An excerpt from the chapter, “The Polycarbonate Problem.”
BPA, Benzene, Phenols, & Carbonyl Chloride (also known as Phosgene)
Although it’s only in the past few years that news of bisphenol A’s health impacts began to reach a nonscientific general public–news that has since spread rapidly–it was first recognized as a synthetic estrogen in the 1930s. Papers published in the journal of Nature in 1933 and 1936 describe its estrogenic effects on lab rats. These papers also commented on the possible carcinogenic activity of materials with similar or comparable composition to bisphenol A–specifically materials synthesized from petroleum (from which bisphenol A is ultimately derived) and coal tar.
Some two decades later, bisphenol A was launched into everyday life with the development of commercially produced polycarbonates. Major production of these plastics began in the United States in the late 1950s after a General Electric engineer named Daniel W. Fox formulated a material based on BPA that GE called Lexan. The invention was not so much deliberately planned as it was the result of what Fox called his ability to take “a few clues and jump to conclusions that frequently panned out.”
While experimenting with different materials that might ultimately make a good moldable polymer, Fox decided to work with bisphenols, compounds derived from petroleum processing that were then being used to make various epoxy resins. As molecules, bisphenols have a structural feature that makes them useful as potential chemical building blocks. Attached to their hydrocarbon ring is what’s called a hydroxyl group, an oxygen and hydrogen that together form a site to which other molecules can bond. This structure is common to both synthetic and naturally occurring compounds, a coincidence that will later turn out to be important to how bisphenol A behaves.
Fox’s interest in the hydroxyl group was as a polymer building site, not for its biological activity. But when attached to a hydrocarbon ring as it is in bisphenol A, the entire chemical grouping becomes a molecule known as a phenol–an aromatic hydrocarbon, a ring made up of six carbon atoms and five hydrogen atoms plus a hydroxyl group. Phenols are commonly made by oxidizing benzene, which essentially means adding oxygen to benzene. Phenols are toxic, but they are also known for their antiseptic properties and so were used to kill germs in the nineteenth century surgical procedures.
This molecular group consisting of six carbon-five hydrogen rings with a hydroxyl group attached, however, is also part of the structure of substances produced naturally by the human body, compounds that include estrogen and thyroid hormones. Introducing a manufactured chemical that includes the phenol group into a cellular environment may therefore pose a problem because the synthetic material may compete biochemically with the similarly structured naturally occurring chemical. Thinking in green chemistry terms, the presence of a phenol group on a synthetic, therefore, should be a sign to investigate that substance’s potential as an endocrine disruptor.
The potential cellular toxicity of phenols has actually been known for decades. Research done in the 1950s, written about by Rachel Carson in Silent Spring, discussed the mechanisms by which pesticides constructed with phenols had the ability to prompt oxidation processes that upset cellular metabolism. These reactive chemical groups can disrupt formation of enzymes vital to energy production, which in turn may interfere with how an organism produces and differentiates cellular material. These processes of cellular reproduction are involved in virtually every bodily system, from how an individual processes sugars and calcium to how its reproductive system functions. Carson described the introduction of xeniobiotic phenols as thrusting “a crowbar into the spokes of a wheel. Had Fox been a green chemist, our current synthetic landscape might look very different.
But because Fox and his colleagues were focused on functional performance and on working with readily available chemical ingredients, bisphenols seemed a good choice. As an additional building block that might combine with the bisphenol molecules’ hydrocarbons to yield a useful polymer, Fox chose a chlorine compound called carbonyl chloride. Carbonyl chloride was then–and is currently–a common ingredient in the synthetics known as isocyanates that are used to make any number of products, including polyurethanes that go into varnishes, paints, and plastic foams. By the 1950s it was known that chlorinated hydrocarbons made useful synthetics so this was a logical route for Fox to follow–but no one had yet made the kind of moldable, shatter-resistant plastic that Lexan turned out to be.
If you’re building a polymer, a linked chemical chain in effect, you need lots of the same repeating pieces; ideally you’ll work with shapes that are easy to find and lend themselves to chemical bonding. It’s here that a Tinkertoy or Lego analogy comes to mind. To add pieces to a chemical structure, you need sites where new sticks and building blocks can be attached. So it was with the choice of bisphenols and carbonyl chloride, which lend themselves to such bonding and were both readily available industrial chemicals. Had Fox been practicing green chemistry, however, he would never–even with what was known in the 1950s–have launched a product that required copious quantities of carbonyl chloride.
Carbonyl chloride is also known as phosgene and is so toxic that it was used as a chemical weapon during World War I. The isocyanates it’s used to make are also highly toxic. One such compound, methyl isocyanate, was the gas involved in the deadly 1984 disaster at the Union Carbide plant in Bhopal, India. Lest anyone wonder if nerve gas is lurking in your bike helmet or CD cases, however, let me quickly explain that no phosgene or even any chlorine ends up in the final bisphenol A polymer; the chlorine compound is simply a reagent, an ingredient that enables the desired chemical bonding to take place.
Yet speaking to an interviewer in 1983, Fox acknowledged that using large quantities of a chemical such as phosgene was indeed hazardous. But, Fox continued, it “was not a totally frightening undertaking because we had good advice. I would say that we have been tightening up our whole phosgene handling ever since, investing in an awful lot of money in trying to make the stuff doubly safe and then triply safe and quadruply safe.” Still, the interviewer pressed, “Has there ever been a problem?” To which Fox responded, “We have had one or two small discharges. To my knowledge, I don’t think GE advertised it, but I think we probably had a ‘casualty’ from phosgene.” Did this give anyone second thoughts about going into business? “I don’t think it did,” Fox replied.
At the time Fox was working, new material inventions like carbonates were just that–inventions that came first, with applications and markets found later. “When we invented polycarbonates in the early 1950s we had a polymer with an interesting set of properties and no readily apparent applications,” Fox said in 1983. But what was known about polycarbonates’ behavior early on that might have hinted at what’s since been discovered about their physical and biological behavior” Could this information have been used to prevent what are clearly problems of chemical contamination? Endocrine-disruption science is relatively new, but some of what was known early on about bisphenol A and polycarbonates would seem to indicate a material perhaps not ideally suited for use, say, with food, heat, and dishwashing detergents.
That polycarbonates built from bisphenol A were vulnerable to certain detergents, solvents, and alkali solutions (household ammonia would qualify) has been known since at least the 1970s. Ammonium hydroxide (essentially a solution of ammonia in water) was discussed as a possible way to break polycarbonates down to its chemical constituents–for materials recovery and reuse and as a way to remove unwanted polycarbonate from another surface. It was also known that various additives used to modify polycarbonate mixtures could leach from the finished plastics when they came into contact with certain liquids. Documents filed with the Federal Register in 1977 list chloroform, methylene chloride, and chlorobenzene among these additives. (The U.S. Department of Health and Human Services considers chloroform and methylene chloride suspected carcinogens, while chlorobenzene is known to cause liver, kidney, and nervous system damage and produce a precancerous condition in lab rats.) Correspondence between GE Plastics Division personnel in the 1970s and 1980s also voiced concern over the presence of chlorobenzene in water stored in polycarbonate bottles (but not bottles made by GE as it happened) and about how the stability of these polymers might affect their ability to be used with food.
A memo circulated within the Lexan division of GE in 1978 also noted that “through reaction with water,” polycarbonate resin can degrade. “The two largest applications of Lexan resin for which hydrolytic stability is critically important are baby bottles and water bottles,” ran the 1978 memo.
In each application the finished parts are subjected to conditions which will cause, after prolonged treatment, molecular weight reduction. However, in each application, actual product failure is usually observed before significant molecular weight reduction is detectable by the usual techniques…..Baby bottles are subjected to autoclaving at 250 degrees F in saturated steam and fail under these conditions by becoming opaque, and sometimes by shrinking and deforming. Milk and water bottles are washed in aqueous solutions of alkaline or caustic cleaning agents and fail by stress cracking. The relationship between practical failure modes and the fundamental physical and chemical processes involved is not fully understood.
That polycarbonates might degrade when heated, washed, or exposed to sunlight was also discussed in company memos in the late 1970s and early 1980s. Three decades later, the plastics industry assures consumers that such wear and tear of polycarbonate baby bottles poses no health concerns for infant users.
Pages 58 – 62
To read more about this extremely informative book and sale information click on the link below.