Document 4JEQZknx4MKjvQ0LJKOZJdN2N

JOINT &unne*t ^ttteMlqeace ^uCetm 2S September 21,1978 VINYL HALIDES CARCINOGENICITY H\ /H Vinyl Bromide /C = CN 2 v H Br URL 13667 Vinyl Chloride H \ C-- H ,H / Cl Vinylidene Chloride H\ H /Cl Cl U. S. DEPARTMENT OP HEALTH, EDUCATION, AND WELFARE Public Health Service Center tor Disease Control National Institute lor Occupational Selety and Health U.S. DEPARTMENT OF LABOR Occupational Safety and Health Administration JOINT NIOSM/OSHA CURRENT INTELLIGENCE BULLETIN: VINYL HALIDES - CARCINOGENICITY Vinyl Bromide, Vinyl Chloride, and Vinylidene Chloride September 21, 1978 The National Institute for Occupational Safety and Health (NIOSH) and the Occ upational Sdfety and Health Administration (OSHA) jointly recommend that vinyl bromide and vinylidene chloride be considered in the workplace as potential carcinogens to humans and controlled with the same degree of prudence as vinyl chloride, another vinyl halide currently regulated as a carcinogen by OSHA. This recommendation is based on the results of recent studies indicating that exposure to vinyl bromide and to vinylidene chloride causes angiosarcoma of the liver and other cancers in laboratory animals. Safe levels of exposure to carcinogens have not been demonstrated, but lowered exposure to carcinogens may in general decrease the probability of cancer development. Vinyl chloride is known to cause angiosarcoma of the liver and cancers of other sites in laboratory animals and in humans. At this time, adequate carcinogenicity studies of vinyl bromide and vinylidene chloride have been conducted only in laboratory animals. 2h view of the present state of knowledge in carcinogenesis, substances that cause cancer in laboratory animals are considered a potential cancer risk to humans. Vinyl chloride is the only vinyl halide for which an OSHA exposure standard currently exists. In light of the recent laboratory animal studies demonstrating carcinogenicity of vinyl bromide and vinylidene chloride, NIOSH and OSHA have jointly prepared this Current Intelligence Bulletin. Its purpose is to advise the occupational health community of the pertinent data and implications for exposed workers. NIOSH and OSHA request that producers, distributors, professional associations, and unions transmit the information in this Bulletin to their customers, employees, associates, and members. LABORATORY STUDIES Carcinogenicity Laboratory studies have demonstrated that exposure by inhalation to vinyl chloride (1,2), vinyl bromide (3), and vinylidene chloride (1,4) all caused angiosarcoma of URL 13669 Mutagenicity Several investigators have reported that vinyl chloride is mutagenic in Salmonella typhimurium and pombe (6-11) and in Escherichia coli (12-14). Vinyl chloride also has been shown to be mutagenic in the yeast mutation assay (11), in the Drosophila recessive lethal test (15,16), and in the host-mediated assay (11). Studies also have shown vinyl chloride to be mutagenic in Tradescantia (17). Three reports bearing on the mutagenicity of vinyl bromide have been noted to date. Bartsch et al., (18) and Simmons (19) have independently reported that vinyl bromide induced mutations in the bacterium, Salmonella typhimurium. La addition, Sparrow (17) has demonstrated a significant increase in mutants in Tradescantia exposed to vinyl bromide vapors. Vinylidene chloride has been shown to induce mutations in Salmonella tvphimurium (10, 20-22), in Escherichia coli (12), and in Tradescantia (17). Other Adverse Effects Other adverse health effects in animals attributed to exposure to vinyl halides include central nervous system (CNS) effects, cardiovascular effects, respiratory effects, skin effects, skeletal effects, and liver or spleen abnormalities (1). HUMAN STUDIES Carcinogenicity Studies of workers exposed to vinyl chloride have demonstrated an excessive risk of death from cancer of the lung, brain, lymphatic system, and angiosarcoma of the liver (1, 23). Cancers of the same sites were previously induced in animals following exposure to vinyl chloride (2). Liver angiosarcoma in humans is a very rare malignant tumor of the blood vessels. Though no clinical signs or symptoms, or laboratory examinations have been found to be specific for the early diagnosis of this cancer, affected individuals may complain of fatigue, abdominal pain, weight loss, anorexia, nausea, vomiting, melena, indigestion, jaundice, hematemesis, or diarrhea. Other manifestations may include liver enlargement and liver function abnormalities. In adults, untreated angiosarcoma of the liver usually is fatal within 8 months. Even with treatment, death usually occurs within 16 months. To date there have been no reported cases of cancer in humans associated with exposure to vinyl bromide or vinylidene chloride. However, vinyl bromide has been in commercial production in the U.S. only since 1971. Due to the long latent period characteristic of occupationally-induced cancers, typically 15-40 years, no unusual risk of cancer among exposed workers would be expected to be be detected at this time. Vinylidene chloride has been in commercial production and use since the early 1940,s. The only study (24) reported to date showed no excessive cancer risk among workers occupationally exposed to vinylidene chloride, but methodologic limitations of this study do not permit an adequate evaluation of the carcinogenic risk of vinylidene chloride to humans. 3 Mutagenicity and Reproductive Effects Cytogenetic studies have demonstrated a significant increase in the frequency of chromosomal aberrations in the lymphocytes of workers exposed to vinyl chloride (25-31). Further evidence for the mutagenicity of vinyl chloride has been provided by investigations showing an increase in fetal wastage among wives of male workers following occupational exposure to vinyl chloride (32,33). No studies addressing mutagenic or reproductive hazards among vinyl bromide or vinylidene chloride exposed populations have been reported. Other Adverse Effects Numerous other adverse health effects have been observed in humans exposed to vinyl chloride, as detailed in Table 2. Reports of effects on workers exposed to vinylidene chloride in combination with other vinyl compounds include liver function abnormalities, headache, vision problems, dizziness, fatigue, weakness, and neurological sensory disturbances. No similar reports for vinyl bromide exposure were found (1). \ J URL 13670 Table 2. Other Adverse Effects of Vinyl Chloride on Humans (1). System Adverse Effect neurologic dizziness, lightheadedness, dulling vision and hearing, drowsiness, headache, loss of memory, euphoria, nervousness, numbness or tingling in fingers or toes gastrointestinal nausea, loss of appetite, abdominal distress, varices of esophagus or stomach, black stools, bloody vomitus cardiovascular increased blood pressure, Raynaud's Syndrome hepatic liver enlargement, liver function abnormalities, increased sulphbromophthalein retention, liver damage, serum enzyme abnormalities respiratory coughing and sneezing, bronchial rales, emphyzema, pulmonary fibrosis, decreased respiratory function, lung function disturbances hematologic anemia, reticulocytosis, leukopenia, throm bocytopenia, splenomegaly dermatologic contact dermatitis, scleroderma-like skin changes musculoskeletal calf and joint pain, acroosteolysis other increased perspiration, cold sensation in fingers and hands, fatigue, weight loss, weakness, impotency ) 4 REFERENCES 1. U.S. Department of Health, Education and Welfare, Public Health Service, Center for Disease Control, National Institute for Occupational Safety and Health. Criteria for a Recommended Standard....Occupational Exposure to Vinyl Halides. (To be published late 1978 - Transmitted to OSHA September 12, 1978). 2. Maltoni, C: Predictive Value of Carcinogenesis Bioassays. N.Y. Acad. Sci. 271:431-447 (1976). 3. Huntingdon Research Center, HRC Project 7511-253. 18-Month Sacrifice Pathology Report, (Vinyl Bromide). New York. (June 26, 1978). 4. Maltoni, C: Recent Findings on the Carcinogenicity of Chlorinated Olefins. Env. Health Perspect. 21:1-5 (1977b). 5. Maltoni, C: Vinyl Chloride Carcinogenicity: An Experimental Model for Carcinogenesis Studies. In: Origins of Human Cancer, H.H. Hiatt, J.D. Watson, and J.A. Winston, eds., Cold Spring Harbor, 4:119-146 (1977a). 6. Bartsch, H., C. Malaveille, R. Montesano: Human, Rat and Mouse LiverMediated Mutagenicity of Vinyl Chloride in S. Typhimurium Strains. Int. J. Cancer 15:429-437 (1975). 7. Rannug, U., A. Johansson, C. Ramel, and C.A. Wachtmeister: The Mutagenicity of Vinyl Chloride After Metabolic Activation. Ambio. 3:194-197 (1974). 8. Garro, A.J., J.B. Guttenpaln, and P. Milvy: Vinyl Chloride Dependent Mutagenesis: Effects of Liver Extracts and Free Radicals. Mutat. Res. 38 (2):81-88 (1976). 9. Andrews, A.W., E.S. Zawistowski, and C.R. Valentine: A Comparison of the Mutagenic Properties of Vinyl Chloride and Methyl Chloride. Mutat. Res. 40:273 (1976). 10. McCann, J., E. Choi, E. Yamasake, and B.N. Ames: Detection of Carcinogens as Mutagens in the Salmonella/Microsome Test: Assay of 300 Chemicals. Proc. Nat. Acad. Sci. 72 (12):5135-5139 (1975). 11. Loprieno, N., R. Barale, and S. Baroncelli: Evaluation of the Genetic Effects Induced by Vinyl Chloride Monomer (VCM) Under Mammalian Metabolic Activation: Studies In Vitro and hi Vivo, Mutat. Res. 40:85-95 (1976). 8 IDENTIFIERS AND SYNONYMS FOR VINYL CHLORIDE Chemical Abstracts Service Registry Number 75-01-4 NIOSH RTECS Number KU96250 Chemical Formula C^H^Cl Chlorethene Chlorethylene Chloroethene Chloroethylene Ethene, ChloroEthylene, ChloroEthylene Monochloride Monochloroethene Monochloroethylene Trovidur VC VCM Vinyl Chloride Vinyl Chloride Monomer Vinyl C Monomer IDENTIFIERS AND SYNONYMS FOR VINYL BROMIDE Chemical Abstracts Service Registry Number 595-60-2 NIOSH RTECS Number KU64000 Chemical Formula C^H^Br Bromoethene Bromoethylene Ethene, Bromo- Ethylene, BromoNCI-C50373 Vinyl Bromide IDENTIFIERS AND SYNONYMS FOR VTNYUDENE CHLORIDE Chemical Abstracts Service Registry Number 75-35-4 NIOSH RTECS Number KV92750 Chemical Formula C^H^Cl^ 1.1,-DCE 1.1,-Dichloroethene 1,1 ,-Dichloroe thylene Ethene, 1,1,-DichloroEthylene, 1,1,-Dichloro NCI-C54262 Sconatex Vinylidene Chloride Vinylidene Chloride (ID 11 CUMULATIVE LIST OF NIOSH CURRENT INTELLIGENCE BULLETINS * I. * 2. * 3* * 4. Chloroprene Trichloroethylene (TCE) Ethylene Dibromide (EDB) Chrome Pigments * 5. * 6. * 7. Asbestos Hexamethylphosphoric Triamide (HMPA) Polychlorinated Biphenyls (PCBs) 8. * 9 10* 11. 12* 13. 14* *15* * 16. * 17. * 18. * 19* * 20. 21* * 22. * 23* * 24. * 25. 26* * 27. * 28. 4,4-Diaminodiphenylmethane (DDM) Chloroform Radon Daughters Dimethylc&rbamoyl Chloride (DMCC) Revised Diethylcarbamoyl Chloride (DECC) Explosive Azide Hazard Inorganic Arsenic - Respiratory. Protection Nitrosamines in Cutting Fluids Metabolic Precursors of a Known Human Carcinogen, Beta-Naphthylamine 2-Nitropropane Acrylonitrile 2,4-Diaminoanisole Tetrachloroethylene (Perchloroethylene) Trimellitic Anhydride (TMA) Ethylene Thiourea (ETU) Ethylene Dibromide and Disulfiram Toxic Interaction Direct Black 38, Direct Blue 6, and Direct Brown 95 BenzidineDerived Dyes Ethylene Dichloride (1,2-Dichloroethane) NIAX ^Catalyst ESN Chloroethanes: Review of Toxicity Vinyl Halides - Carcinogenicity January 20, 1975 June 6, 1975 July 7, 1975 June 24, 1975 October 7, 1975 October 8, 1976 August 8, 1975 October 24, 1975 November 3, 1975 August 20, 1976 January 30, 1976 March 15, 1976 May 11, 1976 July 7, 1976 July 7, 1976 August 16, 1976 September 27, 1976 October 6, 1976 December 17, 1976 April 25, 1977 July 1, 1977 January 13, 1978 January 20, 1978 February 3, 1978 April 11, 1978 April 11, 1978 April 17, 1978 April 19, 1978 May 22, 1978 August 21, 1978 September 21, 1978 NOTE: Bulletins #1 through #18 have been reprinted as a NIOSH publication, #78* 127, for the convenience of those that desire a complete series of Current Intelligence Bulletins. Distribution of this publication and single copies of Bulletins #19 and later are available from NIOSH Publications Dissemination, Division of Technical Services, 4676 Columbia Parkway, Cincinnati, Ohio 45226. Cancer related alerts ,.! Ml--IIIWlWT m-kSl+ IJRL V3673 ) MAR 2 1973 - J- /(Ls 1Hr. i Petrochemical guide--20 Vinyl ghloriuic has a single important end use as monomer in production of pblyvinyl chloride (PVC' pnd vinyl copolymers. The U.S. production of vinyl ehloj-ide expanded at an annual rat ofM4 percent during jthe 1960s, reaching a level of 4 BiHio'rTpounds in 1970 and 1971. This increase has been accompanied by a steady fall in average F.O.B. selling price from 10 ccnts/pound to about 4.5 cents per pound in 1970 and 1971. As a result, the imputed value of production has expanded for the decade at a 6 percent rate to $180 million in 1970. Polyvinyl chloride and other vinyl copolymers are being substituted for older materials on several fronts. Through replacement of steel and iron in piping, wood in construc \ tion and packaging, glass and paper in packaging, and leather in clothing, polyvinyl chloride consumption will continue to expand at a rate of about 10 percent through 1975 and beyond. IVc expect 1975 to sec U.S. production who--futurexa of 5.6 billion pounds of vinyl chloride (VCM). This monomer will be sold at an average price of 4.5 ccnts/pound, and production will therefore be valued at about $280 million (T971 dollars). Production of PVC in Europe is even larger, with 1970 production approxi mating j.3 biliion pounds. European growth rates have v,rMC also been somewhat higher, with one enthusiastic source predicting a 1975 PVC production of 9.5 billion pounds.1 In all likelihood, European PVC consumption will grow less rapidly than in the United States because of the higher per capita consumption prevalent in some major markets in Europe, which makes growth more difficult. Manufacturing processes for VCM have changed rapidly since 1965. Several new ethylene oxychlormation processes have been employed to shut down older acety lene-based production. These large, new processes now account for over 80 percent of U.S. capacity, and the same process should continue to be employed in expan sions through 1975. Recent announcement of an ethane- Here is how manufacturing methods, markets, buyers and sellers and economics affect the future of vinyl chloride based process (TRAXSCAT) has raised the yct-unproved possibility of bypassing ethylene cracking cntirelv, but even a viable TRAXSCAT cannot greatly influence the prc-19"5 VCM business. This process switch has also resulted in large changes in the cast of players. In general, some users of VCM have ceased to produce their own monomer, with pro duction undertaken by large, integrated producers of ethylene and chlorine. Dow and B. F. Goodrich arc the largest producers and have produced VCM for quite, a few years. However, PPG, Shell and Conoco have become David P. Keane, Robert B. Stobaugh and Phillip L. Townsend, Harvard University Graduate School of Busi ness Administration, Boston, Mass, the third, fourth and fifth largest VCM producers despite not having been in the business in 1965. We expec t this trend toward concentration of large VCM plants among feedstock producers to continue through the next .several plants. The PVC business lias also grown at a rapid rate else where in the uotkl. Because ot difficulty and expense of shipping, export of VCM lias usually constituted a limited i 3 Circle 9G on Reader Service Card V.NYL CHLORIDE market for U.S. producers, with exports during the 1960s accounting for only a few percent of U.S. production. In 1970 and 1971, however a lack of adequate European capacity caused U.S. VCM exports to increase to some J5 percent of domestic VCM production. This proportion will probably return to less than 6 percent by 1975. MANUFACTURE Most present-day vinyl chloride monomer (VCM) capacity relies on the so-called balanced oxychlorination process to convert chlorine and ethylene to VCM. There are, nonetheless, several older process routes which are discussed below in the approximate order of their com mercialization. There arc instances where the first two steps of this process--chlorination of ethylene to EDC and cracking of EDC to yield VCM plus hydrochloric acid--have been applied without an acetylene unit. Ethyl Corp. lias applied this technique, with byproduct HC1 used to pro duce ethyl chloride for manufacture of tetraethyl lead antiknock compounds. Most producers, however, cannot justify sufficient end uses of the HC1 byproduct. The result was the balanced cthylcnc/acetylene complex (as employed by Union Carbide and Monsanto at Texas City, Texas). Plants were sized so that the acetylenebased production consumed most excess hydrochloric acid fom EDC cracking to produce VCM directly. This process reduces by half the dependence upon acetylene as a feedstock, but no new examples have been con structed in the United States in the past few years. 1, Catalytic addition of hydrochloric acid to acetylene HC = CH -(- HC1 Act"~atl carbon > ^ _ CHQ -r HgUl- Acetylene Hydrochloric acid VCM The earliest plants of this sort employed carbidederived acetylene and hydrochloric acid derived by com busting hydrogen with chlorine. The catalyst for this addition is mercuric chloride absorbed on a carrier such as activated carbon. The reaction is simple and of high yield when compared to subsequent VCM processes, thereby allowing simple product purification, no sizable waste disposal problem, and low capital and operating costs. Carbide acetylene and purposely produced hydro chloric acid were gradually replaced with less expensive petroleum-derived acetylene and byproduct hydrochloric acid (which also coincided with major relocations to the U.S. Gulf Coast), but raw material costs remained high relative to newer processes. Some VCM is still produced with this process, but these surviving plants are very' tightly integrated with other acetylene-related units and have been gradually disappearing. 2. Balanced ethylenc/acctylcne VCM production CH5 = CHj + Cl------------------- CH- Cl - C'HnCl Ethylene Chlorine Ethylene dichloride (EDC) 900-950F CHj a - CH* Cl CH, = CH Cl + H Cl EDC VCM Hydrochloric acid HC aCH + H Cl------------------ CHi CHC1 Acetylene Hydrochloric acid VCM 3. Balanced oxychlorination process Direct chlorination CH* = CHj + Cl,------------------ CH,C1 - CH,C1 Ethylene Chlorine EDC Oxychlorination CHa = CHj + HC1 + JO, -> CHiCl - CHaCl + H,0 Ethylene EDC Water Cracking QAlXQ Sf)F CHiCl--CHaCl - ----> CHi = CHC1 + HC1 EDC VCM The difference between these processes and the earlier uses of EDC cracking is the second reaction above to (i) prevent exporting a large HC1 byproduct and (2) elimi nate the need for acetylene-feedstock. This reaction is essentially a variant of the Deacon process to convert HC1 to more valuable chlorine. 2 HC1 + } 02-------------- Cl, + HjO While the Deacon process has not been a commercial success because of technical problems with corrosion and product purification expense, oxychlorination has over come these problems by immediately capturing the chlorine in situ to form EDC. The combined EDC streams are then cracked and the resultant HCl recycled to close the loop. As mentioned above, most presently employed processes--including Goodrich, Do\v, Stauffer and PPG--use some variation of balanced oxychlorination. Fig. 1--Vinyl chloride monomer process via ethane as offered by Lummus/Armstrong. 1 AFb .1 .. . 1A .UD OHCL REACTOR OHCL PRIMARY RECOVERY OmCl SCCONOARY RECOVERY URL 13676 DIRECT CHLORINATION REACTOR EDC PURIFICATION CRACKING FURNACE HCI COLUMN VCI COLUMN Fig. 6-2--Vinyl chloride by oxychlorination by B.F. Goodrich. 4. Single-step chlorination and .cracking of ethane (TRANSCAT process) future, should the process prove to be commercially viable. CHj--CHj molten salt Ethane CHS = CHj Ethylene CHj - CH, + Clj------------------ CHj Cl--CHj Cl Ethylene Chlorine EDO CH, Cl--CH, Cl salt. EDC CHj--CHC1 + HC1 VCM Hydrochloric acid2 2 HC1 + \ O, moUcn 53,1 > Cl, + HjO The recently announced TRANSCAT process involves the development to a pilot stage of an ethane-based route which foregoes the separate purification of ethylene (see Fig. 6-1). Ethane, chlorine, air and excess hydrochloric acid (if desired) are fed to a molten salt bath with an extremely short residence time and a high temperature, where all of the above reactions apparently occur simul taneously. The potential licensors (Lummus/Armstrong Cork) claim above 95 percent VCM yield on chlorine and 80 percent VCM yield on ethane fed. Even more chemically revolutionary is the claim that chlorinated wastes which arc produced may be partially recycled to the reactor to result in additional VCM. The indicated economic advantage (see "Economics") of ethane as a feedstock might make this the VCM process of the Popular VCM processes. While oxychlorination has been the process route chosen for all recent VCM plants in the United States and most of the rest of the world, this has not decreased the competition among different processes and licensors. There are several competing processes which have demonstrated their commercial feasibility. The fol lowing three oxychlorination processes--on which signifi cant information has been published--are representative of most manufacturing facilities. 1. B. F. Goodrich oxychlorination (Badger). This process was the first successful oxychlorination process (1965), with the initial 400-million-pound/ycar unit in stalled at Goodrich's plant in Calvert City, Ky. There are about 8 plants which utilize this process, having an aggregate capacity of about 4 billion pounds/year.s As is true of all balanced oxychlorination processes, the over-all material balance involves feeding ethylene and chlorine and production of vinyl chloride (see Fig. 6-2). Yields arc certainly in the 90 percent range (and may exceed 95 percent) on both primary feedstocks. The yield losses involve production of light and heavy ends which boil above and below EDC. These byproducts consist pj'imarily of q and C: chlorinated organics, some of which arc. suitable as feedstocks for production of chlorinated solvents (carbon tetrachloride, pcrchtorocthylenc), while tiic balance require disposal. The first of three segments of the process is direct TT. ... T>, Fi'hritnvv' 1071 101 VINYL CHLORIDE addition of chlorine to ethylene to produce ethylene dichloride (EDC). This reaction is essentially stoichio metric ;in both reactants, with chlorine usually maintained in a slight excess. The reaction is liquid phase with a dissolved catalyst and mildly exothermic. Reactor tem perature and pressure are controlled by cooling water heat removal. Following the reactor, the EDG product is fed to a two-step purification train to remove light and heavy ends (the same train processes two other EDC streams). The second major operation consists of cracking EDG to produce vinyl chloride and anhydrous hydrochloric acid. The cracking occurs at 900-1,000 F in a directfired furnace which is packed with a catalyst (pumice has been a traditional choice). Conversion of EDC is typically in the 50-60 percent range to optimize the costs between coking cycle, utilities cost and yield. The hot product gases are quenched by direct contact with a condensed recycle stream of the same composition, prior to fractionation. In the HC1 column, anhydrous HC1 is removed overhead for use in the oxychlorination unit. The VCM column produces VCM as an overhead prod uct meeting finished product specifications, and produces a recycle EDC stream of unconverted EDC. This stream is purified in the common EDG columns prior to recycl ing to the cracking furnaces. The purification step is required to prevent fouling of reaction surfaces in the cracking furnace. These two previous process sections offer relatively little advantage over older processes--they constitute an unbalanced EDG cracking operation (such as practiced by Ethyl) which produces HCl as a byproduct. The third section of the B.F. Goodrich and other oxychlorination processes involves conversion of this excess HCl to EDC. Air, ethylene and IIC1 are charged to a fluidized catalyst bed at a moderately high temperature and somewhat elevated pressure. The reaction to produce EDC (given above) yields water byproduct, which is rejected as the product vapor is condensed. The first step of condensa tion produces a crude EDC product, while the gases (primarily diluent nitrogen) must be fed to a secondary absorber to recover a second stream of EDC by absorp tion/stripping. The tail gases are vented, while crude EDC is fed to the common EDC finislung train. The oxidation of HC1 to Cl- is highly exothermic and the subsequent addition of Cl- to ethylene is mildly exo thermic. As a result, the oxychlorination reactor, is cooled by generating steam, which brings the whole process closer to self-sufficiency on steam supply. The net effect of this crucial section is thus the con version of HCl, ethylene and air to a VCM precursor. While the process is depicted as balanced, design and operating changes allow any individual plant to produce or accept either HCl or EDC as local conditions warrant. 2. Toyo Soda oxychlorination (see Fig. 6-3). A pro cess very similar to that of B. F. Goodrich has been employed in several Japanese plants. The essential de sign differences appear to be: (1) the use of a fixed bed reactor for oxychlorination, as opposed to the B. F. Good rich fluid bed; (2) use of a hot oil reactor coolant and external steam generation, and (3) incorporation of a specific EDC dehydrator column not shown in the B. F. Goodrich publications. Nunc of these differences are CHLORlNATOR ABSORBER STRIPPER HEAVIES COLOUR CRACKER QUENCHER HCl COLUMN URL 13677 Fig. 6-3--Toyo Soda's oxychlorination vinyl chloride process. 102 February 1973 Hydrocarbon Processing major; the processes should be roughly competitive based upon published data.* 3. Stauffer Chemical oxychlorinntion. This is again very similar to the earlier oxyclilorination processes. Minor changes in the routing of streams, small differences in catalyst performance, and features of the mechanical design arc all that differentiate between the published information for this process and tire previous two. It is noteworthy that Stauffer published an estimate of 3.7 billion pounds/year of vinyl chloride capacity in 17 plants utilizing this process.4 * * 4. Monsanto oxyclilorination. There are several licen sees for this similar process.* 5. Dow oxyclilorination. Only Dow and foreign sub sidiaries have used the Dow VCM process, and no indi cation of a willingness by Dow to license to others has been published. 6. Union Carbide (Lummus) balanced acetylene/ ethylene process. No recent examples of this process have been constructed in the United States, but a Union Car bide venture using a WulfT process acetylene/ethylcne plant in Brazil is employing this process to consume part of the production. Complete startup of this plant is planned for 1972. The basic flow scheme is shown in Fig. 6-4. Ethylene and chlorine produce EDC, which is cracked to yield HC1 and VCM; the process is very similar in this area to die above processes. However, instead of using oxygen and ethylene to convert the byproduct HCl to EDC, the balanced process then involves the older acetylene route. The reaction is a vapor phase reaction over HgCI2/carbon catalyst with a slight HCl excess. A circulating coolant removes reaction heat to control the reaction temperature to about 400 F. Conversion is less than complete in some plants, requiring recycle of acetylene and HCl, but the reaction yields are 95 percent or greater. Therefore, the removal of light and heavy byproducts is simple. This results in low capital and operating costs for the process, but the use of acetylene at current or projected U.S. prices makes this process less than competitive with balanced oxychlorination (sec "Economics"). 7. Others. Pechiney-Saint-Gobain offers combined production of chlorinated solvents and vinyl chloride in flexible proportions. A single plant at Saint Auban, France, has operated since tire spring of 1970 to produce about 120,000 metric tons/year of VCM with this pro cess.* Diamond Shamrock/deNora offer "Dianor," a process to produce and crack EDC which is aimed at developing countries with no ethylene complexes.7 The process op erates on ethylene as low as 60 percent concentration and produces HCl byproduct, which would lease the process uncompetitive under any but the special circum stances of small, developing chemical markets with tariff protection. Feedstock availability. The importance of feedstock cost to VCM producers lias resulted in the entry to the VCM market of large, integrated chlorine and ethylene producers (see "Individual Companies" and "Econom ics"). Future manufacturing efforts will be largely infiu- Flg. 6-4--Author's Interpretation of the balanced VCM process as announced by Union Carbide (Lummus). TABLE 6-1--Properties of VCM* Mol. wt.......................................................................... Specific greviiy.......................................................... Melting point............................................................. Boiline point............................................. ................. Flashpoint.................................................................. Maximum allowable concentration (ppm by volume)................................................. Explosive limits % by volume in air................. 62.30 0.9334 -- I53.S C - 13.81* C -108*F 500 Lower 4 Upper 22 20*/20 *C {-244.6*F) (7.1 *F) enced by trends in both of these related fields. A brief review of trends follows: Ethylene production in the United States has been drifting away from the traditional patterns. Until recently more than 80 percent has been manufactured from ethane and propane cracking and about the same proportion located in the U.S. Gulf Coast. Some recent plants have involved movement toward the north and Puerto Rico and cracking of heavy feedstocks. With developing short ages of natural gas, a continuing erosion of the competi tive advantage of cracking Gulf Coast ethane and propane is likely, and cracking of heavier feedstocks will probably result in higher ethylene prices. With ethylene much more difficult to ship large distances than either vinyl chloride or chlorine, a tendency to locate vinyl chloride production near ethylene plants should continue. Chlorine is the other major feedstock, with VCM accounting for about 15 percent of U.S. chlorine produc tion in recent years. Largely as a result of slumping VCM demand (which declined silghtly in the first half of 1971), chlorine demand has been slack in the 19G9-1971 period. However, as various chlorinated products resume their growtli trends between 1971 and 1975, chlorine demand should once again require expanded capacity. Given the considerable economics of scale in chlorine/caustic pro duction, the largest and most integrated producers will retain their competitive advantage in VCM. Very substantial electrical energy requirements for chlorine production will force locations to sources of lowcost power. While nuclear fuel and coal are in the running as long-term suppliers of low cost power, petroleum and natural gas remain necessary until at least 1900. There fore, availability of petroleum and natural gas fuels will help determine chlorine plant locations until at least 1975. URL 13678 fi/p riin n VINYL CHLORIDE These trends indicate that manufacturing locations for VCM are likely to be heavily Influenced by availability of inexpensive hydrocarbon feedstocks and fuels, with relative labor and construction costs, water and transpor tation facilities acting as less important constraints. Physical properties. Sec Table 6-1. MARKETS U.S. consumption of VCM was about 3.3 billion pounds in 1970, with 664 million pounds being exported. Con- Fig. 6-5--Per capita consumption of polyvinyl chloride.1* APPAREL CONSTRUCTION 9 5 FLOORING <7 ............. HOME FURNISHINGS PACKAGING PIPE ft FITTINGS RECOROS TRANSPORTATION 16 4 5 5 9 ^...... WIRE a CABLE 12 6 . 12 (5 ...... 8 12 4 .......... 7 13 5 10 9 12 II 14 6 10 ALL OTHER 18 1965 17 1970 20 - <975 Fig. 6-6--End use markets for PVC.1* sumption of VCM is almost entirely for production of polyvinyl chloride resins and copolymer resins (propylene, ethylene and vinyl acetate arc commonly copolymerivcd with VCM). Domestic consumption is projected to grow about 10 percent annually from the 1970 level of 3.3 billion pounds to 5.3 billion pounds in 1975. Export markets will decline front the 16 percent of U.S. produc tion experienced in 1970 to the more usual level of 5-6 percent of production, or 300 million pounds, by 1975, End uses of PVC are characterized by more variety than most other commodity resins. Physical properties of PVC vary from the soft and flexible plasticized varieties used in dolls to the strong and rigid PVC pipe invading the construction markets. Fig. 6-5 demonstrates the rapid growth in per capita consumption which has resulted from this variety of properties.10 As a result, PVC con sumption is also less vulnerable to the loss of any single end use market. The more rapidly growing markets for PVC are con struction products, packaging, pipe and fittings. These areas should exceed the 10 percent growth rate of the overall PVC market. Segments which are more mature and will grow at less than 10 percent include apparel, flooring, home furnishings, phonograph records, trans portation equipment; and wire and cable coatings. An estimate of the 1963, 1970 and 1975 market share of each category is given in Fig. 6-6. While some categories will have declined relative to all PVC consumption, ac tual volume sold in 1975 is projected to increase in all categories.1 Construction uses for PVC include vinyl-coated wall coverings, and strips of PVC sheet to serve as water stops in walls and weatherstripping. The largest potential, how ever, probably belongs to PVC siding and window frames, which are rapidly growing competitors of older wooden and aluminum products. The construction market should grow in excess of 20 percent annually for the next five years. The packaging application of PVC has been growing very rapidly in recent years. There are environmental pressures to restrict PVC content of packaging because of HCI released when packaging is incinerated, but such restrictions will act to slow this growth area before 1975, not to reverse the trend. Packaging consumption of PVC grew at above 25 percent annually in the last five years, and should grow at 15-20 percent for the next five years. After 1975, however, look for growth to slow considerably as effective control of HCI emissions causes other materials to replace PVC in some applications. Use of PVC pipe and fittings has benefitted from the accelerating change of U.S. building codes to allow plastic drain, waste and vent piping. A.D. Little has projected a 1975 consumption of 1 billion pounds of plastic piping, with PVC representing a large share.11 The PVC in this application is often blended with chlorinated polyethylene resin. Competition with acrylonitrile-butadicne-styrene (ABS) resins and styrene-acrylonitrile (SAN) resins will be important in determining the actual growth of this PVC application, but 15 percent is a likely growth rate if PVC prices remain below those of ABS and SAN resins. Among the more mature PVC markets, tisc of PVC in transportation equipment should continue at a rela tively high growth rate. Further penetration of the auto mobile market is not a major hope for PVC, since seat 104 February 1973 Hydrocarbon Processing TABLE 4-2--Pr9tulng method* for PVC'* Extrusion (v. ire, film, sheet and general extrusion)................... Colcndcring (film, sheet and coating).................................................. Molding (blow , injection, roto, compression)............................ Coating {dip, knife, roll, spray, lamination)............................. Other........................................................................................... Share of market, % 40 35 10 10 5 URL 13680 fig. 6*7--PVC production by type according to the U.S. Tariff Commission. covers, headliners and dashboards have all been heavily penetrated. As a result, this end use will grow at about the rate automobile production grows. While other trans portation uses will increase, look for this segment to grow at 8-9 percent over-all. Uses of PVC in apparel, flooring, home furnishings and wire coating are mature segments. Growth of these seg ments should keep pace with real GNP growth at about 5 percent annually through 1975. Another possible categorization of PVC uses is by the processing methods being used to meet the above markets. Table 6*2 presents a breakdown by processing methods, again reflecting the diversity of PVC markets. Polymer producers. One reason for the varied properties of PVC resin is the diversity of polymerization methods employed and another is the large number of PVC pro ducers, each with a slightly different product line. While 65 percent of VCM was polymerized captively in 1960, this percentage declined rapidly as older acetylenc-based plants were replaced with larger ethylene-based plants. Today companies such as Dow and Shell are exclusively merchant sellers of VCM, while PVC producers such as Monsanto and Union Carbide have discontinued VCM manufacture. As a result, only about 40 percent of VCM was captively consumed in 1970, and the percentage will probably be closer to 30 percent in 1975. The 21 com panies currently producing PVC are listed in Table 6-3, together with capacity by region. Of these 21 companies, 7 also produce VCM (see Tabic 6-5). A final characterization of PVC production is the type of polymerization employed. Suspension homopolymer resins have been an increasing fraction of PVC production in recent years, mostly at the expense of copolymer resins, while dispersion resins have maintained a relatively con stant share. Pig. 6-7 presents a breakdown of PVC by method of polymerization. World markets. During the 1960s, exports of VCM averaged less than 5 percent of domestic production. VCM must be shipped and stored either under pressure or in a refrigerated tank. The relative difficulty of shipping, low selling price in relation to freight cost, and ready avail ability of VCM technology has created a strong tendency to produce VCM locally rather than import for an ex tended period. lit 1969, 1970 and 1971, however, rapid expansion of European VCM demand and lagging pro duction capacity led to abnormally large imports fiotn the U.S. Exports accounted for a high of 16 percent of domestic VCM production in 1970. As added capacity TABLE 6-3--PVC producers in tho United State*,1* Jon. 1, 1971 Northeast Borden..................... DiamondShamrock............. Firestone.................. B. K. Goodrich Chemical.............. Goodyear Tire ft Rubber................. Great American Plastics................. Hooker...................... Monsanto................. Leominister, Mast. Delaware City. Del. PottscoHa. Pa. Ped ricktown. N. J. Niagara Falls. N. Y. Fitchburg. Mass. 12 companies 13 plants 1.300 MM lbs. capacity Pantawle................. Stauffer..................... Tenneco.................. Southeast Air Products............ Continental Oil___ Firestone.................. Pantatote................. Midwest Air Products............ Allied Chemical,... Borden....................... GB.enFe.rGaloToidrreic..h.......... Chemical.............. UniroyaJ................. Southwest Diamond- Shamrock............. fciihvi......................... Goodyear Tire & Rubber................. Union Carbide. ... Far West American Chemical.............. B. F. Goodrich Chemical.............. Keysor....................... Passaic. N. I. Delaware City. Del. Burlington, N. J. Hemington. N. J, Aberdeen. Miss. Perrwille. Md. Point PJeasant. W. Va. S. Charleston. W. Va. Calvert City. Ky. Painesville. Ohio Illiopolis, 111. Ashtabula. Ohio Henry. III. Avon Lake. Ohio Louisville. Ky. Painesville. Ohio Deer Park, Texas Baton K"u*e, La. Plaauemine, La. Texas City. Texas Long Beach, Calif. Long Beach. Calif. Saugus. Calif 5 companies 5 plants 450 MM lbs. capacity 6 companies S plants 1.000 MM lbs. capacity 4 companies 4 plants 00 MM lbs. capacity 3 companies 3 plants 200 MM lbs. capacity 21 total companies S3 total plants 3,550 MM lbs. total capacity TABLE 6-4--Wtern European PVC consumption (thousands of metric tans) Year........................................ 1972......................................... 1963......................................... 1904......................................... 1966......................................... 1907......................................... 1968......................................... 1970*....................................... Growth rate {1962-1970)....................... 1975 Consumption*............ Projected growth Rate (to 1973)*.............. EEC 506 610 709 843 950 1.061 1.227 1.505 1.598 15% 2,575 10% EFTA 227 2jo 303 339 359 401 472 525 557 12% 820 8% Spain 17 29 33 46 59 71 91 97 24% 220 16% Ocher 40 52 72 56 55 141 140 148 IS% 235 14% Total Western Europe 795 943 1,173 1.271 1.410 1.601 1.925 2/hil 2.400 15% 3.900* 10% Source: European Chemical News and Oil, Paint and Drug Reporter, Oct. 26. 19T0, plus author*' estimate#.* comes onstream in Western Europe and developing coun tries rush to build VCM plants, U.S. exports of VCM will decrease again to 5-6 percent of production or lower. U.S. exports of PVC resin are also minimal; in 1969 and 1970, about 5.5 percent and 6 percent of the U.S. PVC sold was exported. Two reasons for this arc the tn<* UHL I Jbo l VINYL CHLORIDE TABLE 6-5--U.S. produter* of VCM rapidly expanding domestic markets for PVO and the high relative freight costs. Furthermore, tarilT barriers are high--20 percent or more--between European coun tries, as well as between Europe and the United States.12 Despite this, the specific properties required by resin users remain the most difficult barrier to a large world trade in PVC. In a recent report of the Standard Research Institute" projections of the world output of PVC for 1970 and 1980 are compared. The percentages accounted for by each producing area are as follows: North America Western Europe Japan Others 1970 26.5% 41.3 16.5 15.7 1980 26.5% 39.7 17.9 15.9 100.0% 100.0% TroJuw Location Allied................ American Chemical. . , . (AH CO-Stauffcr) Monoehem.. , (Uorden- UnlroyaJ) Conoco................. Baton Rouse, La. Watson, Calif. Geirmar. La. Dow.................... Freeport Texas Ethyl.................... Baton Rouge. La. Coodrich.............. PPG...................... Shell...................... Tenneco............... Houston Nameplate rapacity (MM Ibs./yr.) 300 170 300 Process Oxyctilorinatlon Stauffer oxychlorination Acetylene 600 340 ISO 700 270 ISO 1,000 300 500 SOO (Expanding) SOO (1974) 255 Stauffer oxyclilorination Dow oxychlorination Dow oxychlorination Dove oxychlorination Ethylene/EDC cracking Ethylene/EDC cracking Goodrich oxychlorination (multiple train) Oxychlorination Oxychlorination Stauffer oxychlorination Stauffer oxychlorination Acetylene While special purpose resins sales to developing markets 1972 Total 5.S&S will continue to be an attractive market, PVC resin ex ports will remain a small market at 5 percent or less of Source: Many published e.timatej aa interpreted by the authors. Note that effective capacity probably does not equal nameplate capacity. A reasonable production. capacity ticure would probably be 60% of the above rates. Oil, Point amt Dtut Rtpofttr. Oct. 11, 1971. has the most complete listing. The European PVC market is considerably larger than the U.S. market (see Table 6*4), based upon Europe's large population and greater per capita consumption. rapid, with fully half of the 1965 capacity being shut down This will allow European producers to build large, com- in the same period. Eighty percent of 1971 capacity is ` petitive VCM facilities to serve their own markets. As less than 6 years old. i U.S. producers encounter higher ethylene costs based Among individual companies, Dow has about 19 per ! upon cracking heavier feedstocks and higher chlorine cent of VCM capacity, and about 23 percent of oxy .i costs from more expensive electric power, exporting to chlorination capacity. Dow's position as a capacity leader the European VCM market should cease to be attractive among ethylene, chlorine and chlorinated solvents pro to U.S. produceis.14 ducers leaves no doubt of Dow's long-term position as a Japanese PVC producers also represent a sizable market VCM producer. Raw materials costs are crucial to VCM for VCM, but domestic Japanese production is ample to economics, being roughly equally split between chlorine provide the estimated 2.4 billion pounds which will be and ethylene (see "Economics"). In addition, byproduct required in 1972. Estimated VCM capacity is in the chlorinated hydrocarbons from oxychlorination are rou neighborhood of 3 to 3.3 billion pounds/year (see Table tinely absorbed into Dow's production of pcrchloro- 6-6), indicating either sizable exports or low operating ethylene and carbon tetrachloride. At the present time, rates for Japanese VCM producers. this VCM is largely sold in the merchant market. While A partial listing of foreign VCM producers is given in Dow has extensive experience in production of bulk poly Table 6-6. Since this is not an exhaustive listing, total mers, Dow has chosen to remain a merchant supplier and capacity figures are not available. However, the per capita has discontinued production of PVC. consumption of VCM is even higher in Europe than in The second largest VCM capacity belongs to B. F. the United States and trends toward oxychlorination are Coodrich at Calvert City, Ky. Two or three trains at this i evident. site employ Goodrich's oxychlorination process (see "Man ufacture"), and are notable in being the only sizeable INDIVIDUAL COMPANIES plants located in the northern or eastern United States. Table 6-5 lists U.S. producers of VCM as of 1972. Of the total capacity shown, 10 percent is based on acetylene and 8 percent on ethylene without oxychlorina tion capability. These plants must be considered vulner able to continued construction of large oxychlorination facilities. The remaining 5 billion pounds (82 percent) consists of modem, apparently competitive oxychlorina tion plants. Also noteworthy is the concentration of over 70 percent of VCM capacity on the U.S. Gulf Coast. With the exception of PPG's Puerto Rican plant, all recent capacity additions have been in Texas and Louis iana. The expansion of VCM capacity from 2.4 billion pounds/year in 1965 to about 6 billion pounds/year by 1971 was at an annual rate of 18 percent. The actual rate of new capacity construction has been even more Goodrich has captive use for most or all of this VCM, but has made no recent move to expand beyond their existing 17 percent capacity share. Both ethylene and chlorine for this plant are produced by Goodrich at Cal vert City. PPG has moved into second place with a 14 percent capacity share. Plants in Lake Charles, La., and Puerto Rico benefit from large internal chlorine sources and chlorinated solvents business to absorb byproducts, but ethylene is purchased in both cases. As is the case with most recent expansions, the majority of VCM produced moves in the merchant market. The fourth and fifth largest producers arc oil com* panies with large ethylene capacity but insufficient chlorine-for VCM production. Shell has recently com pleted and started up 12-14 percent of U.S. industry 106 February 1973 Hydrocarbon Processing TABLE 6-6--Foreign VCM producers Country rt'ROI'E Belgium:. Czechoslovakia Finland: .. .. France:.............. Greece: Italy: Netherlands:......... Rumania:................ Spain:..................... Sweden:.................... U. K.i........................ USSR........................... West Germany: ., LATIN AMERICA Argentina:.............. Brazil:....................... Chile:.............................. ............ Mexico:................. ........ Venezuela:........................... AFRICA/MIDDLE EAST ASIA/PACIFIC Australia:........ Japan:................ Korea ? ... Taiwan: .. Thailand r Company BASF Sol vie T.imburgse Vinyl Technoexport Pekemn Oy DAUFAC Pcchiney-Saint-Gabian Solvay Ethyl Hellas ANIC Montedison Ruml.-inca Industrial Import Monsanto Ibcrica Oxycro* Viniclor Kemanord B. P. Chemical* British Geon ICI Techmashimport BASF CbemWerke Huels Hoechst Knapsack AG Wacker Chemie Electrocior SAIC Industriss-Dow Contortio-Paulista Union Carbide Empressa Xationatede Petroleo Petrorjuimica-Dow-ENAP PEMF.X B. F. Goodrich/IVP/others General Organization lor Industr Petkim Petrokimya Goodrich Asahi-Penn Chemical Central Chcm. Chiba VCM Chi'so Pctro-chemical iapanese Geon Lnnestafuchi Kaaei Mlzushima Kureha Mitsubishi-Monsanto Mitsui Mitsui Toatsu Nihon VCM Senpoku Petro-chemkals Toyo Soda Korean Pacific Chemical Chinese Petroleum Taiwan VCM Thai Plastic City Antwerp Jemcppe Tesjrnrierloo NovaIcy Porvoo LaVera Tavern Ravennt Porto Miffherm Cagliari Botlek Pemis Rimnico Vilcea Tarragona Puertollano M&rtorell Stenungsund Bagian Bay, Wale* fliilhouse Roncorn Djerilnski Gorki Marl Knapsack Koln Burghausen Capltan Bermudea Bahia Blanca , Elelor Sao Paulo N. A. Concepcion Concepcion Pamliof El Tablazo Alexandria Izir.it Allaga Yarimea Altona Gol Kawasaki Chiba Minamata Osaka Misuahima Yoklcaichi Noeoya lwakuni Nihon Osaka Tolcoyama Yokkaichi Ulsan KaohMung Toufen Bangkok Annual capacity (thousand metric tons) 120 200 200 50 S3 90 120 J20 200 25 250 2IS5O0 100 (expanded! 300 75 36 SO SO (planned) 65 75 200 140 N. A. N- A. 30 33 300 365 (expanded) 60 (planned) 100 N. A. 50 (planned) 50 (planned) 100 100 85 15 (plan) 15 (plan) 70 50 43 (planned) 30 95 (planned) 27 (planned expansion) N. A. 145 (EDC) 90 (EDC) 160 55 130 120 50 120 60 N. A. 80 120 110 100 GO 64 60 (planned) 40 Process/completion Stauffer Solvay--ICI Goodrich Goodrirh/I iocchst Solvay/ICI Etiiyleue fexpanding) Own (1972) F-lhylcnc-bJ`ed (1972) Balanced accti'lcne/ethylene Ethyl EDC cracking Stauffer (completion)? Goodrich (1971) PPG 0972) Goodrich/Stauffer N. A. Monsanto/Ethylene Goodrich Solvay/ICI (1972) Ethylene Goodrich Acetylene Oxychiorinalion OxycMorination P-S-G (1972?) Stauffer (1973?) Balanced Hub (1972) Goodrich/linechst Goodrich/Hoechst Acetylene (1971) ICI (1973) Dow (1974) Ethyl/Solvay/ICI Solvay/ICI Balanced ethylene*acetyTene (1972) N. A. Dow Scientific Design B. F. Goodrich (1973) N- A. Solvay/ICI 1974 1972 Goodrich PPG Toyo Soda Stauffer Toyo Soda (1972) Balanced. Goodrich Stauffer Monsanto/Scientific Design Balanced Mitsui Scientific Design Stauffer Mitsui Toyo Soda Toyo Soda Dow Monsanto/Tokoyama Union Carbide (1972) Dvnamlt Nobel Source: Authors' estimate based on many published sources. This I ist is not Intended to be comprehensive, but to list most major producers. c xi CO CD CO ro capacity at Houston. Ethylene comes from a gas oil cracker with 1 billion pounds/ycar capacity, but required chlorine is purchased externally. Shell produces chlorine and other chlorinated organics beside VCM, but has apparently chosen not to expand chlorine capacity for use in VCM. "With recent announcement of a similar scale VCM unit in Norco, La., Shell will become the largest VCM producer by 1975. Conoco has 10 percent of in dustry capacity at Lake Charles, with ethylene provided by ethane/propane cracking and chlorine requirements purchased. It is reasonable to watch major ethylene and chlorine producers for the next VCM units. HISTORICAL DATA Tabic 6-7 presents data based on U.S. Tariff Commis sion reports for 1958 through 1969 and the authors' estimates for 1975. These data show a steady rise in the production of VCM, excepting a slump in 1967, when the industry was particularly plagued with overcapacity. Pro duction lias grown at over 17 percent per year for the TABLE 6-7--U.S. historical data: VCM, 1959 to 1975 Year 1959., . 19110... 1961... 1962... 1963... 1964... 1965.. . 1966 . . 1967. . . 196S... 1969. . . 1970... 1975... Number of producer* 10 12 IS 12 13 13 13 13 13 12 11 9 10* Production MM Ib./yr. 9 78 1,037 1.044 1,311 1,435 1.614 2,000 2,500 2.4'>4 2,969 3,i36 4,000 5,COO* Sale* MM Ib./yr. Averace price, f/lb. Total value of prod'af $MM 329 352 424 516 501 598 088 836 952 1,463 >.356 2 520* 4,200* 11 10 8.1 7.5 7.0 6.3 6.1 5.9 5-3 4.6 4.4 4.5* 4.5* 10S 104 85 98 10O 102 122 148 12R 136 164 ISO* 252* Source: U.S. Tariff Commission * Author'* estimates, with price projection in 1971 dollars. past five years. Average sales price has continually dropped because oxychlorinntion of ethylene, which is accounting for an cvcr-incrcasirg proportion of VCM production, is considerably cheaper than the acetylene route, and be cause of economies of scale. Because of these offsetting u VINYL CHLORIDE Fig. 6-8--VCM production in the United States, U.S. Tariff Commission and authors' estimates. trends, the total value of production has increased only 6 percent per year since 1964. However, it is believed that the average sales price (real dollars) will hold steady through 1975 because of rising ethylene and chlorine costs and that the growth rate of VCM production will be closer to seven percent per year from 1970 to 1975. The future of VCM is directly tied to the future of PVC, and PVC consumption in the United States is ex pected to show an annual growth rate of 10 percent from 1970 to 1975. The trends of VCM production, price and value are shown in Figs. 6-8, 6-9 and 6-10. ECONOMICS The price of VCM has exhibited the typical downward trend of maturing petrochemical monomers. The major reasons for this trend have been the consistently improv ing technology, the nmch larger scale of existing produc tion facilities, and the considerable decrease in the price of major feedstocks--ethylene, acetylene, chlorine and hydrochloric acid. As with other petrochemicals such as acrylonitrile, vinyl acetate, and neoprene rubber, the 1960s saw the introduc tion of new processes which replaced acetylene as a raw material with a less expensive feedstock. Even though the new balanced oxychlorination processes (such as Goodrich, Stauffer and Monsanto processes) have a higher capital cost (see Fig. 6-11) than balanced ethylene/acctylenc units of the same size, the raw material advantage of ethylene has been sufficient to more than justify the added capital. At ethylene and acetylene prices of 3 and 8 cents/pound, the switch in feedstocks reduces raw material cost by about 1.1 cents/pound at normal yield ratios. For a 500-million-pound/year plant, the raw material savings for balanced oxychlorination amounts to $5 million/year (90 percent capacity). This in turn is more than enough justification for the approximate $3-$4 million increase in capital cost. While two U.S. vinyl chloride producers (Tenneco and Monochem) have been able to continue operation of highly integrated acetylene complexes, the operation of these units can almost cer tainly only be justified with an out-of-pocket analysis of cost. No new acetylene-based plants will be built in the United States. The economics of a 600-million-pound/year balanced oxychlorination producer are shown in Table 6-8. This analysis presents a single year of the life of the plant at 90 percent of capacity, which can be justified as equally accurate with some of the component cost data. In any case, the economics indicate several notable characteristics of the VCM business. First, the two raw materials com prise two-thirds of the 8 percent profited manufacturing cost of VCM. With the large economies of scale in chlorine and the evident trend to oil company domina- UHL 13683 TABLE 6-8--Estimated cost of VCM production by ethylene chlorination and oxychlorination Component Uaaerate (per lb. VCM) 0.47 lb. 1.5 lb. .0018 MM Btu. 31 gal. 0.1 KWH (e/unit) 2.25 3.00 25 Manufacturing cose (Thousand */yr.) ()f/lb. VCM) $ 7.C50 7.600 550 450 200 550 500 1.42 1.41 o.ioo.os .04 .10 .06 .09 817.600 3.29 1,000 450 350 3.600 200 2 5.000 $23,700 1.09 4.38 Basis: Usage rates ami capital costs are derived from published claims. See especially Spits. Peter. "Vinyl Chloride Economics." Ck/mieal Enti-tetrint Projrerr March t'JiVH. C00-million-|*oun<l/year Gulf Conn plant (F. O. B.) Working capital one month of VCM sales value. Capital com -- 16 on-situs (million dollars) 8 of!-site* Note: 24 Total 11 year taxable life 15 year estimated useful life This is a stc.ul y-;tc calculation for one year at 00% of capacity; a more accurate discounted cash flosv might produce a notkably different price. No by product credits or disposal costs are included: credits for chlorinated solvents feedstocks are assumed to balance heavy and liyht ends disposal costs. t \e i tit t 108 February 1973 Hydrocarbon Processing TABLE A-B--Snl|vlty of VCM menufacrvring cost to assumptions A change In thta Input By this amount Chan|*e cost of \ CM Uy amount /lb.) Energy costs (fuel., steam, electricity).. (without increase in overhead) 0.2*> cents/pound 10 % 10% 0.02 0-07 Source: Derived from previous estimate of F. O. B. manufacturing cost. URL 13684 lion of ethylene production (primarily because of byprod uct handling from heavier feedstock cracking), the vinyl chloride business is the obvious domain of both large chlorine producers (Dow, PPG, Goodrich) and large ethylene producers (Conoco, Shell, Dow and Union Car bide). This definite trend should continue for the fore seeable future. A second notable feature (to the extent that these published data are accurate) is the low profit margin on recent VCM contracts. With the average price of VCM sales (reported by the U.S. Tariff Commission) ap proaching 4J/2 cents/pound, it is evident that new VCM plants must be operated very efficiently and with minimal startup difficulties to yield any reasonable return to VCM producers. The sensitivity of a VCM operation to the various manufacturing cost components is presented in Table 6-9. This indicates the probable upward pressure on VCM pricing in the event of increasing costs for fuel, construction, and ethylene which have been characteristic of 1970-1971. A recent development in VCM manufacture was the announcement ol a process to pioduce VCM directly from ethane, chlorine and hydrochloric acid. The "TRANSCAT' process, as it has been dubbed by its inventors and potential licensors at Armstrong Cork and Lummus, claims to have economics superior to the bal anced oxychlorination process. Yields and raw material prices have been reported as superior to oxychlorination and capital costs seem to be comparable. This process presents a potential reduction in manufacturing cost of about 1 cent/pound if these characteristics are present in commercial-sired plants.1'* No announcements of a commercial TRANSCAT venture have been made public to date. A final, developing factor in VCM economics is the production of 3-5 percent chlorinated byproducts by most oxychlorination processes. While most of these are suit able feedstocks for chlorinated solvents manufacture, a fraction are suitable only for disposal. As environmental laws impinge on the more economic means of disposal-- atmospheric venting, deep wells and deep sea dumping-- die VCM producers will be forced to more expensive disposal means. These costs will exert an unknown, but potentially significant, upward pressure on VCM price. fHE FUTURE Despite production volume which indicates approach ing maturity in its life cycle, vinyl chloride conrimtcs to grow at a rapid rate. This growth has resulted from extension of PVC and copolymers into new materials applications. Replacement of older materials--such as *tccl and iron in pipe, glass and paper in packaging--will continue to provide the major impetus to PVC and, 1959 61 63 65 67 69 71 YEAR 73 1975 Fig. 6-9--U.S. VCM average selling price, authors' estimates and U.S. Tariff Commission. 1959 61 63 65 67 69 71 73 1975 YEAR Fig. 6-10--Value of U.S. VCM production, authors' estimates and U.S. Tariff Commission. Fig. 6-11--Estimated battery limit capital cost lor VCM pro cesses.1* therefore, VCM growth. While these markets are themselves mature and growing only moderately, their very size allows 15-20 percent annual growth of PVC con sumed from only moderate inroads. At the same time, older PVC markets arc maturing noticeably and will exhibit growth of 5-8 percent. The upshot will be PVC consumption growth of 10 percent from 19G9 to 1975 and Hydrocarbon Processing February 1973 109 URL 13685 'W. CHLORIDE: a lower 7 percent growth of VCM production because of a decreasing proportion of exports. VCM manufacturing costs have been rapidly approach ing feedstock values. VCM sells for 4.5-5 cents/pound, ethylene for 3-3*4 cents/pound, and chlorine for 2-2.5 ccnts/pound in large scale contracts. Therefore, nearbalanced oxychlorinntion is not likely to be supplanted as the major VCM process unless the change to even less expensive feedstocks is involved. Not enough information is available to assess the commercial success of the re cently announced TRANSCAT process to convert ethane directly to VCM. If preliminary information is confirmed, watch for TRANSCAT to be employed on a very large scale. Location of new VCM production will be determined by availability of low-cost hydrocarbons, with VCM transportation by waLer, pipe line and rail acting as a constraint. If present erosion of U.S. Gulf Coast ad vantages continues, large VCM markets will draw new producers to the North and East. The companies which build the next few VCM plants will be those with captive supplies of ethylene, chlorine or both, and those with considerable VCM operating knowhow. Paper-thin profit margins almost preclude en try by any producer that lacks more than one of these three advantages. Look for Dow, PPG and Shell to About the authors David P. Keane is a product manage- k- z > Jjivis.on of the Singer Co.. He received an M.B.A. from Harvard Business School in 1970. He also holds an A.B. in economics from Georgetown University, Washing ton, D.C., and the "Certifical" from the University of Fribourg, Switzer land, in French literature. Robert R. Stobavgii is a professor at Harvard Justness School where he teaches a doctoral seminar in interna tional technology and production. He holds a B.S. in. chemical engineering from Louisiana State University and a recent doctorate from Harvard Busi ness School. He has served as a con sultant to a number of chemical and sA oil firms and governments and has en gineering experience with Monsanto, ^3." Caltcx Oil Group and Jersey Standard affiliates. He has written numerous articles and two books, Petrochemical Manufacturing and Marketing Guide, Volume I and II (Gulf Publishing Co.). Piiillip Townsend is an industrial con sultant and working toward a doctorate at Harvard Business School. His spe cial field is production and operations management, particularly t petro chemicals. He has held technical and managerial positions with IT./?. Grace, American OH find Shell Chemical. Mr. Townsend received a B.S. in economics and chemical engineering from M.I.T. and an M.S. in chemical engineering from Purdue. CUMULATIVE PRODUCTION EXPERIENCE, MILLION POUNDS Fig. 6-12--Experience curves for VCM price. PVC price, and value added by polymerizer. Source: U.S. Tariff Commission, Boston Consulting Group, and Manufacturing Chemists Asso ciation figures are combined with the GNP deflator (1958 base year) and authors' estimates of future production. A 3 percent inflation projection through 1975 is incorporated. continue expansion, with Union Carbide and Monsanto considering eventual reentry of VCM production. Intro duction of a successful ethane-based process, however, would cause chlorine producers to dominate future pro duction. Based upon these trends (see Figs. 6-4, 6-5 and 6-6) wc predict that production of "VCM almost exclusively from ethylene will reach 5.6 billion pounds by 1975, with merchant sales at a price of 4.5 cents/pound (1971 dol lars). This will result in an imputed value of $252 million for 1975 production (1971 dollars). These large contracts will be written with escalation clauses to offset increasing power and hydrocarbon costs. Fig. 6-12 presents the history of PVC and VCM pric ing, as well as the value added by polymerizers. Projection of the decreasing polymerization margin (using cumula tive production experience versus price) gives a further estimate that average 1975 PVC prices will approximate 11 cents/pound (in 1971 dollars).18 ii * i LITERATURE CITED 1 European Chemical News: Polymer Intermediate*, Oct. 30, 1070, p. 5. 1 Hydrocarbon Pretesting, November 1967, p. 239; Chemical Week, Auf. 29. 1964. * "Foreign Aid for Vinyl," Chemical Week, Sept. 24, 1966. 4 Hydrocarbon Processing, November 1069, p. 249. * Hydrocarbon Processing, November 1969, p. 248. 4 European Chemical News, April 30, 1971, and Oil and Gat Journal, Nov- 8, 1971. t Chemical Engineering, April 22, 19G8, pp. 142-144. * Freilins, Huson and Summerville. "Which Feedstock for Ethylene," Hydra, carbon Processing, November 1968. p. 149. * Faith, Keyes and Clark, Industrial Ckemicalt, Wiley, 3rd edition, 1965, p. 8U9. u "Polyvinyl Chloride: Outlook and Opportunities," John Auchter (B. C. Goodrich), Chemical Marketing Research Association, New York, May 6, 1971. "Chemical Week, Aug, 10, 1971, p. 26. ** Chemical Week, May 24, 1969, p. 32. uOif, /'aim and Drug Reporter, May 5, 1969, p. 3. >* The Oil and Gas Journal, March 8, 1971*1, p. 53. u For an explanation of the experience curve method of price forecasting, see the various publications of the Button Consulting Group, including, "Per* spcctivrs on Experience," Boston, 1968. " "Vinyl Chloride Lfonom'n.s," Brier Spitz, Chemical Engineering Progress, March 1968, p. 19-26 with appropriate escalation factors to 19711 applied by author, ~ END OF SERIES no February 1973 Hydrocarbon Processing H