FIELD OF THE INVENTION

The present invention relates to methods for predicting the asphaltene precipitation envelope and related parameters. In particular, the present invention relates to methods for predicting the asphaltene precipitation envelope and related parameters of a fluid from a subterranean formation.

BACKGROUND OF THE INVENTION

Hydrocarbon fluid production requires complex subsea and surface production systems which are designed to safely extract hydrocarbons from a hydrocarbon fluid producing reservoir. The fluid is typically extracted under extreme pressure and temperature conditions, particularly when it is being extracted from deepwater reservoirs.

The fluid which is extracted typically contains hydrocarbon solids such as wax, hydrates and asphaltenes. The deposition of these hydrocarbon solids in the production system can create significant disruption to overall operations. For instance, asphaltenes can deposit in any one or all of the well-bore, the manifold, flowlines/risers and topsides.

Asphaltene deposition is largely a composition and pressure driven phenomenon, with temperature playing a secondary role. In particular, under-saturated, high pressure reservoirs with a high gas to hydrocarbon fluid ratio tend to exhibit the highest risk of asphaltene deposition.

Under-saturated hydrocarbon fluid reservoirs are not fully saturated with dissolved gas. As the pressure of the hydrocarbon fluid is reduced on extraction, the gas remains in solution until the oil bubble point of the fluid is reached. As the pressure rises from reservoir pressure to the oil bubble point, dissolved gas components in the hydrocarbon fluid start to expand, resulting in a decrease in the fluid density and increased molar volume of the fluid.

The increasing molar volume results in a reduction in the solvent power (SP) and the solubility parameter (δ) of the hydrocarbon fluid. These are simple metrics used to define the ability of a fluid to dissolve asphaltenes. A higher solvent power and solubility parameter gives better dissolution of asphaltenes.

When the solvent power falls below the asphaltene critical solvent power of the hydrocarbon fluid (CSPa), or the solubility parameter falls below the onset solubility parameter of the hydrocarbon fluid (5onset), asphaltenes from the hydrocarbon fluid begin to precipitate. Increasing quantities of asphaltenes will precipitate out from the fluid with a greater difference between the solvent power and the asphaltene critical solvent power, or the solubility parameter and the onset solubility parameter.

Accordingly, it can be seen that the presence of dissolved gas in a fluid can act as an asphaltene precipitant.

The upper asphaltene onset pressure is the pressure above the oil bubble point at which asphaltenes start to precipitate from the hydrocarbon fluid. The lower asphaltene onset pressure is the pressure below the oil bubble point at which asphaltenes stop precipitating from the hydrocarbon fluid. As the pressure falls during hydrocarbon fluid extraction, asphaltene precipitation starts at the upper asphaltene onset pressure and occurs until the lower asphaltene onset pressure is reached.

It will be appreciated that methods which predict the asphaltene precipitation envelope may be very useful when it comes to designing or operating hydrocarbon fluid extraction systems. The asphaltene precipitation envelope may be used to assess the asphaltene deposition risk. For instance, knowledge of the asphaltene precipitation envelope enables locations to be identified which may be prone to asphaltene instability and thus help devise suitable mitigation and/or remediation strategies.

The ASIST (ASphaltene Instability Trend) method is widely known, and is based on the fundamental assumption that the solubility parameter and the refractive index of non- polar substances such as crude oils are linearly related. However, predictions of the asphaltene onset pressures that are made using the ASIST method generally do not match with the measured asphaltene onset pressures of fluids. The ASIST method is described by Wang et al. : An Experimental Approach to Prediction of Asphaltene Flocculation (SPE 64994, 2001).

Asphaltene onset pressure can also be measured on live fluids by depressurization experiments performed on live fluids in a Solids Detection System (SDS) apparatus.

However, this process is expensive and laborious, requiring both specialist equipment and live downhole fluid samples.

Accordingly, there is a need for a method which reliably predicts the asphaltene precipitation envelope of a fluid. SUMMARY OF THE INVENTION

The present invention provides a method for determining a solubility parameter of a stock tank oil, 6STO, at one or more pressures, said method comprising:

determining a correction factor, F _correction , according to formula (1):

Fcorrection ⁼ 5sTO(physical) / 5sTO(solvent power) (1)

wherein: 6sTO(physkai) is an estimate of the solubility parameter of the stock tank oil based on a physical property of the stock tank oil, and

5sTO(soivent power) is an estimate of the solubility parameter of the stock tank oil based on the solvent power of the stock tank oil,

applying the correction factor, F _cor rection, to an estimate of the solubility parameter of the stock tank oil based on a physical property of the stock tank oil, 6sTO(estimated), at one or more pressures, according to formula (2):

5sTO ⁼ 5sTO(estimated) / F _cor rection (2).

The present invention further provides a method for estimating a solubility parameter of a fluid consisting of stock tank oil and dissolved gas, 6fi„id, at one or more pressures, said method comprising calculating 6fi„id according to formula (3):

5fluid ⁼ V(fraction DG) * &T>G + V(f _rac tion STO) * 5sTO (3)

wherein: V(f _rac tion DG) is a volume fraction of the dissolved gas,

6DG is a solubility parameter of the dissolved gas,

V(fraction STO) is a volume fraction of the stock tank oil, and

6STO is a solubility parameter of the stock tank oil,

wherein 6STO is determined according to a method as defined herein.

The present invention also provides a method for predicting an onset solubility parameter of a fluid consisting of stock tank oil and dissolved gas, 6 ₀ nset(fiuid), at one or more pressures, said method comprising:

titrating the stock tank oil against two or more titrants to determine, for each titrant, a volume fraction of the stock tank oil at the onset of asphaltene precipitation, V( _0nS et fraction STO), a volume fraction of the titrant at the onset of asphaltene precipitation, » (onset fraction T)? and a root molar volume of precipitants at the onset of asphaltene precipitation, V _P ^0'5 ₍ STO+T);

calculating an onset solubility parameter of the stock tank oil with each titrant,

6 ₀ nset(STo+T), according to formula (4):

8 ₀ nset(STO+T) ⁼ V( _on set fraction T) * δτ + V( _on set fraction STO) * &STO (4) wherein: δχ is a solubility parameter of the titrant, and

6STO is a solubility parameter of the stock tank oil;

determining a relationship between 6 _on set(STO+T) and V _P ° ^'5 (STO+T); and

predicting 6 _onS et(fluid) from a root molar volume of dissolved gas in the fluid, Vp ^0'5 (fl _u id), based on the relationship between 6 _on set(STO+T) and V _P ° ^'5 (STO+T),

wherein 6STO is determined according to a method as defined herein.

The present invention further provides a method for predicting an asphaltene precipitation envelope of a fluid consisting of stock tank oil and dissolved gas, said method comprising comparing a solubility parameter of the fluid, 6 _f i„id, and an onset solubility parameter of the fluid, 6 ₀ nset(fiuid), across a range of pressures, to predict pressures at which asphaltene precipitation will be observed, wherein:

a solubility parameter of the stock tank oil, 6STO, is used to determine δπ„ _ίΛ and

5onset(fiuid) across the range of pressures and is determined according to a method as defined herein.

Also provided is a method for mitigating the deposition of asphaltenes from a fluid consisting of a stock tank oil and dissolved gas in a fluid extraction process, said method comprising predicting the asphaltene precipitation envelope of the fluid using a method as defined herein, and modifying the fluid extraction process so that the deposition of asphaltenes is reduced.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. l depicts the asphaltene precipitation envelope (10) for a fluid;

FIG. 2 depicts a graph of 5onset(STO+T) against vp0.5(STO+T) for a fluid from the Gulf of Mexico (fluid A);

FIG. 3 depicts a graph of 5fluid and 5onset(fluid) across the range of pressures measured for fluid A; and

FIG. 4 depicts a graph comparing the direct measurement with the predicted measurement of the onset volumes of a series of commingled stock tank oils with each titrant.

DETAILED DESCRIPTION OF THE INVENTION

By applying the correction factor, F _cor rection, during the methods of the present invention, an improved prediction of the solubility parameter of the stock tank oil may be obtained which, in turn, leads to an improved predictions of the asphaltene precipitation envelope, the solubility parameter of a fluid and the onset solubility parameter of a fluid. In particular, the asphaltene precipitation envelope may be predicted with good agreement with the measured asphaltene precipitation envelope for a fluid. The methods of the present invention represent an improvement on known methods, such as the ASIST method described above.

Moreover, the methods of the present invention enable accurate prediction of the asphaltene precipitation envelope, and related parameters, from just a PVT report (i.e. Pressure- Volume -Temperature data) and a small sample of stock tank oil.

In some instances, the method for predicting the asphaltene precipitation envelope of a fluid comprises carrying out the abovementioned steps of the method for predicting the solubility parameter of a fluid, 6fi„id. In some instances, the method for predicting the asphaltene precipitation envelope of a fluid comprises carrying out the abovementioned steps of the method for predicting the onset solubility parameter of a fluid, 6 _onS et(fluid)- In some instances, the method for predicting the asphaltene precipitation envelope of a fluid comprises carryout the abovementioned steps of the method for predicting the solubility parameter of a fluid, 6fi„id, and the abovementioned steps of the method for predicting the onset solubility parameter of a fluid, 6 _onS et(fluid)- If, at a particular pressure, 6fi„id is lower than 6 ₀ nset(fiuid), then asphaltene precipitation is predicted to occur. If, at a particular pressure, 6 ₀ nset(fiuid) is lower than 6fi„id, then asphaltene precipitation is not predicted to occur. A graph of 6fi„id and 6 ₀ nset(fiuid) across the range of pressures measured may be plotted so that the asphaltene precipitation envelope (if present) may be visualized. The upper asphaltene onset pressure and the lower asphaltene onset pressure may be estimated, for instance from the graph.

In some instances, δηηΐα, 6 _onS et(fluid) and 6S _TO are determined over a range of pressures. For instance, 6fi„id, 6 ₀ nset(fiuid) and 6S _TO may be determined at two or more pressures, such as at 5 or more pressures, or at 10 or more pressures. The pressures may be in the range of from 2,000-140,000 kPa, such as from 3,500-45,000 kPa.

In some instances, δηηΐα, 6 _onS et(fluid) and 6S _TO are determined at reservoir temperature. For instance, 6fi„id, 6 ₀ nset(fiuid) and 6S _TO may be determined at a temperature in the range of from 30-200 °C, such as from 80-130 °C. In other instances, 6fi„id, 6 ₀ nset(fiuid) and 6S _TO may be determined across a range of temperatures, for instance at two or more temperatures, such as 5 or more temperatures. The temperatures may be in the range of from 30-200 °C.

The correction factor, F _correc tion

In order to calculate the correction factor, F _cor rection, it is necessary to determine

5sTO(physkai), an estimate of the solubility parameter based on physical parameters of the stock tank oil, and 6sTO(soivent power), a solubility parameter based on the solvent power of the stock tank oil.

5sTO(physkai) is an estimate of the solubility parameter of the stock tank oil based on a physical property of the stock tank oil. Suitable physical properties include the density of the stock tank oil and the refractive index of the stock tank oil.

In some instances, 6sTO(physkai) may be calculated according to formula (5):

6sTO(ph _yS kai) = 52.042 * ( RISTO ² - 1 ) / ( R + 2 ) + 2.904 (5) where: RIS _TO is the refractive index of the stock tank oil.

The refractive index of the stock tank oil, RIS _TO , ^ma y be measured experimentally using known methods. For instance, RIS _TO may be measured according to ASTM D 1747- 09. RIS _TO may be measured at temperatures falling within the range of from 15-90 °C, such as from 20-60 °C, and at atmospheric pressure, i.e. 100 kPa.

In other instances, 6sTO(physkai) may be calculated according to formula (6):

OSTO(physical) ^— 17.347 * psTo + 2.904 (6)

where: PS _TO is the density of the stock tank oil. The density of the stock tank oil, ps _T o, may be measured experimentally using known methods. For instance, ps _T o may be measured according to ASTM D 4052 or D 5002. PS _TO will typically be measured at room temperature and atmospheric pressure, i.e. 20 °C and 100 kPa, though it may be measured at temperatures of up to 200 °C and pressures of up to 140,000 kPa using a high pressure -high temperature densitometer, such as an Anton-Paar device.

It will be appreciated that formulae (5) and (6) are essentially equivalent, since it is known that the density of hydrocarbons is typically three times ( RISTO ^{2 ~} 1 ) / ( RI _STO ^{2 +} ^ ). Further explanation in this regard can be found in Zuo, J. Y. et al: A Simple Relation between Solubility Parameters and Densities for Live Reservoir Fluids (J. Chem. Eng.

Data, 55 (2010) 2964-2969) and Vargas, F. M. et al: Application of the One-Third rule in hydrocarbon and crude oil systems (Fluid Phase Equilibria, 290 (2010) 103-108), the disclosures of which are incorporated herein by reference.

5sTO(soivent power) is an estimate of the solubility parameter based on the solvent power of the stock tank oil. Any known method may be used to determine the solvent power of the stock tank oil. For instance, the methodology described in Patent US 2004/0121472 (Nemana, S. et al: Predictive Crude Oil Compatibility Model; incorporated herein by reference) may be used, according to which oil solvent power is estimated using the Watson K factor.

The Watson K factor, KS _TO , is calculated according to formula (7):

KSTO = VABPSTO ^{1 3} SGSTO (7)

where: VABPS _TO is the volume average boiling point of the stock tank oil, in degrees

Rankine, and

SGS _TO is the standard specific gravity of the stock tank oil.

The volume average boiling point of the stock tank oil, VABPs _TO , may be determined using known methods. In some instances, VABPS _TO may be determined from the yield profile of the stock tank oil.

The yield profile of the stock tank oil may be determined from physical distillation, for instance according to ASTM D 2892 or ASTM D 5236. The yield profile of the stock tank oil may alternatively be determined using GC and high temperature simulated distillation (HT-SIMDIS). Use of GC analysis allows the hydrocarbon composition of the oil to be determined for components boiling in the Ci_ ₉ hydrocarbon range. GC analysis may be carried according to standard test method IP PM-DL. HT-SIMDIS analysis may be carried out according to standard test method IP 545.

The standard specific gravity of the stock tank oil, SGS _TO , is the ratio of the density of the stock tank oil to that of water at 60 °F (i.e. 15.6 °C). SGS _TO ^ma y be determined using known methods. For instance, as mentioned above, the density of the stock tank oil may be measured experimentally according to ASTM D 4052 or D 5002. The density of the stock tank oil may also be determined from the yield profile of the oil, for instance using a simulation tool (such as HYSYS) which may predict the density of the stock tank oil at 60 °F.

The solvent power of the stock tank oil, SPS _TO , may be determined from the Watson

K factor using linear interpolation. For instance, SPS _TO may be determined from KS _TO based on the relationship between the Watson K factor and the solubility parameter of heptane and toluene. The Watson K factor and the solubility parameter of heptane and toluene are known in the art.

The solubility parameter of the stock tank oil based on the solvent power of the stock tank oil, 6sTO(soivent power), may be determined from the solvent power of the stock tank oil, SPS _TO , also using linear interpolation. For instance, 6sTO(soivent power) may be determined from SPS _TO based on the relationship between the solvent powers and solubility parameters of heptane and toluene. The solvent powers and solubility parameters of heptane and toluene are known in the art.

Fcorrection is a coefficient which is assumed to be substantially independent of pressure and temperature. Accordingly, a similar value is assumed to be obtained, regardless of the pressure or temperature at which F _CO rrection is determined.

Fcorrection may be determined at a single pressure. In other instances, for greater accuracy, F _CO rrection may be determined at more than one pressure. F _CO rrection may be obtained at a single pressure such as at atmospheric pressure, i.e. 100 kPa, or at one or more pressures up to 140,000 kPa.

Where Fcorrection is determined at one or more pressures, it is calculated as the mean average of values determined at more than one pressure. For instance, where 6sTO(physkai) and 6sTO(soivent power) are determined at a first pressure to an nth pressures, F _CO rrection = [

5sTO(physical at PI) / 5sTO(measured at PI) + 5sTO(physical at P2) / 5sTO(measured at P2) + · · · 5sTO(physical at Pn)

/ 6sTO(measured at Pn) ] / n, where PI is the first pressure, P2 is the second pressure and Pn is the nth pressure, n is preferably from 2-5. Generally, determination of 6sTO(pnysicai) and 5sTO(soivent power) at a single pressure provides a correction factor, F _cor rection, with sufficient accuracy for use in the methods of the present invention.

Fcorrection may be determined at a single temperature. In other instances, F _correction may be determined at more than one temperature. F _cor rection may be obtained at room temperature, i.e. 20 °C, or at one or more temperatures up to 200 °C. Where F _cor rection is the mean average of values determined at more than one temperature, it will be understood that, as with pressure, each value is determined at a single temperature.

Accordingly, the skilled person will appreciate that although the value obtained for Fcorrection is assumed to be substantially independent of pressure and temperature, when calculating F _corr ection, 6 _S TO(ph _yS icai) and 6 _S TO(soivent power) should be determined at the same temperature and pressure.

Since the value obtained for F _correction is substantially independent of pressure and temperature, it may be applied across the range of pressures or temperatures at which 6fi„id and 6 ₀ nset(fiuid) are estimated, irrespective of the one or more pressures and temperatures at which 6sTO(physkai) and 6sTO(soivent power) are determined. The skilled person will appreciate that the temperature and pressure should be kept consistent for parameters described herein other than F _cor rection, i.e. only those parameters obtained at the same pressure and temperature should be combined.

The solubility parameter of the stock tank oil, 5STO

The solubility parameter of the stock tank oil, 6S _TO , is calculated from F _correction , the correction factor, and 6sTO(estimated), the estimate of the solubility parameter of the stock tank oil based on a physical property of the stock tank oil. Suitable physical properties include the density of the stock tank oil and the refractive index of the stock tank oil. For instance, 6sTO(estimated) may be calculated according to formula (8):

OSTO(estimated) ^— 17.347 * PSTO + .904 (8)

where: PS _TO is the density of the stock tank oil.

At some pressures, i.e. those close to atmospheric pressure, the density of the stock tank oil, PS _TO , may simply be measured using the methods mentioned above.

However, across a wider pressure range, such as that typically encountered in reservoirs, PS _TO at one or more pressures may be predicted by determining the yield profile of the stock tank oil, and using the yield profile to predict the density of the stock tank oil. The yield profile of the stock tank oil may be analysed using GC and HT-SIMDIS. GC and HT-SIMDIS analysis may be carried according to standard test methods mentioned above, i.e. standard test methods IP PM-DL and IP 545, respectively.

A simulation tool (such as HYSYS) may be used to predict the density of the stock tank oil, ps _T o, at a wide range of pressures and temperatures. Typically, the simulation tool will slice the yield profile into groups of components with similar boiling points, which then enables the prediction of the stock tank oil density at a wide range of pressures and temperatures. It will be appreciated, that by using GC, HT-SIMDIS and HYSYS, the density of the stock tank oil may be estimated, and so it does not need to be measured at high temperature and high pressure.

Alternatively, 6sTO(estimated) may be calculated according to formula (9):

6sTo _(est ima _te d) = 52.042 * ( RISTO ² - 1 )/( R + 2 ) + 2.904 (9) where: RIS _TO is the refractive index of the stock tank oil.

The refractive index of the stock tank oil, RIS _TO , may be measured experimentally using known methods. For instance, RIS _TO may be measured as outlined above.

Since it is desirable to assess 6sTO(estimated) across a wide range of pressures, as found in a reservoir, then 6sTO(estimated) will typically calculated based on ps _T o- The solubility parameter of the fluid, 5 _u id

As mentioned above, the solubility parameter of the fluid, 6fi„id, may be calculated according to formula (3):

5fluid ⁼ V(f _r action DG) * δοθ + V(f _rac tion STO) * 5sTO (3)

where: V(f _rac tion _D G) is a volume fraction of the dissolved gas,

6 _D G is a solubility parameter of the dissolved gas,

(fraction S _TO ) is a volume fraction of the stock tank oil, and 6S _TO is a solubility parameter of the stock tank oil.

The solubility parameter of the dissolved gas, 6 _D G, may be estimated based on a physical property of the dissolved gas. Physical properties include the density of the dissolved gas. For instance, 6 _D G may be calculated according to formula (10):

6 _DG = 17.347 * p _DG + 2.904 (10)

where: P _D G is the density of the dissolved gas. The density of the dissolved gas, poo, at one or more pressures may be determined from the composition of the dissolved gas in the fluid. In some instances, the dissolved gas is represented by the Ci_ ₆ paraffin components of the fluid.

The composition of the dissolved gas may be determined by known methods. For instance, the composition of the dissolved gas may be derivable from PVT data, such as single stage flash data.

At pressures lower than 3,500 kPa, the composition of the dissolved gas may change, due to evaporation of the heavier components, such as the C ₄ _ ₆ paraffin components. To determine the composition of the dissolved gas in the live fluid at pressures of less than 3,500 kPa, then a simulator tool may be used, such as MultiFlash or PVTSim.

The density of the dissolved gas, poo, at different pressures may be determined from the composition of the dissolved gas using equations of state, such as the Peng-Robinson or Soave-Redlich-Kwong equations of state.

Alternatively, the density of the dissolved gas may be determined by direct measurement of the fluid, e.g. at temperatures up to 200 °C and pressures up to 140,000 kPa using a high pressure -high temperature densitometer, such as an Anton-Paar device. However, it is preferred that the density of the dissolved gas be determined from the PVT data.

There is no need to apply a correction factor when determining the solubility parameter of the dissolved gas, 6DG-

Methods for measuring the solubility parameter of the stock tank oil, 6STO _? are given above.

The volume fraction of the dissolved gas, V(f _rac tion DG), and the volume fraction of the stock tank oil, V(f _rac tion STO), may be measured using any known method. In some instances, V(f _rac tion DG) will be determined, and V(f _rac tion STO) will be determined according to the relationship V ₍ f _ract ion STO) = 1 - V ₍ f _ract ion DG).

In some instances, V(f _rac tion DG) and V(f _rac tion STO) may be derived from PVT data on the fluid. In particular, V(f _rac tion DG) and V(f _rac tion STO) may be derived at one or more pressures from the differential liberation residual oil density, the gas to oil ratio, the density of the stock tank oil, psTo, and the density of the dissolved gas, PDG- Methods for measuring the density of the stock tank oil and the dissolved gas are provided above. In these instances, V(f _rac tion STO) is calculated assuming that the overall "shrinkage" of the mixture when the dissolved gas and stock tank oil are combined is fully absorbed by the gas phase. This is a reasonable assumption given that the mass of dissolved gas in the live fluid is significantly lower than that of the stock tank oil.

The onset solubility parameter of the fluid, 5 ₀ n _Set (fluid)

As mentioned above , the onset solubility parameter of the fluid, 6 ₀ nset(fiuid), may be predicted, at one or more pressures, by titrating the stock tank oil against two or more titrants.

The titrants may be two or more different n-paraffins. In some instances, at least three different n-paraffins are used. In some instances, the titrants are selected from heptane, undecane and pentadecane.

The period of time for which the stock tank oil and the titrant are equilibrated may be from 20-40 minutes, such as 30 minutes. These equilibration times improve the quality of the data which is obtained, due to minimized heating times and improved test turnaround times. In some instances, the stock tank oil and the titrant are undisturbed during this time, i.e. they are not subjected to any mixing or agitation. Aliquots of stock tank oil and titrant may be prepared so that the precipitation onset volume may be determined to a precision of at least 5 % by volume, such as at least 2 % by volume. The stock tank oil and titrant mixtures may be observed under an optical microscope to determine when asphaltene precipitation occurs.

Accordingly, it can be seen that the volume fraction of the stock tank oil at the onset of asphaltene precipitation, V( _0nS et fraction STO), and a volume fraction of the titrant at the onset of asphaltene precipitation, V( _0nS et fraction τ) are determined from the titrations. In some instances, V( _0nS et fraction τ) will be measured, and V( _0nS et fraction STO) will be determined based on the relationship as V _(onset fraction STO) = 1 - V _(onset fraction i

The root partial molar volume of precipitants at the onset of asphaltene precipitation, Vp° ^'5 (STo+T), may be determined using known methods. For instance, V _P ° ^'5 (STO+T) may be determined using a simulation tool, such as HYSYS, and equations of state, such as the Peng-Robinson equations of state.

The onset solubility parameter of the stock tank oil with each titrant, 6 _on set(STO+T), is calculated according to formula (4):

8 ₀ nset(STO+T) ⁼ V( _on set fraction T) * δτ + V( _on set fraction STO) * §SIO (4). The solubility parameter of the titrant, δχ, may be determined at one or more pressures experimentally, or may be known in the art. Where δχ is determined

experimentally, it may be determined based on the density or the refractive index of the titrant. For instance, δτ may be calculated according to formula (1 1):

δ _τ = 17.347 * p _T + 2.904 (11)

where: ρτ is the density of the titrant.

Densities of titrant are known in the art, or may be determined using standard methods.

Alternatively, δχ may be calculated according to formula (12):

δ _τ = 52.042 * ( RI _T ² - 1 )/( RI _T ² + 2 ) + 2.904 (12) where: RI _T is the refractive index of the titrant.

The refractive index of the titrant, RI _T , may be known in the art, or may be determined experimentally using standard methods.

0S _TO is the solubility parameter of the stock tank oil and is determined as described above, USing Fcorrection-

In some instances, the method for predicting the onset solubility parameter of the fluid, 0onset(fiuid), is carried out at a temperature which is close to that of the reservoir temperature.

However, in most instances, it will be necessary to modify the method so that the findings can be extrapolated to reservoir temperature. In these instances, the method involves titrating the stock tank oil against two or more titrants at two or more

temperatures with each titrant. In other words, at least four separate titrations are performed (two titrants, at two temperatures each).

The test temperatures should be above the Wax Appearance Temperature (WAT) of the titrant. Typically, titrations may be carried out with each titrant at three temperatures. In some instances, the temperatures are selected from 40, 50 and 60°C.

Determination of 0 _on set(STO+T) and V _P ° ^'5 (STO+T) at two or more temperatures enables, by extrapolation, 0 _on set(STO+T) and V _P ° ^'5 (STO+T) to be determined at reservoir temperature. The relationships between 0 _onS et(S _TO + _T ) and temperature, and between 0 _onS et(S _TO + _T ) and

Vp° ^'5 (STo+ _T ), are assumed to be linear. As mentioned above, reservoir temperature typically falls within the range of from 30-200 °C, such as from 80- 130 °C. Once 6onset(STO+T) and V _P ° ^'5 (STO+T) are known for two or more titrants, for instance at reservoir temperature, a relationship between 6 _on set(STO+T) and V _P ° ^'5 (STO+T) may be determined. As mentioned, the relationship is assumed to be a linear relationship. In some instances, it may be desirable to plot a graph of 6 _onS et(STO+T) against V _P ° ^'5 (STO+T though the relationship can also be determined without the need to plot a graph.

The onset solubility parameter of the fluid, 6 ₀ nset(fiuid), may then be predicted from the root partial molar volume of dissolved gas in the fluid, v _p ° ^'5 _(f n,id), based on the relationship between 6 _onS et(STo+T) and V _P ° ^'5 (STO+T)- This is because the relationship between 6 _onS et(STo+T) and v _p ° ^'5 (STo+T) is assumed to be the same as the relationship between 6 ₀ nset(fiuid) and

V _p ° ^'5 (fluid).

The root partial molar volume of dissolved gas in the fluid, v _p ° ^'5 (fn,id), may be derived from PVT data on the fluid. In particular, v _p ° ^'5 (fl _u i _d) may be derived at one or more pressures from the differential liberation residual oil density, the gas to oil ratio and the density of the stock tank oil, ps _T o- The fluid

The fluid referred to herein is typically a downhole fluid, such as a hydrocarbon fluid which is present in a subterranean formation (commonly referred to as a live fluid). The fluid will typically be extracted from the subterranean formation as crude oil.

The fluid consists of stock tank oil and dissolved gas. Accordingly, removal of the dissolved gas from the fluid gives oil which is considered, for the purposes of the present invention, to be stock tank oil. Stock tank oil may be obtained by bringing the fluid to atmospheric conditions, for instance of 20 °C and 100 kPa.

The stock tank oil is preferably free from any asphaltene inhibitors. The stock tank oil is preferably free from any dispersants. The stock tank oil is preferably free from drilling mud, and any other contaminants.

Typically, 400 cm ³ of stock tank oil will be suitable for carrying out the analysis required by the method of the present invention. The stock tank oil may be obtained from surface separators, or from down-hole fluid that has been depressurized and returned to ambient pressure.

Comingled systems

In some instances, the method of the present invention is used to predict the asphaltene precipitation envelope of a single fluid. In other instances, the fluid may be a comingled fluid which is formed from two or more separate fluids. Comingled fluids are common where an oil reservoir has multiple wells producing from different "sands". The properties, e.g. composition, density, asphaltene content and gas to oil ratio, of fluids from each producing sand may be very different, and asphaltene precipitation may vary between separate fluids. The mixing of two or more separate fluid streams may serve to increase, decrease or have no impact on the asphaltene precipitation for the commingled system.

Commingled fluids may be assessed by carrying out the methods outlined above on the comingled fluid, for instance by using PVT data for the comingled fluid (or predicting it using a tool such as PVTSim) and a stock tank oil sample from the comingled fluid.

In other instances, comingled fluids may be assessed by carrying out the methods outlined above on the separate fluids that combine to make the comingled fluid. The pressure and temperature at which the comingling occurs may be readily determined from operating data.

As before, the correction factor, F _cor rection, for a comingled fluid may be determined from 6sTO( _P hyskai) and 6 _S TO(soivent power). However, with a comingled fluid, 6sTO( _P h _y skai) and 5sTO(soivent power) are determined by % blending, such as volume % blending, for each of the separate fluids which form the comingled fluid.

Accordingly, where the comingled fluid is formed from n separate fluids, 6sTO(pnyskai)

= [ 6sTO(physkal of Fl) * Volume % Of Fl + 6 _S TO(physkal of F2) * Volume % Of F2 + ... 6 _S TO(physkal of Fn) * volume % of Fn ], where Fl is the first fluid, F2 is the second fluid and Fn is the nth fluid which forms the comingled fluid. Similarly, 6 _STO (solvent power) = [ 6 _STO (soivent power of FI)

* Volume % Of Fl + 6 _S TO(solvent power of F2) * Volume % Of F2 + ... 6 _S TO(solvent power of Fn) volume % of Fn ], where Fl is the first fluid, F2 is the second fluid and Fn is the nth fluid which forms the comingled fluid.

It will be appreciated that the volume % blending may be carried out at any appropriate stage during the calculation of F _cor rection. For instance, in the case of 6sTO(soivent power), % blending calculations may be carried out in order to determine the solvent power of the comingled fluid, from which 6STO (solvent power) may be determined directly without further % blending considerations.

The volume % blending of the separate fluids which form the comingled fluid may be determined using known methods. For instance, the volume % blending may be readily determined from operating data. As before, the solubility parameter of the stock tank oil, 6STO, for a comingled fluid is calculated from F _cor rection, the correction factor, and 6sTO(estimated), which may be calculated based on the density of the stock tank oil, ps _T o, for the comingled fluid.

PSTO may be determined for the comingled fluid using known methods. In some instances, PSTO may be determined by determining the yield profile for each of the separate fiuids which form the comingled fluid, and using a blend assay tool (such as CrudeSuite). The density of the comingled fluid can be predicted at a wide range of pressures and temperatures using a tool such as HYSYS.

As before, the solubility parameter of the comingled fluid, 6 _f i„id, may be calculated from V(fraction DG), δοο, V(f _rac tion STO), and 6STO- doomay be estimated based on the density of the dissolved gas, poo, in the comingled fluid. This may be determined % blending, such as volume % blending, the composition of the dissolved gas in each of the separate fluids which form the comingled fluid. The density of the dissolved gas, poo, in the comingled fluid at one or more different pressures may then be determined using equations of state.

(fraction DG) and V(f _rac tion STO) for the comingled fluid may be derived from the PVT data on each of the separate fluids which form the comingled fluid. An equations of state tool, such as PVTSim, may be used to determined V(f _rac tion DG) and V(f _rac tion STO) for the comingled fluid.

Methods for determining 6STO for a comingled fluid are discussed above.

As above, the onset solubility parameter of the comingled fluid, 6 _onS et(fluid), may be predicted from a root partial molar volume of dissolved gas in the comingled fluid, Vp° ^'5 (fluid), based on the relationship between 6 ₀ nset(STO+T) with V _P ^0'5 ( _S TO+T).

The root partial molar volume of dissolved gas in the fluid, v _p ^0'5 (fi„id), may be derived from PVT data on the separate fluids which form the comingled fluid.

Vp° ^'5 (STo+T) for the comingled fluid is simply a function of the titrants used during the experiment and do not vary when a commingled fluid is used.

As above, the onset solubility parameter of the comingled stock tank oil with each titrant, 6 _onse t(STO+T), may be calculated from V ₍₀ „ _set fraction τ), δ _τ , V ₍₀ „ _set fraction STO) and 6 _S TO- Methods for determining 6STO for a comingled stock tank oil are discussed above.

Methods for determining δχ are discussed above, and do not vary when a comingled stock tank oil is used. (onset fraction STO) may be determined for the comingled stock tank oil from V( _0nS et fraction T) for the comingled stock tank oil. Methods for determining V( _0nS et fraction τ) for the comingled stock tank oil are slightly more complicated, since it is not appropriate to merely use % blending of the values for the separate stock tank oils which form the comingled stock tank oil.

In some instances, V( _0nS et fraction τ) for the comingled stock tank oil with each titrant may be determined from the asphaltene critical solvent power for the comingled stock tank oil with each titrant, CSP( _b i _en d STO+T ₎ , for instance according to formula 13 :

» (onset fraction T) - ( 1 - ( CSP (blend STO+T) / SPblend STO ) * 100 (13) The solvent power of the comingled stock tank oil, SPbiend STO, is calculated by volume % blending of the solvent powers of the separate stock tank oils that form the comingled stock tank oil.

Where the comingled stock tank oil is made of up of n separate stock tank oils,

CSP (biend STO+T) for the comingled stock tank oil with each titrant may be determined from the critical solvent power of each of the separate stock tank oils with each titrant,

CSP (separate STO+T), according to formula (14): ¾ _p - ⁿ Asp contribution( _{SEPARATE S} TO) „ _¾P

(blend STO+T) ^_ ^ »r (separate STO+T) ( 14 )

^ n=STO 1 Asp content ₍ biend STO)

The asphaltene contribution from each of the separate stock tank oils, Asp

contribution( _sep arate STO), may be determined using known methods. For instance, the asphaltene content of each stock tank oil may be determined from the PVT data, or it may be determined by carrying out a crude oil assay on the stock tank oil. The asphaltene contribution may then be calculated by multiplying the asphaltene content by the weight % for each of the separate stock tank oils that form the comingled stock tank oil.

The asphaltene content of the comingled stock tank oil, Asp content(bi _en d STO), may be determined by summing the asphaltene contributions from each of the separate stock tank oils that form the comingled stock tank oil.

CSP (separate STo+T) for each of the separate stock tank oils may be determined according to formula (15):

CSP(separate STO+T) (100 - » (onset fraction T) ) * SPseparate STo / 100 (15)

SP separate STO is the solvent power of the separate stock tank oils, which may be determined based on the Watson K factor. (onset fraction T) may be determined experimentally by titrating the separate stock tank oils against titrant, as previously.

Methods for mitigating asphaltene precipitation

The asphaltene precipitation envelope may be used to identify locations in a system in which asphaltenes may precipitate. Accordingly, the method of the present invention enables mitigation and/or remediation strategies to be devised for areas which are prone to asphaltene precipitation.

In some instances, the present invention provides a method for mitigating the deposition of asphaltenes in a fluid extraction process, said fluid consisting of a stock tank oil and dissolved gas, said method comprising predicting the asphaltene precipitation envelope of the fluid using the methods described herein, and modifying the fluid extraction process so that the deposition of asphaltenes is reduced.

In some instances, asphaltene deposition may be reduced in at least one of the well- bore, the manifold, flowlines/risers and topsides. Deposition may be reduced by preventing asphaltene precipitation. For instance, pressure could be applied in the extraction system so as to maintain asphaltenes in their dissolved form. Alternatively, deposition may be reduced by modifying the system so that any precipitated asphaltene does not forms deposit. Deposition may also be reduced by modifying the comingling of fluids e.g. by modifying which separate fluids are comingled, the ratios in which the separate fluids are comingled, or the location at which separate fluids are comingled.

Figure 1 depicts the asphaltene precipitation envelope (10) for a fluid. The lower asphaltene onset pressure (12) is the pressure below the oil bubble point at which asphaltenes start to precipitate from the oil. The upper asphaltene onset pressure (14) is the pressure above the oil bubble point at which asphaltenes start to precipitate. Asphaltene precipitation starts at the lower asphaltene onset pressure and occurs up to the higher asphaltene onset pressure.

It can be seen from Figure 1 that asphaltene precipitation from the oil starts at when the pressure is reduced to around 5000 psia, i.e. the upper asphaltene onset pressure. At this point, the 6 _f i _u i _d and 6 _onS et(fluid) profiles first intersect. Precipitation continues until the pressure reached around 1000 psia, i.e. the lower asphaltene onset pressure.

EXAMPLES

Example 1 - determination of F _C n _rrPC tinn for a fluid from the Gulf of Mexico (fluid A) Fluid A is a down-hole fluid from an oil reservoir in the Gulf of Mexico. The fluid was assessed in order to determine the correction factor, F _cor rection.

Stock tank oil was obtained from fluid A. Basic measurements were performed on the stock tank oil, as shown in Table 1 :

Table 1: Physical properties of the stock tank oil of fluid A

An estimate of the solubility parameter based on the refractive index of the stock tank oil, 6sTO(pnysicai), was determined using formula (5):

OSTO(physical) ^— 52.042 * ( RISTO ² - 1 )/( ISTO ² + ) + 2.904 (5) At 20 °C, 6 _STO (phy _S kai) = 52.042 * ( 1.4904 ² - 1 )/( 1.4904 ² + 2 ) + 2.904 = 17.96 MPa ^{0 5}

An estimate of the solubility parameter based on the solvent power of the stock tank oil, 6sTO(soivent power), was determined using the Watson K factor. The solvent power of the stock tank oil, SPS _TO , was measured as 33. It is known that toluene has a solvent power of 51 and a solubility parameter of 18.2, and that heptane has a solvent power of 0 and a solubility parameter of 15.2. Using linear interpolation, 6sTO(soivent power) was determined at 20 °C to be 17.18 MPa ^{0 5} .

The correction factor, F _CO rrection, was determined according to formula (1):

Fcorrection ⁼ 5sTO(physical) / 5sTO(solvent power) (1)

At 20 °C, Fcorrection = 17.96 / 17.18 = 1.045

Example 2 - determination of 5 _u id for the fluid from the Gulf of Mexico

Fluid A was further assessed in order to determine the solubility parameter of the fluid, dfluid, across a range of pressures.

The solubility parameter of the stock tank oil, 6S _TO , was calculated across a range of pressures from F _CO rrection and 6sTO(estimated), the estimate of the solubility parameter of the stock tank oil based on a physical property of the stock tank oil, according to formula (2):

5sTO ⁼ 5sTO(estimated) / F _CO rrection (2). Since F _cor rection is independent of pressure, the value determined in Example 1 was applied across the range of pressures at which 6sTO(estimated) was determined.

5sTO(estimated) was calculated based on the density of the stock tank oil, psTo, according to formula (8):

OSTO(estimated) ^— 17.347 * PSTO + 2.904 (8)

The density of the stock tank oil, psTo, was predicted across a range of pressures from the yield profile of the stock tank oil, as analysed using GC and high temperature simulated distillation (HT-SIMDIS), using HYSYS.

The predicted values for the density of the stock tank oil, ps _T o, and the solubility parameter of the stock tank oil, 6STO, are given in Table 2:

Table 2: Predicted values for the density of the stock tank oil, psro, and the solubility parameter of the stock tank oil, SSTO

The solubility parameter of the dissolved gas, 6 _D G, was estimated based the density of the dissolved gas, poo, according to formula (10):

6DG = 17.347 * poo + 2.904 (10)

The density of the dissolved gas, PDG, was determined from the composition of the dissolved gas in the live fluid, with the dissolved gas taken to be the Ci_ ₆ paraffin components of the live fluid. The composition of the dissolved gas was derived from single stage flash data in a PVT report on fluid A, and is shown in Table 3 :

Table 3: Single stage flash data on fluid A (components are normalized to a total of 100%or use in calculations)

Analysis of the separator flash data indicates that the C1-C6 light ends composition is very stable, except at pressures of lower than about 200 psi. The density of the dissolved gas, PDG, at different pressures was determined using Peng-Robinson equations of state based on the composition shown in Table 3.

The predicted values for the density of the dissolved gas, poo, and the solubility parameter of the dissolved gas, 6DG, are given in Table 4:

Table 4: Predicted values for the density of the dissolved gas, poo, and the solubility parameter of the dissolved gas, SDG

The volume fraction of the dissolved gas, V(f _rac tion DG> and the volume fraction of the stock tank oil, V(f _rac tion STO), at different pressures were derived from PVT data on the live fluid, using the differential liberation residual oil density, the gas to oil ratio, the density of the stock tank oil, and the density of the dissolved gas. V(f _rac tion STO) was calculated assuming that the overall "shrinkage" of the mixture when the dissolved gas and stock tank oil are combined was fully absorbed by the gas phase. The derived values for V(f _rac tion DG) and V(f _rac tion STO) are given in Table 5 :

Table 5: Predicted values for the volume fraction of the dissolved gas, V (fraction OG), and volume fraction of the stock tank oil, V(f _ra ction STO)

Once V(f _rac tion DG), δοο, V ₍ f _rac tion STO) and 6 _S TO had been determined, 6 _a „ _id was calculated across a range of pressures according to formula (3):

6fluid = V(f _rac tion DG) * δοθ + V(f _rac tion STO) * 5sTO (3)

The predicted values for 6fi„id are given in Table 6:

Table 6: Predicted values for δ flui

Example 3 - determination of 5 _onS etffluid) for the fluid from the Gulf of Mexico

Fluid A was further assessed in order to determine the onset solubility parameter of the fluid, 6onset(fiuid), at one or more pressures.

Fluid A was titrated against each of three n-paraffin titrants: heptane (C7), undecane (CI 1) and pentadecane (C15) at three different temperatures: 40 °C, 50 °C and 60 °C.

The stock tank oil and the titrant were equilibrated for 30 minutes. The precipitation onset volume was determined to a precision of at least 2 % by volume. The stock tank oil and titrant mixtures were observed under an optical microscope to determine when asphaltene precipitation occurs.

From the titrations, the volume fraction of the stock tank oil at the onset of asphaltene precipitation, V( _0nS et fraction STO), the volume fraction of the titrant at the onset of asphaltene precipitation, V( _0nS et fraction τ), and the root molar volume of precipitants at the onset of asphaltene precipitation, V _P ° ^'5 (STO+T), were determined.

The solubility parameters of the titrants, δχ, were determined experimentally based on the refractive index of each titrant at each temperature. The solubility parameter of the stock tank oil, 6STO, was also determined experimentally based on the refractive index of the stock tank oil. The correction factor, F _cor rection, was applied according to formula (2) in the determination of 6STO- The measured solubility parameters of the titrants, δχ, and the solubility parameter of the stock tank oil, δ το, are shown in Table 7:

Table 7: Solubility parameters of the titrants, ST, and the solubility parameter of the stock

Once the volume fraction of the stock tank oil at the onset of asphaltene

precipitation, V( _0nS et fraction STO), the volume fraction of the titrant at the onset of asphaltene precipitation, V( _0nS et fraction Ί the solubility parameter of the stock tank oil, 6STO, and the solubility parameters of the titrants, δχ, had been determined, the solubility parameter of the stock tank oil and titrant, 6 _on set(STO+T), was determined according to formula (4):

8 ₀ nset(STO+T) ⁼ V( _on set fraction T) * δτ + V( _on set fraction STO) * §SIO (4).

The solubility parameter of the stock tank oil and the different titrants, 6 _on set(STO+T at the different temperatures is shown in Table 8:

Table 8: Solubility parameter of the stock tank oil and titrant, d _ONSET (STO+T), at different temperatures

As shown in Table 9, the solubility parameter of the stock tank oil and the different titrants, 6 _on set(STO+T), and the root molar volume of precipitants at the onset of asphaltene precipitation, V _p ° ^'5 (STO+T), were determined at a reservoir temperature of 93 °C by extrapolation: ύδΜΊ

6irC sre 93X 9TC

15.47 -1.33E-02 12.40 12 78 5.03

C11 I S .95 •1 20E-G2 14.77 15.02 5.S6

CIS 16.28 -1. 14E-02 16 36 17.07 5.90

Table 9: Predicted values of the solubility parameter of the stock tank oil and the different titrants, 0 _on set(STO+T), and the root molar volume of precipitants at the onset of asphaltene precipitation, v _p ^0'5 (STO+T), at reservoir temperature

Once 6onset(STO+T) and V _P ° ^'5 (STO+T) at reservoir temperature had been determined for the three titrants, a relationship between 6 _onS et(STO+T) and V _P ° ^'5 (STO+T) was determined. Figure 2 shows a graph of 6 _onS et(STO+T) against V _P ^0'5 ₍ STO+T)- It can be seen from the graph that, at reservoir temperature, the relationship between 6 _on set(STO+T) and V _P ° ^'5 (STO+T) is:

6o„ _SET( STO ₊ T) = 0.20269 * V _P °- ⁵ ₍ STO ₊ T) + 12.468.

The root partial molar volume of dissolved gas in the fluid, v _p ° ^'5 (fl _u id), was derived from across a range of pressures from the differential liberation residual oil density and the gas to oil ratio change. The onset solubility parameter of the fluid, 6 ₀ nset(fiuid), was then predicted from the root partial molar volume of dissolved gas in the fluid, v _p ^0'5 (fi„id), based on the relationship between 6 _on set(STO+T) with V _P ° ^'5 (STO+T)-

The predicted onset solubility parameter of the fluid, 6 ₀ nset(fiuid), is shown in Table 10:

Table 10: Prediction of onset solubility parameter of the fluid, S _onset (fluid)

Example 4 - determination of the asphaltene precipitation envelope for the fluid from the Gulf of Mexico

The asphaltene precipitation envelope of fluid A was predicted by comparing the solubility parameter of the fluid, 6fi„id, with the onset solubility parameter of the fluid, 5onset(fiuid), across a range of pressures. Figure 3 shows a graph of 6fl _u id and 6 _onS et(fluid) across the range of pressures measured. From the graph, the upper asphaltene onset may be estimated as approximately 6,500 psi and the lower asphaltene onset pressure may be estimated as approximately 2,750 psi at a reservoir temperature of 93°C.

From SDS experimentation on a high quality live fluid sample, Fluid A is known to have an asphaltene onset pressure of 6,500 psi. The ASIST method predicted that there was no asphaltene precipitation method. Accordingly, it can be seen that the methods of the present invention may be used to predict the asphaltene precipitation from live fluids with greater accuracy than the prior art ASIST method.

Example 5 - other fluids

The methods outlined in Examples 1-4 were carried out on a further four fluids: fluid B, fluid C, fluid D, and fluid E. The upper asphaltene onset pressure and lower onset pressure as predicted using the method of the present invention are shown in Table 11 , together with the upper asphaltene onset pressure as predicted using the ASIST method, and as measured directly from live oil (or estimated based on direct measurements on similar fluids):

Table 11: Upper and lower asphaltene onset pressure as predicted using the method of the present invention, as compared to upper asphaltene onset pressure as predicted using the ASIST method and as measured (or estimated*)

Accordingly, it can be seen that the method of the present invention provides a better estimate of the asphaltene precipitation behavior of a fluid than the prior art ASIST method.

Example 6 - comingled fluids

Validation of the described approach for comingled fluids was carried out by direct measurement of the onset volumes of a series of commingled stock tank oils and comparison of the predicted with the measured onset volumes with each titrant. The comparison is shown in graph form in Figure 4 for mixtures of fluids B and C, with nC7 used as the titrant. It can be seen that there is good agreement between the predicted and measured onset volumes.