1
Green Chemistry Metrics

Frank Roschangar1 and Juan Colberg2

1 Boehringer Ingelheim Pharmaceuticals, Inc., Ridgefield, Connecticut, USA

2 Pfizer Global Research and Development, Pfizer Inc., Groton, Connecticut, USA

1.1 Business Case

Green chemistry is an integral, strategic component for pharmaceutical firms to inspire development of drug manufacturing processes with optimal environmental impact, process safety, and energy consumption, all of which bring about improved economics. Manufacturing contributes a substantial part of industry expenditures that has been estimated at one-third of total costs to one-third of total sales, or about $200 billion worldwide in 2008 [1,2]. This figure includes about 10 billion kg of annual drug manufacturing waste treatment with costs of $20 billion [3]. Therefore, if effectively utilized, green chemistry represents a significant opportunity for industry to increase drug development and manufacturing efficiencies that could translate to trillions of dollars in social value for the public health consumer surplus [4]. This is precisely the reason why industry should optimally utilize green chemistry. In this context, metrics become vital as a reflection of corporate priority, in line with the proven management adage “you can't manage what you don't measure.” Unless improvements are defined, quantified, and measured, we cannot establish clear objectives that allow us to estimate manufacturing improvements. We must, therefore, measure green chemistry by carefully choosing metrics that matter. Ideally, those selected metrics are standardized and aligned within the industry, and also leveraged within the firms with key stakeholders, namely company leadership, technical staff, and suppliers, thereby promoting a culture of continuous ambition and improvement. It was not until 23 years after introduction of the E factor [5] that the first standardized and unified green manufacturing goal metric became available that will be detailed vide infra [6,7].

1.2 Historical Context

The origins of metrics date back to 1956 when Nobel laureate Woodward questioned how to create the best possible synthesis, and invented the concept of synthetic design [8]: “synthesis must always be carried out by a plan, and the synthetic frontier can be defined only in terms of the degree to which realistic planning is possible, utilizing all of the intellectual and physical tools available.” In 1989, Corey leap-frogged the field of synthetic design by introduction of retrosynthesis methodology, in which the chemist starts planning from the product backward via the most efficient bond dissection to arrive at simple and readily available raw materials [9]. For these contributions, he was awarded the 1990 Nobel Prize in Chemistry. The initial considerations for environment in synthetic planning, and thus the first environmental green chemistry metrics, can be traced to Trost and Sheldon who went beyond synthesis design and assessed efficiency through Atom Economy (AE) [10] and Environmental impact factor (E factor) [11] in 1991 and 1992, respectively, with the implied goal to consider waste as a criterion for molecular design and thereby minimize it. AE measures what proportion of the reactants becomes part of the product, and as such addresses a shortcoming of chemical yield (CY). For example, we can have a step with 100% CY that produces more waste than product weight, as was the case with the key step of the first commercial process of phenol via pyrolysis of sodium benzenesulfonate that was developed in Germany in the 1890s (Equation 1.1). Trost received the Presidential Green Chemistry Challenge 1998 Academic Award for development of the AE concept [12].

Equation 1.1 Key step of commercial phenol process.

Unlike AE, the E factor considers CY and selectivity of a process by measuring the amount of waste, excluding water, that is co-produced with 1 kg of the target molecule. A high E factor indicates more waste and greater negative environmental impact. The ideal E factor is 0. Typical E factors for various chemical industries were estimated by Sheldon in 1997 and indicate that pharmaceuticals face substantially elevated waste burden compared to the allied chemical industries (Table 1.1) [13].

Table 1.1 E factors, waste and process complexity across chemical industries.

Industry Segment (Examples)

Annual Product Tonnage

E-Factor (kg waste/ kg product)

Total Annual Waste Tonnage

No. of Steps

Years of Development

Petrochemicals (Solvents, Detergents)

1,000,000– 100,000,000

∼0.1

10,000,000

“Separations”

100+

Bulk Chemicals (Plastics, Polymers)

10,000– 1,000,000

<1–5

5,000,000

1–2

10–50

Fine Chemicals (Coatings, Electronic Parts, Pharmaceutical Raw Materials)

100–10,000

5–>50

500,000

3–4

4–7

Pharmaceuticals (Antibiotics, Drugs, Vaccines)

10–1,000

25–>100

100,000

6+

3–5

The primary cause for the high E factors of pharmaceutical manufacturing is the greater molecular complexity of drugs and the resulting larger step number count to produce them. In addition, the industry faces internal and external barriers that may obstruct optimal manufacturing efficiencies as summarized in Table 1.4 vide infra.

1.3 Metrics, Awards, and Barriers

1.3.1 Mass-Based Metrics

Efficiency and productivity metrics conceived after AE and E factor focused on the amount of generated waste with respect to the product, and for simplicity, assumed that all waste had the same environmental impact, independent of its nature. The ACS GCI PR compiled drug manufacturing waste data and showed that solvents and water make up the majority, or 86% of waste for the processes studied, and should therefore be included in comprehensive waste analysis (Figure 1.1) [14,19]. Thus, the Pharmaceutical Roundtable consequently introduced the Process Mass Intensity (PMI) metric that does consider all materials used in the process and workup, including water.

Figure 1.1 Typical pharmaceutical drug manufacturing waste composition.

For a comprehensive overview, we summarize the common mass-based metrics and their consideration for resources in Table 1.2.

Table 1.2 Mass-based environmental process waste metrics.

From the above group of diverse green chemistry mass metrics, both E factor and PMI emerged as the most utilized in industry. Recently, the complete E factor or cEF was introduced, combining the advantages of PMI that is the inclusion of water and solvents in analysis, with E factor that is step mass balance, as a well-suited metric for multi-step manufacturing process analysis [6].

However, while mass-based metrics can measure process improvements and thereby aid route design to a specific drug target, they do not allow for comparison of manufacturing processes between different drugs, and thus by themselves cannot deliver a standardized green process goal.

1.3.2 Life-Cycle Assessment

Accurately measuring the greenness of a manufacturing process unquestionably goes beyond quantifying co-produced waste, and includes assessing sustainability of process inputs such as metals, reagents, and solvents, evaluating overall environmental impact including eco-toxicity and carbon footprint, energy consumption, as well as occupational health and risk factors, all of which are integral part of the comprehensive life-cycle assessment (LCA) (Figure 1.2) [24,25].

Figure 1.2 Comprehensive green metrics categories for life cycle assessment.

LCA methodology encompasses cradle-to-grave impact analysis starting from sources and upstream processes for process inputs, the processes themselves to manufacture intermediates and the drug, including equipment cleaning and waste handling, all the way to pharmaceutical manufacturing, packaging, and eventually drug disposal and recycling over the useful life of the drug. However, there are several hurdles to overcome with LCA [26]. A significant challenge is the lack of life-cycle inventory (LCI) input data and standardization [27], as well as the difficulty to allocate energy consumption to a particular process within pharmaceutical multi-purpose plants. A further barrier is that analysis remains time-consuming, and thereby inhibits widespread use, particular during early phases of drug development where LCA is expected to have the biggest impact during the synthesis design phase, despite efforts to simplify the methodology via fast life-cycle assessment of synthetic chemistry (FLASC)...