Toward substitution with no regrets

See allHide authors and affiliations

Science  13 Mar 2015:
Vol. 347, Issue 6227, pp. 1198-1199
DOI: 10.1126/science.aaa0812

Embedded Image
Replacement troubles.

In many plastic baby bottles, bisphenol A (BPA) has been replaced with bisphenol S (BPS), but the latter compound may also have adverse health effects.


Vast numbers of synthetic chemicals are used in everyday consumer products. Many are safe, but some can have unintended biological or environmental effects. For example, phthalates are widely used to increase the flexibility of plastics but also disrupt hormonal balance (1). Organophosphates are highly effective insect repellents but cause severe neurotoxicity to mammals (2). In many cases, chemicals of concern have been replaced by other chemicals that are functionally equivalent and believed to be of less concern (see the photo). However, the need for expedient substitution can lead to the use of chemicals that are no less harmful than those they replace. How can such “regrettable substitutions” be avoided?

There have been many examples of regrettable substitution over the years. The example gaining most attention recently is the substitution of bisphenol S (BPS) for bisphenol A (BPA) (see the figure, panel A). BPA is widely used to make clear and rough plastics, used in water bottles, sports equipment, and CDs and DVDs; it is also used in food and beverage can coatings (where it helps to protect food from contamination and spoilage) and in thermal paper (such as till receipts). Numerous studies suggest that BPA may act as an endocrine disruptor—a substance that mimics hormonal function and could cause developmental and reproductive problems in animal and human offspring, perhaps even affecting future generations (3). Rising public concern led to a voluntary withdrawal of BPA from many children's products even before regulatory action was taken (see the photo). One substance among many emerged as a replacement: BPS could expediently serve as a drop-in replacement for products that could be labeled “BPA free.” Yet recent studies have identified toxicological concerns with BPS, suggesting that one hazardous chemical was substituted for another (4).

Regrettable substitutions.

Because of toxicological concerns, BPA has been replaced with BPS in some applications (A). Similarly, diacetyl has been replaced with 2,3-pentanedione (B). However, these replacement compounds are chemically similar to the ones they replace, causing concerns about possible toxicological effects.

Several other examples are similarly attracting attention, including the case of diacetyl (2,3-butanedione), a naturally occurring chemical that gives butter its characteristic aroma and taste and that is produced industrially for use as a butter flavoring. Diacetyl has been linked to lung cell damage via inhalation exposure, as evidenced by epidemiological studies of factory workers diagnosed with obliterative bronchiolitis (5). In response to the publicity deriving from these reports, a number of major food companies voluntarily stopped using diacetyl and began to use alphadiketone (2,3-pentanedione) as a flavoring substitute for diacetyl (see the figure, panel B). However, toxicology results indicate that acute inhalation exposure to 2,3-pentanedione can cause airway epithelial damage that is similar to that caused by diacetyl (6). Further studies have raised concerns that the toxicological effects of diacetyl may be shared with this and other alpha-diketones, which are close structural analogs (7).

Given the growing number of regrettable substitution examples, as exemplified by BPA and diacetyl, the National Research Council (NRC) recently released a report calling for “informed substitution” (8). As the Environmental Protection Agency notes, practicing informed substitution is meant to “minimize the likelihood of unintended consequences, which can result from a precautionary switch away from a chemical of concern without fully understanding the profile of potential alternatives, and to enable a course of action based on the best information—on the environment and human health—that is available or can be estimated” (9). In practice, this can be achieved through an alternatives assessment, a process of comparing functionally equivalent chemicals based on a variety of factors, including performance, costs, potential adverse effects to human health and the environment, and societal impacts (8). However, alternatives assessments tend to only consider chemical substitutes that have been commercialized, can be readily obtained, and, typically, have known physiochemical properties or information about their effects (8).

Cases in which viable alternatives are available can often be attributed to chemical design innovations that view reduced hazard as part of the performance criteria (10). An example of informed substitution is the replacement of chromated copper arsenate (CCA), used for wood preservation, with alkaline copper quaternary (ACQ). CCA contains both arsenic and chromium and is associated with increased cancer risks following environmental exposure (11). ACQ instead relies on a bivalent copper complex and a quaternary ammonium compound and offers equivalent performance against biological impacts, such as decay and termite attack. In North America, CCA has been widely replaced with ACQ and other newer, more benign wood preservatives for most residential applications. ACQ wood products are expected to reduce soil contamination and limit impacts from construction and debris landfills, but questions remain about the amount of copper released from ACQ-treated wood, with potential implications for aquatic ecotoxicity (12). Thus, even though this example is largely regarded as successful, it shows that design decisions must be revisited and continuously improved as new toxicity information and assays become available.

In many other cases, alternatives that can serve the desired function and have the desired toxicity profile have not been invented, discovered, or even imagined. Alternatives assessment then merely identifies an alternative that meets minimal functional requirements while offering a satisfactory profile in terms of cost and potential hazards.

One route to chemical products and processes that reduce or eliminate the use and generation of hazardous substances is provided by green chemistry (13). This approach uses insights into the critical stages of biological activity—absorption, distribution, metabolism, and excretion (ADME), and molecular initiation—as an important piece of the design framework. By using physiochemical properties and structural motifs along with an understanding of mechanistic toxicology, it is possible to devise chemical design rules for reduced toxicity. There are many examples of chemicals designed to avoid specific undesirable or hazardous aspects, such as persistence (14), use of depleting feedstocks, and cytotoxicity (15); however, we will not have plentiful examples of systematic design for reduced hazard until we routinely view the task as a systems issue. Safer substitutions require scientific advances in toxicity data generation and curation, computational models, and systems insight that lead to intentional design.

As stated in the NRC report (8), rational design of our next-generation molecules will not be easy. The report calls for more research on property-toxicity relationships and greater elucidation of toxicological mechanisms of action at the molecular level. Substantial resources are devoted to identifying and measuring the problem of hazardous chemicals in society, but at least the same resources need to be available for the molecular designs of safer chemicals. It is only through this process that we can hope to avoid future regrettable chemical substitutions.


  1. U.S. Environmental Protection Agency, EPA's DfE Standard for Safer Products (2012);

Navigate This Article