Membrane Process for Separating H,S from Natural Gas

Authors:

Richard Baker

Contractor:

Membrane Technology and Research, Inc. 1360 Willow Road, Suite 103 Menlo Park, CA 94025

Contract Number:

DE-AC2 1 -92MC28 133

Conference Title:

Natural Gas RD&D Contractor’s Review Meeting

Conference Location:

Baton Rouge, Louisiana

Conference Dates:

April 4 - 6, 1995

Conference Sponsor:

Co-Hosted by Department of Energy (DOE) Morgantown Energy Technology Center Morgantown, West Virginia and Southern University and Agricultural and Mechanical College Baton Rouge, Louisiana

E

DISCLAIMER

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

This report has been reproduced directly from the best available copy.

Available to DOE and DOE contractors from the Office of Scientific and Technical Information, 175 Oak Ridge Turnpike, Oak Ridge, TN 3783 1; prices available at (615) 576-8401.

Available to the public from the National Technical Information Service, U.S. Department of Commerce, 5285 Port Royal Road, Springfield, VA 22 16 1 ; phone orders accepted at (703) 487-4650.

DISCLAIMER

Portions of this document may be illegible in electronic image products. Images are produced from the best available original document.

3B.3 Membrane Process for Separating H,S from Natural Gas

CONTRACT INFORMATION

Contract Number DE-AC2 1 -92MC28 13 3

Contractor Membrane Technology and Research, Inc. 1360, Willow Road, Ste 103 Menlo Park, CA 94025 (4 15) 328-2228 (telephone) (415) 328-6580 (telefax)

Contractor Project Managers Kaaeid A. Lokhandwala Johannes G. Wijmans

Principal Investigator Richard W. Baker

METC Project Manager Harold Shoemaker

Period of Performance September 29, 1992 to October 29, 1995

Schedule and Milestones

FY95 Program Schedule

S O N D J F M A M J J A

Prepare Membraneshlodules

DesignKonstruct Test System - Evaluate System in Laboratory - Prepare System Manuals

Select Field Site I

Install and Operate System at Site

Technical and Economic Evaluations

Program Management/Reporting

OBJECTIVES

The overall objective of this program is to develop a membrane process for the separation of hydrogen sulfide and other impurities (carbon dioxide and water vapor) from low-quality natural gas. The specific objectives of the program are as follows:

Develop membranedmembrane processes for H,S removal from natural gas Optimize membranes Optimize spiral-wound modules Study suitable process designs Perform economic analysis Address scale-up issues in membrane and module production Perform bench-scale demonstrations Construct a field test system Conduct field test

BACKGROUND INFORMATION

Production of natural gas in the Lower-48 states is expected to increase significantly to meet the rising demand. Natural gas supply is projected to increase by 25% between 1991 and 20 10 with 70% of the increased production after 1995 coming from the Lower-48 gas fields''). Recent studies of U.S. natural gas reserves have shown that an increasing fraction of the gas produced will be from smaller gas fields at remote locations, and will be subquality on hydrogen sulfide, carbon dioxide, and nitrogen specifications. More than 13% of current reserves are known to be contaminated with hydrogen sulfide(2). It is projected that, when these subquality reserves are brought into production, between $30-40 million will be invested every year in the new treatment facilities that will be req~ired'~). Alternatives to currently available absorption-based technologies are being sought

for processing gas streams to pipeline specifications in an environmentally and economically acceptable manner.

Conventional amine processes are energy- intensive, and, because they are prone to problems such as corrosion in the reboiler, and foaming and solvent losses in the absorption column, require constant supervision and maintenance. Also, spent amines must increasingly be disposed of as hazardous wastes, which increases operating costs. Membrane processes are simple and modular, and offer greater operational reliability and lower maintenance costs. Membrane processes require almost no operator supervision and are, therefore, ideally suited for operation in remote locations. Membrane processes would be very cost-competitive with conventional amine processes, especially for gas streams with high acid gas concentrations and low flow rates.

PROJECT DESCRIPTION

Phase I Status

The Phase I portion of the project was completed during the 1994 Fiscal year. All the milestones of the Phase I project were hlly achieved. During the Phase I program, we selected an appropriate membrane material from a family of copolymers. Membrane screening tests were performed at three feed pressures, 400,600 and 1,000 psia, and with a feed gas containing 0.1% H,S, 4% CO, and balance methane. Based on the screening test results, we selected two membrane materials for hrther evaluation in a parametric test plan.

The two membrane materials were tested as 3-inch-diameter stamps over a wide range of feed compositions, pressures and temperatures. The test plan covered the wide range of feed compositions anticipated in produced natural gas.

The effect of water vapor in the feed gas was also evaluated. Based on the tests of these two materials, we selected one polymer, Pebax 40 1 1, for hrther development in the subsequent Phase I1 program.

In Phase I we also conducted an extensive survey of H,S-prone formations, utilizing existing databases and reports by the Gas Research Institute. Further, we made a preliminary technical and economic evaluation of the membrane process. This included evaluation of the feasibility of different membrane process designs, and membrane hybrid processes. Economic evaluations were also performed for a conventional DEA amine absorption process for comparison with the membrane process. The results of this analysis are discussed in a subsequent section. A Phase I final report was prepared and submitted.

Phase II Status

In Phase I1 of the program, we scaled up the production of the Pebax 401 1 membrane to commercial-size 40-inch-wide membrane rolls. For commercial applications involving large gas flow rates, a substantial membrane area is required; MTR incorporates flat-sheet membrane into spiral-wound modules. Pebax 401 1 membrane modules were prepared and tested. The issues involved in the scale-up process are discussed in subsequent sections, and the module test results are also discussed.

During Phase 11, we have also designed and constructed a field test system incorporating two 3-inch module housings. This unit is currently being tested in the laboratory and will be sent to a field site in June 1995.

3

MTR's Composite Membrane

Figure 1 shows a diagram of the membrane produced by MTR. The composite membrane has at least three layers. The optimized membrane for the separation of hydrogen sulfide from natural gas consisted of the following components:

Support fabric Polymeric microporous support layer Rubbery gutter layer (not shown) Selective Pebax layer Rubbery top protective layer (not shown)

Figure 1. MTR's Composite Membrane

Each component plays an important part in the overall hnction of the membrane. The support fabric characteristics affect the support membrane structure. The choice of appropriate material for the support layer is important because this layer provides the mechanical strength of the membrane. The selective layer is responsible the separation of hydrogen sulfide and methane. The top layer provides protection during handling and module rolling. Each of these layers was optimized during Phase 11.

MTR's Spiral-Wound Module

A schematic diagram of MTR's spiral- wound module is shown in Figure 2. The module

consists of a central product tube, around which are wrapped a number of membrane envelopes. Each membrane envelope consists of two sheets of membrane separated by a feed spacer. The permeate side of the membrane envelopes are separated by a permeate spacer. The entire assembly is rolled tightly around the central product tube and locked in place with adhesive tape.

Figure 2. MTR's Spiral-Wound Module

The spiral-wound module has one inlet for the feed gas and two outlets. One outlet is for the permeate stream, which is enriched in the faster permeating component, in this case H2S. The other stream, which is depleted in H2S, is removed through the residue stream outlet. driving force for the separation is the difference between the feed and permeate pressures.

The

In natural gas applications, the feed gas is typically at high pressure, while the permeate is held at a lower pressure. MTR's commercial modules usually operate at pressures up to 200'psig. For this program we developed modules capable of withstanding pressures up to 1,200 psig. This involved optimizing various parts of the module, including:

feed spacer material product pipe material, and membrane envelope dimensions.

Another important issue that was addressed in Phase I1 was the protocols used in

manufacturing the spiral-wound module. The result of the optimization effort was a high- pressure spiral-wound module able to withstand high pressure operation and pressure cycling, and at the same time have separation performance matching that of the membrane stamps.

Bench-Scale Test System

The modules have been continuously tested in our bench-scale test system, This test system was built specifically for this project, and contains a two-stage diaphragm compressor to recompress the residue and permeate gas to feed pressure. The system operate in a hll recycle mode, i.e., the residue and permeate streams are hlly recycled. The system can be used for feed pressures as high as 1,500 psig and at feed flow rates of up to 11 scfin.

RESULTS

Bench-Scale Module Tests

Figure 3 shows the results obtained with several spiral-wound modules. In Figure 3, the module selectivity is plotted as a hnction of feed pressure. The module selectivity is defined as the ratio of the permeation flux of H2S to that of CH, through the membrane in the module.

100 -

10 I , I a I I 0 200 400 800 100 1000 l200

Foed Pm*suro l ~ t l a l

Figure 3. Bench-Scale Test Results

4

For comparison, Figure 3 shows the intrinsic H,S/CH4 selectivity of the optimized membrane. Ideally the module selectivity will match the membrane selectivity, provided the module design is such that mass transfer effects on the feed side are completely eliminated and there is no permeate side pressure drop. Figure 3 shows that progressively better modules were developed during the program as optimization of the module internals and the manufacturing protocols was carried out, so that the H,S/CH4 selectivity in the module approached that of the membrane.

Figure 3 also shows the data for the final optimized module, at low and high feed flow rates. At higher flow rates and lower pressures, the module selectivity matches that of the membrane. At lower flow rates, the module selectivity is somewhat lower. This difference can be explained by the concentration polarization effect common in membrane separations. If an adequate velocity cannot be developed at the membrane surface in the module, and the membrane is highly selective for the faster permeating component, a boundary layer is formed adjacent to the membrane. In this layer the concentration of the faster permeating component is greatly depleted. This effect leads to a reduction in the separation performance of the module. At higher flow rates, such as those expected in commercial operation, the problem of inadequate flow velocity is much less.

Overall Process Design

The membranes designed in this project will be used to separate hydrogen sulfide from natural gas. A schematic of the acid gas removal process is shown in Figure 4. The overall process consist of an initial separation step in which the H,S is removed from the feed and concentrated in the acid gas stream. The sweetened natural gas then enters an optional polishing step to remove

the H,S to pipeline specifications of 4 ppm or less. The H,S-enriched acid gas stream then enters a sulhr recovery stage. This will be a process for converting the H,S to elemental sulfur, such as the Claus process, or a redox process depending on the composition and volume of the feed stream entering the sulfbr plant.

Bulk removal 4 1 + Polishing

Polishing step i i (ifnecessary) . . - - - - - -

Hydrogen sulfide I ~weetnaturaigas . . I -.-.'.:kgee;;;;;j ' r--I .-.-.-.-

Sulfur Sulfur recovery

Sulfur

Figure 4. An Overall Schematic Diagram For Treating QS-Bearing Natural Gas

A membrane process would be used in the initial separation step to remove hydrogen sulfide from sour natural gas. The membrane process design will also remove the CO, to pipeline specification. The membrane process design used in this portion of the plant will depend on the specific requirements of the operator. Three such designs are shown in Figure 5.

Figure 5. Membrane Process Designs For Separation of Hydrogen Sulfide From Natural Gas

5

Economic Analysis

A preliminary economic analysis of the membrane/Sulfatreat hybrid process was performed. The module permeation properties used for this calculation were those obtained from the optimized modules in the bench-scale tests. A cost comparison of this process was made with amine absorption. The cost of the amine system was obtained from a GRI. study based on calculations performed by M. W. Kellogg. For a 2-MMscfd stream containing 6.3% CO, and 2.7% H2S, the capital costs of the membrane hybrid system were $430,000, whereas those of the amine system as reported by GRT were $1.5 million. The processing cost for the membrane process is $0.57iMMJ3tuf, compared to $0.83/MMBtuf for the amine absorption process. Although the costs of the membrane process and the amine process are dependent on number of site specific variables, a few general observations can be made. Membrane processes are most economical for treating feed gas at low flow rates and high acid gas concentrations. Based on our economic evaluations, for a wide range of feed H,S and CO, concentrations we prepared an applications envelope for the region of feasibility of membrane processes. This application envelope is shown in Figure 6.

As can be seen in Figure 6, the Pebax membranes developed in this program would be suitable for treating subquality natural gas for which the economics are not viable or marginal. Also, the Pebax membranes are best used for low to medium gas flows and for H,S contents in the feed gas exceeding 1,000 ppm up to about 5%. Interestingly, a significant proportion of H2S contaminated subquality natural gas fits this category. The applications envelope also shows that absorption processes would be economical at flow rates more than about 10-15 MMscfd, mainly because they show good economies of scale compared to membrane processes. Also, at high acid gas concentrations, a combination of the amine and the membrane process may be the most economical, because the membrane process removes most of the acid gases and the amine plant removes the remainder separation. This allows both processes to operate in their most effective range.

Hydrogen Sulfide Concentration (ppm)

Figure 6. Application Envelope for the Membrane-Based H,S and CO, Removal Process

ACCOMPLISHMENTS REFERENCES

The major accomplishments of the program to date are listed below:

Identified membrane material exhibiting very high H,S/CH, selectivity in the range of 40- 60. Scaled up membrane production to commercial size rolls. Completed high-pressure membrane and module development and optimization. Achieved a membrane permeation flux for CH, of about 4 x lod cm3/scm2-cmHg, which is about twice as high as state-of-the-art cellulose acetate gas separation membranes. Demonstrated separation at bench-scale with laboratory-scale modules. Developed process designs and economics. Prepared applications envelope. Constructed field test system.

FUTURE WORK

The following work is planned within the project period:

Conduct field test Test pilot skid in laboratory

Update economics based on field data

We also anticipate a subsequent cot c- shared Phase I11 commercial demonstration program. In this phase, we expect to perform the following work.

1 , The Clean Fuels Report, J.E. Sinor Consultants, Inc., Nivot, CO 80544, 5(3), June 1993.

2. "Chemical Composition of Discovered and Undiscovered Natural Gas in the Lower-48 - Executive Summary," R.M. Hugman, E.H. Vidas, and P.S. Springer, Energy and Environmental Analysis, Inc., prepared for the Gas Research Institute, Chicago, IL March 1993.

3. "Gas Research Institute in Natural Gas Processing," D. Leppin and H.S. Meyer, presented at SPE Gas Technology Symposium, Houston, TX January 23-25, 1991.

Scale up modules to 8-inch Build demonstration system Conduct demonstration field tests

7