i I

I I

- SOURCE REDUCTION RESEARCH PARTNERSHIP

Metropolitan Water District of Southern California Environmental Defense Fund

e

SOURCE REDUCTION AND RECYCLING OF HALOGENATED

SOLVENTS IN THE TEXTILE INDUSTRY

TECHNICAL SUPPORT DOCUMENT

A REPORT ON RESEARCH PERFORMED BY THE SOURCE REDUCTION RESEARCH PARTNERSHIP

FOR THE METROPOLITAN WATER DISTRICT AND THE ENVIRONMENTAL DEFENSE FUND

F

SOURCE REDUCTION AND RECYCLING OF

HALOGENATED SOLVENTS IN THE

TEXTILE INDUSTRY

TECHNICAL SUPPORT DOCUMENT

A REPORT ON RESEARCH PERFORMED BY THE SOURCE REDUCTION RESEARCH PARTNERSHIP

FOR THE METROPOLITAN WATER DISTRICT AND THE ENVIRONMENTAL DEFENSE FUND

Prepared By

JACOBS ENGINEERING GROUP INC. 251 SOUTH LAI(E AVENUE

PASADENA, CA 91101 (818) 449-2171

DescriDtion of Proiect

This document is part of a 12-volume report on research performed by the Source Reduction Research Partnership (SRRP) over the past five years. The report as a whole, entitled Potential for Source Reduction and Recycling of Halogenated Solvents, covers a wide range of industries.

The Summary Report of this project, which is available separately, gives an overview and the results of the research as a whole. This is one of ten Technical Support Documents covering each of ten solvent-using industries and operations (listed below). The full report also includes a separate Lifecycle Inventory and Tradeoff Analysis, covering issues that arise in the comparison of existing halogenated solvent uses with potential alternatives. The ten categories of solvent-using industries and operations are:

0 Adhesives 0 Flexible Foam 0 Aerosols 0 Paint Stripping

Chemical Intermediates 0 Parts Cleaning 0 Dry Cleaning 0 Pharmaceuticals 0 Electronics 0 Textiles

The Source Reduction Research Partnership was formed and jointly managed by the Metropolitan Water District of Southern California (Metropolitan) and the Environmental Defense Fund (EDF). Metropolitan is a public agency, obtaining and supplying water for some 15 million consumers in Southern California. EDF is a national, not-for-profit, public interest organization with more than 200,000 members nationwide.

The research leading to this 12-volume report took place in two phases. The first phase consisted of multi-year field research, primarily in the Southern California area, involving on- site visits, site-specific data gathering, and research into individual processes for all of the affected industries, using a full-time staff employed directly by SRRP. Dr. Kathleen Wolf served as Project Manager during this phase, and staff included Richard Holland, &ita Yazdani, Pamela Yates, and Fidelia Fulmore.

The second phase involved development of a methodology to quantifj potential reductions, additional research and data assessment, analysis of lifecycle and tradeoff issues, derivation of results, and preparation of all report documents. Jacobs Engineering Group performed the work of this phase as consultants under contract. Michael Callahan served as Principal Investigator and Hector Ortk served as Project Manager for Jacobs during this phase, assisted by Carl F r o m , Dr. Rajeev Sane, Harry Van Den Berg, David Shoemaker, Ross Teneyck, and Dr. Arthur Purcell.

_. -Dr. Timothy Quinn of Metropolitan and David Roe of EDF served as co-administrators of the project throughout. A formal advisory committee, including over 40 representatives of solvent-using industries, industry associations, government agencies, and environmental groups, oversaw the design and development of the project.

E

Acknowledgments

Major funding for this project came from the Metropolitan Water District of Southern California, the Switzer Foundation, the U.S. Environmental Protection Agency, CALEPA (formerly the California Department of Health Services), and the City of Los Angeles through its Department of Water and Power. Both Metropolitan and EDF also contributed in-kind professional services, and EDF provided administrative and financial services for the project.

Additional funding to help set up the unusual partnership that made this project possible, and to help design the research format, came from the Michael J. Connell Foundation, the Andrew Norman Foundation, and Southern California Edison Company.

Metropolitan and EDF warmly thank all the sponsors of the Source Reduction Research Partnership for their support, patience, and faith in this unconventional and unprecedented research effort.

Even as originally conceived, this projects goals were ambitious; and while it was underway, unanticipated difficulties in meeting those goals regularly emerged. The results would not have been possible without the extraordinary effort of everyone involved in both phases of the project. Metropolitan and EDF gratefully acknowledge their special debt to the projects dedicated participants including staff, consultants, advisers, and supporters.

Disclaimer

In a project Iike this, covering a vast and quickly evohing field, no warranties and no express or implied representations can be offered as to the completeness, usefulness, or accuracy of any of the information presented in this document or in any of the companion documents. No such warranties and no such representations are made by any party.

In addition, no warranty and no representation is made by any party that the use of any information, cost estimate, apparatus, method, or process disclosed in this document or in any of the companion documents will not infringe privately owned rights.

Any person or entity making any use of this report or its contents or relying thereon does so at its own risk and in doing so releases all parties named above, including but not limited to Jacobs Engineering Group, the Metropolitan Water District of Southern California, and the Environmental Defense Fund, from any liability, damage, cost, or expense such person or entity may incur as a result, including consequential or other indirect or contingent liabilities whether due to the negligence of Jacobs Engineering Group Inc., the Metropolitan Water District, the Environmental Defense Fund, or otherwise.

iii

Readers Note

Note on Terminoloar: The definition of key terms in this field, including the term %ource reduction'' itself, has been subject to rapid evolution and nearly continuous debate during the period in which this study was carried out.

In 1986, when SRRP was first taking shape, the sponsors chose to focus the SRRP study on "source reduction" defined to include recycling and recovery of solvents, as well as substitution of alternative materials, equipment and process modifications, housekeeping measures, and the like. In 1990, midway through the study, Congress passed the Pollution Prevention Act which defined "source reduction" to exclude recycling.

Although this report tries to use the terminology of the 1990 Act, the fact that the scope of this study is larger than the 1990 definition of "source reduction" might be confusing to the casual reader. Definitional issues and the reasons for choosing the scope of this study are discussed at greater length in the Summary Report.

iv

. TABLE OF CONTENTS

Page 1.0 INTRODUCTION ............................................................................................................. 1

2.0 BACKGROUND ................................................................................................................ 3

2.1 MARKET SEGMENTS ....................................................................................... 3

2.2 SOLVENT USAGE .............................................................................................. 3

2.3 NUMBER OF FACILITIES IN OPERATION .............................................. 4

2.4 PROCESS DESCRIPTION AND SOURCES OF RELEASE .................... 4

2.5 REGULATORY TRENDS .............................................................................. 13

2.6 INDUSTRY TRENDS ....................................................................................... 15

3.0 SOURCE REDUCTION OPPORTUNITIES ............................................................ 17

3.1 SCOURING ......................................................................................................... 17

3.2 DYEING OPERATIONS ................................................................................. 29

4.0 ANALYSIS OF SOURCE REDUCTION OPTIONS ............................................... 31

5.0 ESTIMATION OF SOURCE REDUCTION POTENTIAL ................................... 33

5.1 QUANnmCATION .......................................................................................... 33

5.2 METHODOLOGY ............................................................................................. 35

.. 6.0 REFERENCES ................................................................................................................. 41

V

TABLE OF CONTENTS (Continued)

Page

List of Tables

Table 2.1

Table 2.2

Table 2.3

Table 3.1

Table 3.2

Table 5.1

Number of Textile Establishments by SIC Code ............................................... 5

Solvent U s e h s s Profile for the Textile Industry ........................................... 12

Health and Environmental Characteristics of Halogenated Solvents .......... 14

Summary of Source Reduction Options Investigated ..................................... 18

Low Molecular Weight Organic Solvents Used for Scouring ........................ 21

Source Reduction Potentials for the Textile Industry .................................... 34

..

vi

F

,

SOURCE REDUCTION AND RECYCLING

OF HALOGENATED SOLVENTS

IN THE

TEXTILE INDUSTRY

1.0 INTRODUCTION

The focus of this Technical Support Document is on the textile industry and the production of woven and knit fabrics. Most of the perchloroethylene (PERC) and 1,1,2-trichloro-l,2,2- trifluoroethane (CFC-113) used in textile processing is used for scouring @e., cleaning) polyester and double knits. Minor quantities of l,l,l-trichloroethane (TCA), trichloroethylene (TCE), and CFC-113 are used for the scouring of wool. Other uses include PERC as a solvent for applying water repellents and dyes, TCE as a swelling agent in disperse dyeing of polyesters, and both TCA and PERC as cleaning fluid for the removal of spinning oils and lubricants from equipment. The use of halogenated solvent to clean finished goods (i.e-, clothing) is discussed in the companion Technical Support Document for the dry cleaning industry.

The textile industry used 10,500 metric tons (MT) of halogenated solvent in 1988. About 7,000 MT of TCA, 2,OOO MT of PERC, 1,OOO MT of TCE, and less than 500 MT of CFC-113 were employed. Viable ways to reduce solvent use, while not increasing the generation of wastewater, include better housekeeping, installation of vapor controls for recovery and

reuse of solvent, and solid waste recycling systems for recovery and reuse of solvent.. By implementing these measures, solvent use reductions of 0 to 23 percent short-term (less than 5 years), 0 to 26 percent medium-term (5 to 10 years), and 12 to 40 percent long-term (10 to 20 years) are projected. Conversion to aqueous scouring, which could eliminate solvent use, was not included in this projection since wastewater discharges from this process are quite -high and have been an ongoing problem for the industry.

Section 2 of this study provides a general discussion of the textile industry, a description of the solvent using processes employed, and other characteristics of the industry structure.

1

Section 3 describes the source reduction options that could be implemented for the reduction of solvent use. No detailed cost analysis of source reduction options is presented in Section 4 since most facilities that could afford to install air emission controls or solvent recycling equipment have already done so. Section 5 presents the methodology and results of the estimation of potential solvent use reduction in the Unites States by the textile industry.

2

2.0 BACKGROUND

The textile industry, operating under Standard Industrial Classification (SIC) Code 22, includes the conversion of natural and man-made fibers into fabrics and other textile products. Fibers are processed by mechanical methods into yarns that are then woven, knit, or otherwise processed into fabric. The fabrics are then dyed, printed, and/or finished into final goods. Solvent use is mainly associated with the cleaning, dyeing, and printing of the yam and fabric. The following subsections discuss market segments, solvent usage, number of mills, processes, regulatory trends, and industry trends affecting the use of halogenated solvent in the textile industry.

2.1 MARKETSEGMENTS

Chnversion of fabrics into final goods is often performed by other industries that convert them into apparel such as hosiery, underwear, or outerwear. A large number of household, commercial, and industrial products are also produced, the most notable being floor covering (carpets and rugs), upholsteq, sheeting, and draperies. Other specialized segments of the industry produce sewing thread, tire cords, twine, and rope.

The raw material consumption of all textile mills in the United States has slowly increased although individual sections have experienced large fluctuations both up and down. In the past, cotton was the principal raw material. In the late 196Os, however, man-made fiber consumption surpassed that of cotton and has claimed a progressively larger share of the raw material market. In the mid-l970s, 70 percent of the fiber utilized was man-made. The usage of wool has steadily declined to 1 percent of all fibers utilized in 1975 (Cooper, 1978).

2.2 SOLVENT USAGE

Halogenated solvents are used in a number of cleaning and dyeing operations in the textile industry. The actual extent to which each operation involves the use of solvents or alternative chemicals is not known, although the majority of halogenated solvent used is believed to be for cleaning. Industry sources estimate that approximately 10,500 MT of 'halogenated solvents are used. Solvent use in 1988 amounted to 7,000 MT of TCA, 2,000 MT of PERC, 1,OOO MT of TCE, and 500 MT of CFC-113. The use of METH by the industry is believed to be insignificant.

3

The use of PERC, which was reported at 16,000 to 18,000 MT in 1974, and 18,000 to 20,000 MT in 1977 (SRI, 1982) has dropped substantially to only 2,000 MT in 1988. This drop could be attributed to regulation of PERC as a photochemical smog contributor and its status as a suspect carcinogen. Many users who employed PERC in scouring operations converted to aqueous systems. In spotting and cleaning operations, many users of PERC switched to TCA

The use of CFC-113 has also declined in recent years because of its regulation as an ozone depleter. According to a recent study (Ramsey, 1990), CFC-113 is used at only five facilities nationwide. Each employs approximately 150 gallons annually. This represents a substantial decrease from the 500 MT reported in 1988. TCA, recently included in the Montreal Protocol as an ozone depleter, may experience similar reductions in use.

23 NUMBER OF FACILITIES IN OPERATION

The textile industry is one of the most fragmented industries in the United States. According to the 1986 Census of Manufacturers, more than 6,100 establishments are part of the industry. However, few carry out all of the functions associated with textile manufacturing. The manufacturing processes are conducted by a number of specialists that carry the process one or more steps toward completion before transferring (selling) the semi-finished product to another plant where further processing is conducted. The textile industry contains 10 major SIC classifications and 30 subclassifications. A breakdown of the number of establishments by SIC classification is listed in Table 2.1.

2.4 PROCESS DESCRIPTION AND SOURCES OF RELEASE

Major activities involved in the production of textile goods include yam preparation, textile construction, dyeing and printing, and finishing. Yam preparation and textile construction, mainly by means of weaving or knitting, are mechanical operations and do not involve the use of halogenated solvents. However, chemicals applied to the yam during preparation and construction can play a role in the choice of cleaning methods subsequently employed. Cleaning operations are often referred to as scouring or desizing, depending on the types of fabric being cleaned and contaminants being removed. The following sections discuss the cleaning, dyeing, printing, and finishing of both woven and knitted goods. Information regarding sources of release is also presented.

4

Table 2.1

Number of Textile Establishments by SIC Code

SIC Code Classification Number Of Establishments Nation California

221 1 2221 223 1 2241

2251 2252 2253 2254

* 2257 2258 2259

2261 2262 2269

2271 2272 2279

2281 2282 2283 2284

2291 2292 2293 2294 2295 2296 2297 2298 2299 2200

Weaving Mills - Cotton Weaving Mills - Man-made Fiber Weaving and Finishing Mills - Wool Narrow Fabric Mills

Womens Hosiery Hosiery, NEC Knit Outerwear Mills Knit Underwear Mills Circular Knit Fabric Mills Warp Knit Fabric Mills Knitting Mills, NEC

Finishing Plants - Cotton Finishing Plants - Man-made Fiber Finishing Plants, NEC

Woven Carpets and Rugs Tufted Carpets and Rugs Carpets and Rugs, NEC

Yam Mills, except wool Throwing and Winding Mills Wool Yarn Mills Thread Mills

Felt Goods Lace Goods Padding and Upholstery Filling Processed Textile Waste Coated Fabrics Tire Cord and Fabric Non-Woven Fabrics Cordage and Twine Textile Goods, NEC Textile Goods (Total)

222 456 113 262

164 372 875 66

307 162 70

248 241 168

57 311 77

33 1 149 76 65

38 49 92 87

179 20

113 164 189

6,152

17 29 6

11

2 2

78 2

13 3 0

30 15 7

8 37 10

8 2 0 0

4 0

11 5

16 0 5

13 15

679 ~~

_-

Source: USDC (1986a and 1986b).

5

Woven Fabric Finishing

The basic processes in a woven fabric finishing plant are fabric preparation, dyeing, printing, and finishing; each is followed by a drying step. In certain cases, some processes may be omitted and the order of the processes in a particular plant may vary depending on the particular end product produced. The preparation process is discussed briefly since it may involve the use of halogenated solvents.

Preuaration

Before woven fabric can be properly dyed, the sizing chemicals applied to the yam prior to weaving must be removed. Sizing increases the strength of the yam and allows it to withstand abrasion during the weaving process. Starch, polyvinyl alcohol (PVA), and carboxymethyl dllulose (CMC) are the most common sizing chemicals used. Starches are removed with enzyme systems or sulfuric acid solutions. PVA is removed by washing the fabric with a hot (190F) detergent solution, using a multiple dip operation with counter-current flow. The desizing solution is padded onto the fabric to achieve a 90 percent uptake and then steamed to increase the uptake rate. After an exposure time varying from a few minutes to 12 hours (depending on the type of equipment and the type of fabric), the desize solution is rinsed from the fabric. Desizing is an aqueous based process.

Fabric scouring, where halogenated solvents may be used, is performed to remove foreign substances such as hydrocarbon oils, corning oils, silicone oils, and fluorocarbon oils used to lubricate the fiber during production of the fabric. The continuous dyeing process, which is now in widespread use, only allows a short contact time between the cloth and dye liquid. Hence, the cloth must be adequately clean to absorb the dye rapidly and uniformly. In addition, scouring can lighten the color of the base cloth and remove motes and impurities that may spoil the appearance of the finished cloth.

Aqueous scouring of fabric may be performed with a solution of caustic (including some solvent to remove waxes from man-made and some blended fabrics) or detergent. After allowing an adequate time for the solution to remove impurities from the fabric, the fabric is whhed in hot water. This may then be followed by drying.

Solvent scouring may be performed either in a batch or continuous operation. In the batch process, the solvent is heated and the machine has a system to hold clean solvent, rinse

6

P

liquor, and finished liquors in four to six tanks. The fabric is contained in a perforated drum that rotates in the solvent, which is circulated through the drum. The solvent liquors are removed from the fabric by centrifugation. Fabrics are dried by hot air circulation and solvent is recovered by condensation and carbon adsorption. Contaminated liquid solvent is recycled on site by distillation (USEPA, 1983).

In continuous solvent scouring, fabric is fed on rollers to the scouring unit. Here the fabric is dipped or sprayed with cold solvent or immersed in solvent vapor. Following the scouring process, the fabric is wrung and fed to the drying operation. The solvent vapors from the drylng operations are condensed and recovered. The spent solvent and the condensate are distilled and reused. Following drymg, the fabric is folded, rolled, or sent to a dyeing process.

During solvent scouring with halogenated solvents, hydrochloric acid can be formed in the fi6er. A 5 to 10 percent moisture content is common in fibers, particularly cotton and polyester/cotton fabrics. This moisture can react with the solvent, resulting in the formation of hydrochloric acid. The solvent remaining in the fabric cannot be completely recovered. About 8 percent of the solvent remains in the fabric. The ratio of solvent consumption to fabric scoured is about 1 pound of solvent to 10 pounds of fabric (Camp, 1990).

Both PERC or CFC-113 may be used in solvent scouring. Solvent scouring can often achieve soil removal efficiencies approaching or exceeding 99 percent. Use of CFC-113 is reported to have some advantages over PERC in that the oil recovered by distillation as a part of the scouring process can sometimes be directly reused for sizing of yarns. The oil recovered from PERC scouring commonly contains some PERC residue (because of PERC's high boiling point) and cannot be reused for sizing operations (Ramsey, 1990).

Fabrics that are typically solvent scoured include 100 percent synthetics, bast fibers, glass fibers, silk fibers, and metallic fibers. Occasionally, fabrics made of wool and other animal hair fibers or blends of wool and synthetics (both woven and knit) are solvent scoured. Cotton and cotton synthetic blends are mostly scoured in aqueous caustic solutions though some solvent scouring processes have been patented.

Skuring of raw wool has been a concern because of the 30 to 50 percent non-wool impurities that have to be removed before processing (Cui et al., 1988). Compositions of these impurities are quite complex and may be divided into the following components: wool wax or

7

lanolin, water soluble material including dried sweat, urine, and protein degradation products, and insoluble particulate material including sand, soil, vegetable, and fecal matter.

If an aqueous scouring process is employed, the water insoluble wool grease has to be emulsified to remove it from the fiber. The subsequent destabilization to recover grease and minimize water pollution is difficult. Wool scouring of a typical 1 MT per hour train can produce an effluent load equivalent to that of a town of 30,000 people. Not only PERC but other hydrocarbon solvents such as kerosene, hexane, and petroleum distillates have been used in scouring of raw wool material (Cui et al., 1988).

The majority of woven fabric is continuously dyed with pad-squeeze type equipment. In this prbcess, the dye is transferred to the fabric by passing the fabric across rollers, which are partially submerged in the dye solution. The liquid content of the fabric is then reduced with squeeze rollers in order to conserve dye liquid and to reduce drying time. With water-based dyes, an infrared gas-fired predrier is used to begin the drying process. The fabric is then heat-fired in a thermos01 oven (for polyesters), dried, and cured. Excess dye is sometimes removed by rewetting and drying the fabric.

Solvent dyeing is commonly employed for 100 percent polyester material, and is reported to have a lower yield than aqueous dyeing. For example, solvent dyeing has a 60 to 70 percent color yield, while aqueous dyeing has a color yield of 85 to 95 percent (Camp, 1990). The two solvents most commonly used in solvent based dyeing are PERC and methanol.

Printing is normally accomplished by a semicontinuous flat screen process, a continuous rotary screen process, or a roller printing process. Either non-halogenated solvent or water- based pigments or a mixture of the two can be used. Solvent-based printing is normally more energy consumptive than water-based printing; a larger volume of heated air must be exhausted to prevent a dangerous build-up of volatile hydrocarbon vapor in the drier. _-

In both flat screen and rotary printing, pigment is applied to fabric in a process similar to silk screening. In roller printing, the pigment is transferred to the fabric with an engraved roller. Following application to the fabric, the pigment is fixed in a brief high temperature curing

8

process and then dried. Synthetic fabrics also require washing with hot water and detergents to remove excess dye materials followed by drying.

Finishing

Finishing operations normally involve the padding of a specific chemical onto the fabric, drying, and curing to fix the finish. Finishes such as antistatics, fire retardants, softeners, water repellents, and durable press resins may be applied. The chemicals stay on the fabric while the camer (usually water) is evaporated when the fabric is dried. Drying is accomplished with predriers and drymg ovens, and the fabric is then cured in order to stabilize the finish. Levels of waste or wastewater from fabric finishing operations are small because the chemicals are applied by padding.

One industry source indicated that METH, which is not commonly used in the textile industry, was used to coat weather balloons with plastic (Ostrowski, 1990). Other halogenated solvents are also used in fabric finishing. Some other finishing operations that could involve solvents are application of Scotchgard, &pel, and flame retardants. Dow has marketed solvents for these applications (Oakes, 1987).

Knit Fabric Finishing

The operations normally utilized in knit fabric finishing are similar to those encountered in woven fabric finishing: mainly fabric preparation, dyeing, printing, and finishing. Solvent usage patterns may differ due to the different types of fabrics produced and chemical agents used in their production. The following sections discuss these differences.

PreDaration

Scouring of knit fabrics is employed to remove impurities and it can be accomplished by kier boiling, open width or rope form scouring, or solvent scouring. Kier boiling was the original knit fabric cleaning operation. In this process, several tons of clothes are impregnated with a heated solution of caustic, soap, chelating agents, and sodium silicate for 6 to 12 hours, after which the fabric is rinsed with cool water. Use of the process is primarily limited to lightweight fabrics since heavy weight fabrics show dye creases if scoured in this manner.

9

Open width and rope form scouring use a more concentrated caustic solution with added antioxidants. Steaming of the fabric ensures even swelling and rapid destruction of impurities, which are removed during the subsequent rinsing of the fabric. In the case of fabrics knitted from synthetic yarns, scouring consists of washing with a detergent solution to emulsify the knitting oils, followed by rinsing. Knitting oils may be based on mineral oils, vegetable oils, synthetic ester-type oils, or waxes and may also contain antistatic agents, antioxidants, bacteriostats, and corrosion inhibitors.

Solvent scouring of synthetic knits has been used in other countries but has not found widespread application in the United States. The solvent used is PERC or CFC-113 and the equipment must be airtight to prevent air pollution and the escape of the solvent. The solvent may be applied by either high velocity jets or by an immersion method. Solvent scouring has advantages over aqueous scouring in that wastewater loads from the scouring operation are eliminated and that the cost and energy consumption of subsequent drying operations are significantly reduced. However, solvent purchase costs are high and the use of some solvents can present worker exposure hazards. Some solvent scouring of 100 percent cotton and cottodsynthetic blends is performed in this country.

Dveing

Large lots of knit fabric may be continuously dyed in a manner similar to that described for woven fabrics. Recently, batch jet dye machines have been introduced for dyeing sensitive knit fabrics. In this process, the fabric is placed in the jet dye machine in a continuous loop, which is propelled by a jet of dye liquor. The advantage of this process is that the relaxed fabric allows development of a full hand. The venturi jet turbulence ensures good dye penetration, and the process uses less energy and water than other methods of dyeing.

Printing

Printing of knit goods is less common than for woven goods, but may be accomplished with any of the same processes. Heat transfer printing is more common for knit goods than for wovens. The relative importance of the various printing processes in knit goods production is not yet well defined since the knit industry is relatively new.

10

Finishing can involve final washing and drying of the fabric, or it can involve the application of different finishes such as permanent press or fire retardant. Mechanical finishing operations are less commonly used on knits than for woven fabric.

Equipment Cleanup

Cleaning equipment with halogenated solvents is a common practice in the textile industry. Solvents such as PERC, TCA, and TCE are used to remove oil, wax, grease, and lubrication fluids from equipment. The solvents are applied as cold cleaning agents and they are popular because of their good solvency for organic materials and low heat of evaporation. They are nonflammable, noncorrosive, and relatively stable. Overall solvent use and the level of releases is highly dependent on individual operations. The reader is referred to the companion Technical Support Document on parts cleaning for information on cleaning and possible source reduction measures that may reduce the use of halogenated solvents.

Sources of Release

In 1988, 2,000 MT of PERC were consumed by the textiles industry. A review of Toxic Release Inventory Data for SIC Codes 2221,2231,2253,2257,2258,2262,2269,2287,2284 shows a range of PERC use among the reporting plants. PERC was used for a variety of reasons: in ancillary uses, as a manufacturing aid, as a formulation component, and as a chemical processing aid. The releases of PERC, however, were reported inconsistently. Some firms reported no waste generation or water releases, while others reported releases to all media. The amount of releases varied greatly. Since reliable waste generation data was not available to SRRP staff, a release profile for the industry was assumed.

The two major halogenated solvent waste generating processes employed by the textile industry are scouring and dyeing. Solvent used for equipment cleaning is assumed to be minor compared to these two uses. Scouring uses 100 percent of the TCE and CFC-113 used by the industry and 70 percent of the PERC and TCA The remaining PERC and TCA is %ed in dyeing processes. While the actual profile of solvent usage by the industry is unknown, SRRP staff believe this profile to be reasonable. Impacted media profiles for these two operations are presented in Table 2.2 and are discussed below.

\

11

Table 2.2

Industry and Operations

scouring

Usage Usage

Dyeing Usage Usage

Total Use / Loss

Solvent Use/Loss Profile for the Textile Industry

Media (a)

Input

S(50%) Input

S(20 5%) Input

A(50 W )

A(80R)

Emission Emission

Air 1 Solid

1988 Solvent Use (Thousand Metric Tons) TCE PERC METH TCA CFC-113

1.0 1.4 0.0 4.9 0.5 0.5 0.7 0.0 2.4 0.3 0.5 0.7 0.0 2.4 0.3 0.0 0.6 0.0 2.1 0.0 0.0 0.5 0.0 1.7 0.0 0.0 0.1 0.0 0.4 0.0 1.0 2.0 0.0 7.0 0.5 0.5 1.2 0.0 4.1 0.3 0.5 0.8 0.0 2.9 0.3

Total 7.8 3.9 3.9 2.7 2.2 0.5

10.5 6.1 4.4

Source: CMR, 1989; SRRP staff estimates; and industry sources. Data shown above excludes

Notes: solvent used in dry cleaning.

a) Percent shown represents amount of solvent used lost to air (A), water 0, or as solid (S) waste. Solvent lost to water is assumed to be negligible.

Although most solvent equipment used for scouring is equipped with vapor control and the units are totally enclosed, some solvent may still escape. Causes of emissions include carryout of the solvent on fabric, loss of solvent from equipment due to diffusion fillingdraining losses, solvent recovery still losses, storage tank vent losses, and fugitive equipment leaks from pipes, valves, and pumps. The level of these emissions depends on the type, design, and size of the equipment, the number of hours of operation, and the operating techniques. According to one reference (USEPA, 1979), solvent scouring is similar to a conveyorized cold cleaning system. If so, then air emissions and solid waste generation can both be assumed to be equal to 50 percent of solvent use.

The solvents used as dye carriers contain various dyestuffs, which are complex organic compounds that are refractory (nonbiodegradable) and hazardous. Dyestuffs also contain heavy metals, such as chromium, copper, and zinc. Only about 50 percent (by weight) of commercial dye is dyestuff. The remainder is usually nonhazardous filler (such as sugar) and Siirfactant. The dyestuff ends up in the waste solvent, which may be recovered on site or sent for off-site recycling. Since the majority of solvent applied to the fabric eventually evaporates, it is assumed that 80 percent of the solvent is emitted to the atmosphere and 20 percent ends up as solid waste.

12

P

Releases to water may occur on a routine basis but are assumed to be relatively minor for the industry as a whole. Halogenated solvents may be released to water when the fabric is rinsed with water after scouring, as a result of vapor control and waterholvent separation, or when the dyestuff and printing chemicals containing halogenated solvents are rinsed from the fabric. Wastewater concentrations from dyeing and printing are dependent on the process and the various add-ons used. In addition to color, the waste can contain high concentrations of BOD and dissolved solids. Printing pigments also introduce suspended solids into the waste stream.

2.5 REGULATORY " D S

Several of the halogenated solvents have, for some time, been associated with human carcinogenesis. As ozone depletes of the stratospheric ozone layer, CFC-113 and TCA contribute to enhanced ultraviolet radiation at the earth's surface, with resulting increases in skin cancers. While a number of specific points of disagreement remain in regard to the health and environmental impacts of halogenated hydrocarbons, a consensus has emerged that these substances can pose major problems when they are released into the environment, and that significantly reducing their use can consequently reduce the health and environmental threats associated with them. Table 2.3 summarizes the health and environmental characteristics of the five halogenated solvents.

Under mandates to curb emissions, the most internationally si@cant is the Montreal Protocol, signed by nations producing and using the bulk of ozone-depleting chlorofluorocarbons (CFCs), as well as TCA Under the Protocol, to which the United States is a signator, both CFC-113 and TCA production are to be phased out - CFC-113 by the end of the century and TCA by 2005. During the second renegotiation to be held in November 1992, it is anticipated that the phase-out schedule may be accelerated. The phase- out of CFC-113 may be moved ahead to 1995 and TCA to 1997 (Morehouse, 1991).

\

The 1990 amendments to the Federal Clean Air Act point toward strict regulations of METH, PERC, TCE, TCA, and other halogenated solvents identified as toxic air pollutants. Under the emergency provisions and accelerated rulemaking provisions of the Clean Air Act, 5everal environmental organizations have petitioned the EPA to phase-out production of CFC-113 by January 1,1995 and HCFCs by January 1,2000 (Morehouse, 1991).

13

* 0

_-

0 E

0

E:

hl 0 E 0 B

a E 3 0 v)

L W

0 s z m

0 c L 0

Y 5 v: e 3 0

ci > 0

.-

.- L

v-

;/j c 0

14

In California, METH and PERC were recently identified as toxic air contaminants. As a result of this determination, METH and PERC use and emissions will be strictly controlled throughout the state. The Air Resources Board (ARB) is in the process of developing regulations that are likely to severely curtail METH use in California. Similar regulations for PERC are likely to follow.

METH, PERC, and TCE are listed by the State of California as known carcinogens. Under the state's landmark toxic control law, passed by the voters in 1986 and commonly known as Proposition 65, any business that exposes people to a listed chemical must give clear and reasonable warning to the individuals exposed. In addition, businesses may not discharge a listed chemical into any source or potential source of drinking water. Exceptions to both requirements exist if the business can show that the amount in question was within "no significant risk levels (currently defined as 50 micrograms per day for METH, 14 micrograms pix day for PERC, and 60 micrograms per day for TCE).

Provisions of Title ID of the Superfund Amendments and Reauthorization Act (SARA) have resulted in public pressure on users of air toxics, including halogenated solvents. By mandating a national toxic chemical emissions inventory and delineating mechanisms for public disclosure of all toxic and hazardous chemicals in use by industry, SARA Title ID is providing both a database on which to build source reduction programs and effective stimuli for generators to "ize their inclusion in this publicly scrutinized database.

The Federal Pollution Prevention Act of 1990 increases EPA visibility in source reduction activities and provides increased resources to state programs aimed at source reduction. EPA's new 33-50 cooperative government-industry program calls for voluntary emission reductions of 33 percent in 17 toxic and hazardous chemicals by the end of 1992 and 50 percent by 1995. Two of the five major halogenated solvents, TCA and TCE, are targeted in the 33-50 effort.

\

2.6 INDUSTRY " D S

Halogenated solvents have been and likely will continue to be used to some extent in the -textile industry. The use of PERC, which is exclusively used for polyester double knit scouring, has declined largely because of a shrinking knit market in the last decade. Several factors, including the flood of imports into the market and the shutdown of some operations and their movement overseas to the sources of fabrics, has encouraged domestic mills to

15

become more efficient. Plant modifications have resulted in faster and cleaner production of goods that require less cleaning and scouring (Dow, 1987).

Prior to 1970, TCE was the halogenated solvent most widely used in the textile industry. Initially, many users moved to TCE as a means of avoiding wastewater discharges associated with aqueous scouring processes. However, the use of TCE, a suspect carcinogen that was also linked to photochemical smog, has declined considerably in recent years. TCE was initially replaced by PERC, which was then replaced by TCA In recent years, some users have converted back to aqueous-based processes in cases where it is technically feasible to do so. Aqueous scouring generally requires pretreatment prior to sewer discharge and appropriate sludge disposal. Aqueous scouring is also more energy intensive.

16

3.0 SOURCE REDUCTION OPPORTUNITIES

This section examines a variety of source reduction options for the scouring and dyeing operations performed by the textile industry. Most options fall into the general categories of solvent substitution, process modification, improved operating practices, and improved housekeeping measures. A summary of the options discussed is presented in Table 3.1 along with a brief mention of advantages and disadvantages of each option.

It is important to point out that options involving solvent substitution do not necessarily constitute true source reduction. Switching from CFC-113 to non-halogenated solvent may reduce the risk of ozone depletion but does so at the expense of increased photochemically reactivity, flammability, and worker safety. When substituting one solvent material for another, many different issues come into play. Meaningful source reduction occurs when pioducts and materials are used more efficiently, independent of their composition. Material substitution should be viewed as true source reduction only when there is a clear environmental advantage to the substitution.

3.1 SCOURING

Seven options for reducing or eliminating the use of halogenated solvents in scouring are discussed below. Options include use of alternative solvents, use of aqueous scouring, use of emulsion scouring, monitoring and maintaining solvent quality, housekeeping, vapor recovery, and waste solvent recycling.

1) Use of Non-Halogenated Solvents

There are a number of chemicals that are technically feasible substitutes for halogenated solvents in textile applications. While HCFCs were at one time viewed as being potentially viable, their accelerated phase-out by the Montreal Protocol will make these short-term or interim solutions at best. Viable non-halogenated solvents include aliphatic, aromatic, and oxygenated compounds.

-Ikw molecular weight organic solvents comprise one of the possible categories of substitutes that might be used to replace halogenated solvents in the scouring process. Table 3.2 lists some low molecular weight solvents currently used in scouring operations. These chemicals can be used as substitutes in scouring conducted at room temperature. A standard batch

17

Ir

0 --

M 0 a

5 i

(l

(l

0

0

55

$ M El

9

c

.r(

2:

s * 0 e, s 0

.r(

- cu 2 m c s

0

9

.d

3

m

e,

> 9 k B

a

a m 5 5 c

c 0 .d

Y

5 2 0 WJ C

0

.r( c

5

% U a 0

m

3 El a e, U

cn

s .

.

e, 2 E u

_- e, e

2 0 C

0

h

c

c5 c

I4

0

M

c 0 .C

( -.. 9 9:

d I' r-:

19

CrJ E

.r(

i a C

la

*3

.r(

1 e

.r(

20

Table 3.2 Low Molecular Weight Organic Solvents Used for Scouring

Solvent PEL Flash Point Boiling Point Latent Heat

(PPm) (C) (C) Evaporation (caVgm) Flammable

N-Hexane 50 -21 68.5 80 Y Kerosene NIA 38-74 175-325 80 Y Stoddard Solvent 100 3849 220-300 80 Y Methanol 200 11 65.5 28.3 Y Acetone 750 - 18 37.1 131.9 Y

Source: Cui, et al. (1988).

process using non-halogenated solvent is the Maertens process while the Sover process is a continuous one (Cui et al., 1988).

2) Use of Aqueous Scouring

Aqueous scouring is widely used in the textile industry to scour fabrics. The process employs alkali reagents and anionic or nonionic surfactants to remove impurities. Alkali compounds commonly used are sodium hydroxide, potassium hydroxide, lithium hydroxide, ammonia, and alkali metal alkoxides.

The substitution of aqueous scouring for solvent scouring necessitates equipment replacement which is dependent on the processes to which the fabric will be subject after scouring. In solvent scouring, the fabric leaves the process dry, while after aqueous scouring the fabric remains wet. When the next step in the processing of the fabric is aqueous-based, as in the dyeing process, the wet fabric coming from the aqueous scouring machine need not be dried. In this case, the solvent scouring machine would be replaced only by a machine designed for aqueous scouring. However, if the scoured fabric needs to be dry, then a dryer must be added.

-Aqueous scouring is, generally, less effective than solvent scouring in removing oils. Aqueous scouring uses alkali reagents to remove oils via breakdown of the oil and formation of soap. The soap then behaves as an emulsifying agent and removes those substances on which the alkali has no chemical action. Aqueous scouring, on the average, results in residual oil

21

concentrations of 0.3 to 0.4 percent by weight compared to 0.05 to 0.3 percent for solvent scouring. Although residue oil levels after aqueous scouring are higher, a residual oil content under 0.5 percent typically does not pose sigmficant problems. Oil concentrations greater than 0.5 percent may make the fabric resistant to dye or print paste (USEPA, 1983) and result in yellowing and oil "mist" when fabric is heat set. Though less effective at removing oil, aqueous systems can remove impurities in cotton that solvents cannot. Thus, aqueous scouring is often used for scouring cotton-synthetic blends.

A drawback to aqueous scouring is that it is one of the major sources of wastewater from textile operations. Wastewater from wool scouring has historically posed problems for wastewater treatment systems (Cooper, 1978; Cui et al., 1988). In the last 15 years, many aqueous operations were converted to solvent scouring because of the heavy pollution loading of such operations. Raw wool contains wool grease (wax or lanolin), water soluble material and insoluble dirt that has to be removed during scouring. Wool scouring is one of the highest sources of 5-day BOD (BODS) in the raw wastewater discharge. The high concentrations of mineral dirt and proteinaceous matter have led to high effluent disposal costs (Warner, 1985).

Another disadvantage of aqueous scouring is the possibility of wool felting. If alkali agents such as sodium carbonate are used in scouring heavily soiled wool, fiber damage is possible. To avoid these problems, the Sover process and the Toa process both specify the use of halogenated solvents for initial wool scouring followed by an aqueous scour to remove dirt (Christoe and Bateup, 1987). The extent to which these processes are used is not known.

3) Use of Emulsion Scouring

The emulsion scouring process is another possible substitute for halogenated solvent scouring. Emulsion scouring is a combination of solvent and aqueous scouring techniques. In emulsion scouring, an organic solvent is dispersed or suspended as fine droplets in the aqueous scouring solution. In addition to the solvent, emulsifying agents, which disperse the solvent into water, and coupling agents, which control the size of the droplets, are used. The organic components of the emulsion consist of mineral spirits or a similar petroleum fraction sometimes blended with a halogenated, aromatic, or naphthenic solvent to improve solvency. Emulsions have neutral to slightly alkaline pH. Emulsion scouring is sometimes followed by alkaline cleaning to remove the last traces of oil and grease from the fabric.

22

4) Monitor and Maintain Solvent Quality

Although equipment control and operating practices play a major role in reducing solvent emissions and discharges, emission testing and monitoring can also provide helpful information to evaluate these practices. For example, if emissions are greater than average for a scouring unit of certain size, it may be an indication that the system is inadequately or improperly controlled or operated. Routine testing and estimation of emission losses can also be used to set reduction goals and then track progress.

In addition to emission testing, halogenated solvents such as TCA should be routinely checked for acid acceptance. By routinely performing an acid acceptance test, an operator will be able to tell when the solvent is about to go "acid." Once solvent goes acid, it can no longer be reclaimed and extensive procedures must be followed to neutralize the scouring eq-uipment. A common cause of halogenated solvent going acid is excessive water contamination and depletion of the chemicals used to stabilize the solvent. Water should be routinely decanted from the solvent. Removal of insoluble dirt, fiber, and greases by means of filtration can also help to maintain solvent quality.

5) Housekeeping Measures

A variety of different practices adopted on site can minimize solvent losses. Proper procedures to start-up/shutdown the solvent scouring machine include starting the condenser coolant flow prior to turning on the sump heater and keeping the condenser on after the sump heater is turned off. Losses during transfer of solvent into and out of the scouring machine can be minimized if solvent filling, draining, and transfer are performed with enclosed piping systems. Pumping the solvent directly from the drum to the scouring machine could cut down on spills and diffusional losses. If the solvent is pumped into the scouring machine with little or no splashing, such as with submerged fill piping, less solvent would be lost. Leakproof couples can also reduce transfer losses. Finally, other measures such as good planning and inventory control, repair of visible leaks, and conducting routine equipment inspections, can lead to reduction in solvent use.

5 ) Recovery and Reuse of Vapors

Recovery and reuse technologies for halogenated solvent vapors have been available and practiced for years in the textile industry. Several of these techniques, which are used today

23

or that are promising for use in the future by the textile industry, are examined below. These include carbon adsorption, polymeric adsorption, the Brayton Cycle Heat Pump, membrane separation, and the Cryosol process.

Carbon Adsomtion

This technology can be used to recover various halogenated solvents from a vapor stream. The solvent vapor and air stream can be channeled to a carbon adsorption unit for recovery. Carbon adsorption systems may be most appropriate for use on large scouring units where the credit from solvent recovery helps to offset the high capital equipment cost.

In the adsorption step, the solvent is adsorbed using activated charcoal. The desorption step is generally accomplished with steam. For desorption, the bed is initially heated with steam fdr 5 to 15 minutes, and then the solvent vapors leave the bed and are routed to the condenser. The solvent and the water phases are then separated by gravity or by distillation.

Proper operation and maintenance procedures are critical to maintain the efficiency of the carbon bed. Examples of operating procedures that have an effect on the operation of the carbon bed are: 1) dampers that do not open and close properly to allow the solvent vapor/air stream to pass the bed, 2) use of carbon that does not meet specifications, and 3) improper timing of the desorption cycles. Desorption cycles must be frequent enough to avoid breakthrough but not so frequent as to cause excessive energy consumption.

PERC, TCE, and CFC-113 can be recovered using carbon adsorption, but TCA requires special equipment. For TCE and PERC, carbon adsorption equipment is normally constructed of mild steel covered with a solvent-resistant coating. TCA is very susceptible to hydrolysis. Consequently, its recovery equipment is often constructed of Hastelloy C and can cost much more than coated mild steel adsorbers.

Use of steam to regenerate the carbon subjects the recovered solvent to water. This removes the water soluble components, and the recovered solvent may not be directly suitable for recycle and reuse. In the case of TCA, for example, contact with water condensate will remove water soluble stabilizers and the new unstable TCA may decompose and form HC1, which can easily corrode equipment. Water pollution problems also may result from the materials dissolved in the discharged water. Indeed, water soluble stabilizers removed during steam desorption will end up in water, and eventually enter the sewer system.

24

Polvmeric Adsomtion

Polymeric adsorption (PA) is a new process developed by Nobel Industries in Sweden. The process is based on the adsorption of solvent by macroporous polymer particles using fluidized bed technology. Specially developed polymes have been optimized for use as adsorbents; they consist of crosslinked polymer particles of styrene divinyl benzene. The polymer adsorbent has several advantages over carbon including ease of regeneration, long lifetime, and absence of catalytic effect on the degradation of unstable solvents. The polymer is therefore suitable for the recovery of TCA

The solvent is adsorbed by the polymer particles as the air passes through the adsorption beds. The flow of air also causes the polymer adsorbent to fluidize and behave like a liquid. A continuous flow of adsorbent through the bed is maintained as the saturated adsorbent is &moved from the bottom, regenerated, and fed back in at the top of the adsorption section. During regeneration, absorbent is heated to a suitable temperature in the desorber, the released solvent is picked up by stripping air, and the released solvent and air are collected and routed to the condenser. This process continues indefinitely. The process is attractive because it has fewer moving parts and lower energy consumption than fixed bed systems.

According to the manufacturer, the PA process is specially suited for continuous emission airflows from about 300 standard cubic feet per minute (scfm) to 100,OOO scfm and for solvent concentrations between 0.005 to 0.1 pounds per 1,OOO scf. It is especially favorable for humid air, for water soluble solvents, or ones that oxidize in warm humid air. This process is also recommended for solvents with low boiling points (~105F); however, the incoming temperature must be kept below 90F. Presently, a pilot test unit and a full scale PA plant are operating in Sweden.

Brayton Cvcle Heat Pump

The Department of Energy (DOE) has been supporting the development of the Brayton Cycle Heat Pump ( B O ) in conjunction with 3M and Garrett Air Research since 1978. The technique uses a reverse Brayton refrigeration cycle to condense solvents to liquids. It cools gas streams to very low temperatures, as low as -298"F, and directly condenses the solvent components for collection.

_ _

25

Another technique uses adsorbent beds to strip solvent from air and the BCHP to regenerate the beds. The use of inert gas in combination with the BCHP allows direct reuse of the solvent in the process. The use of steam to regenerate activated carbon adsorbent sometimes catalyzes a reaction that produces unwanted by-products and also releases heat, which can lead to carbon fires. The BCHP eliminates the use of steam and uses special adsorbents that ize catalytic reaction. The system is identical to a conventional adsorption system during the adsorption cycle. The organic materials are then regenerated using the BCHP process. Hot, inert gas passes through the adsorbent bed and desorbs the solvent. The solvent laden inert gas is cooled, compressed, cooled further, and sent through the compressor side of the free-spindle turbo unit. It is further cooled in an interchanger and enters the expander side of the BCHP where it is cooled to as low as -80F.

To prevent ice formation, water must be eliminated from the regeneration gas. Use of either diying steps or appropriate water rejecting absorbents can eliminate this problem. Because it operates in low pressure vessels, the capital and operating costs of the BCHP are substantially lower than for other systems, according to the manufacturer. This method has been demonstrated at 3M for solvent recovery on commercial sized magnetic tape manufacturing facilities. A current project is investigating the use of on-site adsorption vessels regenerated by means of a mobile BCHP System (Marr, 1991). This would allow for small facilities to take advantage of this technology without having to invest in expensive regeneration equipment.

Membrane Seuaration

Synthetic membranes have recently been used to separate aqueous and gas mixtures and for recovery of hydrogen from petrochemical purge gas streams and other chemical production processes. Since membranes can be tailored to fit a certain mass separation task, they seem to be appropriate for separation problems that are difficult to handle, such as removal of organic solvent from waste air streams. A semipermeable composite membrane is used to separate the organic solvent from air. The membrane modules, which are composite structures made by coating a tough, relatively open, microporous support membrane with a very thin, dense fdm, allow a large membrane surface area to be packed into a small volume. The support membrane provides mechanical strength and the thin dense coating performs the separation. Organic solvents are preferentially drawn through the membrane by a vacuum pump and the solvent is condensed and removed as a liquid.

26

The firm manufacturing the membranes claims that comparison with carbon adsorption shows that the membrane process is more cost-effective if the solvent concentration is relatively high, 0.5 percent or higher, and the air stream to be treated is small, between 100 and 1,OOO scfm (Wijmans et al., undated). Capital costs of the systems are in the range of $400 to $l,OOO per scfm of airflow with an operating cost of $0.5 to $1.00 per 1,OOO scfin treated (MIXI, undated).

Crvosolv Process

The Cryosolv Process developed by Meisser Griesheim, a German company, is a method for purifying air containing high solvent content and can handle up to 1,OOO cubic meters or 35,000 cubic feet per hour (Meisser, 1988). The process is a low temperature condensation process, which allows for the direct reuse of recovered solvent. The process uses liquid niirogen as a cooling agent, and it is believed that the process involves the direct contact of the Iiquid nitrogen with the solvent laden air stream. Condensation temperatures range from -100C (-150F) to -140C (-220F). The system has been demonstrated on a coating line in West Germany. Liquid Carbonic also offers a similar process for sale in the United States (Liquid Carbonic, 1989).

7) Recycling of Waste Solvent

There are several methods of reducing the use of virgin solvent by recycling contaminated liquid solvent. A number of methods for treating contaminated solvent for reuse are discussed. The techniques evaluated include on-site and off-site recycling.

On-Site Recvcling

Solvent reclamation is one of the most common methods of handling waste solvents on site. All solvent scouring machines are commonly designed with a distillation system in place. On- site recycling has become attractive because of the high cost of virgin solvent, especially for expensive solvents like CFC-113. The use of on-site stills to recover solvent can result in up to a 20 percent reduction in solvent use overall.

Halogenated solvent reclamation is technically feasible because of the stability of most halogenated solvents and their relatively low boiling points. While there are many on-site reclamation techniques that can be employed to reduce solvent use, the most common is

27

single plate distillation. Single plate distillation involves heating the waste, vaporizing the solvent, and condensing the solvent into a clean drum or returning it to the scouring unit. The lowest boiling point materials distill first followed by the higher boiling constituents. The high boiling point impurities and sludges remain at the bottom of the still. Distillation may be performed in either a batch or continuous mode.

In-house distillation is appropriate for pure solvents and for azeotropes; it is less appropriate for blends or formulations with stabilizers because the solvent must be reformulated before reuse. Stills can be steam-heated, electrically-heated, or gas-fired. Still bottoms that result from the distillation process can contain up to 50 percent solvent; they can be processed further by using steam stripping or thin film evaporation. They can also be sent to off-site recyclers who will reclaim additional solvent. Capital costs for distillation systems ranging in capacity from 0.5 to 100 gallons per hour begin at $5,000. The operating costs of the dGtillation apparatus includes labor, energy, cooling water, and maintenance.

Another possible method of reducing the amounts of waste being shipped off site is use of a mobile on-site solvent reclaimer. Mobile on-site recycling is simply a transportable solvent distillation system. The recycler pays regular visits to the plant to perform the recycling operation and the recycled solvent is returned to the unit or storage, while arrangement is made to dispose of the still bottoms. Most on-site solvent reclaimers charge a single fixed price for their service, which depends on the type and quantity of the solvent to be processed. The charges include the cost of sludge disposal, a payment for processing the contaminated solvent, and the replacement of virgin solvent. Mobile on-site reclamation offers a number of benefits including no capital outlay for equipment and elimination of training and other labor costs associated with on-site distillation.

Off-Site Recvcling

For generators who are not able either economically or technically to use in-house recovery techniques, off-site reclamation may be an option. The recycler, under a contractual agreement, picks up the generators contaminated solvent, which can contain 50 to 90 percent by weight recoverable solvent, recycles it, and delivers purified solvent back to the generator. The sludges that result from the off-site reclamation contain small concentrations of solvents and are usually sent for destructive incineration.

28

In areas with large numbers of users, solvent reclamation is practiced profitably. In rural areas where large distances limit collection and transportation of the waste solvent, off-site recycling is less profitable. Since the RCRA land disposal ban was implemented in November 1986, off-site solvent reclamation has gained widespread use because of the high cost of other disposal options such as incineration.

3.2 DYEING OPERATIONS

Aqueous dyeing processes can replace solvent-based processes, but can also contribute substantially to textile wastewater generation. Color is a visible problem and a high level of dissolved solids is expected. Carriers, which are essential for dyeing polyester, tend to create high levels of biological oxygen demand (BOD). With thermos01 dyeing of cottodpolyester blends, carriers may be avoided thus reducing BOD loads. The following sections discuss use df liquid ammonia dyeing and air lift dyeing. These two methods can be used to reduce potential wastewater generation. Many of the solvent vapor and solid waste reduction methods discussed under scouring also apply to dyeing.

1) Use of Liquid Ammonia Dyeing

This is a new and patented system that uses anhydrous ammonia to dye textiles and other fibers. It seeks to eliminate problems associated with water quality, temperature control, long dyeing cycles, steam for heating dye baths, and close pH control. As compared to the aqueous system of dyeing, this system uses liquid ammonia instead of water as a medium to carry dyestuffs to the fiber. Liquid ammonia is an excellent solvent, which dissolves the dyestuff and allows it to penetrate into the fiber and fabric. At this point, the dye particles have little or no chemical affinity for the fiber at low temperatures. The fiber is then subjected to live steam, which drives off the ammonia and leaves the dye chemically entrapped in the fiber.

Due to the low boiling point of liquid ammonia (28F below zero), the ammonia is driven off very quickly, resulting in extremely short dyeing times. The fiber or fabric is then given a light washing to remove excess dye. This system is said to be adaptable to cellulosic and

--noncellulosic fibers. It is compatible with 95 percent of existing conventional dyestuffs and has the possibility of eliminating most of the preparation processes for dyeing. The ammonia has also been found to impart a mercerizing and shrinking effect on cellulosic fibers, which is very similar to caustic mercerization.

29

The processing equipment necessary for this system includes a unit for the recovery of ammonia. The ammonia recovery system would require a large capital investment, but with the increasing cost of all chemicals including anhydrous ammonia, and the need to control ammonia emissions, it would be essential to include this recovery unit. While this system requires further experimentation and acceptance by the industry, it appears to have potential for replacing solvent-based dyeing. It could also result in increased productivity and decreased costs for dyes, chemicals, and energy compared to solvent dyeing methods.

2) Use of "Air Lift" Dye System

The traditional system of dyeing uses a basic liquid jet machine. In the air lift system, the fabric is camed through the machine in a jet of air instead of in a jet of water. This necessitates a large blower system but eliminates the need for a large water pump. Another major change is that the jet machine has to be made larger since the fabric is relatively dry and bulky instead of floating in the aqueous dye solution. The method of mixing and applying the dye to the fabric must also be altered. The dye is mixed with a foaming chemical and applied onto the fabric in the form of a mist.

In the air lift system, the basic dye cycle for polyesters, wool, and other synthetics consists of an application cycle and rinse cycle. In the application cycle, the dyestuff is mixed with the foaming chemical and sprayed on the fabric so that a total of approximately 15 percent moisture is applied to the fabric. Because of the mechanical action of the foaming agent, heat and a frothy foam is created, which allows the dye to distribute evenly throughout the fabric. In the rinse cycle, rinsing is performed in the same manner as the application of dye, that is, by injecting a water spray into the air stream rather than using a high volume of water.

Total water consumption in this process is approximately 8 to 10 pounds of water per pound of polyester (1 gallon per pound) as compared with 40 to 50 pounds of water per pound of polyester (5 to 6 gallons of water per pound) on the regular jet machine. This represents a water reduction in the range of 80 percent. Besides eliminating the need for halogenated solvent in the dyeing process, this method of dyeing appears to be very important in reducing water and hence, heat requirements for drying. -.

30

P

4.0 ANALYSIS OF SOURCE REDUCTION OPTIONS

The flood of imports into the U.S. textile market, and the movement of industry out of this country to the sources of fiber has forced the remaining US. mills to become more efficient and reduce their use of solvent. In recent years, some users have converted back to water in cases where it is technically feasible to do so. While aqueous scouring generally requires pretreatment prior to sewer discharge, appropriate sludge disposal, and is more energy intensive, these conversions are expected to continue as regulatory scrutiny over halogenated solvent use increases.

For this study, no detailed cost analysis of proposed source reduction options was performed. Halogenated solvent use by the industry is relatively minor and at those facilities using halogenated solvents, controls are in place. The most common means of control are use of water cooled or refrigerated condensers for air emissions and use of on-site distillation for solid waste (i-e., du ty solvent from scouring). Readers interested in an economic analysis on the use of carbon adsorption for the recovery of PERC from air streams can refer to the companion Technical Support Documents for the dry cleaning industry and parts cleaning.

31

I

i

i

_-

(intentionally blank)

32

5.0 ESTIMATION OF SOURCE REDUCTION POTENTIAL

A standardized methodology was devised to determine ranges of source reduction potential for each option and then an overall reduction for the industry. The use of ranges, as opposed to fixed values, was deemed more appropriate since the surveys conducted in this study, and from which the range estimates are derived, were limited in geographic range and number (a fuller discussion of limitations is presented in the Summary Report for these Technical Support Documents). The results of this exercise are presented in Section 5.1 followed by the discussion of methodology in Section 5.2.

5.1 QUANTIFICATION

To quantify potential source reduction, each option discussed in Section 3 was assigned a ringe of effectiveness and implementation potential in accordance with the methodology discussed in Section 5.2. The top one or two methods for each impacted medium were then combined and the medium reductions combined to yield reductions for a given operation (i.e., scouring or dyeing). Operation reductions were then combined to yield an overall reduction for the industry. Since only the top one or two options were selected, as opposed to selecting all options, the estimated reductions are conservative. The results of this exercise are presented in Table 5.1.

The highly competitive nature of the textiles industry, and declining U.S. market share, have generally meant that the industry has been capital-limited in efforts to pursue source reduction. Analysis, nevertheless, indicates that additional implementation of solvent vaporhaste recovery along with improved housekeeping could achieve reductions of 0 to 23 percent short-term (less than 5 years), 0 to 26 percent medium-term (5 to 10 years), and 12 to 40 percent long-term (10 to 20 years). This represents reductions of 0 to 2,400 MT, 0 to 2,700 h4T, and 1,200 to 4,200 MT, respectively.

These reductions should also be viewed in terms of the reductions already achieved by the industry. Use of PERC alone was estimated to be 18,000 to 20,000 MT in 1977 (SRI, 1982). Comparing this to the total usage of halogenated solvent in 1988 of 10,500 MT, the industry

l k s already achieved a 50 percent reduction. This reduction was most likely achieved by closing of out-dated facilities, consolidation of existing facilities, conversion to aqueous processes, and installation of solvent vapor controls and waste recycling systems. The actual role each one of these actions played in reducing solvent use is not known.

33

- 3 5 2 .- 8 Y Y Y Y 3 E

0000

34

t

5.2 METHODOLOGY

The methodology used to estimate nationwide source reduction potential is comprised of the following major steps: assign source reduction effectiveness ranges to each option, assign implementation potential ranges to each option, compute the range of source reduction potential for each impacted medium, and then compute the range of source reduction potential for the industry. Definitions and examples are provided in the following sections.

Source Reduction Effectiveness

Source Reduction Effectiveness (SRE) measures the pure technical effectiveness of a source reduction option for a unit operation, for example, the extent to which the use of the target substance can be reduced by application of a given source reduction option in an industrial process or operation at a typical facility. The ranges of SRE used in this study were:

Reduction

None Small Moderate Medim High Very High Full Elimination

Range (percent]

0 0-20 20 - 40 40-60 60-80 80 - 100

100

For example, the use of water in place of halogenated solvent would be assigned an SRE of 100 percent since halogenated solvent use is fully eliminated. Likewise, the substitution of one halogenated solvent with another would be assigned an SRE of 0 percent because the option does not reduce halogenated solvent use. The assignment of SRE ranges ignores the effect of non-technical constraints on reducing source reduction potential.

Implementation Potential

The Implementation Potential (IP) is a percentage range estimate of the extent to which a particular source reduction option could be implemented on an industry-wide scale, taking into consideration many technical and nontechnical constraints. The ranges used for IP were identical to those used for SRE above. To account for time effects, IP ranges were

_-

35

determined for short-term (0 to 5 years), mid-term (5 to 10 years), and long-term (10 to 20 years) periods of time.

Estimating the IP for each source reduction option is a complex task, involving the meshing of environmental, technical, and socioeconomic variables into workable scenarios. Since this study has generally looked at source reduction technologies that have been technically demonstrated, then from a technical standpoint the IP of most options studied is relatively high. Other environmental and socioeconomic variables prominent in determining IP are:

Technology, however, is only one component of IP.

0

0

0

0

0

0 .

0

0

extent of current use economic considerations time customer attitudes regulatory compliance management capabilities and commitment information and training resources monitoring and maintenance capabilities

Extent of Current Use

Implementation potential for a given source reduction alternative depends on the extent to which this alternative has been implemented in the studied industry. For example, easy and accessible techniques such as improved maintenance or operator training have largely been implemented, especially by larger, more sophisticated firms. Hence, the implementation potential ratings for these alternatives would be lower, because a fraction of facilities where they can still be applied is probably small.

Economic Considerations

A contemplated source reduction project often finds itself as only one of many competitors for available capital. As with other activities requiring capital expenditures, two basic economic considerations govern source reduction: availability of capital, and the potential return on investment. Capital availability varies considerably among halogenated solvent users, with smaller operations generally at a disadvantage. To account for this effect, low cost measures were rated higher than high cost measures.

36

Time

Time can markedly change feasibility and potential of source reduction measures. Technologies can move from novel to standard; what may be considered economically infeasible today may be judged cost-effective tomorrow. Generally, well established technologies have been given a higher short-term IP rating than the emerging ones.

Customer Attitudes

If a service or a product developed through source reduction methods is less marketable than one from standard practices, the source reduction option will likely fail. Customer attitudes are an important influence in source reduction. SRRP research found that rigid customer specifications, particularly military specifications, pose significant source reduction barriers in many industries. These barriers tend to keep the IP rating for measures involving material or process changes low.

Regulatorv Comuliance

Complexities in meeting regulatory requirements for end-of-pipe pollution control generally spur source reduction. Both technological complexities in achieving specific emission reductions, and administrative complexities in obtaining required permits and approvals (e.g., obtaining of air permits), influence the rating of the implementation potential of source reduction measures.

Manapement - Capabilities and Commitment

Successful implementation of source reduction depends on a) the management level at which it is actively endorsed and b) the technical and administrative capability of that management in regard to implementation. Source reduction usually requires a reorientation of management perceptions and priorities away from traditional environmental management approaches, and toward pollution prevention. Some of the common barriers to such reorientation can be expressed by the following viewpoints:

_-

e We already invested in an expensive treatment system. Why reduce?

37

e We will not be the first to try this out. Let others do it, then we will see. 0 Product quality will definitely be affected by substitutions. We will not do anything

until we have to.

The rating of IP must consider resistance to change of managements perceptions. In deriving the rating, higher scores were assigned to industries where process and product changes are frequent than to more static industries.

Information and Training Resources

The best source reduction technologies available, supported by top management, can fail if sufficient information and training for effective implementation, running, and monitoring are not available to managers who must direct source reduction efforts and to line employees who must make them work, and adapt them to changing conditions. Information and training needs in the constantly evolving source reduction field are many. They range from simple instruction manuals to sophisticated databases. In general, the greater the requirement for information and training resources, the lower the IP rating.

Monitoring - and Maintenance Capabilities

Unless source reduction programs are carefully monitored, they can rapidly become suboptimal, posing a number of technical, economic, and other problems. Similarly, without proper maintenance geared specifically to source reduction goals, source reduction measures can falter. For small firms, the need for extensive monitoring and maintenance will lower an options IP rating.

Source Reduction Potential

The Source Reduction Potential (SRF) for each implementation time frame was calculated as the product of Source Reduction Effectiveness and Implementation Potential. As an example, an SRE of 20 to 40 percent and an IP of 60 to 80 percent would yield an SRP of 12 to 32 percent. The assignment of short-term, medium-term, and long-term IPs was done to account for reductions in constraints over time. While it could be argued that short-term, medium-term, and long-term SRE values should have been assigned to account for technical improvements over time, the added complexity and the movement away from a conservative stance did not appear warranted.

38

i

After calculation of individual SRPs, the top one or two options for each impacted medium were selected from among the non-solvent based alternatives. While the use of hydrocarbon solvents in place of halogenated solvents could achieve sizable reductions in halogenated solvent use, these options were not viewed as true source reduction since the overall environmental benefit is unclear. Also excluded were options that could pose major impacts on other media (i-e., air versus water). In general, most selected options dealt with use of similar products with less solvent content, more efficient use of products, and recyclinglreuse of solvent vapors and waste. The SlW for an impacted medium was calculated as follows:

where SRPl2 are the two leading calculated source reduction potentials for each medium. S W m d i , is the combined range of source reduction.

It should be noted that selection of one or two top ranking options in calculating an SRP for an impacted medium is arbitrary. If there are 10 source reduction methods available, and there are appropriate driving forces, then it is likely that more than two will be implemented, in which case the reduction in halogenated solvent use will exceed the estimate. Therefore, this technique provides a conservative estimation of SRP.

Overall Source Reduction Potential

Once the SWs for each impacted medium have been determined, the overall SRP for each industrial sector (e.g., formulator and user) and the industry as a whole are determined. The SRP for each industrial sector is determined by the equation:

where Wain Wsolid, and W,, k the weight of halogenated solvent released into the environment from a given sector. The industry-wide SRP is then determined by weighing each SRPsmtor by the ratio of solvent released from that sector by the total amount of solvent used by the industry.

39

(intentionally blank)

40

Ostrowski, P. February 12,1990 (Ostrowski, 1990).

Occidental Chemical Corp., Personal communication with SRRP staff,

Paul, H., et al., "Removal of Organic Vapors from Air by Selective Membrane Permeation," Daimler Benz AG, Stuttgart, West Germany, 1988 (Paul et al., 1988).

Ramsey, R. "Fabric Scouring with Trichlorotrifluoroethane," Technical Report RP-10 Freon Solvent and Chemicals, DuPont Chemical Company, May 1980 (Ramsey, 1980).

Ramsey, R. February 14 and 21,1990 (Ramsey, 1990).

DuPont Chemical Company, Personal communication with SRRP staff,

SRI International, "Chemical Economics Handbook - C2 Chlorinated Solvents," February 1982 (SRI, 1982).

TRI. U.S. Environmental Protection Agency, Toxic Release Inventory Data, "SIC Codes 2221, 2231, 2253,2257, 2258, 2269, 2262, 2281, 2284 and Tetrachloroethylene," 1987 (nU, 1987).

USDC. U.S. Department of Commerce." "County Business Patterns - US.," Bureau of Gnsus, 1986 (USDC, 1986a).

USDC, "County Business Patterns - California," 1986 (USDC, 1986b). USDE, U.S. Department of Energy, "Brayton Cycle Heat Pumps for Solvent Recovery and Recycling," (USDE, undated).

USEPA, U.S. Environmental Protection Agency, ttSource Assessment, Solvent Evaporation - Degreasing Operations," PB80-128812. Prepared by Monsanto Research Corp. for the Industrial Environmental Research Laboratory, Cinchatti, OH. August 1979 (USEPA, 1979).

USEPA, "Preliminary Analysis of Possible Substitutes for l,l,l-Trichloroethane, Tetrachloroethane, Dichloromethane, Tetrachloromethane, Trichloroethylene, and Trichlorotrifluoroethane" EPA 68-02-3 168, Prepared by GCA Corporation for the Office of Policy and Resource Management, Washington, D.C., May 1983 (USEPA, 1983).

U.S. Patent, 4,076,500, June 18,1976 (US. Patent, 1976).

Warner, J.J., "The Recovery of Dirt From Wool-Scouring Effluent by Treatment in a Decanter Centrifuge," Australian Textile, May 1985 (Warner, 1985).

Whitaker, N.T., and Stewart, R.G., "Treatment of Wool Scour Effluent by Anaerobic Digestion," Proceeding of the 7th International Wool Textile Research Conference, Tokyo, Japan, Vol. V. 1985 (Whitaker and Stewart, 1985).

Wijmans, J.G. et al., "Removal of Volatile Organic Compounds from Air Streams Using a Membrane System," MTRI, (Wijmans, et al. undated).

Wolf, IC, Institute for Research and Technical Assistance, "HCFC's: Are They Still Viable in Solvent Applications?," Second Annual International Workshop on Solvent Substitution held at Phoenix, AZ, December 10-13, 1991. Managed by Weapons Complex Monitor Forums (Wolf, 1991).

--

42

6.0 REFERENCES

Barnhard, E., "Textile Industries," EPA Waste Minimization Workshop, Atlanta, GA, May 1989 (Barnhard, 1989).

Camp, J., Gaston County Machines, Personal communication with SRRP staff, February 5, 1990 (Camp, 1990).

Christoe, J.R., and Bateup, B.O., "Developments in Wool Scouring - An Australian Perspective," Wool Science Review, December 1987 (Christoe and Bateup, 1987).

CMR. Chemical Marketing Reporter, "Production Profiles for TCE, TCA, PERC, and METH," January 23 and 30, February 6 and 20,1989 (CMR, 1989).

Cooper, S.G., The Textile Industry - Environmental Control and Energy Conservation," Noyes Data Corporations, Park Ridge, N.J., 1978 (Cooper, 1978).

Cui, KH., et al., "The Scouring of Raw Wool Using Freon TF Solvent," Magazine of Textile Chemist and Colorist, October 1988 (Cui et al., 1988)

Doshi, S.M. and Pillai, G.R., "Water Pollution in Textile Industry - I," The Indian Textile Journal, September 1982 (Doshi and Pillai, 1982).

Dow Chemical U S A , "Chlorinated Solvent Consumption in the U.S. Textile Market - Summary," May 18,1987 (Dow, 1987).

F", Federal Register Notice, "Advanced Notice of Proposed Rulemaking - Protection of Stratospheric Ozone," Vol. 54, No. 72, p.15228, April 17,1989 (FRN, 1989).

Liquid Carbonic. Product Literature on Liquid Carbonic's Environmental Vapor Recovery System. Form 6898,1989 (Liquid Carbonic, 1989).

Marr, A, Southern California Edison, "A Utility's VOC Recovery Program," Second Annual International Workshop on Solvent Substitution held at Phoenix, AZ, December 10-13, 1991. Managed by Weapons Complex Monitor Forums (Marr, 1991).

Meisser Griesheim, "Focus on Gas - Clean Air from Exhaust Air: Condensation of Solvents by Means of the Cryosolv Process," May 1988 (Meisser, 1988).

MTRI. Membrane Technology and Research, Inc., Vapor Separator Systems Product Literature (MTRI, undated).

Morehouse, Maj. Tom. HQ USAF/CEW. "Update on the Montreal Protocol and the Air Force Pollution Prevention Program." Second Annual International Workshop on Solvent Substitution held in Phoenix, AZ, December 10-13, 1991. Managed by Weapons Complex Monitor Forums (Morehouse, 1991).

_. Oakes, D.W., "Practical Applications of Solvent Emission Control Using Activated Carbon," Journal of Coated Fabrics, Vol. 16, January 1987 (Oakes, 1987).

Occidental Chemical Corp., "Chlorinated Solvents," Product Brochure, 1988 (Occidental, 1988).

41

1.0 INTRODUCTION2.0 BACKGROUND2.1 MARKET SEGMENTS2.2 SOLVENT USAGENUMBER OF FACILITIES IN OPERATIONPROCESS DESCRIPTION AND SOURCES OF RELEASE2.5 REGULATORY TRENDS2.6 INDUSTRY TRENDS

SOURCE REDUCTION OPPORTUNITIES3.1 SCOURING3.2 DYEING OPERATIONS

ANALYSIS OF SOURCE REDUCTION OPTIONSESTIMATION OF SOURCE REDUCTION POTENTIAL5.1 QUANnmCATION5.2 METHODOLOGY

6.0 REFERENCESNumber of Textile Establishments by SIC CodeSolvent Usehss Profile for the Textile IndustryHealth and Environmental Characteristics of Halogenated SolventsSummary of Source Reduction Options InvestigatedLow Molecular Weight Organic Solvents Used for ScouringSource Reduction Potentials for the Textile Industry