Hydraulic fracturing includes injecting a fluid at high volume or high pressure or both to facilitate flow in a fracture in a subterranean formation. Fluid is mixed with propping agent (proppant) that keeps a fracture open when injection is stopped and fluid leaks off into formation or flows back on the surface. By this means a homogeneous proppant pack is created. Hereinafter this type of the treatment is referred as conventional fracturing treatment.
Heterogeneous proppant placement (HPP) is an additional approach for hydraulic fracturing. Heterogeneous proppant packs, also referred to as proppant pillars, prevent fracture closure and provide highly conductive channels around pillars serving as flow path for hydrocarbons.
Fracture conductivity is a parameter that relates to a wellbore's production rate. For a fracture with a conventional fracturing treatment, conductivity is directly related to the conductivity of the proppant pack and thus to permeability of the proppant used for the treatment.
Herein, we discuss methods of materials selection to be used as a proppant for HPP fracturing treatments. We also discuss a procedure for designing a HPP treatment.
For a fracture with HPP, conductivity is determined by geometry of the channels created and does not depend on proppant permeability. This phenomenon changes requirements for the proppant in use. Also for HPP fracturing treatments it gives possibility to use materials as the proppant that are not suitable for conventional fracturing treatments.
Some embodiments pump slugs of a fluid with different rheological and/or other properties.
Conductivity of the HPP fracture is much higher than that of conventional one provided that the channels remain open. Conductivity of HPP fracture Cch in case of open channels will be referred further as “Infinite” and can be estimated by the following equation:
C
ch
=w
3/12 eq. 1
Here w is a minimum distance between fracture walls. As can be seen fracture conductivity does not depend on permeability of the proppant that is used for pillars creation. Hence, material used during pulsing steps can be qualified as a proppant for HPP hydraulic fracturing treatments regardless of proppant pack permeability.
Although Infinite conductivity of HPP fracture is independent of proppant pack permeability, it strongly depends on whether channels are open or not. The criterion of open-close channels is correlated with distance L between pillars inside the fracture. Channels are considered open if the length L does not exceed critical value Lcr. Critical length can be estimated from the following equation:
L
cr
=L
cr(Pcl,E, v, whyd, H, D, Hfr, Ap) eq. 2
Here closure stress (Pcl), Young modulus (E) and Poisson ratio (v) are formation geomechanical properties; hydraulic width (whyd) and fracture height (Hfr) are fracture geometrical properties; and pillar height (H), characteristic length of the pillar (D) and propping ratio (Ap) are pillar geometrical properties.
Formation geomechanical properties are inherent properties of the formation. Fracture and pillar geometrical properties can be to some extent controlled by the treatment design (pumping rate, fluid rheology, etc.). Pillar geometrical properties can be controlled by treatment design and selection of material used as proppant. With other parameters being fixed, the critical distance Lcr depends on pillar geometrical properties.
Pillar height H depends on proppant concentration in proppant pulses ω and pillars compaction (porosity of proppant pack, φp) under exerted stress. Characteristic length D in non-limiting example of a pillar round in cross section is the pillar diameter. Propping ratio Ap is defined as ratio between propped and unpropped surface area of a fracture face.
Propping ratio can be calculated as
A
p
=V
pr
/V
cl*Disp*Squash* eq. 3
Here Vpr is volume of proppant slug, Vcl is volume of clean slug, Disp is coefficient due to proppant slug dispersion or consolidation from pumping start to fracture closure start. Squash is coefficient due to pillar squashing during fracture closure. Erosion is a coefficient due to pillar erosion.
Non-limiting example of Disp, Squash and Erosion definition is following: Disp=V′pr/Vpr; Squash=SS/S and Erosion=SE/SS. Here V′pr is volume of proppant slug after dispersion; SS is the pillar area after squashing/spreading; SE is the pillar area after the “erosion time”, ET. The ET is as a period defined from the instant the pillar is exposed to the fluid flow until the time after which further erosion is negligible on the scale of the original area of pillar; This ET can only be introduced for a limited set of flow conditions. Typically, it can be introduced for low density/viscosity fluids in the low linear flow velocity ranges. For flow conditions out of this range, the propping ratio will be time dependent. S is a surface area of placed pillar exposed neither to squashing nor to erosion. It should be accounted that erosion usually occurs only after squashing, which explains form of Squash and Erosion coefficients.
Herein, we provide a workflow of measuring material properties critical for HPP treatments, selection of appropriate material, and creating treatment schedule for HPP treatments.
HPP treatment with a given material is to be designed in such a way that channels remain open after fracture closure. To achieve this, pillar geometrical properties H, D, Ap are to be determined. Pillar geometrical properties can be calculated provided that material critical parameters are known.
The workflow for the material parameters determination is provided below. The following assumptions have been taken into account in order to simplify the said workflow:
(asmp. i) The squashing/spreading caused by fracture closure is negligible for large enough pillars. Hence Squash=1.
(asmp. ii) Coefficient Disp does not depend on the type of material used as a proppant. If viscosity of carrier fluid is above 200 cp at 170 1/s and fibers are added to the fluid then we can assume that Disp=1.
1. Absolute density ρ of material should be measured.
2. Erosion rate of proppant should be measured at a testing fluid linear velocity, which can be derived from the produced fluid properties and its actual velocity in the fracture. An approach of measuring erosion rate of proppant pack is discussed in more detail below.
3. One should calculate Erosion coefficient based on erosion rate measured in 2.
4. Propping ratio Ap is calculated using eq. 3 and assumptions i and ii.
5. Effective stress on the proppant pack is calculated based on formation stress and propping ratio.
6. Porosity at effective stress (φp) calculated in 5 should be measured.
Following parameters are provided as result of the experimental procedures listed above: ρ, Erosion, φp. Knowing these values for a particular proppant in combination with formation geomechanical properties and fracture geometrical properties the geometrical properties of pillars can be calculated, and, thus, the treatment design can be elaborated.
Parameters ρ, Erosion, φp can be used as input parameters to fracture design software such as the software services including MANGROVE™ commercially available from Schlumberger Technology Corporation of Sugar Land, Tex. to calculate pillar geometrical properties D, H and Ap. Then, these properties can be used to create the HPP treatment design.
Some embodiments substitute proppant material in HPP treatment design that was validated by production results in a particular formation. To substitute propping agent currently used (proppant 1) with new propping agent (proppant 2) both proppants have to be tested according to the procedure listed above. Then, there are 2 options to obtain the treatment design: using reference proppant or direct modeling.
If reference proppant approach is used, then new HPP treatment design with use of proppant 2 is created in which all the treatment parameters remain the same except for proppant concentration. Proppant concentration is:
ω2=ω2(ρ1, ρ2, φ1p, φ2p, ω1) eq. 4
Here ω is concentration of propping material in the slug in PPA units, φp is proppant pack porosity (i.e. ratio of the void space to the total volume of the pillar) at a certain stress, ρ is absolute density of proppant, indexes 1 and 2 correspond to materials 1 and 2. One should note that φp is a function of effective stress (stress on the proppant pack), which in turn is function of propping ratio and the closure stress; by these means Disp, Squash and Erosion coefficients are implicitly included in eq. 4.
For a direct modeling approach, first the treatment design is created using proppant 1. Then, results of this design (examples include, but not limited to fracture conductivity, width, half-length and open channels flag) are used as design objectives to design treatment using proppant 2.
Example workflow for the proppant substitution using fracture design software for the wells where no erosion is expected during production is the following:
1. Samples of both proppant 1 and proppant 2 are sent to the lab;
2. Lab tests are performed to measure proppant porosity vs. stress, critical erosion velocity (at which pillar erosion starts), proppant absolute density, proppant mesh size distribution, proppant mean diameter, proppant bulk density;
3. Test results are provided to the field in the following format:
4. Field user prepares the treatment design in the fracture design software using proppant 1.
5. Field user opens the prepared design and enters abovementioned test results for both proppants into the fracture design software. All other proppant parameters as Non-Darcy flow coefficients, proppant friction coefficients, permeability vs. stress etc. are taken from the database supplied with the fracture design software;
6. Field user selects one of two options for proppant substitution:
7. As a result, the user obtains a fracture design using proppant 2 and also recommended maximum production/flowback rate for particular well. If the treatment will be pumped as per this design and the recommended maximum production rate will not be exceeded during the well lifetime than the well production should not be compromised by the proppant substitution.
Non-ISO proppants. Examples include river sand; soil excavated close to the well site; Frac-sands and ceramic proppant that do not meet ISO specification (such as standards ISO 13503-2 and ISO 13503-5). Using of non-ISO material for conventional fracturing will result in low conductivity of the fracture. For HPP fracturing treatment given material still can be used as proppant even if it does not meet ISO criteria in such parameters as permeability, grain size distribution, and resistance to crush. The material size has to be small enough to provide proppant admittance to the fracture.
Residuals of construction and other industrial processes. Examples are glass, set cement, bricks, marble, concrete, mortar, masonry, and drilling cuttings. Such solid materials after grinding can also be considered as proppant if can be pumped and form pillars of desired geometrical properties under bottomhole conditions. Some of these materials can be categorized as industrial byproduct.
Material with selected transport properties. Generally, the lower the mesh size and the lower the density of proppant material are the better transport properties that material has. In many cases, a requirement for high proppant permeability contradicts the requirements for good proppant transport. To achieve required proppant pack conductivity using of high mesh size (e.g. 20/40) proppant with high density (e.g. 3.2 g/cm3) is often required. Since conductivity of HPP fracture does not depend on proppant permeability, one can intentionally use material with relatively low mesh size (e.g. 40/70) and low density (e.g. 2.6 g/cm3). That will help to deliver proppant further towards the tip of the fracture achieving longer propped fracture half-length. Longer propped fracture half-length, in turn, will lead to increased production of the well treated.
Measuring Pack Settling
Conductivity of a HPP fracture is much higher than that of the conventional one while the flow passages for hydrocarbons (channels) remain open. This is so until the distance between pillars inside the fracture exceeds a critical length. Exceeding this length leads to channel closure and drastic drop of HPP fracture conductivity. Thus, change of the pillar/channel geometry due to the settling of proppant pillars during or after the treatment can affect fracture conductivity. Settling during treatment and before the fracture closes on proppant is different from settling after fracture closure. After fracture closure, there is a formation closure stress which compresses the pillar and creates friction force which acts against gravity force and can, therefore, prevent settling. In the case of HPP treated fractures, settling of proppant pillars in the formation might have a drastic impact on fracture conductivity and, therefore, on well production. Thus measurement of settling (especially under stress and at formation temperature) and characterization of its influence on well production is meaningful for HPP treated wells.
One technique to study settling of the proppant pillar uses a vertically aligned cell where a vertical dimension is high enough to allow proppant to settle under the force of gravity. Also, to apply stress to the pillar in such configuration, equipment which can apply force in the horizontal direction is needed.
Herein we discuss a technique of measurement of proppant settling which can be applied to heterogeneous proppant packs. Embodiments include modeling the force of gravity by another force acting in the horizontal direction which will allow using a horizontally aligned cell and equipment applying force in a vertical direction which is more elegant.
Embodiments test proppant/proppant pack sedimentation in horizontal geometry, instead of vertical geometry of the real fracture.
This is influenced by the following factors:
The main principles of this approach include the following:
Experimental Considerations.
The drag force of the fluid is used to model gravity. Testing methodology is analogous to characterizing erosion rate of proppant pack, described in more detail below. However, several points have to be taken into account:
Test can be performed both in “pillar” and “channel” configuration.
The goal of the experiment is not to measure sample erosion, but to test its sedimentation. Flow rate, sample and cell dimensions have to be adjusted to correctly simulate forces, acting on the proppant pack in vertical geometry. Drag force is increasing with linear velocity of the fluid and it is possible to find the velocity at which drag force acting on a particle is equal to gravity force acting on the same particle (velocity V in
If the sample erosion occurs (and corresponding changes in the sample and cell dimensions), then the flow rate has to be increased respectively to ensure that the drag force of the fluid is matching the forces acting on the proppant pack in case of vertical geometry.
Experimental conditions should be selected to be as close as possible to the conditions of a real fracture and real proppant pack. This might include but not limited to:
Results Implementation:
Sedimentation impact. If model force (e.g. drag force) is set to match forces acting on the pack in vertical geometry it can be determined, whether sedimentation is an issue or not. Indicator of the fact that sedimentation has no impact on proppant is no continuous pillar destruction at constant model force. It should be noted, that the outer rim border of the proppant pack is formed with unconsolidated (free) proppant, which will be washed out at minimal force applied (e.g. minimal flow rates if drag force is used).
Threshold stress on the proppant pack to prevent sedimentation. The operation mentioned above can be repeated for different stresses on the pack to estimate minimal stress at which sedimentation is not an issue. The minimal stress value is important to design a fluid in order to maintain viscosity for a sufficient period of time. Basically, sedimentation has to be prevented by fluid viscosity until the stress on the pact increases higher than threshold stress. Threshold not necessary corresponds to the formation closure stress, and proppant pack stability might be reached earlier.
Proppant grain size. The impact of gravity is proportional to the linear dimension of the proppant grain in magnitude of three (
Compaction Test
The overall performance of an HPP treated fracture directly depends on geometry of the proppant pack created: diameter and height of the pillars, and distance between them beside the rock properties. It is assumed that diameter and distance between pillars is determined by volumes of proppant and clean pulses during the treatments. The height of proppant pack depends on the concentration and physical properties of the proppant pumped. The set of proppant properties affecting proppant pack height includes but not limited to: crush resistance, grain size distribution, roundness and sphericity, amount of non-ceramic (non-silica) content, proppant grain porosity, mechanical properties of the propping material (Poisson ratio, Young's modulus), crystallinity and uniformity of each grain. Most importantly the pack height is dependent on the stress too. Combining these properties in the analytical approach seems to be non-trivial and tedious task. That is why a simple, reliable and efficient procedure for measuring proppant pack height is required.
Proppant pack height is registered in ISO conductivity measurements (ISO 13503-5). The compaction test procedure discussed herein is related to ISO crush test (ISO 13503-2). The ISO crush test itself (without modifications suggested below) cannot provide data on proppant pack height.
In an HPP treated fracture, the proppant pack is usually exposed to the effective stress greater than that for homogeneous proppant pack. Under such high stress, proppant crushing will most likely occur and the height of HPP proppant pack will be reduced. Reduction of the pack height will lead to a decrease of channels cross-section (or to the channel closure in the worst case) and will compromise well production (or even reduce well production rate to that of untreated formation in the worst case).
Embodiments herein measure a dimensionless quantity of the proppant pack height, which can be scaled up to proppant pack of realistic size. Scaling up can be done for both homogeneous and heterogeneous proppant. However, to scale up experimental results to a realistic heterogeneous proppant pack, the pack dimensions should be large enough (it is suggested, that procedure can be scaled up to a proppant pillar which not less than 1 meter in diameter).
Implementation of such measurement includes but is not limited to: comparison of proppant performance; modeling implementation (general understanding of hydraulic fracturing and proppant pack behavior under formation stress); software implementation (more reliable and realistic job design; proppant substitution design)
One embodiment of the workflow for proppant pack height measurement procedure is as follows:
1. Equipment
2. Chemicals
3. Calibration procedure
4. Sample preparation
5. Testing procedure
6. Result processing
7. Data representation (examples)
Here Hinf is sample height at “infinite” pressure, which can be easily calculated by knowing sample mass (m), absolute density (ρ) and area occupied with sample (Ssample); Hp is sample height at certain pressure (data calculated in Section 6.2). In other Hinf words is height of sample fully composed of solids (i.e. φ for such sample is equal to zero). (Here the “infinite” only means that the sample is reorganized by such a way that there is no void space whatsoever among the grains; it is not implied that the sample absolute density is not dependent on the pressure, however, this dependence is neglected in this formula.) Equation 1 is deduced on the basis of equal surface occupied with sample for any stress (confined stress) and the change of cell area with pressure is neglected.
Here Hp is sample height at certain pressure (data calculated in Section 6.2), m is sample mass, Ssample is area occupied by sample, H0 is sample height when no stress is exerted on the sample, and ρb is sample bulk density. Equation 2 is deduced on the basis of equal surface occupied with sample for any stress (confined stress)
8. Data implementation
9. Alternatives and modifications
9.1.6.1. Characteristic stresses on the proppant pack have to be determined as described in Section 3.5
Characterizing Erosion Rate of Proppant Pack
Conductivity of the HPP fracture is higher than that of the conventional one because the flow passages for hydrocarbons (channels) remain open. This is so until the distance between pillars inside the fracture exceeds a critical length. Exceeding this length leads to channel closure and drop of HPP fracture conductivity. Thus, change of the pillar/channel geometry due to the erosion of proppant pillars during the flow-back and/or well production can affect fracture conductivity. In case of HPP treated fractures erosion of proppant pillars might have an impact on well production. Erosion of proppant pillars might lead to proppant back-production and drop of fracture conductivity. Thus, erosion measurement and characterization of its influence on well production is of importance for HPP treated wells.
Embodiments herein relate to proppant pack erosion measurement. The key point of the technique is to create scale independent fluid flow patterns, which allows us to collect experimental data, which would support modeling and simulating fluid flow through a channel or channeling network in the reservoir. This data can be used for: modeling input, proppant characterization and selection, well performance evaluation, or proof of reliable long-term performance of the well in some embodiments.
In some embodiments, proppant pack erosion measurements include the following: a proppant pillar placed in the testing environment as shown in
Alternative idea for test configuration (
As it can be seen in
To obtain representative and repeatable results, the test procedure may include but is not limited to one or multiple following stages: (i) forming channel with dissoluble material (e.g. sugar, salt etc.) to keep initial dimensions of proppant pack(s) similar from test to test; (ii) forming proppant pack with binding agent (“forming” additive) (e.g. glycerol, polymeric gel, sugar syrup etc.) for the same purpose as in (i), (iii) forming channel and proppant pack with die-bar for the same purpose as in (i); (iv) installation of shields which prevent direct hitting of proppant pack with the fluid. If any chemicals are used to form channel and/or proppant pack, a pre-test procedure should carried out to ensure that this chemicals were washed out prior to the erosion data collection; on the other hand if for the test purpose erosion properties of proppant pack have to be tested in presence of chemicals, this condition is irrelevant.
General testing procedure is the following (some variations might occur depending on techniques described in previous paragraph) but not limited to:
1. Sample of desired composition should be prepared
2. Sample should be placed in erosion cell and shaped (if required) to form open channel (one-side loading/multiple channels/angular loading)
3. Desired stress and fluid flow should be applied to the sample
4. Sample erosion should be measured
5. Fluid linear velocity should be calculated
Here S and m are sample area and mass at the moment when eroded solid were started being collected, Send is resulting sample area, mend is resulting sample mass and Δm is mass of washed out solids collected in the considered time period.
One (not limiting) of possible representation of such test results is plotting the eroded mass (g) normalized to fluid-to-pack contact area (cm2) and time (s). This normalized erosion rate (g/[s*cm2]) can then be plotted against the linear fluid velocity. Alternatively, the eroded mass is normalized with the contact area first and then this normalized value is plotted against the time at a constant linear velocity of the fluid. In the second step, it can be further normalized with the time by determining the slope of the curve. This latter introduced two-step procedure should lead to the results introduced initially. Results represented this way can be: (i) used as a specific proppant characteristic; (ii) scaled up regardless of system size; (iii) implemented to computation models and software; (iv) applied for job design of HPP treatment; (v) used for easy data presentation.
Extension of (iv): One of the erosion test results is the value of critical erosion velocity, below which no erosion occurs. This value can be used as input data to fracture design software to calculate the recommended maximum production/flowback rate for particular well to avoid pillar erosion.
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/RU2015/000130 | 3/3/2015 | WO | 00 |