MC² Market & Competitive Convergence

Parker Hannifin Corporation Process Filtration Division Lebanon, Indiana 46052 USA

ABSTRACT: Selecting the appropriate filter cartridge for a completion or well flooding application is a complex process involving the consideration of many factors. For oil field applications, understanding the filters ability to remove particles of a certain particle size distribution is a key consideration in narrowing down the filter choices. Particle removal efficiency is discussed in discussed in detail. Consideration is given to factors affecting results. correlation with field performance and differences in test methods.

KEYWORDS: Particle removal efficiency, beta ratio, single pass, multipass, filtration, well completion, well flooding.

The filter selection process involves the consideration of many factors and often several trial and error tests are necessary. Understanding a filter cartridge's ability to remove particles of a certain particle size distribution will minimize the complexity of the selection process for completion and well flooding projects. Because the filter's particle removal efficiency is determined by fluid and particle characteristics, and time, it is essential to clearly define the conditions of the test. Failure to include these conditions can result in the selection of filters that could prevent the well from producing at its full potential.

Test Objective

One of the objectives of the particle removal efficiency test described here is to provide a repeatable and objective procedure for evaluating a filter's ability to retain test particles in a particular size distribution throughout its useful service life.

Another test objective is to provide data which are representative of the filter's future performance. Care must be taken to select filters for testing which are representative of the filters which will be used day in and day out in actual field use. Statistical methods are used in the selection process and the results represent the compilation of data from a significant number of tests. In fact. the test should be a key quality assurance step for the filter manufacturer. With this data, the manufacturer can be sure that the quality checks at each stage of the manufacturing process actually correlate to performance objectives.

An additional objective is to simulate, as much as possible, the actual field conditions in which the filter will be operating. By selecting test particles, fluid conditions and a methodology that closely approximates field conditions, the complexity of filter selection will be reduced.

Test Limitations

The particle removal efficiency data are not applicable to all applications. For instance, in applications requiring the reduction of total suspended solids, gravimetric efficiency data can stand on its own as a performance parameter.

A major limitation to any standardized test is that it cannot perfectly simulate a filter's performance under actual oil field conditions. This is due to variations within the key factors which affect filtration:

• Particle Characteristics
• Fluid Characteristics
• Time

However, standardized tests, in conjunction with a complete suspension analysis (particle size distribution, viscosity, etc.) and a clear definition of the filtration objective, are useful in narrowing down the possible filter choices.

Factors That Affect Retention

Filters can remove particles smaller than the smallest pore by forces of attraction. The extent of adherence to the filter surface (adsorption) is dependent on the size and the chemical nature of the particles, and the characteristics of the fluid and the filter. For instance, the density and polarity of their charge at a given pH will determine the degree of attraction the particles have for each other and for the filter. While adsorption is a significant mechanism of small particle removal, direct filter interception is the governing mechanism for larger particles.

The number of particles challenging the filter will also influence retention efficiency. Particle retention due to adsorption is most evident when low levels of small sized particles are filtered than when higher levels are encountered; as the adsorptive sites are occupied (quenched), this mechanism of retention diminishes in importance. Beyond this, higher particle concentrations increase the extent of pore bridging and caking. Additionally, there is more opportunity for flocculation (aggregation) to occur which effectively increases particle size. The tendency of the particles to adhere to each other is dependent on how and to what extent they chemically express themselves in a given fluid.

The distribution of particle sizes in the fluid also affects retention efficiency. For instance, the addition of larger size particles to a suspension of monosized particles will increase the filter's retention efficiency. The larger particles contribute to filter efficiency by partially blocking pores and by cake formation which, in effect, reduces the pore size and increases the depth of the filter.

In order to have a realistic standardized test, challenge particles must be chosen which simulate the characteristics of the field particles as much as possible. Some of the most commonly used test particles are listed below.

Because AC Test Dust has characteristics common to so many applications, its low cost and availability not withstanding, it is often used for testing. The particle size distributions are listed below.

A constant concentration of test dust is achieved by maintaining a constant flow rate and controlling the test particle introduction by computer-directed turbidity monitoring coupled to an automatic feeder mechanism. The test particle concentration is predetermined by taking into account the dirt holding capacity of the filter and the length of time necessary to collect an adequate number of samples.

Moreover, the extent of particle flocculation and adsorption is also dependent on other characteristics of the fluid such as pH, temperature, ionic strength, and chemical composition. Changes in pH or ionic strength, for instance, can alter the interaction of particles relative to other particles, the fluid, and the filter medium.

Although many fluids can be used for testing, water is most often used because so many applications are aqueous based. To ensure a constant viscosity, the fluid is temperature controlled. A flow rate is chosen which simulates the field conditions and is maintained at that rate throughout the test.

For the particle removal efficiency test. samples are taken upstream and downstream of the filter at timed intervals over its useful life. The particle counts are time-averaged to give a true representation of the filter's performance.

If, on the other hand, changes in differential pressure were used as the sampling interval, more weight would be given to performance towards the end of filter life where efficiency is usually higher. Thus, filter performance would be exaggerated. This is the case because significant changes in differential pressure only occur at the end of the filter's useful life when filtration efficiency is normally peaking. See Figures  3  and  4 .

Test Apparatus

The test filter is exposed to the challenge particles in the liquid stream only once (single-pass) which closely simulates completion and well flooding systems. Refer to Figure 5. This is contrasted to a multipass method where the test fluid is looped back to the filter. The multipass technique simulates a closed looped filtration process one would find in a hydraulic or lubricating system.

Particle Analysis

Samples are analyzed using an electronic particle counter capable of measuring 16 particle diameters. The counter is interfaced to an IBM PC computer which is programmed to report differential and cumulative particle counts. Differential particle counts refer to the number of particles counted in a given channel. The cumulative particle counts represent the number of particles counted in a given channel plus the particle counts of the higher channels. Refer to Table 1.

Performance Reporting

A performance efficiency over the filter's useful service life is often expressed as the Beta Ratio (ß).

Therefore, over the service life of the cartridge, on the average, one particle out of every 1000 challenging the filter ³ 2µm passes through the filter. Inversely, 999 particles out of every 1000 are retained for a removal efficiency rating of 99.9%
 

The following chart illustrates the relationship between Beta Ratio and % Cumulative Removal Efficiency.
Cumulative Beta Removal Efficiency.

Specification Caveats

In standardized tests for hydraulic filters (multipass), it is common practice to state performance for particles larger than a specified particle size: For particles larger than 5µm, ß₅ = 100. Unfortunately, this practice is also seen in the oil field. Because it is a matter of interpretation how much larger the particles are, it can lead to overstating the filter's true performance. A more precise specification is to state the Beta Ratio for particles at a specified size and larger: For particles 5µm and larger, ß₅ = 100.

When comparing filter performance specifications, one must know whether or not the fluid was recirculated (multipass) or a single-pass design was used. For oil field applications, multipass data is ambiguous and has no correlation to single-pass performance results.

Because the filter's particle removal efficiency is determined by fluid and particle characteristics, and time, it is essential to clearly define the conditions of the test. An example specification is given below.

Example Specification:

Filter must achieve ß₂ = 100 under the following test conditions:

Fluid condition: Constant Flow rate of 2.5 gpm per 10" cartridge

Test Particles: Aqueous dispersion of ACFTD, l00 mg/L

Test Protocol:

- Single pass

- Sample at time intervals

- Terminate test at 35 psid

Test Results:

- Time average particle counts

- Express Beta Ratio for particles 2µm and larger.

Absolute Versus Nominal Filtration

In principle, all filters have nominal retention efficiency. Even membrane filters used to sterilize injectable pharmaceutical products cannot be considered absolute. There is always a possibility that particles can pass through a filter because retention is dependent on so many variables (concentration, pH, temperature, flow, viscosity, etc.). Therefore, test data cannot be extrapolated to "Absolute Filtration Efficiency."

Conclusions

Particle removal efficiency is dependent on many factors including: particle and fluid characteristics, and time. Because of all the variables that affect particle retention, no filtration process can be considered absolute. These variables also prevent standardized particle removal test results from being used to perfectly forecast similar performance in actual applications. For this reason, test results cannot be substituted for evaluating filters in the process. However, the test, in conjunction with suspension analysis and with clearly defined filtration objectives, can be useful in narrowing down the filter choices. In addition, the test results can be useful when comparing filter cartridge performance provided that the test conditions are the same. It should also be a key quality assurance step in the filter cartridge manufacturing process.

Bibliography
 

(1) ASTM Practice for Determining the Performance of a Filter Medium Employing a Single-Pass, Constant-Pressure, Liquid Test, F 795-82, 1985.

(2) ASTM Practice for Determining the Performance of a Filter Medium Employing a Single-Pass, Constant-Rate, Liquid Test, F796-82, 1985.

(3) ASTM Practice for Determining the Performance of a Filter Medium Employing a Multi-Pass, Constant-Rate, Liquid Test, F797-82, 1985.

(4) Bensch, L.E. and E.C. Fitch, "An Examination of Filter Rating Methods", Paper No. P74-52, The Eighth Annual Fluid Power Research Conference, Oklahoma State University, Stillwater, Oklahoma, 1974.

(5) Harris, C. and Odom, C., "Effective Filtration in Completion and Other Wellbore Operations Can Be a Good Investment," Oil and Gas Journal, September 20, 1982, pp 148 - 165.

(6) Hong, I. T. "The Beta Prime - A New Advanced Filtration Theory", Filtration and Separation, July/August, 1985.

(7) Hong, I. T. and E.C. Fitch, "An Innovative Technique in Filter Rating", Paper No. 851590, 1985 International Off-Highway & Powerplant Congress & Exposition MECCA, Milwaukee, Wisconsin, 1985.

(8) International Organization for Standardization, "Hydraulic fluid power - Filters - Multipass method for evaluating filtration performance", ref. No. ISO 4572-1981(E).

(9) Johnson, A.I., "Some Factors Contributing to Decreased Well Efficiency During Fluid Injection," ASTM STP 735, J.L. Johnson, J.R. Stanford, C.C. Wright and A.G. Ostroff, Eds., American Society for Testing and Materials, 1981, pp 89-101.