank (such as measurements with permanent sensors, for example tank pressure) assumed to be in place near the top of the storage.

Faced with this situation, the main object of the invention is to grant considerably increased time to the operators for the purpose of analyzing the situation and implementing remedies intended to protect against possible risks arising from the damage to the integrity of the cover.

The two situations involving or not implementing the invention are compared below in terms of the speed and the propagation time of the CO2 towards the outside of the storage tank.

To simplify the analysis of the problem, it is considered that a vertical crack, of hydraulic opening e, was initiated at the roof of the tank and spreads through the cover 4 and overlying terrain. The compressibility of the fluid is neglected and it is assumed that the flow in the crack is laminar. The velocity of a viscosity fluid μ _g circulating within this hydraulic connection is then expressed by means of Poiseuille's law:

AP where - is the average pressure gradient prevailing within the fluid in

flow over crack length L.

1) Previous case:

Reference is made to Figure 1 illustrating the progression of CO2 in the cover and overlying terrain. A formula is developed that gives the speed and time of CO2 migration in the crack. The pressure of the aqueous fluid saturating the ground is always assumed to be hydrostatic.

We consider average (fixed) values of the properties (densities and viscosities) of water and CO2 in the depth range studied, the density of CO2 being also assumed to be little different from that which exists within the tank. This approximation is justified as long as the point B, at a distance L from the roof, remains fairly close to the roof, that is to say L "Z. These hypotheses are introduced only to simplify the demonstration of the interest of the invention. The capillary pressure is neglected at the front of progression of the CO2 within the crack. The pressure drop within the tank due to the leak is also negligible compared to the pressure drop in the hydraulic path (assumption justified given the contrast of the flow section between the reservoir and the hydraulic path).

Given these approximations or hypotheses: the hydrostatic pressure at the front of progression of the CO2 reached at a vertical distance L (point noted B in FIGS. 1 and 2) of the roof of the tank (L remaining small in front of Z) is written:

P _B = P _w g (Z -L)

the pressure at the roof of the tank (noted point A in FIGS. 1 and 2) is:

_Λ P = p _w + gZ (ρ _w -p _g) gH

- and the difference in driving pressure of the flow of CO2 in the crack between A and B, has for expression:

^ P = P _A - P _B - P _g gL = (p _w -p _g ) g (L + H) from which we deduce the expression of the rate of progression of CO2 in the crack:

(NB: the speed, invariant along a path of section and length L fixed, however varies with the value of L, because the pressure gradient,

ΛP

-), decreases with L)

From this expression, we can calculate the CO2 travel time, t (L), to reach B from the roof (position A):

Considering a crack of hydraulic opening e = 10 μm, with intrinsic permeability equal to 8.4 Darcys, which propagated from the top of the reservoir towards the overlying ground, by taking again the storage conditions given above, and taking into account a viscosity of CO2 equal to 0.03 Cp at 100 bar and 50 ⁰ C, we deduce that the CO2 from a "bubble" of 100 m height travels 265 m in a day.

The CO2 velocity, as formulated above, is compared to that employed using the present invention.

2) Case according to the invention: (FIG. 2)

The storage bubble control devices having revealed the loss of integrity of the cover, in particular by fracturing, the procedure which is the subject of the present invention consists in injecting a volume ("stopper") of aqueous solution comprising a surfactant. specific to the top of the storage tank. In Figure 2, there is shown a conduit 5 installed in a well 6 drilled from the surface. All means known to those skilled in the art of drilling and production can be used for the establishment of this conduit. The particularity of this surfactant is that it is chosen to generate a foam in the presence of CO2. This foaming character results in a very strong increase in the apparent viscosity of the fluid. By means of foaming agents and foam stabilizers well chosen, the state of the art shows that the viscosity of the foam thus obtained can reach a value of the order of 100 to 1000 times that of the initial fluid lacking foaming agent. Such an increase in viscosity has been observed during experiments of flows performed in porous media of high permeability (greater than 1 Darcy), to which the cracks can be assimilated. This operation of placing a surfactant in the area where a leak is presumed to the roof is designed and sized to obtain a near-stop of the leak. To do this :

- A surfactant is selected to generate a foam in contact with gas. In practice, all the families of surfactants, whether they are anionic, nonionic or amphoteric, make it possible to obtain a foam with a low concentration of additive in solution, in particular of the order of 0.001 to 0.5%;

the surfactant is chosen as a function of its good stability under the conditions of the reservoir;

the surfactant can be injected in its initial commercial form, or more or less concentrated;

it can also be selected to have a low adsorption on the rocks in the presence, in order to limit the quantities of solution to be injected; the density of the injected surfactant solution is optimized so as to minimize the gravitational segregation phenomenon of the plug towards the base of the reservoir. However, given the practical constraints related to the aqueous nature of the injected solution and the density of the fluid in place (CO2 of density, at most, close to 0.6 g / cm ³ under reservoir conditions), tries to protect itself from this risk of segregation by integrating in the protocol of the storage operation the possibility, if necessary, to carry out periodic injections of the surfactant solution in order to ensure the presence of foaming agent at the potential points leak.

The foaming nature of the solution injected in the presence of CO2 is all the more asserted that the CO2 has, or tends towards, the properties of a gas. During a migration of CO2 from the top of the reservoir to the surface, the properties of the latter, stored in depth at a pressure and a temperature above the critical conditions (critical pressure = 74 bar and critical temperature = 31 ° C), actually evolve towards those of a gas, which amplifies the foaming behavior of CO2 in the presence of the weak amounts of surfactant solution that it causes in the hydraulic path of leakage.

Taking into account such thermodynamic properties specific to CO2, consider the case shown schematically in Figure 2, where the foam properties are only acquired at the point M located at the vertical distance L _g of the roof of the tank

(L _g <L <Z), and not from the leak point at the roof of the tank, the latter situation being optimal from the point of view of the slowing down of the leak.

According to this general situation, the pressure gradient linked to the flow is no longer uniform along the path AB but increases considerably on the segment MB in which the fluid flowing in the crack has a foam behavior. By neglecting the effects of compressibility (constant speed on the length L remaining small in front of Z), and treating the foam as a single fluid with increased apparent viscosity, the expression of the pressure losses on the segments AM (of length Lg) and MB (of length Lm = L-Lg), makes it possible to formulate the new velocity u _mod , of progression of CO2 (in the form of foam). As a first approximation, the density μ _g of the foam is assumed not to differ from that of the original CO2.

Otherwise :

AP AB = AP AM + AP MB = (p _w - P _g ) g ( ^L + H) From these 3 equalities, we deduce the expression of the velocity U _1n ^:

μ _g and μ _m being respectively the average viscosity of the initial fluid (CO2) and the apparent average viscosity of the CO2 foam.

The previous expression shows that the leakage rate through the crack, unchanged from the previous situation as long as L <L _g , slows down considerably as soon as the CO2 continues its progression under foam shape (ie when L> L _g ). The speed ratio at the distance L, L> L _g , in the presence and in the absence of foam is indeed:

u (L) μ _g L _g + μ _m L _m

again, by setting μ = - (apparent viscosity ratio (s) (therefore

mobility) of the original CO2 and CO2 in the form of foam), and by normalizing the path distance of CO2 in the form of foam (L γm = - L ¹ m ^{1 •} ):

"Mod with μ. 1 u (l -L _m ) + μ.L _n

In the optimal case where the foaming properties of the mixture (CO 2 + surfactant solution) are acquired as soon as the mixture enters the crack initiated at the top of the tank, that is to say in the case where the distance

L _g is zero, the rate of progression of CO2 is reduced in the ratio of the viscosities of the gas (CO2) and the foam, that is:

W mod u μ. As before, the transit time of CO2, t (L), is expressed as long as L <L _g , t (L) is expressed according to the formula given above; for L> L _g : t (L) = _tg (L _g ) + t _m (L _m ) with

and

UiU +. U,. _g - -μ (, L, _g +, H ττ)) L _r n (, L ^L _e ^g + ⁺ H + ⁺ L ^L _n ^m )]

(If L _g = 0, L = L _m , of course we find the expression (1) given above, with the viscosity μ _g replaced by the viscosity μ _m = μ.μ _g ) We finally get:

By repeating the application data used above, and assuming that the foam properties are acquired 100 meters above the roof of the tank (ie by pressure drop of the order of 10 bar), and that this foam increases the apparent viscosity of the gas by a factor of 500, the following travel times are obtained:

- CO2 in its original form takes 5 to 6 hours to reach 100 meters from the roof, from which side the viscous nature of the foam is assumed,

- from this coast, its progression is slowed down considerably and 71 days are then required to reach 100 meters higher, that is to say 800 meters deep. - after one year, the CO2 is still only 630 meters deep, whereas it would have reached the surface in 6 days in the absence of foam (assuming the hydraulic path established to the surface and neglecting variations of CO2 properties as a first approximation).

In the optimum case where, in the presence of the foaming agent solution, the CO2 behaves like a foam under the conditions of pressure and temperature of the tank, the CO2 leak takes place in the form of foam as soon as it penetrates into the crack and then the speed ratio ^Mmo is u (L)

constant and equal to = = 1/500 over its entire flight path. The leak times

M are then increased in the same ratio viscosities of the foam and the original fluid. In this situation, CO2 (in the form of foam) is found in 790 meters deep after a year and takes almost 8 years to reach the surface.

Thus, during the loss of integrity, suspected or proven, of a storage tank by fracturing, the procedure according to the invention of setting up a foaming agent at the top of the tank allows to slow down about two or even three orders of magnitude the speeds and migration times of CO2 in the hydraulic path created from the top of the tank. The present invention thus provides operators with considerably increased time to analyze the situation and take the necessary measures.

Claims

1) Method for controlling the integrity of a geological storage containing CO2, characterized in that the following steps are carried out:

2) Method according to claim 1, wherein the mobility agent is a foaming agent chosen to foam the CO2 that leaks through said blanket.

3) Method according to claim 2, wherein said foaming agent is a surfactant or a mixture of surfactants, in aqueous solution.

4) Method according to one of the preceding claims, wherein the injection point of the mobility reducing agent is in the vicinity of the roof of the tank.

5) Method according to one of the preceding claims, wherein sensors detect the embrittlement of the cracking cover to trigger the injection of a determined volume of CO2 mobility reducing agent.

6) Method according to one of the preceding claims, wherein said agent is injected through a well opening to the roof of the tank.