Solvents account for more than 80% of the waste generated during pharmaceutical manufacture, but their use remains largely governed by precedent and familiarity rather than by considerations of sustainability or safety. That’s according to an analysis of US patent data from 1976 to 2016 by Maarten Dobbelaere at Ghent University and Helen Sneddon at the University of York (Angew. Chem., Int. Ed. 2026, DOI: 10.1002/anie.1677569).
The pair say that they found no net reduction in the proportion of hazardous solvents used throughout industrial syntheses, despite an increased focus on health, safety, and sustainability over the same period.
In an effort to tackle the ecological footprint of solvent usage and promote sustainable alternatives, industry players and public consortia have together developed a series of solvent selection guides that classify the different options as red (hazardous or problematic), amber (some issues), or green (preferred), according to set criteria. Governments and international bodies have also introduced regulation to restrict the most harmful of these chemicals, notably the Montreal Protocol in 1987, which banned ozone-depleting chlorofluorocarbons (CFCs) and, more recently, controls on the use of methylene chloride (DCM) across Europe and the US. But the long-standing cumulative impact of these interventions is a complicated story to unpack.
Dobbelaere and Sneddon used a reaction classification algorithm to analyse more than 1.3 million reactions reported in 40 years of data from the US Patent and Trademark Office. They found that between 1976 and 2016, the proportion of hazardous solvents used across all stages of synthesis (reaction, workup, purification, and analysis) remained almost constant, with some problematic examples—such as DCM growing in prevalence.
The researchers also note that regulations restricting specific chemicals (for example, benzene) tended to result in lateral substitution, in which one problematic solvent was simply replaced by another (like DCM, tetrahydrofuran, or dioxane).
These suboptimal swaps weren’t simply driven by a lack of other options, Sneddon says. While specific solvent properties such as lipophilicity undoubtedly played a role, analysis across different reaction classes revealed limited correlation between substrate requirements and solvent choice, thus suggesting that compatibility was not necessarily the deciding factor. Instead, “people follow precedent,” she says.
According to Sneddon, the standard conditions for a particular reaction type, repeated time and time again throughout the literature, can result in chemists unconsciously developing a heavy bias toward established solvent choices rather than sustainable ones. “That’s going to be increasingly important in future as AI and computer-based reaction condition prediction systems will also follow precedent,” she says. “If you put benzene and chlorinated solvents into those training models, that’s what you’re going to get out.”
Industry in particular often requires a legislative push to shift away from these norms. Hannes Sels, a sustainable industry researcher at the Karel de Grote University of Applied Sciences and Arts, says that price and availability also play a part in solvent choice and that many greener solvents simply can’t compete with existing petrochemical supply chains. What’s more, implementing a solvent switch in an approved process isn’t as simple as swapping A for B, Sels says. “There’s always some property that’s different, and it can have a huge impact on your process design and costs.”
Sels notes that there are many examples of good industry practice and that for new processes sustainability is taken into account from the start. He adds, however, that modernizing the existing drug pipeline will require a clear legal mandate.
But Dobbelaere says any regulation needs to be balanced with incentives promoting the use of greener alternatives. This could help avoid the widespread lateral substitutions seen in the wake of restrictions on the use of benzene in the 1980s.
Looking forward, Sels sees past regulatory successes as a promising indication. “After the Montreal Protocol, they needed more than 10 years to phase out all the CFCs, so I’m optimistic about what post-2016 data could show in the wake of recent legislative changes,” he says. “I think it’s really great research, and the way they’ve done the analysis makes targeted regulation and solvent guides possible.”
