Process Mass Intensity (PMI) in Green Chemistry: A Strategic Guide for Sustainable Drug Development

Abstract

This article provides a comprehensive overview of Process Mass Intensity (PMI), a pivotal green chemistry metric for researchers and drug development professionals. It explores PMI's fundamental principles and its relationship to other metrics like E-Factor. The content details practical methodologies for calculating PMI, supported by real-world case studies from the pharmaceutical industry, including the use of predictive apps and optimization algorithms. It further addresses common challenges in PMI application and strategies for process optimization. Finally, the article critically examines the correlation between PMI and broader environmental impacts, validating its role while discussing its limitations and the emerging context of simplified Life Cycle Assessments (LCA) for a more holistic sustainability assessment.

Understanding Process Mass Intensity: The Cornerstone of Green Chemistry Metrics

Process Mass Intensity (PMI) is a pivotal green chemistry metric that quantifies the efficiency of a manufacturing process by measuring the total mass of materials required to produce a unit mass of a desired product [1]. It is defined as the ratio of the total mass of inputs (including reactants, solvents, catalysts, and auxiliary materials) to the mass of the final product obtained [2] [3]. Mathematically, PMI is expressed as:

PMI = (Total Mass of All Inputs) / (Mass of Product) [3]

A lower PMI value indicates a more efficient process that utilizes resources judiciously, yielding more product with less waste and a reduced environmental footprint [1]. Conversely, a high PMI signals significant material losses and a heavier burden on planetary systems [1]. The ideal, theoretical limit for PMI is 1, indicating that every gram of input is incorporated into the final product [3]. PMI has been identified by the American Chemical Society Green Chemistry Institute Pharmaceutical Roundtable (ACS GCIPR) as a key mass-related metric and an indispensable indicator of the overall greenness of a process [4].

PMI in Context: Comparison with Other Metrics and Modalities

PMI vs. Other Green Chemistry Metrics

While several metrics exist to evaluate process efficiency, PMI offers a holistic assessment by including all materials used in synthesis, purification, and isolation [4]. The table below compares PMI with other common green chemistry metrics.

Table 1: Comparison of PMI with Other Green Chemistry Metrics

Metric	Definition	Key Focus	Advantages/Limitations
Process Mass Intensity (PMI)	(Total mass of inputs) / (Mass of product) [3]	Total material consumption across the entire process [4]	Holistic; includes reactants, solvents, purification materials. Does not account for environmental impact of material manufacture [4].
Atom Economy (AE)	(MW of product) / (Σ MW of reactants) [4]	Efficiency of a reaction's design	Simple; based on molecular weights. Assumes 100% yield and excludes solvents/reagents [4].
E-Factor	(Total mass of waste) / (Mass of product) [3]	Total waste generated	Directly measures waste. A relation exists: E-Factor = PMI - 1 [3].
Reaction Mass Efficiency (RME)	(Mass of product) / (Σ Mass of reactants utilized) x 100% [3]	Mass efficiency of reactants only	Incorporates actual yield. Scope is limited to reactants, excluding solvents and other materials [3].

PMI Across Pharmaceutical Modalities

The environmental efficiency of producing different types of pharmaceutical agents varies significantly, as reflected in their PMI values. The following table summarizes typical PMI ranges for major modalities, highlighting the particular resource intensity of synthetic peptides.

Table 2: Comparison of PMI Values Across Different Pharmaceutical Modalities [4]

Modality	Reported PMI (kg material/kg API)	Context and Notes
Small Molecules	Median: 168 - 308	Represents the most efficient modality due to mature, optimized synthetic processes.
Biopharmaceuticals	Average: ~8,300	Includes monoclonal antibodies and other biotechnology-derived molecules.
Oligonucleotides	Average: ~4,299 (Range: 3,035 - 7,023)	Traditionally assembled via solid-phase processes similar to peptides.
Synthetic Peptides (SPPS)	Average: ~13,000	Does not compare favorably with other modalities; warrants more sustainable processes.

Experimental Protocol: Determining PMI for a Peptide Synthesis Process

This protocol outlines the steps for calculating the Process Mass Intensity for a peptide synthesized via Solid-Phase Peptide Synthesis (SPPS), based on industry methodology [4].

Principle

The PMI for a peptide active pharmaceutical ingredient (API) is determined by dividing the total mass of all materials used in its manufacture—across the stages of synthesis, purification, and isolation—by the mass of the final, isolated peptide product. This provides a comprehensive view of the process's material efficiency [4].

Materials and Equipment

• SPPS Reactor: Automated or semi-automated peptide synthesis reactor (e.g., 250 mL to 6000 L capacity systems) [4].
• Resin: Solid support (e.g., Wang resin, Rink amide resin).
• Amino Acids: Fmoc-protected amino acids (Fmoc-AA).
• Solvents: (N,N)-Dimethylformamide (DMF), (N)-Methyl-2-pyrrolidone (NMP), Dichloromethane (DCM), Diethyl ether (DEE), Trifluoroacetic acid (TFA) [4].
• Reagents: Coupling agents (e.g., HATU, HBTU), deprotecting agents (e.g., Piperidine).
• Purification System: Preparative High-Performance Liquid Chromatography (HPLC) system.
• Lyophilizer: For final product isolation.
• Analytical Balance: Precision of 0.1 mg or better.

Procedure

Step 1: Solid-Phase Synthesis

• Weigh Inputs: Accurately record the mass of the resin, all Fmoc-protected amino acids, coupling reagents, deprotection reagents, and all solvents used for washing and swelling.
• Perform Synthesis: Carry out the automated SPPS cycle (deprotection, coupling, washing) according to the established peptide sequence.
• Cleave from Resin: After chain assembly, treat the resin-bound peptide with a cleavage cocktail (typically containing TFA and scavengers) to release the crude peptide. Record the mass of all materials used in the cleavage step.

Step 2: Purification

• Purify Crude Product: Purify the cleaved crude peptide using preparative HPLC.
• Record Solvents: Accurately measure and record the masses of all solvents and buffers used in the purification process (e.g., acetonitrile, water with 0.1% TFA).

Step 3: Isolation

• Isolate Product: Lyophilize the purified peptide fractions to obtain the final, solid API.
• Weigh Final Product: Precisely weigh the mass of the isolated, dry peptide product.

Data Analysis and PMI Calculation

• Sum Total Input Mass: Add the masses of all materials recorded in Steps 1 through 3. This includes resins, amino acids, reagents, and all solvents (even those used in washing and purification).
• Record Final Product Mass: Use the mass of the final, lyophilized peptide from Step 3.3.
• Calculate PMI: Use the formula below. The workflow for this calculation is also presented in Figure 1.

PMI = (Total Mass of All Inputs) / (Mass of Final Peptide Product)

Figure 1: Workflow for Calculating Process Mass Intensity

The Scientist's Toolkit: Key Reagents and Materials in SPPS

The following table details critical reagents used in Solid-Phase Peptide Synthesis, many of which contribute significantly to the overall PMI and environmental footprint [4].

Table 3: Research Reagent Solutions for Solid-Phase Peptide Synthesis

Reagent/Material	Function in Synthesis	Environmental & Safety Notes
Fmoc-Protected Amino Acids	Building blocks for peptide chain assembly.	Poor atom-economy; the Fmoc protecting group constitutes a significant portion of the mass that is later cleaved as waste [4].
DMF, DMAc, NMP	Primary solvents for swelling resin and dissolving amino acids/reagents.	Classified as reprotoxic; use may be restricted or banned. A major contributor to PMI due to large volumes used [4].
Coupling Agents (e.g., HATU)	Activate carboxylic acid groups for amide bond formation.	Can be potentially explosive or sensitizing [4].
Trifluoroacetic Acid (TFA)	Cleaves the peptide from the resin and removes side-chain protecting groups.	Highly corrosive and generates hazardous waste [4].
Dichloromethane (DCM)	Used for resin washing and in cleavage mixtures.	Toxic solvent with significant environmental and health concerns [4].

Advanced Concepts: System Boundaries and Correlation with LCA

Expanding System Boundaries: From Gate-to-Gate to Cradle-to-Gate

The most commonly used PMI calculation employs a gate-to-gate system boundary, which only considers materials directly used in the manufacturing process from factory entrance to exit [2]. To gain a more complete picture of environmental impact, the system boundary can be expanded to cradle-to-gate, accounting for the mass expenditures of raw materials and intermediates throughout the upstream value chain. This broader metric is sometimes termed Value-Chain Mass Intensity (VCMI) [2]. The relationship between these system boundaries is illustrated in Figure 2.

Figure 2: System Boundaries for Mass Intensity Calculations

PMI as a Proxy for Environmental Impact

Recent research systematically analyzes whether mass intensities like PMI can reliably approximate the results of a full Life Cycle Assessment (LCA) [2]. Key findings include:

• Expanding the system boundary from gate-to-gate (PMI) to cradle-to-gate (VCMI) strengthens the correlation with LCA environmental impacts for most impact categories [2].
• However, a single mass-based metric cannot fully capture the multi-criteria nature of environmental sustainability [2]. For example, PMI does not distinguish between the environmental impacts of different waste types (e.g., chlorinated vs. non-chlorinated solvents) or account for energy usage and renewable feedstocks [4] [2].
• The reliability of mass-based assessment is highly time-sensitive, as the environmental impact of producing a specific input (e.g., coal) can change over time, especially during the transition to a defossilized chemical industry [2].

Process Mass Intensity serves as a crucial, easily calculable metric that provides an initial gauge of the material efficiency and environmental footprint of pharmaceutical manufacturing processes, particularly in the context of peptide synthesis [4]. Its strength lies in its simplicity and ability to direct chemists and engineers toward waste reduction and resource conservation [1]. However, to make fully informed decisions for a sustainable future, researchers and drug development professionals should use PMI as a screening tool and not a standalone verdict. It is most powerful when complemented with other metrics and, for final process evaluation, more comprehensive methods like Life Cycle Assessment [2]. The ongoing pursuit of lower PMI values will continue to drive innovation in green chemistry, process intensification, and the adoption of circular economy principles in the pharmaceutical industry.

In the pursuit of sustainable chemical manufacturing, particularly within the pharmaceutical and fine chemical industries, metrics are indispensable for quantifying environmental impact and driving process improvement. Two of the most prominent mass-based metrics are Process Mass Intensity (PMI) and the Environmental Factor (E-Factor). Both metrics serve to evaluate the efficiency of chemical processes by focusing on material usage and waste generation, aligning with the fundamental green chemistry principle that it is better to prevent waste than to treat or clean it up after it is formed [5]. While often discussed in tandem, a clear understanding of their distinct definitions, mathematical relationship, and appropriate application is crucial for researchers and drug development professionals seeking to optimize synthetic routes and reduce the environmental footprint of their processes.

Definitions and Core Concepts

Process Mass Intensity (PMI)

Process Mass Intensity (PMI) is defined as the total mass of materials input into a process per unit mass of the product obtained. It provides a comprehensive measure of the material efficiency of a process [6].

PMI = (Total Mass of Input Materials) / (Mass of Product)

The "input materials" encompass all substances including reagents, solvents, catalysts, and work-up or purification materials used in the reaction and isolation steps [6]. A lower PMI indicates a more efficient process with less material consumption. The ideal PMI is 1, signifying that every gram of input is converted into a gram of product, resulting in zero waste.

Environmental Factor (E-Factor)

The E-Factor, pioneered by Roger Sheldon, quantifies the waste generated by a process. It is defined as the mass of total waste produced per unit mass of product [7] [8] [5].

E-Factor = (Total Mass of Waste) / (Mass of Product)

Waste is rigorously defined as "everything but the desired product" [5]. This includes by-products, side products, unreacted excess reagents, catalysts, solvents, and work-up and purification materials [6]. The ideal E-Factor is 0, representing a theoretical zero-waste process.

The Mathematical Relationship and Industry Benchmarks

Fundamental Equation

PMI and E-Factor are intrinsically linked. Since the total mass of inputs in a process equals the mass of the product plus the mass of all waste produced (conservation of mass), the relationship between the two metrics is straightforward [9]:

PMI = E-Factor + 1

Conversely, E-Factor = PMI - 1. This simple relationship means that for any process, the PMI will always be exactly 1 greater than its E-Factor. This also highlights why their ideal values differ (PMI=1, E-Factor=0) [6] [9].

Industry Sector Benchmarks

The application and typical values of PMI and E-Factor vary significantly across different sectors of the chemical industry, largely due to the complexity of products and the number of synthesis steps involved. Table 1 summarizes benchmark values for key industry sectors, illustrating the waste generation challenge in pharmaceuticals [10] [8] [5].

Table 1: PMI and E-Factor Benchmarks Across the Chemical Industry

Industry Sector	Annual Production (Tonnes)	Typical E-Factor (kg waste/kg product)	Equivalent PMI (kg input/kg product)
Oil Refining	10⁶ – 10⁸	~0.1	~1.1
Bulk Chemicals	10⁴ – 10⁶	<1 – 5	<2 – 6
Fine Chemicals	10² – 10⁴	5 – 50	6 – 51
Pharmaceuticals	10 – 10³	25 – >100	26 – >101

The pharmaceutical industry's high PMI and E-Factor result from multi-step syntheses, stringent purity requirements, and frequent use of complex protective groups and chromatographic purifications [10]. A survey of 97 Active Pharmaceutical Ingredient (API) commercial scale syntheses found an average complete E-Factor (cEF) of 182, indicating a massive average PMI of 183 [5].

Calculation Protocols and Methodologies

Protocol for Calculating PMI and E-Factor for a Single Reaction

This protocol provides a step-by-step methodology for determining the PMI and E-Factor of a chemical reaction, essential for environmental impact assessment.

I. Materials and Data Recording

• Electronic Lab Notebook (ELN): Use an ELN configured to automatically capture masses of all input and output materials [9].
• Balances: Calibrated analytical and mass balances.
• Inventory Log: A standardized template for recording all materials.

Table 2: Research Reagent Solutions and Essential Materials for Metric Calculation

Item	Function in Protocol
Electronic Lab Notebook (ELN)	To automatically and accurately record the mass of all input materials (reagents, solvents) and the mass of the isolated final product.
Calibrated Mass Balance	To provide the fundamental mass measurement data required for all calculations.
Solvent Inventory Log	To track the volume and mass of all solvents used in the reaction, work-up, and purification stages.

II. Experimental Procedure

• Mass Inputs: Precisely weigh and record the mass of all reagents, catalysts, and solvents introduced into the reaction vessel.
• Execute Reaction: Carry out the synthetic reaction according to the established procedure.
• Work-up & Purification: Record the masses of all additional materials used during work-up (e.g., quenching solutions, extraction solvents) and purification (e.g., chromatography solvents, recrystallization solvents).
• Mass Product: After isolation and drying, accurately weigh the final, pure product.

III. Data Analysis and Calculations

• Sum Total Input Mass: Σm_input = m_reagents + m_catalysts + m_solvents + m_work-up_materials + m_purification_materials
• Calculate PMI: PMI = Σm_input / m_product
• Calculate E-Factor: E-Factor = PMI - 1 OR E-Factor = (Σm_input - m_product) / m_product

IV. Interpretation and Reporting

• Report both the PMI and E-Factor values.
• Clearly state the system boundary (e.g., whether water and recycled solvents are included; see Section 5.1).
• For a more nuanced assessment, the E-Factor can be deconstructed into contributing factors from by-products, solvents, etc. [6].

Protocol for Multi-Step Synthesis: Cumulative PMI and E-Factor

For multi-step syntheses, such as those common in API development, the cumulative PMI (cPMI) and cumulative E-Factor (cE) across all steps provide the most accurate picture of overall material efficiency. The step metrics are not additive [6]. The following diagram illustrates the recursive calculation workflow for a linear 3-step synthesis plan.

Diagram Title: Cumulative Metric Calculation Flow

The logic for this recursive calculation, where m_i is the mass of the intermediate product from step i, is defined by the following generalizable formula [6]:

cPMI_(1→n) = cPMI_(1→n-1) + (PMI_n - 1) × (m_(n-1) / m_n)

This methodology accurately accounts for the changing mass of intermediate products throughout the synthesis, preventing the over-inflation of waste metrics that would occur from simple addition of step-level data [6].

Critical Considerations for Effective Application

System Boundaries and Definitions

A critical step in applying PMI/E-Factor is defining the system boundaries, as inconsistent definitions lead to incomparable results.

• Inclusion of Water: There is ongoing debate. Often, simple E-Factor (sEF) and PMI exclude water, while complete E-Factor (cEF) and PMI include it, as its treatment has an environmental cost [5].
• Solvent Recycling: In a commercial context, solvents are often recycled. The original E-Factor assumed 90% solvent recovery. The true commercial metric lies between sEF (no solvent) and cEF (all solvent counted as waste) [5].
• Defining "Starting Material": For multi-step syntheses, the E-factor can be reduced overnight by purchasing an advanced intermediate. An agreed-upon definition (e.g., a readily available chemical <$100/kg) is essential. The E-factor of earlier steps (the "intrinsic E-factor") should be added for a fair comparison of full synthetic routes [5].

Advantages, Limitations, and Complementary Metrics

Both PMI and E-Factor are powerful but have specific strengths and weaknesses. Table 3 provides a comparative analysis.

Table 3: Advantages and Limitations of PMI and E-Factor

Aspect	Process Mass Intensity (PMI)	Environmental Factor (E-Factor)
Core Focus	Inputs / Resource Consumption [9]	Outputs / Waste Generation [5]
Key Advantage	Easier to communicate to management (focus on cost-saving via reduced material use) [9]. Simpler to calculate from standard lab data [9].	Conceptually aligns directly with the goal of waste prevention. The ideal of zero is a powerful, clear target [5].
Key Limitation	Does not directly measure waste, though it is directly related.	Requires specific measurement of waste streams, which can be more complex than tracking inputs [9].
Shared Major Limitation	Mass-based: Neither metric differentiates between benign waste (e.g., water, NaCl) and hazardous waste (e.g., heavy metals, toxic organics) [8] [5].

To address the limitation of being purely mass-based, PMI and E-Factor should be used in conjunction with other metrics:

• Environmental Quotient (EQ): EQ = E-Factor × Q, where Q is a hazard factor assigned to the waste. However, quantifying Q is challenging [7] [5].
• Life Cycle Assessment (LCA): Provides a broader cradle-to-grave environmental impact assessment, including factors like energy use and global warming potential [8].
• Solvent Selection Guides: Traffic-light coded guides (Green=preferred, Red=undesirable) are vital for selecting safer solvents, which often constitute the majority of process mass [5].

PMI and E-Factor are foundational, complementary metrics that provide a crucial lens for evaluating the material efficiency and waste profile of chemical processes. Their simple mathematical relationship, PMI = E-Factor + 1, underscores their interdependence. For the pharmaceutical industry and related fine chemical sectors, these metrics have served as a wake-up call, revealing the significant environmental footprint of complex syntheses and providing a clear framework for improvement. While their mass-based nature requires them to be part of a broader toolkit that includes environmental impact assessment and solvent selection guides, the rigorous application of PMI and E-Factor calculation protocols—from single reactions to complex, multi-step syntheses—remains an essential practice for any research and development organization committed to sustainability and green chemistry principles.

In the pharmaceutical industry, the acronym PMI carries dual significance, both aspects of which are crucial for driving cost reduction and operational excellence. In the context of green chemistry and sustainable development, PMI stands for Process Mass Intensity, a key metric for evaluating the environmental and economic efficiency of chemical processes [2] [11]. Concurrently, within operational and project management frameworks, PMI references the Project Management Institute and its methodologies, which provide structured approaches for optimizing drug development pipelines [12] [13]. This document explores how both interpretations of PMI contribute to creating a more efficient, cost-effective, and sustainable pharmaceutical industry.

The application of green chemistry metrics, particularly Process Mass Intensity, has become essential as the industry addresses the United Nations Sustainable Development Goals, specifically Goal 12 concerning responsible consumption and production [11]. PMI represents a simple yet powerful metric that quantifies the mass expenditures required to produce one kilogram of a chemical product, thereby providing insights into resource efficiency and waste generation [2]. When implemented within robust project management frameworks, PMI tracking enables researchers, scientists, and drug development professionals to make data-driven decisions that simultaneously reduce environmental impact and operational costs.

Process Mass Intensity: A Foundational Green Chemistry Metric

Definition and Calculation

Process Mass Intensity (PMI) is calculated as the total mass of materials used in a process divided by the mass of the final product [2]. Mathematically, this is represented as:

PMI = Total Mass of Inputs (kg) / Mass of Product (kg)

Unlike simpler metrics such as atom economy (AE) which focus solely on the reaction step, PMI provides a more comprehensive assessment by including all mass inputs across the process, including solvents, reagents, and catalysts [11]. A lower PMI value indicates a more efficient process with less waste generation and better resource utilization. Ideal green chemistry processes aim for a PMI approaching 1, indicating nearly all input materials are incorporated into the final product.

System Boundaries and Environmental Correlation

The definition of system boundaries significantly impacts PMI calculation and interpretation. Recent research demonstrates that expanding system boundaries from gate-to-gate (traditional PMI) to cradle-to-gate (Value-Chain Mass Intensity or VCMI) strengthens the correlation between mass intensity and environmental impacts [2]. Table 1 compares PMI system boundaries and their applications.

Table 1: PMI System Boundaries and Applications in Pharmaceutical Development

System Boundary	Scope	Materials Included	Primary Application
Gate-to-Gate (Traditional PMI)	Factory entrance to exit	Intermediate products, solvents, reagents	Internal process optimization, reaction step evaluation
Cradle-to-Gate (VCMI)	Resource extraction to factory exit	Natural resources, all value chain inputs	Holistic environmental impact assessment, supply chain optimization
"Commonly Available Materials"	Commercially available materials to factory exit	Raw materials available from suppliers (e.g., Sigma-Aldrich)	Practical industrial assessment, preliminary design decisions

The correlation between mass intensity and environmental impacts varies significantly depending on the system boundary applied. A 2025 study by Eichwald et al. found that expanding the system boundary from gate-to-gate to cradle-to-gate strengthened correlations for fifteen of sixteen environmental impact categories [2]. This finding underscores the importance of clearly defining and consistently applying system boundaries when using PMI for environmental or business case assessments.

Quantitative Case Studies: PMI Driving Cost and Waste Reduction

Catalytic Process Optimization in Fine Chemicals

Recent case studies in fine chemical production demonstrate how PMI optimization directly correlates with improved sustainability and cost reduction. Table 2 summarizes green metrics from catalytic processes that have been optimized for efficiency [14].

Table 2: Green Metrics for Optimized Catalytic Processes in Fine Chemical Production

Process	Catalyst	Atom Economy (AE)	Reaction Mass Efficiency (RME)	Process Mass Intensity (PMI)*	Key Improvement
R-(+)-limonene epoxidation	K–Sn–H–Y-30-dealuminated zeolite	0.89	0.415	2.41	High AE with moderate yield
Florol via isoprenol cyclization	Sn4Y30EIM	1.0	0.233	4.29	Perfect atom economy
Dihydrocarvone from limonene-1,2-epoxide	dendritic ZSM-5/4d	1.0	0.63	1.59	Excellent overall green characteristics

*PMI calculated as inverse of RME (1/RME) for comparison purposes.

The dihydrocarvone synthesis exemplifies outstanding catalytic performance, achieving perfect atom economy while maintaining high reaction mass efficiency, resulting in a PMI of 1.59 [14]. This represents approximately 63% mass efficiency, significantly reducing raw material requirements and waste generation compared to conventional processes.

Protocol for PMI Assessment and Optimization

Objective: To systematically measure, analyze, and optimize Process Mass Intensity for pharmaceutical manufacturing processes.

Materials and Equipment:

• Detailed process flow diagram with all input streams
• Mass balance data for all process steps
• Laboratory or production batch records
• Life cycle inventory databases (for cradle-to-gate assessment)
• Green metrics calculation software or spreadsheet

Experimental Workflow:

• Process Mapping: Document all process steps from raw material inputs to final product isolation and purification.

• System Boundary Definition: Determine appropriate boundaries (gate-to-gate, cradle-to-gate) based on assessment goals.

• Mass Inventory: Quantify all material inputs, including reagents, solvents, catalysts, and processing aids.

• PMI Calculation: Compute PMI using standardized formula: PMI = Total Mass Input / Mass of Product.

• Benchmarking: Compare calculated PMI against industry benchmarks or alternative processes.

• Improvement Identification: Identify high-mass intensity steps for optimization focus.

• Iterative Optimization: Implement process modifications and recalculate PMI.

The following workflow diagram illustrates the standardized protocol for PMI assessment and optimization:

Project Management Integration for Operational Excellence

Facilitated Planning for Drug Development

The Project Management Institute (PMI) methodologies provide structured frameworks for implementing green chemistry metrics in pharmaceutical development. A proven approach involves facilitated planning sessions that bring together cross-functional teams including discovery scientists, biology, chemistry, legal, marketing, toxicology, metabolism, regulatory, medical, clinical pharmacology, and product development [12].

Protocol: Three-Day Facilitated Planning Session

Day 1: Foundation and Strategy

• Team introductions and ground rules establishment
• Scientific overview presentation by discovery scientist
• Draft package insert development to define target product profile
• Clinical plan alignment with target indications
• Work Breakdown Structure (WBS) completion

Day 2: Operational Planning

• Network chart development depicting task dependencies
• Team operating agreements establishment
• Project issue identification and resolution planning
• Critical path analysis using project management tools

Day 3: Finalization and Commitment

• Critical path review and optimization
• Program strategy validation
• Planning document assignment distribution
• Management approval action plan development

These intensive sessions typically reduce planning time from the traditional 100-day approach to just three days while creating more robust, cross-functionally aligned plans [12]. The integration of green chemistry metrics like PMI into these plans ensures that sustainability considerations are embedded early in the development process.

Clinical Trial Optimization

Project management principles applied to clinical trials, particularly bioequivalence studies (BES), demonstrate significant efficiency gains. A seven-year implementation effort managing 30 clinical studies revealed that structured project management approaches could optimize trial conduct, harmonize activities, and lower expenditures [13].

The following diagram illustrates the clinical trial management process optimized through project management principles:

The Scientist's Toolkit: Research Reagent Solutions

Table 3: Essential Research Reagents and Materials for PMI-Optimized Pharmaceutical Development

Reagent/Material	Function	PMI Optimization Consideration
K–Sn–H–Y-30-dealuminated zeolite	Epoxidation catalyst for terpene substrates	Enables high atom economy (0.89) in limonene epoxidation [14]
Sn4Y30EIM zeolite	Cyclization catalyst for isoprenol conversion	Achieves perfect atom economy (1.0) in florol synthesis [14]
Dendritic ZSM-5/4d zeolite	Multifunctional catalyst for epoxide rearrangement	Provides excellent green characteristics (AE=1.0, RME=0.63) [14]
Recoverable solvent systems	Reaction media with recycling potential	Reduces PMI by minimizing waste solvent mass
Immobilized reagents & catalysts	Facilitates separation and reuse	Decreases PMI through multiple reusability cycles
Bio-based starting materials	Renewable feedstocks	Lowers environmental impact while potentially reducing PMI

Limitations and Future Directions

While PMI provides valuable insights into process efficiency, it has limitations as a standalone metric. Mass intensities do not directly reflect environmental impacts or consider the origin of input materials, such as renewable versus fossil-based feedstocks [2]. Additionally, PMI does not account for energy consumption or waste properties, which can significantly influence environmental footprint.

Future research should focus on integrating PMI with more comprehensive assessment methods, particularly simplified Life Cycle Assessment (LCA) tools that can provide multidimensional environmental impact evaluation while maintaining practicality for chemists and engineers [2]. The transition toward a defossilized chemical industry further emphasizes the need for metrics that capture both mass efficiency and environmental sustainability.

The pharmaceutical industry's continued adoption of green chemistry metrics, supported by robust project management frameworks, positions it to achieve significant cost reductions while advancing operational excellence and environmental stewardship. By embracing both interpretations of PMI—Process Mass Intensity and Project Management Institute methodologies—organizations can drive sustainable innovation throughout the drug development pipeline.

Linking PMI to the 12 Principles of Green Chemistry

In the pursuit of a more sustainable chemical industry, the Twelve Principles of Green Chemistry provide a foundational framework for designing safer, more efficient chemical processes [15] [16]. To translate these qualitative principles into quantitative practice, researchers and industry professionals rely on green chemistry metrics, among which Process Mass Intensity (PMI) has emerged as a particularly influential measure [2] [17]. PMI is defined as the total mass of materials used to produce a unit mass of a target product, calculated as PMI = Total mass in a process (kg) / Mass of product (kg) [18]. A lower PMI indicates a more efficient process with less waste, directly supporting the foundational principle of waste prevention [15] [18].

This application note explores the critical linkages between PMI and the 12 Principles, providing drug development professionals with structured protocols for implementing PMI measurement to advance sustainability goals within a broader thesis on green chemistry metrics. As the pharmaceutical industry faces increasing pressure to improve environmental performance, PMI offers a practical tool for quantifying progress toward greener synthesis [17]. Recent research underscores that while expanding PMI system boundaries strengthens its correlation with environmental impacts, mass-based metrics alone cannot fully capture the multi-criteria nature of sustainability, highlighting the need for complementary assessment methods [2].

Quantitative Framework: PMI and Related Green Metrics

Comparative Analysis of Mass-Based Green Metrics

Table 1: Key Mass-Based Green Chemistry Metrics for Process Evaluation

Metric	Calculation Formula	Ideal Value	Primary Green Principle Addressed
Process Mass Intensity (PMI)	Total mass of inputs (kg) / Mass of product (kg) [18]	1 (theoretical minimum) [18]	Waste Prevention (Principle 1) [15]
E-Factor	Total mass of waste (kg) / Mass of product (kg) [18] [10]	0 [18]	Waste Prevention (Principle 1) [10]
Atom Economy (AE)	(FW of desired product / Σ FW of all reactants) × 100% [18]	100% [15]	Atom Economy (Principle 2) [15]
Reaction Mass Efficiency (RME)	(Mass of product / Σ Mass of reactants) × 100% [14]	100%	Atom Economy (Principle 2)

The relationship between PMI and E-Factor is mathematically defined: E-Factor = PMI - 1 [10]. This interconnection allows researchers to calculate either metric when one is known. Industry data reveals significant variation in typical PMI values across sectors, with pharmaceutical manufacturing often exhibiting higher intensities (PMI = 25-100+) compared to bulk chemicals (PMI <1 to 5) [10], highlighting substantial opportunity for improvement in fine chemical and API synthesis.

Documented PMI Reductions in Pharmaceutical Applications

Table 2: Documented PMI Reduction Case Studies in Pharmaceutical Synthesis

API/Target Molecule	Initial PMI	Optimized PMI	Key Green Chemistry Improvements
Sertraline (Zoloft)	Not specified (E-Factor significantly reduced) [10]	E-Factor = 8 [10]	Process redesign, catalyst implementation, solvent optimization [10]
Sildenafil Citrate (Viagra)	E-Factor = 105 (discovery) [10]	E-Factor = 7 (production) [10]	Solvent recovery, elimination of volatile solvents [10]
Simvastatin	High legacy waste >100 kg/kg API [15]	~10-fold reduction [15]	Biocatalytic synthesis method reducing solvent use [15]
Dihydrocarvone from Limonene Epoxide	Not specified	RME = 0.63 (inversely related to PMI) [14]	Use of dendritic zeolite catalyst under green conditions [14]

These case studies demonstrate that catalysis (Principle 9) and safer solvent design (Principle 5) drive significant PMI reductions. For example, Pfizer's redesign of sertraline manufacturing eliminated or replaced several reagents and solvents, substantially reducing mass intensity while maintaining product quality [15].

Experimental Protocol: PMI Measurement and Application

Step-by-Step PMI Determination Protocol

Objective: Quantify the Process Mass Intensity for a chemical synthesis process to evaluate and improve its greenness.

Materials and Equipment:

• Analytical balance (precision ±0.001 g)
• Laboratory notebook or electronic data recording system
• PMI calculation spreadsheet or ACS GCI PMI Calculator [17]

Procedure:

• System Boundary Definition: Clearly establish the analysis scope:

  • Gate-to-Gate: Includes only materials used within the production facility [2]
  • Cradle-to-Gate: Extends to include upstream value chain materials [2]

• Material Input Inventory: Precisely measure and record all mass inputs for the process:

  • Reactants and reagents (stoichiometric and excess amounts)
  • Solvents (for reaction, workup, and purification)
  • Catalysts and processing aids
  • Water used in the process (Note: Some metrics exclude water [18])

• Product Mass Determination: Accurately measure the mass of the final isolated and purified product.

• PMI Calculation: Apply the standard PMI formula using the collected mass data.

• Data Interpretation and Optimization:

  • Compare calculated PMI against industry benchmarks (Table 1)
  • Identify major contributors to mass intensity
  • Develop process modifications to reduce high-mass inputs
  • Recalculate PMI for optimized process

Notes: For convergent syntheses, use the Convergent PMI Calculator to account for multiple synthesis branches [17]. The ACS GCI Pharmaceutical Roundtable has developed specialized PMI prediction tools for early-stage route selection [17].

Research Reagent Solutions for PMI Reduction

Table 3: Key Reagents and Materials for Green Process Optimization

Reagent/Material Category	Example Substances	Function in PMI Reduction
Heterogeneous Catalysts	Sn-dealuminated zeolites [14], dendritic ZSM-5 zeolites [14]	Enable higher atom economy, reduced reagent use, recyclability (Principle 9)
Biocatalysts	Enzymes for specific transformations [15]	High selectivity, mild conditions, reduced protection/deprotection steps
Green Solvents	2-Methyltetrahydrofuran, ethyl acetate, water [15]	Replace hazardous solvents, enable recycling, reduce auxiliary mass
Renewable Feedstocks	Plant-based biomolecules [16], R-(+)-limonene [14]	Implement Principle 7 (Renewable Feedstocks), often with better atom economy

Visualization: PMI Calculation and Principle Integration

PMI Calculation and Application Workflow

PMI Integration with Green Chemistry Principles

Advanced Implementation: Beyond Basic PMI

System Boundary Expansion for Comprehensive Assessment

While traditional gate-to-gate PMI (considering only factory inputs) remains valuable for internal process optimization, recent research demonstrates that expanding to cradle-to-gate Value-Chain Mass Intensity (VCMI) significantly improves correlation with Life Cycle Assessment (LCA) environmental impacts [2]. This approach accounts for upstream resource consumption across seven product classes, providing a more holistic sustainability perspective [2].

Implementation Protocol for VCMI:

• Map complete supply chain for all input materials
• Classify inputs according to Central Product Classification (CPC) system [2]
• Calculate cumulative mass intensity including upstream resource requirements
• Interpret results in context of specific environmental impact categories

Integrated Assessment Framework

For comprehensive environmental evaluation, PMI should be integrated with complementary metrics:

• Eco-Scale: Penalty-based system evaluating yield, safety, technical setup, and purification [18]
• Life Cycle Assessment (LCA): Comprehensive environmental impact assessment across multiple categories [2] [19]
• Toxicity Metrics: Evaluation of process hazards and environmental impacts [11]

The ACS GCI Pharmaceutical Roundtable has developed a PMI Life Cycle Assessment Tool that integrates mass efficiency with environmental impact assessment, enabling smarter sustainable decision-making [19].

Process Mass Intensity serves as a critical bridge between the conceptual framework of the 12 Principles of Green Chemistry and practical, measurable process improvements in pharmaceutical development and fine chemical synthesis. By providing a standardized methodology for quantifying material efficiency, PMI enables researchers to systematically reduce waste, optimize resource utilization, and demonstrate tangible progress toward sustainability goals.

The protocols and applications detailed in this document provide drug development professionals with implementable strategies for PMI integration throughout the research and development lifecycle. As the chemical industry continues its transition toward a defossilized, circular economy, PMI and its expanded derivatives will play an increasingly important role in guiding innovation toward genuinely sustainable outcomes [2]. Future methodology development should focus on integrating mass-based metrics with hazard assessment and life cycle impact evaluation to provide comprehensive sustainability guidance.

Process Mass Intensity (PMI) has emerged as a foundational green chemistry metric for quantifying the environmental efficiency of chemical processes, particularly in the pharmaceutical and fine chemical industries. PMI is defined as the total mass of materials used to produce a unit mass of the desired product, providing a comprehensive measure of resource utilization efficiency. This metric is calculated using the formula: PMI = Total Mass of Materials Input (kg) / Mass of Product (kg) [20] [10]. A lower PMI value indicates a more efficient and environmentally favorable process, as it signifies less waste generation and better resource management.

The strategic importance of PMI tracking extends beyond simple waste accounting. As the chemical industry faces increasing pressure to adopt sustainable practices, PMI provides a standardized measurement framework for benchmarking performance, setting improvement targets, and demonstrating environmental responsibility [11]. In pharmaceutical manufacturing, where multi-step syntheses often generate substantial waste, PMI optimization offers significant economic and environmental benefits through reduced raw material consumption, lower waste treatment costs, and improved process economics [10].

Industry Benchmark Data and Comparative Analysis

PMI and E-Factor Benchmarks Across Chemical Sectors

Extensive industry data reveals significant variation in PMI and E-Factor values across different chemical sectors, reflecting fundamental differences in process complexity, purity requirements, and technological maturity. The table below summarizes typical benchmark ranges for key industry segments:

Table 1: PMI and E-Factor Benchmarks Across Chemical Industry Sectors [10]

Industry Sector	Production Scale (tonnes/year)	E-Factor (kg waste/kg product)	PMI (kg input/kg product)
Oil Refining	10⁶-10⁸	<0.1	<1.1
Bulk Chemicals	10⁴-10⁶	<1.0 to 5.0	<2.0 to 6.0
Fine Chemicals	10²-10⁴	5.0 to >50	6.0 to >51
Pharmaceutical Industry	10-10³	25 to >100	26 to >101

The pharmaceutical industry consistently demonstrates the highest PMI values across chemical sectors, primarily due to multi-step synthetic pathways, stringent purity requirements, and complex purification processes [10]. These elevated PMI values translate to substantial waste generation, with E-Factors typically ranging from 25 to over 100 kg waste per kg of active pharmaceutical ingredient (API). The fine chemicals sector shows intermediate PMI values, reflecting its position between bulk chemicals and pharmaceuticals in terms of product complexity and purity specifications.

Pharmaceutical Fine Chemicals Market Context

The global fine chemicals market, valued at USD 210.7 billion in 2025 and projected to reach USD 391.8 billion by 2035 (CAGR 6.4%), provides important context for PMI benchmarking efforts [21]. Pharmaceutical applications dominate this market, accounting for approximately 35% of fine chemicals demand by product type and 38% by application [21]. This substantial market share underscores the critical importance of PMI optimization in pharmaceutical fine chemicals manufacturing, where small improvements in process efficiency can yield significant economic and environmental benefits across global supply chains.

Captive manufacturing (in-house production) accounts for approximately 60% of pharmaceutical fine chemicals production, while merchant manufacturers serve the remaining 40% [21]. Both segments are increasingly adopting green chemistry principles and PMI tracking, though captive facilities often have more direct control over process optimization and continuous improvement initiatives.

Experimental Protocols for PMI Determination

Standard Operating Procedure for PMI Calculation

Objective: To establish a standardized methodology for calculating Process Mass Intensity (PMI) in pharmaceutical and fine chemical processes.

Scope: This protocol applies to all chemical development and manufacturing activities involving API synthesis, intermediate production, and process optimization studies.

Materials and Equipment:

• Analytical balance (precision ±0.1 mg)
• Laboratory notebook or electronic data capture system
• Process flow diagram with identified input and output streams
• Material Safety Data Sheets (MSDS) for all process materials

Procedure:

• Process Boundary Definition: Clearly define the system boundaries for PMI assessment, including all process steps from raw material input to final product isolation and purification.

• Material Mass Documentation:

  • Record the mass of all input materials, including reactants, reagents, catalysts, and solvents used in the reaction and work-up steps.
  • Document the mass of all output materials, including the target product, by-products, and waste streams.
  • For multi-step syntheses, maintain separate mass balances for each discrete step and for the overall process.

• PMI Calculation:

  • Sum the total mass of all input materials within the defined system boundaries.
  • Determine the mass of the final isolated product meeting quality specifications.
  • Calculate PMI using the formula: PMI = Total Mass Input (kg) / Mass of Product (kg).

• Data Quality Assurance:

  • Verify mass balance closure (Inputs - Outputs = Accumulation) to ensure data integrity.
  • Reconcile any significant mass discrepancies (>5%) before finalizing PMI values.
  • Document all assumptions and methodological considerations for future reference.

• Reporting: Report PMI values with complete context, including process boundaries, calculation methodology, and any excluded materials with justification.

Table 2: Research Reagent Solutions for PMI Determination

Reagent/Material	Function in PMI Analysis	Application Context
Analytical Balance	Precise mass measurement of inputs and products	Quantification of all process materials
Process Solvents	Reaction medium, extraction, purification	Major contributor to total PMI
Catalysts	Reaction rate enhancement without consumption	Consideration for PMI allocation
Drying Agents	Product purification and isolation	Inclusion in total mass input
Reference Standards	Method validation and verification	Ensuring calculation accuracy

Advanced PMI Assessment Protocol

For comprehensive environmental assessment, the basic PMI calculation can be extended to provide additional insights:

Differentiated PMI Analysis:

• Calculate solvent PMI separately from reaction mass PMI to identify key improvement opportunities.
• Determine catalyst and reagent PMI to guide substitution with more efficient alternatives.
• Compute water-specific PMI for aqueous processes to assess water utilization efficiency.

Normalization and Benchmarking:

• Normalize PMI values against reaction theoretical yield to calculate PMI efficiency.
• Compare experimental PMI against theoretical minimum PMI based on reaction stoichiometry.
• Benchmark process PMI against industry standards for similar chemical transformations.

Workflow for PMI Implementation and Optimization

The following diagram illustrates the systematic approach to PMI implementation and process optimization in pharmaceutical and fine chemical development:

Case Studies and Industry Applications

Pharmaceutical Industry Success Stories

Several pharmaceutical companies have demonstrated substantial PMI improvement through targeted process optimization:

• Sildenafil Citrate (Viagra) Manufacturing: Process optimization efforts, including solvent substitution and recovery implementation, reduced the E-Factor from 105 during initial development to 7 in commercial production, with a target of 4 for future improvements [10]. This corresponds to a PMI reduction from 106 to 8, approaching the target of 5.

• Sertraline Hydrochloride (Zoloft) Synthesis: Redesign of the synthetic pathway achieved an E-Factor of 8 (PMI = 9) through improved catalyst efficiency and solvent selection [10].

These case studies demonstrate that systematic process optimization can achieve 80-90% reduction in PMI values, translating to significant environmental and economic benefits through reduced raw material consumption, waste generation, and processing costs.

Integration with Green Chemistry Principles

PMI optimization aligns directly with multiple green chemistry principles, particularly Atom Economy and Waste Prevention [20] [11]. The relationship between these metrics provides a comprehensive framework for environmental assessment:

• Atom Economy focuses on maximizing the incorporation of starting materials into the final product [20].
• PMI provides a practical measure of actual material utilization in laboratory or industrial settings.
• E-Factor quantifies waste generation, calculated as PMI - 1 [10].

This integrated metrics approach enables researchers to identify specific improvement opportunities, track progress toward sustainability goals, and demonstrate environmental performance to regulators, customers, and other stakeholders.

Process Mass Intensity has established itself as a critical performance indicator for environmental efficiency in pharmaceutical and fine chemical manufacturing. The industry benchmarks presented in this document provide valuable reference points for organizations seeking to assess and improve their processes. Typical PMI values ranging from 26 to over 101 for pharmaceutical operations highlight the substantial improvement potential available through systematic application of green chemistry principles [10].

Future developments in PMI optimization will likely focus on continuous manufacturing technologies, biocatalytic processes, and artificial intelligence-driven process design [21]. As regulatory pressure for sustainable manufacturing intensifies and life cycle assessment methodologies become more sophisticated, PMI tracking will evolve from a voluntary improvement tool to an essential component of chemical process design and operation. The protocols and benchmarks provided in this document offer a foundation for organizations to build robust PMI assessment capabilities and drive continuous improvement in their environmental performance.

Implementing PMI: From Calculation to Real-World Application in API Synthesis

A Step-by-Step Guide to Calculating PMI for a Chemical Process

Process Mass Intensity (PMI) is a key green chemistry metric used to benchmark the environmental efficiency and sustainability of chemical processes, particularly in the pharmaceutical industry [22]. It provides a comprehensive measure of the total mass of materials used to produce a unit mass of the final product. Within the broader context of green chemistry metrics research, PMI offers a holistic view that encompasses all inputs—including solvents, reagents, catalysts, and process chemicals—enabling scientists and engineers to identify opportunities for reducing waste, lowering costs, and minimizing environmental impact [22]. The adoption of PMI has helped the pharmaceutical industry focus attention on the main drivers of process inefficiency and environmental impact, making it an essential tool for developing more sustainable manufacturing processes [22].

Principles and Calculation Methodology

Definition and Formula

Process Mass Intensity is defined as the total mass of materials entering a process divided by the mass of the final product. The formula for calculating PMI is:

PMI = Total Mass of Inputs (kg) / Mass of Product (kg)

A perfect PMI value is 1, indicating that all input materials are incorporated into the final product with no waste. In practice, however, PMI values are always greater than 1 due to solvents, catalysts, and other materials that are not incorporated into the final product but are necessary for the reaction and purification processes. Lower PMI values indicate more efficient and environmentally friendly processes.

Relationship to Other Green Metrics

PMI does not exist in isolation but forms part of a suite of complementary green chemistry metrics. The table below summarizes key green metrics and their relationship to PMI:

Table 1: Key Green Chemistry Metrics and Their Relationship to PMI

Metric	Calculation Formula	Relationship to PMI
Atom Economy (AE)	(Molecular Weight of Product / Molecular Weight of Reactants) × 100%	Focuses on molecular efficiency; high AE often contributes to lower PMI but doesn't account for solvents or process materials [14].
Reaction Mass Efficiency (RME)	(Mass of Product / Total Mass of Reactants) × 100%	Directly related to PMI; RME = 1/PMI when considering only reactants, but PMI includes all process materials [14].
Reaction Yield (ɛ)	(Actual Mass of Product / Theoretical Mass of Product) × 100%	High yield typically contributes to lower PMI but doesn't guarantee low PMI if excessive solvents or reagents are used [14].
Material Recovery Parameter (MRP)	Measure of solvent and material recovery efficiency	Directly impacts PMI; higher MRP values significantly reduce PMI by enabling material reuse [14].

Experimental Protocols for PMI Determination

Data Collection Protocol

• Define Process Boundaries: Clearly establish the start and end points of the process being evaluated (e.g., from starting materials to isolated API).
• Catalog All Input Materials: Record the masses of all materials used, including:
  • Starting materials and reagents
  • Catalysts
  • Solvents (for reaction, workup, and purification)
  • Process aids and additives
• Measure Product Output: Accurately weigh the final isolated and purified product.
• Document Reaction Conditions: Note temperature, time, and any special processing conditions that might affect material usage.
• Account for Material Recovery: Track and quantify any solvents or materials recovered for reuse.

Calculation Workflow

The following diagram illustrates the step-by-step workflow for PMI calculation:

Case Study Application

Recent research on catalytic processes for fine chemicals demonstrates the practical application of PMI and related metrics. In the synthesis of dihydrocarvone from limonene-1,2-epoxide using dendritic zeolite d-ZSM-5/4d, excellent green characteristics were achieved with AE = 1.0, ɛ = 0.63, 1/SF = 1.0, MRP = 1.0, and RME = 0.63 [14]. This corresponds to a PMI of approximately 1.59 (1/0.63), making it an outstanding catalytic material for biomass valorization research [14].

Table 2: PMI and Green Metrics Comparison for Fine Chemical Processes

Chemical Process	Atom Economy (AE)	Reaction Yield (ɛ)	Reaction Mass Efficiency (RME)	Calculated PMI
Epoxidation of R-(+)-limonene	0.89	0.65	0.415	2.41
Synthesis of florol via isoprenol cyclization	1.0	0.70	0.233	4.29
Synthesis of dihydrocarvone from limonene-1,2-epoxide	1.0	0.63	0.63	1.59

Advanced PMI Considerations and Tools

Convergent Synthesis Calculations

For complex multi-step syntheses, particularly convergent routes common in pharmaceutical manufacturing, the original PMI calculator was enhanced to create the Convergent PMI Calculator by the ACS GCI Pharmaceutical Roundtable [22]. This tool uses the same fundamental calculations but allows for multiple branches in single-step or convergent synthesis, enabling more accurate assessment of complex reaction sequences.

Material Recovery Scenarios

The sustainability of chemical processes improves significantly with better material recovery [14]. When calculating PMI, it's essential to consider different recovery scenarios:

• No Recovery: Base case PMI without any solvent or material recycling.
• Partial Recovery: PMI accounting for recovery of high-volume solvents.
• Full Recovery: Theoretical PMI with comprehensive material recycling systems.

Research shows that evaluating all three scenarios provides crucial insights for process optimization and highlights the potential benefits of implementing recovery technologies [14].

The Researcher's Toolkit for PMI Analysis

Table 3: Essential Research Reagent Solutions for PMI-Optimized Chemistry

Tool/Resource	Function/Application	Relevance to PMI Reduction
ACS GCI PMI Calculator	Web-based tool for calculating Process Mass Intensity	Enables quick determination of PMI values by accounting for raw material inputs relative to API output [22].
Convergent PMI Calculator	Enhanced tool for complex synthetic routes	Accommodates multiple branches for single-step or convergent synthesis, providing accurate PMI for pharmaceutical development [22].
Heterogeneous Catalysts	Reusable catalysts for various transformations	Reduces PMI by eliminating or reducing catalyst waste; examples include Sn4Y30EIM and d-ZSM-5/4d zeolites [14].
Solvent Recovery Systems	Equipment for distillation and purification	Directly improves PMI by enabling solvent reuse, significantly reducing material consumption [14].
Radial Pentagon Diagrams	Graphical evaluation tool for multiple green metrics	Visualizes all five key green metrics simultaneously, helping assess overall process greenness and identify improvement areas [14].

Strategic Implementation and Continuous Improvement

Effective PMI management requires integrating PMI tracking throughout the development lifecycle, from initial route selection to commercial manufacturing. The relationship between various green metrics can be visualized as follows:

Regular PMI benchmarking against industry standards and tracking PMI trends throughout process optimization enables continuous improvement. The pharmaceutical industry's ongoing PMI benchmarking exercises since 2008 have demonstrated significant progress in developing more efficient and sustainable processes [22]. By making PMI a key performance indicator throughout development and manufacturing, organizations can systematically drive innovations that reduce environmental impact while improving economic performance.

In the pursuit of a more sustainable chemical industry, accurately evaluating the environmental performance of processes, particularly in pharmaceutical and fine chemical synthesis, is paramount. Life Cycle Assessment (LCA) is the recommended holistic method for this evaluation but faces practical barriers due to its extensive data requirements and the time-consuming nature of studies [2]. Consequently, simple mass-based green chemistry metrics, especially the Process Mass Intensity (PMI), have been widely adopted as practical proxies for environmental performance [2] [23].

The reliability of these metrics, however, is fundamentally governed by the system boundaries chosen for their calculation. The selection between a gate-to-gate and a cradle-to-gate boundary is not merely a technicality; it is a critical decision that determines the scope of environmental impacts considered and the validity of the resulting conclusions. This application note delineates the distinctions between these system boundaries, provides protocols for their consistent application, and presents recent findings on their efficacy in approximating full life cycle environmental impacts, framed within contemporary green chemistry metrics research.

Defining the System Boundaries

The terms "gate-to-gate" and "cradle-to-gate" define the scope of analysis for assessing a product's environmental footprint, as illustrated in the following diagram.

Comparative Analysis of Boundaries

The choice of system boundary directly influences the insights and conclusions drawn from an environmental assessment. The table below summarizes the key characteristics of each approach.

Table 1: Comparative Analysis of Gate-to-Gate and Cradle-to-Gate System Boundaries

Feature	Gate-to-Gate	Cradle-to-Gate
Definition	Assesses impacts only from the factory entrance to the factory exit [2].	Assesses impacts from raw material extraction ("cradle") to the factory gate [24].
Scope Included	Mass and energy flows directly within the manufacturing process [2].	Raw material extraction, transport, and production of all input materials, plus the gate-to-gate manufacturing process [2] [25].
Primary Use Case	Internal process efficiency tracking and optimization [24].	Evaluating the total upstream resource footprint of a product; basis for Environmental Product Declarations (EPDs) [24] [25].
Data Requirements	Lower; requires detailed process mass balance data from internal operations [2].	Higher; requires lifecycle inventory (LCI) data for all raw materials and precursors [2] [26].
Advantages	Faster, simpler, and useful for comparing internal process variations [24].	Provides a more complete picture of upstream impacts and supply chain dependencies [2].
Limitations	Fails to capture significant environmental burdens in the supply chain, offering an incomplete picture [2] [27].	Does not consider impacts from product distribution, use, or end-of-life phases [24].

Quantitative Correlation with LCA Impacts

Key Research Findings

Recent systematic research has quantitatively evaluated the ability of mass intensity metrics with different boundaries to approximate the multi-criteria environmental impacts of a full LCA. A 2025 study analyzed the correlation between sixteen LCA environmental impact categories and eight mass intensities with varying system boundaries for 106 chemical productions [2].

The core finding was that expanding the system boundary from gate-to-gate (PMI) to cradle-to-gate (Value-Chain Mass Intensity, VCMI) strengthened the correlation for fifteen of the sixteen environmental impacts [2]. This confirms that a gate-to-gate PMI is not a robust proxy for environmental performance, while a cradle-to-gate perspective is significantly more reliable.

The same study revealed that the strength of correlation is highly dependent on the specific environmental impact category and the specific product classes included in the VCMI boundary. This is because each environmental impact is approximated by a distinct set of key input materials, which serve as proxies for the processes they imply (e.g., coal consumption correlates with climate change impact due to associated CO₂ emissions) [2].

Table 2: Correlation of Mass Intensity Metrics with Selected LCA Impact Categories (Based on Eichwald et al., 2025) [2]

LCA Environmental Impact Category	Correlation with Gate-to-Gate PMI	Correlation with Cradle-to-Gate VCMI	Key Influential Input Materials/Processes
Climate Change	Weak	Strengthened	Fossil fuels (e.g., coal, natural gas)
Water Usage	Weak	Strengthened [2]	Water-intensive agriculture and mining
Resource Depletion	Weak	Strengthened	Metal ores, rare earth elements
Toxicity-Related Impacts	Weak	Varies	Specific chemicals and solvents used in the supply chain

Experimental Protocols

Protocol 1: Calculating Gate-to-Gate Process Mass Intensity (PMI)

1. Goal and Scope Definition:

• Objective: To determine the mass efficiency of an internal chemical manufacturing process.
• System Boundary: The physical confines of the chemical plant, from the input of purchased raw materials at the factory gate to the output of the finished product at the factory gate [2].

2. Data Collection:

• Compile a total mass balance for the process. Accurately record the masses (in kg) of all input materials entering the process. This includes:
  • All reactants and reagents.
  • All solvents (including water).
  • Catalysts and other auxiliaries.
• Record the mass (in kg) of the final product produced.

3. Calculation:

• The PMI is calculated using the following equation. All masses should be consistent units (typically kg).

PMI = (Total Mass of Input Materials) / (Mass of Product)

• The result is a dimensionless number representing the mass of input required per unit mass of product. A lower PMI indicates higher mass efficiency.

Protocol 2: Conducting a Cradle-to-Gate Life Cycle Assessment

1. Goal and Scope Definition (ISO 14040):

• Define the product and functional unit (e.g., 1 kg of Active Pharmaceutical Ingredient - API).
• Define the system boundary: from resource extraction (cradle) to the final product leaving the manufacturing facility (gate) [28] [24].

2. Life Cycle Inventory (LCI) Compilation:

• Create a detailed bill of materials for the gate-to-gate process.
• For each input material, collect lifecycle inventory data quantifying all relevant resource extractions and emissions (e.g., CO₂, SO₂, waste) associated with its production. This data is typically sourced from commercial LCA databases (e.g., ecoinvent) [2] or supplier-specific information.
• Model the gate-to-gate manufacturing process, including energy consumption, waste treatment, and direct emissions.

3. Life Cycle Impact Assessment (LCIA):

• Use LCIA methodology to translate the LCI data into potential environmental impacts. Common categories include Global Warming Potential, Acidification Potential, and Water Consumption [28].

4. Interpretation:

• Analyze the results to identify environmental "hotspots" within the supply chain and manufacturing process to guide sustainable development efforts.

The Scientist's Toolkit: Key Research Reagent Solutions

When transitioning from simple metrics to more comprehensive assessments, researchers rely on a suite of tools and data sources.

Table 3: Essential Materials and Tools for Environmental Assessment Research

Item / Tool Name	Function / Description	Application Context
Life Cycle Inventory (LCI) Databases	Databases containing pre-compiled environmental flow data for thousands of materials and processes (e.g., ecoinvent) [2].	Essential for building cradle-to-gate models without primary data for every input.
Fast LCA Software Tools	Simplified LCA tools (e.g., FLASC [23], Streamlined PMI-LCA Tool [26]) tailored for rapid assessment in pharmaceutical development.	Allows for quick, iterative environmental screening of synthetic routes during process development.
Chemical Tree Visualization	A systems-thinking tool to map all inputs, outputs, and emissions for each synthesis step [27].	Provides a visual gate-to-gate inventory and aids in identifying hotspots for improvement.
iGAL 2.0 Metric	A metric that compares PMI and other factors to benchmark data, expressing the result as a Relative Process Greenness (RPG) index [23].	Provides an industry benchmark for evaluating the greenness of a new process compared to existing ones.

The choice between gate-to-gate and cradle-to-gate system boundaries is a critical one with significant implications for the perceived environmental performance of a chemical process. While the gate-to-gate PMI is a useful tool for internal mass efficiency benchmarking, it is an unreliable proxy for the broader multi-criteria environmental impacts assessed by LCA. A cradle-to-gate boundary, which accounts for upstream supply chain effects, provides a significantly more robust approximation for most environmental impacts, including climate change and resource use.

The ongoing transition towards a defossilized chemical industry will further challenge the reliability of mass-based metrics, as the environmental impact of materials becomes decoupled from their mass. Therefore, the future of accurate environmental assessment in green chemistry lies not in refining mass intensities alone, but in the continued development and adoption of simplified, accessible, and streamlined LCA methods that more directly and accurately reflect environmental performance [2].

In the pursuit of sustainable pharmaceutical manufacturing, Process Mass Intensity (PMI) has emerged as a pivotal green chemistry metric. PMI measures the total mass of materials used to produce a specified mass of active pharmaceutical ingredient (API), providing a comprehensive assessment of process efficiency and environmental impact [11]. The drive to reduce PMI aligns with the United Nations Sustainable Development Goals, particularly Goal 12 concerning responsible consumption and production [11]. This case study examines the application of PMI reduction strategies within the context of a clinical candidate's development, demonstrating how green chemistry principles can be systematically implemented to create more efficient and environmentally responsible pharmaceutical processes.

The pharmaceutical industry faces increasing pressure to minimize its environmental footprint while maintaining cost-effectiveness and regulatory compliance. Traditional drug development processes often involve multi-step syntheses with low atom economy and substantial waste generation. Through targeted PMI reduction strategies, including catalyst optimization, solvent selection, and process intensification, significant improvements in mass efficiency can be achieved while maintaining product quality and safety profiles.

Green Chemistry Metrics in Pharmaceutical Development

Foundational Green Metrics

Green chemistry metrics provide quantitative frameworks for evaluating the environmental performance of chemical processes. The most widely adopted metrics in pharmaceutical development include:

• Atom Economy (AE): Atom Economy, introduced by Trost, calculates the theoretical efficiency of a chemical reaction by comparing the molecular weight of the desired product to the total molecular weight of all reactants [11]. It represents the ideal scenario where all atoms are incorporated into the final product.

• Reaction Mass Efficiency (RME): RME measures the practical efficiency by accounting for reaction yield, stoichiometry, and auxiliary materials [14]. It provides a more realistic assessment of process performance compared to AE.

• Process Mass Intensity (PMI): PMI offers the most comprehensive assessment by including all mass inputs—reactants, solvents, catalysts, and processing aids—relative to the product output [11]. This holistic approach makes PMI particularly valuable for evaluating overall environmental impact and resource utilization.

• E-Factor: Developed by Sheldon, the E-Factor quantifies waste generation by measuring the mass of waste produced per unit mass of product [11]. Pharmaceutical processes typically exhibit higher E-Factors compared to bulk chemicals due to complex synthesis pathways and purification requirements.

Table 1: Core Green Chemistry Mass Metrics

Metric	Calculation	Application Focus	Industry Standard
Atom Economy (AE)	(MW product / Σ MW reactants) × 100%	Theoretical maximum efficiency	Fundamental reaction design
Reaction Mass Efficiency (RME)	(Mass product / Σ Mass reactants) × 100%	Actual reaction performance	Laboratory-scale optimization
Process Mass Intensity (PMI)	Total mass in process / Mass product	Overall process efficiency	Manufacturing and scale-up
E-Factor	(Total mass waste / Mass product)	Environmental waste impact	Environmental impact assessment

Advanced Metric Frameworks

Beyond foundational metrics, comprehensive green chemistry assessment requires multi-parameter evaluation. Radial pentagon diagrams serve as powerful graphical tools for simultaneous visualization of five key metrics: Atom Economy (AE), reaction yield (ɛ), stoichiometric factor (SF), material recovery parameter (MRP), and Reaction Mass Efficiency (RME) [14]. This visualization technique enables rapid identification of process limitations and optimization opportunities.

For the synthesis of fine chemicals and pharmaceutical intermediates, recent case studies demonstrate excellent green characteristics with specific metric values. For instance, the synthesis of dihydrocarvone from limonene-1,2-epoxide using dendritic zeolite d-ZSM-5/4d exhibited outstanding metrics: AE = 1.0, ɛ = 0.63, 1/SF = 1.0, MRP = 1.0, and RME = 0.63 [14]. These values represent a benchmark for efficient catalytic processes in pharmaceutical manufacturing.

Case Study: PMI Reduction for Clinical Candidate GX-247

Compound Background and Synthetic Challenges

Clinical candidate GX-247 is a novel kinase inhibitor targeting inflammatory pathways, with a complex molecular structure featuring multiple stereocenters and functional groups. The initial synthetic route developed during discovery phase exhibited a PMI of 387, indicating significant resource consumption and waste generation. Key challenges included:

• Low-yielding stereoselective steps (Step 3: 45% yield)
• High solvent usage for chromatography and purification
• Toxic heavy metal catalysts (Pd, Cr) requiring extensive removal
• Cryogenic conditions (-78°C) demanding substantial energy input
• Multiple protection/deprotection sequences increasing step count

The development team established a target PMI of 85 for the commercial process, representing a 78% reduction from the initial route while maintaining >99.5% chemical purity and meeting all regulatory requirements for API quality.

PMI Reduction Strategy and Implementation

A systematic approach to PMI reduction was implemented across the synthetic route, focusing on the highest mass contributors identified through material flow analysis.

Table 2: PMI Reduction Strategy Implementation

Strategy	Initial Approach	Optimized Approach	PMI Impact
Catalysis	Stoichiometric metal oxidants (CrO₃)	Heterogeneous catalytic oxidation (2 mol% TiO₂-ZrO₂)	-22 PMI points
Solvent System	Multiple solvent changes (DMF, CH₂Cl₂, THF)	Ethanol-water biphasic system	-41 PMI points
Workup & Purification	Column chromatography (3 steps)	Crystallization-driven purification	-118 PMI points
Reaction Mass Efficiency	Low-yielding steps (45-65%)	Improved kinetics and catalyst design (82-91%)	-47 PMI points
Reagent Selection	Protecting groups (Boc, Cbz)	Protecting-group free synthesis	-74 PMI points

The implementation of these strategies resulted in a final PMI of 79, exceeding the initial target while reducing hazardous waste generation by 68% and eliminating all heavy metal catalysts from the process.

Experimental Protocols for PMI Assessment

PMI Calculation Methodology

Objective: To standardize the calculation of Process Mass Intensity for consistent evaluation and comparison of synthetic routes.

Materials:

• Analytical balance (precision ±0.1 mg)
• Process flow diagram with all input streams
• Mass records for all process steps

Procedure:

• Document all mass inputs to the process, including:
  • Starting materials and intermediates
  • Solvents (reaction, extraction, washing, crystallization)
  • Catalysts and reagents
  • Processing aids (filter aids, drying agents)

• Exclude process water from mass calculations if it is fully treated and returned to the environment without contamination.

• Measure the mass of isolated and dried final product using calibrated analytical balances.

• Calculate PMI using the formula: PMI = Total mass of inputs (kg) / Mass of product (kg)

• For multi-step syntheses, calculate both overall PMI and individual step PMI to identify improvement opportunities.

• Record PMI values alongside reaction yield, purity data, and E-Factor for comprehensive environmental assessment.

Validation: Perform mass balance closure within ±5% for the entire process to ensure accurate PMI calculation.

Catalyst Screening Protocol for PMI Reduction

Objective: To identify and optimize catalytic systems that maximize atom economy while minimizing environmental impact.

Materials:

• Library of heterogeneous and homogeneous catalysts
• High-throughput screening platform
• Analytical HPLC with UV/Vis detection
• Green Solvent Selection Guide

Procedure:

• Conduct initial screening of 20-50 catalyst candidates using high-throughput experimentation under standardized conditions.

• Evaluate catalysts based on multiple parameters:

  • Conversion (%) by NMR or HPLC
  • Selectivity to desired product (%)
  • Turnover Number (TON) and Turnover Frequency (TOF)
  • Catalyst loading (mol%)
  • Separation and recyclability potential

• For promising candidates (>90% conversion, >95% selectivity), optimize reaction parameters:

  • Temperature (30-100°C range)
  • Pressure (1-20 bar if applicable)
  • Solvent system (prioritizing water, ethanol, or 2-MeTHF)
  • Catalyst loading (0.1-5 mol%)

• Assess catalyst removal and recycling:

  • For heterogeneous catalysts: filtration efficiency and activity after 5 cycles
  • For homogeneous catalysts: extraction efficiency and metal contamination in API

• Calculate PMI contribution of catalytic system including:

  • Catalyst mass
  • Ligands and additives
  • Solvents for catalyst removal

• Select optimal catalyst system based on balanced assessment of performance, PMI impact, and environmental factors.

Visualization of PMI Optimization Workflow

Diagram 1: PMI Optimization Workflow. This flowchart illustrates the systematic approach to reducing Process Mass Intensity through iterative analysis and improvement.

The Scientist's Toolkit: Research Reagent Solutions

Table 3: Essential Reagents for Green Process Development

Reagent/Category	Function in PMI Reduction	Green Characteristics	Application Notes
Heterogeneous Catalysts (TiO₂-ZrO₂, supported metals)	Enable catalyst recovery and recycling	Reduced metal leaching, >5 reuses possible	Implement fixed-bed reactors for continuous processing
Bio-based Solvents (2-MeTHF, Cyrene, limonene)	Replace halogenated and toxic solvents	Renewable feedstocks, improved EHS profile	Assess water miscibility for waste stream management
Immobilized Reagents (polymer-supported catalysts, scavengers)	Simplify workup and purification	Filtration removal, reduced metal contamination	Monitor loading capacity and swelling behavior
Water as Reaction Medium	Dramatically reduce solvent PMI	Non-flammable, non-toxic, inexpensive	Requires surfactant or phase-transfer catalysts for organics
Switchable Solvents (CO₂-triggered polarity changes)	Combine reaction and extraction steps	Reversible properties, recycling capability	Useful for homogeneous catalysis with product isolation

Results and Discussion

PMI Reduction Achievement and Analysis

The implementation of systematic PMI reduction strategies for clinical candidate GX-247 resulted in a final PMI of 79, representing a 78% reduction from the initial process PMI of 387. This achievement demonstrates that substantial improvements in process efficiency are attainable through targeted optimization strategies.

The most significant PMI reductions came from crystallization-driven purification (-118 PMI points) and the development of a protecting-group-free synthesis (-74 PMI points). These findings align with established green chemistry principles that emphasize waste prevention and atom economy [11]. The optimized process also eliminated all hazardous solvents and heavy metal catalysts, significantly improving the environmental, health, and safety (EHS) profile without compromising product quality.

Analysis of mass distribution across the optimized process revealed that solvents remained the largest mass contributor at 64% of total inputs, despite a 73% reduction in solvent usage compared to the initial process. This highlights the ongoing challenge of solvent selection and recovery in pharmaceutical manufacturing and indicates a focus area for future improvements.

Environmental and Economic Impact

The PMI reduction translated to substantial environmental and economic benefits:

• Reduced waste generation: 68% decrease in total waste mass, with elimination of 3,450 kg of hazardous waste per 100 kg API produced
• Lower energy consumption: 42% reduction in process energy requirements through elimination of cryogenic conditions and reduced distillation demands
• Cost savings: 31% reduction in raw material costs despite higher catalyst expenses
• Improved safety profile: Elimination of all mutagenic reagents and heavy metals

The case study demonstrates that green chemistry metrics like PMI serve as effective proxies for broader environmental impact assessment, correlating with reduced resource consumption and waste generation across multiple categories.

This case study establishes that systematic PMI reduction is both achievable and economically advantageous in pharmaceutical development. The strategies implemented for clinical candidate GX-247—particularly catalyst optimization, solvent system redesign, and purification method improvements—resulted in a 78% reduction in PMI while maintaining stringent API quality standards.

The successful application of green chemistry metrics throughout the development process provided quantitative guidance for decision-making and prioritization of optimization efforts. The radial evaluation of multiple metrics, including Atom Economy, Reaction Mass Efficiency, and material recovery parameters [14], proved invaluable for balanced process assessment beyond single-parameter optimization.

As the pharmaceutical industry continues to embrace sustainability goals, PMI reduction represents a critical methodology for aligning environmental responsibility with economic objectives. The framework presented in this case study provides a replicable model for implementing green chemistry principles across diverse synthetic pathways, contributing to the development of more sustainable pharmaceutical manufacturing practices.

In the pursuit of sustainable pharmaceutical manufacturing, green-by-design strategies have become paramount. These strategies rely on robust metrics to quantify environmental impact and guide decision-making from the earliest stages of process development [26]. Among these metrics, Process Mass Intensity (PMI) is widely recognized as a key indicator, measuring the total mass of materials used per unit mass of active pharmaceutical ingredient (API) produced [20] [22]. This application note explores the integration of predictive analytics with PMI assessment through specialized tools that enable researchers to forecast the probable efficiencies of synthetic routes prior to laboratory evaluation. By leveraging historical manufacturing data and predictive algorithms, these tools facilitate more sustainable route selection during the scouting phase, potentially reducing the environmental footprint of pharmaceutical development [29] [30].

The ACS Green Chemistry Institute (GCI) Pharmaceutical Roundtable has championed this approach, developing computational tools that empower chemists and engineers to incorporate sustainability considerations during molecular design and route selection [30] [31]. This paradigm shift toward in-silico sustainability assessment represents a significant advancement in green chemistry, allowing for identification of potential waste hotspots and efficiency opportunities before resources are committed to laboratory experimentation [31].

Green Chemistry Metrics and PMI Fundamentals

The Centrality of Process Mass Intensity

Process Mass Intensity has emerged as a comprehensive metric for evaluating the efficiency of chemical processes, particularly in pharmaceutical manufacturing. PMI is calculated as the total mass of all materials entering a process divided by the mass of the final product [22]. This encompasses reagents, solvents, catalysts, and process aids, providing a holistic view of resource utilization efficiency. The pharmaceutical industry has extensively adopted PMI as its primary green metric because it directly correlates with waste generation, environmental impact, and production costs [20] [22].

Unlike simpler metrics such as atom economy which only consider the fate of reactant atoms in the final product, PMI accounts for all materials used throughout the synthetic process, including those employed in workup and purification steps [20] [11]. This comprehensive perspective makes PMI particularly valuable for benchmarking and optimizing complex multi-step synthetic sequences common in API manufacturing [22]. The relationship between PMI and the older E-Factor metric is straightforward: E-Factor = PMI - 1, highlighting their fundamental connection in quantifying process waste [10].

Comparative Analysis of Green Chemistry Metrics

Table 1: Comparison of Key Green Chemistry Metrics

Metric	Calculation	Scope	Advantages	Limitations
Process Mass Intensity (PMI)	Total mass inputs / Mass of product [22]	Comprehensive (all materials)	Holistic; covers entire process; industry standard [20]	Doesn't differentiate material types [26]
Atom Economy	(MW of product / Σ MW of reactants) × 100% [20]	Molecular structure	Theoretical maximum efficiency; simple calculation [11]	Ignores yield, solvents, auxiliaries [11]
E-Factor	(Total waste / Mass of product) [10]	Waste production	Simple; highlights waste generation [10]	Doesn't account for waste nature/toxicity [10]
Effective Mass Yield	(Mass of product / Mass of non-benign reagents) × 100% [11]	Hazardous materials	Focuses on hazardous material reduction [11]	Subjective classification of "non-benign" [11]

The PMI Prediction Calculator: Architecture and Predictive Methodology

Tool Development and Data Foundation

The PMI Prediction Calculator represents a significant innovation in predictive green chemistry tools. Developed through leadership from Bristol-Myers Squibb under the ACS GCI Pharmaceutical Roundtable, this web application leverages a substantial historical dataset of pharmaceutical manufacturing processes to generate PMI predictions for proposed synthetic routes [30]. The underlying database contains approximately 2,000 data points collected from multiple pharmaceutical companies, representing diverse synthetic methodologies and development phases [30]. This extensive dataset undergoes continuous refinement through annual data collection initiatives focused on populating under-represented reaction subtypes, enhancing the model's predictive accuracy across chemical space [30].

The tool employs Monte Carlo simulations to estimate probable PMI ranges based on the structural and procedural characteristics of proposed synthetic routes [30]. This probabilistic approach acknowledges the inherent variability in process development and provides researchers with a realistic efficiency range rather than a single-point estimate. The application of this predictive framework enables quantitative sustainability assessment during the ideation phase of route design, bringing environmental considerations to the forefront of synthetic planning [29].

Integration with Green-by-Design Workflows

The PMI Prediction Calculator exemplifies the green-by-design philosophy by facilitating early sustainability assessment rather than retrospective optimization [26]. This proactive approach stands in contrast to traditional development timelines where environmental considerations often occurred after process establishment. The tool integrates within a broader ecosystem of green chemistry resources developed by the ACS GCI Pharmaceutical Roundtable, including solvent selection guides, reagent guides, and PMI calculation tools [32].

A notable advancement in this field is the integration of PMI prediction with subsequent experimental optimization using Bayesian optimization (BO) approaches [31]. This combined methodology was recognized with the 2024 Data Science & Modeling for Green Chemistry Award, highlighting its innovative approach to sustainable process development [31]. In one documented case, this integrated approach achieved an optimized process with 80% yield and 91% enantiomeric excess in only 24 experiments, surpassing results obtained through traditional one-factor-at-a-time optimization that required 500 experiments [31].

Figure 1: PMI Prediction and Optimization Workflow. This diagram illustrates the integrated workflow combining PMI prediction with subsequent experimental optimization.

Experimental Protocols and Application Methodology

Protocol for Predictive PMI Assessment in Route Scouting

Objective: To evaluate and compare the predicted Process Mass Intensity of proposed synthetic routes during early-stage route scouting using the PMI Prediction Calculator.

Materials and Input Requirements:

• Defined target molecular structure
• Proposed synthetic route with complete reaction sequence
• Identification of reaction types and subtypes for each transformation
• Knowledge of intended development phase (early vs. commercial)

Procedure:

• Route Characterization: Deconstruct the proposed synthetic route into individual steps and classify each according to the reaction taxonomy available in the PMI Prediction Calculator.
• Data Input: Access the web application at https://acsgcipr-predictpmi.shinyapps.io/pmi_calculator/ and input the required route information.
• Simulation Execution: Execute the predictive algorithm which applies Monte Carlo methods based on historical data for similar reaction types.
• Result Interpretation: Analyze the output PMI range in the context of industry benchmarks, considering both the median prediction and probability distribution.
• Comparative Analysis: Repeat the process for alternative synthetic routes to enable data-driven route selection.
• Sensitivity Assessment: Identify steps contributing disproportionately to the overall PMI prediction for targeted optimization efforts.

Interpretation Guidelines:

• Compare predicted PMI values against industry benchmarks: early development (typically 100-500), commercial processes (typically 50-150), and theoretical minima [10].
• Consider both the median prediction and the probability distribution when making route selection decisions.
• Use the identification of high-PMI steps to guide focused optimization efforts or alternative chemistry investigation.

Protocol for Streamlined PMI-LCA Assessment

Objective: To conduct a combined environmental assessment integrating PMI with life cycle considerations using the Streamlined PMI-LCA Tool.

Background: While PMI provides a mass-based efficiency metric, it does not differentiate between materials based on their environmental impact profiles [26]. The Streamlined PMI-LCA Tool addresses this limitation by incorporating environmental footprint data for synthesis raw materials while maintaining manageable data requirements [26].

Procedure:

• PMI Determination: Calculate the total Process Mass Intensity using standard PMI calculators [22].
• Material Classification: Categorize all input materials according to their environmental impact profiles using embedded databases.
• Impact Weighting: Apply environmental impact weighting factors to different material classes.
• Composite Scoring: Generate a combined score reflecting both mass efficiency and environmental impact considerations.
• Iterative Re-evaluation: Conduct periodic reassessments throughout process development to guide prioritization of optimization efforts.

Application Context: This methodology was successfully applied in the development of the MK-7264 active pharmaceutical ingredient, contributing to a reduction of PMI from 366 to 88 over the course of process optimization [26].

Table 2: Research Reagent Solutions for Predictive Green Chemistry

Tool/Resource	Function	Access Information
PMI Prediction Calculator	Predicts probable PMI ranges for proposed synthetic routes prior to laboratory evaluation [29] [30]	https://acsgcipr-predictpmi.shinyapps.io/pmi_calculator/ [29]
Process Mass Intensity Calculator	Calculates actual PMI from experimental data for benchmarking developed processes [32] [22]	Available through ACS GCI Pharmaceutical Roundtable resources [22]
Convergent PMI Calculator	Enhanced PMI calculation for convergent syntheses with multiple branches [32]	Available through ACS GCI Pharmaceutical Roundtable resources [32]
Streamlined PMI-LCA Tool	Combines PMI with environmental footprint data for raw materials [26]	Methodology described in literature; implementation varies [26]
EDBO/EDBO+ Platform	Bayesian optimization platform for accelerated reaction optimization with reduced experimentation [31]	Open-source platform available at no charge [31]
ACS GCI Solvent Selection Guide	Guides selection of greener solvents based on health, safety, and environmental criteria [32]	Available through ACS GCI resources [32]

Case Studies and Performance Validation

Pharmaceutical Implementation and Efficiency Gains

The implementation of PMI prediction tools in pharmaceutical development has demonstrated significant value in guiding sustainable process selection. In a documented case involving a real clinical candidate, researchers employed the PMI Prediction Calculator to evaluate multiple synthetic approaches prior to laboratory investigation [31]. This quantitative prediction method enabled the selection of the most efficient synthetic strategy based on sustainability metrics alongside traditional considerations such as complexity and cost.

The subsequent application of Bayesian optimization using the EDBO/EDBO+ platform demonstrated how predictive analytics can further enhance sustainability gains during process optimization [31]. In the highlighted case, traditional one-factor-at-a-time optimization requiring 500 experiments achieved 70% yield and 91% enantiomeric excess. In contrast, the Bayesian optimization approach identified conditions providing 80% yield and 91% enantiomeric excess in only 24 experiments [31]. This represents an order-of-magnitude reduction in experimental burden while simultaneously improving key performance metrics.

Industry Adoption and Impact Assessment

The pharmaceutical industry has increasingly adopted PMI as a standard benchmarking metric since the ACS GCI Pharmaceutical Roundtable conducted its first PMI benchmarking exercise in 2008 [22]. Regular benchmarking has helped the industry focus attention on the main drivers of process inefficiency, cost, and environment, safety and health impact [22]. The progression in tool development—from simple PMI calculators to predictive applications—reflects the growing sophistication of green chemistry implementation within the sector [22].

The predictive tools have proven particularly valuable in addressing the characteristically high E-Factors (25 to >100) and PMI values of pharmaceutical manufacturing compared to other chemical sectors [10]. These elevated values result from multi-step syntheses, rigorous purity requirements, and complex separation protocols [10]. By front-loading sustainability considerations to the route scouting phase, the PMI Prediction Calculator helps mitigate these inherent challenges of pharmaceutical manufacturing.

Figure 2: Traditional vs. Predictive Route Development. Comparative workflow demonstrating efficiency gains through predictive approaches.

The integration of predictive analytics with green chemistry metrics represents a transformative advancement in sustainable pharmaceutical development. PMI prediction tools enable a proactive approach to environmental considerations, moving sustainability from a retrospective assessment to a forward-looking design criterion. By leveraging historical manufacturing data and statistical modeling, these tools provide quantitative guidance during route scouting that was previously unavailable.

The continued evolution of these methodologies—including integration with life cycle assessment principles and machine learning optimization—promises to further enhance their utility and impact [26] [31]. As these tools become more sophisticated and widely adopted, they will play an increasingly vital role in reducing the environmental footprint of pharmaceutical manufacturing while maintaining the economic and technical requirements of drug development.

For researchers engaged in synthetic route development, embracing these predictive tools offers the opportunity to embed sustainability deeply within the research and development workflow, truly realizing the promise of green-by-design chemistry in pharmaceutical applications.

Incorporating PMI into Early-Stage Drug Development and Process Design

The drive towards a more sustainable chemical industry has made evaluating the environmental impact of chemical processes, particularly in pharmaceutical development, more critical than ever [2]. While Life Cycle Assessment (LCA) is the recommended method for holistic environmental evaluation, practical barriers like extensive data requirements and confidentiality often limit its application in early-stage development [2]. In this context, "green chemistry metrics," and specifically the Process Mass Intensity (PMI), have emerged as indispensable practical tools for chemists and engineers to quantify and improve the environmental profile of their processes during development phases [14] [2].

PMI is defined as the total mass of materials used to produce a unit mass of the target product. It is calculated as: PMI = Total Mass Input (kg) / Mass of Product (kg). A lower PMI indicates a more efficient and less wasteful process. Unlike simple yield or atom economy, PMI provides a comprehensive view of the total mass balance, including solvents, reagents, and processing aids, offering a more realistic picture of the resource efficiency of a drug synthesis pathway [2]. Its application in early-stage development allows for the rapid comparison of alternative synthetic routes and the identification of key areas for sustainable process optimization before significant capital investment is committed.

Quantitative Green Metrics for Process Assessment

While PMI is a cornerstone metric, a holistic evaluation requires a suite of complementary green metrics to provide a balanced view of process efficiency. These metrics, when used together, offer insights into different aspects of reaction and process performance. The following table summarizes the key green metrics used in conjunction with PMI for a comprehensive assessment [14].

Table 1: Key Green Chemistry Metrics for Process Evaluation

Metric Name	Abbreviation	Calculation Formula	Optimal Value	What It Measures
Atom Economy	AE	(MW of Product / Σ MW of Reactants) × 100%	100%	Efficiency of incorporating reactant atoms into the final product.
Reaction Yield	ɛ	(Moles of Product Obtained / Moles of Limiting Reactant) × 100%	100%	Efficiency of the chemical conversion itself.
Stoichiometric Factor	SF	Σ (Moles of All Reagents / Moles of Limiting Reactant)	1.0	Excess mass of reagents used beyond stoichiometric requirements.
Material Recovery Parameter	MRP	Mass of Recovered Solvents & Reagents / Total Mass Input	1.0	Effectiveness of solvent and reagent recovery and recycling.
Reaction Mass Efficiency	RME	(Mass of Product / Σ Mass of Reactants) × 100%	100%	Overall mass efficiency of the reaction step, combining yield and atom economy.
Process Mass Intensity	PMI	Total Mass Input (kg) / Mass of Product (kg)	1.0 (Lower is better)	Total mass footprint per unit of product, including all inputs.

Radial pentagon diagrams are a powerful tool for the graphical evaluation of these five key green metrics (AE, ɛ, 1/SF, MRP, and RME), providing an immediate visual assessment of a process's "greenness" and helping to pinpoint specific areas for improvement [14].

Case Study Data: Green Metrics in Fine Chemical Synthesis

Recent research into catalytic processes for fine chemicals demonstrates the practical application of these metrics. The following table compiles quantitative data from case studies, illustrating how different chemical transformations perform against these green metrics [14].

Table 2: Green Metrics from Catalytic Process Case Studies in Fine Chemicals [14]

Case Study	Catalytic System	Atom Economy (AE)	Reaction Yield (ɛ)	1/SF	MRP	RME	Implied PMI*
Epoxidation of R-(+)-limonene	K–Sn–H–Y-30-dealuminated zeolite	0.89	0.65	0.71	1.0	0.415	~2.41
Synthesis of Florol	Sn4Y30EIM	1.0	0.70	0.33	1.0	0.233	~4.29
Synthesis of Dihydrocarvone	Dendritic zeolite d-ZSM-5/4d	1.0	0.63	1.0	1.0	0.63	~1.59

Note: PMI is approximated as 1/RME for these single-step examples for comparative purposes. Full process PMI includes all auxiliary materials.

The data reveals significant variations. For instance, the dihydrocarvone synthesis exhibits excellent characteristics with high atom economy, stoichiometric efficiency (1/SF=1.0), and the best RME of 0.63, making it a standout candidate for further research [14]. In contrast, the Florol synthesis, while having perfect atom economy, suffers from a poor stoichiometric factor (1/SF=0.33), drastically reducing its RME and increasing its PMI. This highlights the critical importance of optimizing reagent excesses and work-up procedures, not just the core reaction chemistry.

Defining System Boundaries for PMI Calculation

A major challenge in using PMI is the lack of standardized system boundaries, which can lead to misleading comparisons and an inaccurate picture of environmental impact [2]. The system boundary defines which materials and processes are included in the "Total Mass Input."

Gate-to-Gate vs. Cradle-to-Gate

The most fundamental distinction lies between gate-to-gate and cradle-to-gate boundaries [2]:

• Gate-to-Gate PMI: This is the traditional Process Mass Intensity (PMI). It considers all materials that enter the factory gate where the chemical is produced, including reactants, solvents, catalysts, and processing aids. It ends with the final product leaving the gate. This boundary is useful for internal process optimization but ignores the upstream footprint of the materials used.
• Cradle-to-Gate PMI (VCMI): This expanded metric, which can be termed Value-Chain Mass Intensity (VCMI), includes the gate-to-gate PMI plus the mass of all natural resources extracted from the environment ("the cradle") required to produce the input materials [2]. This provides a more complete picture of the total resource use associated with the product.

Recent systematic analyses have shown that expanding the system boundary from gate-to-gate to cradle-to-gate strengthens the correlation between mass intensity and a wider range of LCA environmental impacts (fifteen out of sixteen were improved) [2]. This underscores the importance of a broader boundary for meaningful environmental assessment.

A Framework for Systematic Boundary Expansion

To operationalize the cradle-to-gate PMI (VCMI), a structured approach is needed. One method involves categorizing all value-chain products into distinct classes and systematically including them. The following workflow outlines a protocol for defining and calculating PMI across different system boundaries, crucial for consistent reporting in drug development.

Diagram 1: A framework for defining PMI system boundaries.

This systematic expansion allows researchers to understand how including different upstream product classes (e.g., basic chemicals, energy carriers, agricultural products) affects the total mass intensity and its reliability as a proxy for environmental impacts like climate change [2]. The findings indicate that a single mass-based metric cannot fully capture the multi-criteria nature of environmental sustainability, as each environmental impact is approximated by a distinct set of key input materials [2].

Experimental Protocols for PMI Determination

Protocol 1: Determining Gate-to-Gate PMI for a Reaction Step

Objective: To accurately calculate the gate-to-gate Process Mass Intensity for a single chemical reaction step in a drug synthesis pathway.

Principles: This protocol focuses on the mass efficiency of a discrete chemical transformation as performed in the laboratory or pilot plant. It is the foundational measurement for internal benchmarking and initial route scouting.

Materials and Reagents:

• The target API intermediate or drug substance.
• All reactants, reagents, catalysts, and solvents used in the reaction.
• All solvents, acids, bases, and processing aids used in the work-up and purification stages.

Procedure:

• Perform the Reaction: Carry out the synthetic step according to the established experimental procedure on a defined scale (e.g., 1-100g).
• Isolate and Dry Product: Isolate the final product after work-up and purification (e.g., crystallization, distillation). Dry the product to constant weight to determine the net mass of pure product obtained.
• Record Mass Inputs: Accurately record the masses of ALL materials used from the point they enter the reaction vessel. This includes:
  • Masses of all reactants.
  • Masses of all catalysts and reagents.
  • Volumes and densities (to calculate masses) of all solvents for the reaction, work-up, and purification.
  • Masses of any other materials used, such as drying agents, chromatography media, or filtration aids.
• Calculate Total Mass Input: Sum all masses recorded in Step 3.
• Calculate PMI: Divide the Total Mass Input by the Mass of Dry Product obtained in Step 2.

PMI = (Total Mass of Inputs) / (Mass of Product)

Reporting: Report the PMI value for the step. Clearly state the system boundary (e.g., "includes reaction and work-up solvents but excludes chromatography"). Report the reaction yield and purity of the product alongside the PMI.

Protocol 2: Defining System Boundaries for Cradle-to-Gate VCMI

Objective: To establish a standardized methodology for expanding the PMI system boundary to include upstream mass inputs from the value chain, creating a Cradle-to-Gate Value-Chain Mass Intensity (VCMI).

Principles: This protocol provides a systematic approach to account for the mass footprint of input materials, recognizing that the environmental impact of a process depends on its entire supply chain.

Procedure:

• Identify Key Mass Inputs: From the Gate-to-Gate PMI calculation (Protocol 1), identify the input materials that contribute the most to the total mass. Typically, solvents and large-mass reagents are prioritized.
• Map Upstream Production Pathways: For each key input, research or model its industrial production pathway. For example, the common solvent methanol may be produced from syngas (derived from natural gas), which in turn comes from natural resource extraction.
• Categorize Inputs into Product Classes: Classify each key input into a standardized product classification system, such as the Central Product Classification (CPC) [2]. Example classes include:
  • CPC 1: Agriculture, Forestry, and Fishery Products.
  • CPC 2: Ores and Minerals; Electricity, Gas, and Water.
  • CPC 3: Food, Beverages, and Tobacco; Textiles, Apparel, and Leather Products.
  • CPC 4: Basic Chemicals.
  • ... and other classes.
• Calculate Cumulative Mass from Upstream: Using LCA databases (e.g., ecoinvent) or literature values, determine the cumulative mass of natural resources (e.g., crude oil, natural gas, metal ores, water) required to produce one kilogram of each key input material. This is often represented as the Material Footprint or similar indicator.
• Calculate VCMI for Each Boundary: Systematically calculate the VCMI by expanding the system boundary to include different sets of product classes. For example:
  • VCMICPC4: PMI + upstream mass from Basic Chemicals (CPC 4).
  • VCMICPC4+2: PMI + upstream mass from Basic Chemicals and Ores/Minerals/Energy (CPC 4 & 2).
• Correlate with LCA Impacts: For a portfolio of processes, analyze the Spearman correlation coefficients between the different VCMIs (with their varying boundaries) and a suite of LCA environmental impacts (e.g., climate change, water use, land use) [2].

Reporting: Report the VCMI values for the defined system boundaries (e.g., VCMI_CPC4 = X). Discuss how the correlation with specific environmental impacts changes as the boundary expands. This analysis reveals which system boundary is most appropriate for approximating a particular environmental concern.

The Scientist's Toolkit: Key Reagents and Materials

Successful implementation of green chemistry principles and accurate PMI tracking relies on specific materials and tools. The following table details essential items for a lab focused on sustainable process development.

Table 3: Essential Research Reagent Solutions and Materials for Green Chemistry and PMI Studies

Item Name	Function/Application	Green Chemistry Rationale
Heterogeneous Catalysts (e.g., Zeolites, Supported Metals)	Facilitate chemical reactions without being consumed. Can be filtered and reused.	Reduces reagent waste (improving 1/SF and PMI), avoids heavy metal contamination from homogeneous catalysts, and simplifies purification [14].
Alternative Solvents (e.g., 2-MeTHF, Cyrene, Water)	Replace traditional, more hazardous or high-PMI solvents (e.g., DCM, DMF, THF).	Can be derived from renewable resources, often have better recycling potential, and reduce overall process mass and toxicity [14].
Atom-Efficient Reagents	Reagents designed with high molecular weight incorporation into the product (e.g., cascade reaction substrates).	Directly improves Atom Economy (AE) and Reaction Mass Efficiency (RME), leading to a lower PMI by minimizing byproduct formation.
In-Line Analytical Tools (e.g., FTIR, ReactRaman)	Real-time monitoring of reaction progress and endpoints.	Enables optimization of reagent stoichiometry and reaction time, preventing excess reagent use and improving yield (ɛ), which lowers PMI.
LCA Database Subscription (e.g., ecoinvent)	Provides secondary data on the environmental impacts and material footprints of chemicals and energy.	Essential for performing credible cradle-to-gate VCMI calculations and for correlating mass-based metrics with broader environmental impacts [2].

Strategies for PMI Reduction in Drug Development

Reducing PMI requires a multi-faceted approach targeting the largest contributors to mass waste in pharmaceutical processes. The following diagram illustrates the primary strategic pillars for PMI reduction and their logical interplay.

Diagram 2: A strategic framework for reducing PMI.

• Route and Molecule Design: The most significant reductions in PMI are often achieved at the earliest stage by designing synthetic routes with high inherent atom economy and minimizing protective group manipulations. This is a direct application of the first principle of green chemistry: Prevention [14].
• Reaction Optimization: Replacing stoichiometric reagents with catalytic alternatives is a powerful lever for reducing PMI. Furthermore, using in-line analytical tools to precisely determine reaction endpoints allows for the minimization of reagent excess, directly improving the stoichiometric factor (1/SF) and RME [14].
• Solvent Selection and Recovery: Solvents frequently constitute 80-90% of the total mass of a pharmaceutical process. A rigorous solvent selection guide favoring safer, recoverable options, coupled with investment in distillation and recovery systems, dramatically reduces the PMI. The Material Recovery Parameter (MRP) is a key metric here [14].
• Process Integration and Intensification: Telescoping multiple reaction steps without intermediate isolation and purification eliminates the mass of solvents and reagents associated with those steps. Continuous processing often enables more efficient mixing, heat transfer, and smaller equipment, leading to lower overall material and energy footprints compared to batch operations.

The integration of Process Mass Intensity into early-stage drug development and process design represents a critical step towards a more sustainable and economically viable pharmaceutical industry. While the simple gate-to-gate PMI is a valuable tool for internal benchmarking and rapid decision-making, this review underscores the necessity of expanding the system boundary to a cradle-to-gate perspective (VCMI) for a more environmentally meaningful assessment [2]. The provided protocols for PMI determination, the toolkit of reagents, and the strategic framework for reduction offer researchers and scientists a practical roadmap for implementation. However, it is crucial to recognize that mass intensity is a proxy metric. Future research should focus on developing and validating simplified LCA methods that can be deployed early in development, ensuring that mass-based improvements genuinely translate into reduced environmental impacts across a multi-criteria spectrum [2]. By systematically applying these principles, the pharmaceutical industry can significantly advance its green chemistry goals.

Optimizing PMI: Strategies and Advanced Tools for Greener Processes

In the pursuit of sustainable pharmaceutical manufacturing, Process Mass Intensity (PMI) has emerged as a crucial green chemistry metric for assessing environmental performance. PMI measures the total mass of materials required to produce a unit mass of an active pharmaceutical ingredient (API), providing a comprehensive indicator of resource efficiency and waste generation [2]. The pharmaceutical industry faces significant environmental challenges, with API manufacturing accounting for nearly half of the sector's emissions in the purchased goods and services category [33]. The average PMI for API manufacturing ranges from 70 to 433 kg of materials per kg of API produced, with typical synthesis yields between 30-60% for small molecules and as low as 5-10% for complex syntheses [33]. This application note examines the primary contributors to high PMI—solvents, reagents, and water—within the context of green chemistry metrics application research, providing drug development professionals with standardized protocols for identification, measurement, and reduction strategies.

Quantitative Analysis of PMI Contributors

Relative Contribution of Input Materials

Comprehensive analysis of pharmaceutical manufacturing processes reveals consistent patterns in mass distribution across input categories. The following table summarizes the typical contribution of each category to the overall PMI based on industry data:

Table 1: Typical Percentage Contribution of Input Categories to Total PMI

Input Category	Typical Contribution to PMI (%)	Range Observed in Industry	Key Factors Influencing Variability
Solvents	56-65%	45-80%	Process design, recovery efficiency, selection criteria
Water	20-28%	15-35%	Cooling requirements, extraction methods, purification needs
Reagents & Raw Materials	10-15%	5-25%	Synthetic route complexity, atom economy, catalysis strategy
Catalysts & Auxiliaries	3-8%	1-12%	Catalyst loading, recycling capability, ligand design

Solvents dominate the PMI profile of most pharmaceutical processes, accounting for over 60% of all processed materials in many cases [34]. This predominance stems from their extensive use in reaction media, extraction procedures, and purification steps, particularly in traditional batch processes. Water represents the second most significant contributor, with usage patterns varying substantially based on synthetic biology approaches versus traditional chemical synthesis [33] [34].

Environmental Impact Correlation

While PMI serves as a mass-based efficiency metric, its correlation with environmental impact measures varies significantly based on the nature of the materials. Recent research demonstrates that expanding system boundaries from gate-to-gate to cradle-to-gate strengthens correlations between mass intensity and fifteen of sixteen environmental impact categories in Life Cycle Assessment (LCA) [2]. However, the relationship remains complex due to several factors:

• Solvent waste treatment emissions: Incineration of solvents emits 2-4 kg of CO₂ per kg of solvent, contributing significantly to the carbon footprint [33]
• Energy-intensive reagents: Certain reagents, particularly those requiring complex synthesis or precious metals, carry disproportionate environmental burdens relative to their mass [35]
• Water sourcing impacts: The environmental impact of water varies significantly based on local scarcity and treatment requirements, not adequately captured by mass alone [2]

Experimental Protocols for PMI Assessment

Comprehensive Material Inventory Analysis

Objective: To establish a standardized methodology for quantifying all material inputs throughout API synthesis and identifying major contributors to PMI.

Materials:

• Process flow diagrams with mass balance data
• Batch production records
• Supplier documentation for all input materials
• Analytical balances (precision ±0.001 g)
• Solvent recovery efficiency data

Procedure:

• Define system boundaries: Establish cradle-to-gate boundaries encompassing all value chain processes from raw material extraction to API isolation [2]
• Document mass inputs: Record masses of all materials entering the process, categorized as:
  • Solvents (reaction, extraction, washing, purification)
  • Reagents (starting materials, catalysts, oxidizing/reducing agents)
  • Water (process, cooling, purification)
  • Auxiliary materials (filter aids, drying agents)
• Calculate category-specific mass intensities:
  • Solvent Intensity (SI) = Total solvent mass (kg) / API mass (kg)
  • Water Intensity (WI) = Total water mass (kg) / API mass (kg)
  • Reagent Intensity (RI) = Total reagent mass (kg) / API mass (kg)
• Normalize results: Express all intensities per 1 kg of final API product
• Identify hotspots: Rank contributors by mass percentage and environmental impact potential

Data Interpretation: Calculate overall PMI using the formula: PMI = Total mass of inputs (kg) / Mass of API product (kg) Category-specific PMI contributions should be expressed as percentages of the total PMI. Processes with solvent contributions exceeding 60% or water contributions exceeding 30% warrant immediate investigation for reduction opportunities.

Reagent and Solvent Assessment Protocol

Objective: To evaluate the environmental profile and efficiency of reagents and solvents used in API synthesis.

Materials:

• Life Cycle Inventory databases (e.g., ecoinvent)
• Safety Data Sheets
• Green chemistry assessment tools (SANESS, CHEM21)
• Solvent selection guides (ACS GCI Pharmaceutical Roundtable)

Procedure:

• Characterize reagent efficiency:
  • Calculate Atom Economy for each synthetic step: AE = (MW product / Σ MW reactants) × 100%
  • Determine Reaction Mass Efficiency: RME = (Mass product / Σ Mass reactants) × 100%
• Assess solvent environmental impact:
  • Categorize solvents using GSK or CHEM21 solvent selection guide criteria
  • Calculate solvent recovery efficiency: % Recovery = (Mass recycled / Mass total used) × 100%
  • Determine waste treatment emissions: CO₂ emissions = Mass incinerated × Emission factor (2-4 kg CO₂/kg)
• Evaluate water usage profile:
  • Differentiate between process, cooling, and purification water
  • Identify opportunities for recycling and reuse
  • Assess water source sustainability factors

Data Interpretation: Reagents with atom economy below 40% or reaction mass efficiency below 50% represent priority optimization targets. Solvents categorized as "red" in selection guides or with recovery rates below 30% should be targeted for substitution or recovery improvement.

Figure 1: Systematic PMI Assessment Workflow

The Scientist's Toolkit: Research Reagent Solutions

Essential Materials for PMI Reduction

Table 2: Key Research Reagents and Solutions for PMI Reduction

Reagent/Solution Category	Specific Examples	Function in PMI Reduction	Application Notes
Green Solvents	2-MethylTHF, Cyrene, dimethyl isosorbide	Replace hazardous solvents with bio-based alternatives	Select based on ACS GCI Pharmaceutical Roundtable solvent guide; assess recyclability
Catalytic Systems	Pd nanoparticles, immobilized enzymes, biphasic catalysts	Reduce stoichiometric reagent usage through catalysis	Focus on recoverable and reusable systems; evaluate metal scarcity
Alternative Reagents	Polymer-supported reagents, flow chemistry compatible agents	Enable purification without solvent-intensive chromatography	Particularly valuable for peptide synthesis and complex molecule assembly
Process Monitoring Tools	In-line IR/UV sensors, PAT (Process Analytical Technology)	Enable real-time reaction monitoring and endpoint detection	Minimize excess reagent use and solvent washes through precise control
Separation & Recovery Aids	Molecular sieves, membrane filters, distillation apparatus	Facilitate solvent and reagent recovery for reuse	Implement automated stripping/distillation systems for solvent recovery

Case Studies: Successful PMI Reduction Strategies

Merck's Antibody-Drug Conjugate Linker Synthesis

Merck's award-winning approach to ADC linker manufacturing demonstrates substantial PMI reduction through strategic process redesign. The team achieved a 75% reduction in PMI by reengineering their synthetic route from a 20-step sequence to a more convergent approach that eliminated seven steps and reduced the Process Mass Intensity by approximately 75% [36]. Key strategies included:

• Solvent intensity reduction: Decreased energy-intensive chromatography time by >99% through alternative purification methods
• Feedstock optimization: Sourced starting materials from a widely available natural product
• Process simplification: Reduced synthetic steps from 20 to 13 while maintaining product quality

This approach transformed a manufacturing bottleneck with production limited to less than 100 g per month into a scalable process, highlighting how PMI-focused process redesign simultaneously improves sustainability and supply reliability.

Olon S.p.A.'s Peptide Synthesis Platform

Olon's microbial fermentation platform for therapeutic peptide production represents a paradigm shift in PMI reduction for complex biomolecules. Their novel Fermentation Platform employs recombinant DNA technology and chimeric protein expression to achieve significant improvements over traditional Solid Phase Peptide Synthesis (SPPS) [36]:

• Solvent reduction: Eliminates extensive solvent use associated with SPPS protection/deprotection cycles
• Material efficiency: Minimizes excess building blocks by requiring no protecting groups
• Process intensification: Master Cell Bank implementation enables logarithmic cell proliferation to maximize product yield in fermenters

This platform approach, currently being applied to GLP-1 peptides and other therapeutic targets, demonstrates how biological synthesis routes can dramatically reduce PMI compared to traditional chemical synthesis.

Advanced Methodologies for PMI Optimization

Life Cycle Assessment Integration

While PMI provides valuable mass-based efficiency data, integration with Life Cycle Assessment (LCA) enables more comprehensive environmental impact evaluation. Recent advances demonstrate:

• Iterative LCA approaches: Bridging life cycle assessment with multistep synthesis development provides holistic sustainability evaluation [35]
• Expanded system boundaries: Cradle-to-gate assessments strengthen correlations between mass intensity and environmental impacts [2]
• Hotspot identification: LCA reveals that certain process steps, particularly metal-mediated couplings and asymmetric catalysis, contribute disproportionately to environmental impacts despite moderate mass contributions [35]

The synthesis of Letermovir illustrates this approach, where LCA identified Pd-catalyzed Heck cross-coupling as a sustainability hotspot despite its moderate PMI contribution, guiding development of alternative synthetic strategies [35].

Data Science and Modeling Approaches

Computational tools are increasingly enabling PMI reduction through predictive optimization:

• Algorithmic Process Optimization (APO): Merck and Sunthetics developed technology using Bayesian Optimization to locate global optima in complex operational spaces, minimizing material use and selecting non-toxic reagents [36]
• Predictive modeling: Tools leveraging machine learning guide design of sustainable chemical processes with reduced process mass intensity [37]
• Multi-objective optimization: Advanced algorithms simultaneously optimize for cost, material efficiency, and environmental impact, identifying solutions that traditional approaches might miss [36]

These computational approaches enable sustainable process design while minimizing experimental resource consumption, creating a virtuous cycle of PMI reduction.

Systematic identification and reduction of major PMI contributors—particularly solvents, reagents, and water—represents a critical pathway toward sustainable pharmaceutical manufacturing. The protocols and methodologies outlined in this application note provide researchers with standardized approaches for quantifying PMI contributions, identifying optimization opportunities, and implementing reduction strategies. Successful case studies demonstrate that PMI reduction of 75% or more is achievable through strategic process redesign, alternative synthesis routes, and advanced technologies. As the pharmaceutical industry advances toward its net-zero emissions commitments, PMI optimization will remain an essential component of comprehensive sustainability programs, delivering both environmental and economic benefits through more efficient resource utilization.

The Role of Catalysis and Atom Economy in PMI Reduction

In the modern pharmaceutical industry, the drive toward sustainability is increasingly guided by the principles of Green Chemistry. Central to this effort is the reduction of the Process Mass Intensity (PMI), a key metric defined as the total mass of materials used to produce a specified mass of product. High PMI values indicate inefficient processes that generate significant waste, increasing environmental impact and manufacturing costs. Within this framework, catalysis and atom economy emerge as fundamental strategies for PMI reduction. Catalytic processes enable more efficient transformations, while atom economy measures the incorporation of starting material atoms into the final product, minimizing waste generation at the molecular level. Together, these approaches facilitate the development of more sustainable synthetic pathways that align with the 3Rs (Reduction, Refinement, and Replacement) principles now being advocated through New Approach Methodologies (NAMs) in regulatory science [38].

The pharmaceutical sector faces particular challenges in implementing green chemistry principles due to the complex multi-step syntheses often required for active pharmaceutical ingredients (APIs). These processes traditionally involve stoichiometric reagents, protective groups, and purification steps that dramatically increase PMI. The industry is therefore increasingly focusing on catalytic methodologies that offer superior selectivity and reduce the need for excess reagents and solvents, directly addressing PMI at its source. This application note details practical protocols and metrics for researchers to implement these principles, supported by experimental data and visualization of key concepts.

Quantitative Green Chemistry Metrics Framework

Core Metric Definitions and Calculations

Green chemistry performance is quantified through several interconnected metrics that provide complementary perspectives on process efficiency. The most relevant for PMI reduction include:

• Process Mass Intensity (PMI): Total mass of all materials used in the process (kg) divided by the mass of product (kg). PMI is always ≥ 1, with lower values indicating greater efficiency.
• Atom Economy (AE): Molecular weight of the desired product divided by the sum of molecular weights of all reactants, expressed as a percentage. Ideal reactions have 100% atom economy.
• E-Factor: Total mass of waste (kg) divided by the mass of product (kg). Directly related to PMI through the equation: PMI = E-Factor + 1.
• Reaction Mass Efficiency (RME): Mass of product divided by the total mass of reactants, expressed as a percentage.

These metrics enable researchers to benchmark processes and identify opportunities for improvement, particularly through catalytic strategies that enhance atom utilization [11].

Comparative Metric Analysis of Methodologies

The following table summarizes key green chemistry metrics for different synthetic approaches, highlighting the advantages of catalytic systems:

Table 1: Comparative Green Chemistry Metrics for Synthetic Methodologies

Synthetic Methodology	Typical Atom Economy	Typical PMI Range	E-Factor Range	Key Advantages
Stoichiometric Reactions	30-50%	40-100	39-99	Simple execution
Traditional Catalysis	50-80%	20-50	19-49	Improved selectivity, reduced waste
Advanced Catalysis (C-H Activation)	80-95%	5-20	4-19	Atom-economical, step-efficient
Biocatalysis	70-90%	10-30	9-29	High selectivity, mild conditions
Ideal Green Synthesis	100%	1-5	0-4	Minimal waste, maximum efficiency

Catalytic C-H activation methods represent particularly significant advances, as demonstrated by the synthesis of 2-arylquinolines through palladium-catalyzed intramolecular C-H activation. This approach achieves up to 64% yield with significantly improved atom economy compared to classical synthetic routes such as the Skraup, Friedländer, and Doebner-Miller reactions, which typically involve multi-step processes, harsh conditions, and stoichiometric oxidants with consequently lower atom economy and higher PMI [39].

Experimental Protocols for Catalytic PMI Reduction

Protocol: Palladium-Catalyzed Synthesis of 2-Arylquinolines via C-H Activation

Principle and Objective

This protocol implements a palladium-catalyzed intramolecular C-H activation for the sustainable synthesis of pharmaceutically relevant 2-arylquinolines from simple allyl amines. The methodology demonstrates superior atom economy and reduced environmental impact compared to conventional approaches, directly contributing to PMI reduction goals [39].

Reagents and Materials

Table 2: Research Reagent Solutions for Catalytic Quinoline Synthesis

Reagent/Material	Specification	Role/Function	Handling Precautions
Pd(OAc)₂	99%, anhydrous	Palladium catalyst precursor	Moisture-sensitive, store under inert atmosphere
1,10-Phenanthroline (1,10-phen)	99%	Nitrogen-based ligand	Air-stable, but store in cool, dark location
Substrate 1a	>95% purity	Allyl amine starting material	Standard chemical handling
Acetic Acid (AcOH)	Glacial, 99.8%	Reaction solvent	Corrosive, use in fume hood
Deionized H₂O	HPLC grade	Co-solvent (10% v/v)	Standard laboratory grade
Air atmosphere	-	Green oxidant	-
Nitrogen atmosphere	Inert gas	For sensitive manipulations	-

Equipment and Instrumentation

• Round-bottom flask (25-50 mL)
• Reflux condenser
• Magnetic stirrer with heating capability
• Oil bath or heating mantle
• Temperature controller
• Schlenk line or nitrogen/vacuum manifold (optional)
• TLC plates (silica)
• NMR spectrometer for yield determination

Step-by-Step Procedure

• Reaction Setup: Charge a 25 mL round-bottom flask with substrate 1a (0.25 mmol, 1.0 equiv), Pd(OAc)₂ (20 mol%), and 1,10-phenanthroline (20 mol%).

• Solvent Addition: Add the solvent mixture AcOH:H₂O (9:1 v/v, 0.025 M concentration) to the reaction vessel.

• Reaction Initiation: Heat the reaction mixture to 90°C with continuous stirring under air atmosphere (serves as the terminal oxidant).

• Progress Monitoring: Monitor reaction progress by TLC analysis at 4-hour intervals until completion (typically 24 hours).

• Product Isolation: After completion, cool the reaction mixture to room temperature and dilute with ethyl acetate (10 mL).

• Purification: Transfer to a separatory funnel, wash with saturated NaHCO₃ solution (2 × 5 mL) followed by brine (5 mL). Dry the organic layer over anhydrous Na₂SO₄, filter, and concentrate under reduced pressure.

• Characterization: Purify the crude product by flash column chromatography (silica gel, hexane/ethyl acetate gradient) to afford the desired 2-arylquinoline product. Characterize the final compound by ( ^1 \text{H} ) NMR, ( ^{13} \text{C} ) NMR, and mass spectrometry.

Optimization Notes

Critical parameters for success include:

• Temperature Control: Maintaining 90°C is essential; higher temperatures (130°C) promote competing aza-Cope rearrangement side reactions, reducing yield to 26%.
• Solvent System: The AcOH:H₂O (9:1) mixture is optimal; pure AcOH gives 60% yield, while other solvents (DMSO, HFIP, PivOH) give significantly lower yields (0-11%).
• Oxidant Selection: Air is the most effective and greenest oxidant; other oxidants like Cu(OAc)₂ or Ag₂CO₃ give lower yields (21-22%) with higher E-factors.
• Catalyst Loading: 20 mol% Pd(OAc)₂ provides optimal yield (57%); lower loading (10 mol%) reduces yield to 49%.

Protocol: Green Chemistry Metrics Calculation

Principle and Objective

This protocol provides a standardized methodology for calculating key green chemistry metrics to quantitatively evaluate process efficiency and environmental impact, enabling objective comparison between different synthetic approaches.

Calculation Procedures

• Atom Economy Calculation:

• PMI Calculation:

• E-Factor Calculation:

• Reaction Mass Efficiency:

Example Calculation

For the palladium-catalyzed quinoline synthesis producing 64% yield of 2a:

• Atom Economy: Calculate based on molecular structures of reactants and product
• PMI: Account for all reagents, catalysts, solvents, and workup materials
• Process E-Factor: Include all waste streams from reaction and purification

Visualization of Catalytic PMI Reduction Concepts

Workflow for Catalytic PMI Reduction Strategy

Diagram 1: Strategic workflow for implementing catalytic PMI reduction in synthetic processes

Atom Economy in Catalytic C-H Activation

Diagram 2: Comparative impact of catalytic versus traditional synthesis on atom economy and PMI

Application in Pharmaceutical Development and Regulatory Context

The implementation of catalytic methodologies with improved atom economy aligns with regulatory initiatives promoting sustainable pharmaceutical development. Japan's Pharmaceuticals and Medical Devices Agency (PMDA) has established specific initiatives to promote New Approach Methodologies (NAMs) that go beyond simply replacing animal testing and aim to improve the quality and success rate of scientific decision-making in medical product development [38]. These approaches utilize innovative science and technology, including in chemico, in vitro, and in silico methods, with the goal of increasing predictability of human safety, efficacy, and pharmacokinetics while reducing reliance on animal testing.

The PMDA's regulatory science initiatives include the establishment of an internal NAMs Working Group with cross-functional membership from various review departments, collaboration with domestic stakeholders including the National Institute of Health Sciences and Japanese Center for the Validation of Alternative Methods, and active engagement with international regulatory authorities to ensure consistency in acceptance criteria for sustainable methodologies [38]. This regulatory framework creates a supportive environment for the adoption of catalytic processes with improved green chemistry metrics.

Within this context, the demonstrated palladium-catalyzed synthesis of 2-arylquinolines represents a model system for pharmaceutical development. The 2-arylquinoline scaffold appears in numerous biologically active compounds including antimalarials, antitumor agents, anti-inflammatory, and analgesic drugs, with specific examples including quinine, camptothecin, and the cancer therapeutic agent linsitinib [39]. The application of sustainable catalytic methodologies to such privileged structures demonstrates the practical integration of green chemistry principles with pharmaceutical development objectives.

The strategic integration of catalysis and atom economy principles provides a powerful approach for PMI reduction in pharmaceutical development. The experimental protocols and metrics framework presented in this application note offer researchers practical tools for implementing these strategies, with the palladium-catalyzed C-H activation serving as a demonstrated example of their successful application. The resulting processes align with regulatory trends emphasizing sustainable methodologies while maintaining efficiency and productivity in drug development.

Future advancements in this field will likely focus on expanding the scope of catalytic transformations with inherent atom economy, particularly C-H activation and other methodologies that minimize functional group manipulation. Additionally, the integration of continuous flow systems with catalytic processes may further enhance PMI reduction by improving mass and heat transfer while reducing solvent consumption. As green chemistry metrics become increasingly embedded in pharmaceutical development and regulatory evaluation, the implementation of catalytic strategies with high atom economy will continue to grow in importance for achieving both environmental and economic objectives in drug development.

Machine Learning and Bayesian Optimization for Rapid Process Optimization

In the pursuit of sustainable pharmaceutical manufacturing, the adoption of green chemistry principles is paramount. Process Mass Intensity (PMI), defined as the total mass of materials used to produce a unit mass of active pharmaceutical ingredient (API), has emerged as a key green metric for assessing environmental impact within the industry [20]. Lowering PMI directly correlates with reduced waste, lower energy consumption, and more efficient resource utilization. However, optimizing chemical processes for minimal PMI is a complex, multi-dimensional challenge that traditionally requires extensive, resource-intensive experimentation.

Machine Learning (ML), and specifically Bayesian Optimization (BO), presents a transformative approach to this problem. These data-driven technologies enable the rapid identification of optimal reaction conditions with far fewer experiments, accelerating the development of greener synthetic processes [31]. This document provides application notes and detailed protocols for integrating ML and BO into chemical process development, with a specific focus on PMI reduction.

Theoretical Foundations and Key Metrics

Green Chemistry Metrics: Focusing on PMI

While several green metrics exist, the ACS GCI Pharmaceutical Roundtable identifies Process Mass Intensity (PMI) as a primary metric for pharmaceuticals [20]. Unlike atom economy, which is a theoretical calculation based on the reaction equation, PMI is a practical metric that accounts for all material inputs—including reagents, solvents, and catalysts—used in a specific process. It is calculated as:

PMI = Total Mass of Materials Used in the Process (kg) / Mass of Final Product (kg)

A lower PMI indicates a more efficient and environmentally friendly process. Recent research underscores that while expanding system boundaries (e.g., to cradle-to-gate analysis) can improve the correlation between mass-based metrics and full Life Cycle Assessment (LCA) environmental impacts, no single mass-based metric can fully capture the multi-criteria nature of environmental sustainability [40]. Therefore, PMI is best used as a powerful, readily calculable indicator within a broader suite of sustainability assessments.

Other relevant metrics include [14] [20]:

• Atom Economy (AE): Measures the incorporation of reactant atoms into the final product.
• Reaction Mass Efficiency (RME): The percentage of reactant mass converted into the product.
• E-Factor: Total waste produced per unit of product.

Machine Learning and Bayesian Optimization Basics

Machine Learning provides a set of tools for making data-driven decisions and predictions. In drug discovery and development, ML applications range from target validation to predictive modeling of chemical properties [41].

Bayesian Optimization is a sample-efficient ML strategy ideal for optimizing expensive-to-evaluate "black-box" functions, such as chemical reactions. Its core strength lies in intelligently selecting which experiment to perform next by balancing the exploration of uncertain regions of the parameter space with the exploitation of known promising areas [42] [43]. The BO framework consists of two key components:

• A surrogate model (often a Gaussian Process) that probabilistically models the objective function (e.g., reaction yield or PMI) based on available data.
• An acquisition function that uses the surrogate's predictions to decide the most informative next experiment by balancing exploration of uncertain regions and exploitation of known promising areas [42].

Application Notes: Implementing ML/BO for Process Optimization

The following workflow integrates PMI prediction and Bayesian optimization to achieve "greener-by-design" synthesis.

The typical optimization cycle involves four key stages: prediction, design, execution, and analysis, ultimately leading to an optimized process with a lower environmental footprint. The following diagram visualizes this iterative workflow.

Protocol 1: PMI Prediction for Route Selection

Objective: To quantitatively predict the Process Mass Intensity (PMI) of proposed synthetic routes for a target molecule prior to laboratory experimentation, enabling the selection of the most efficient and sustainable pathway.

Principles: This approach utilizes predictive analytics and historical data from large-scale syntheses to forecast process efficiency [31].

Methodology:

• Route Proposals: Propose 2-3 distinct synthetic routes to the target molecule.
• Data Collection & Input: For each route, compile a comprehensive bill of materials for every reaction and work-up step. This includes masses of all starting materials, reagents, catalysts, and solvents.
• PMI Calculation: Use a predictive PMI application or spreadsheet model to calculate the total mass intensity. The model should sum the mass of all inputs and divide by the mass of the final product for each step and for the entire sequence.
• Decision Point: Compare the predicted PMI values across the proposed routes. Select the route with the lowest overall PMI for further experimental development.

Protocol 2: Bayesian Optimization for Reaction Condition Optimization

Objective: To rapidly identify the set of reaction conditions that optimizes one or more objectives (e.g., yield, selectivity, PMI) for a given transformation with a minimal number of experiments.

Principles: Bayesian optimization uses a probabilistic model to guide the experimental sequence, efficiently navigating a complex multi-parameter space [31] [42].

Methodology:

• Define Optimization Goal:
  • Objective(s): Define primary and secondary objectives. Examples: Maximize yield, maximize enantiomeric excess (ee), minimize PMI of the step.
  • Key Performance Indicators (KPIs): Specify how objectives will be measured (e.g., HPLC yield, chiral HPLC for ee).

• Select Process Variables:

  • Identify critical continuous and categorical variables to optimize (e.g., temperature, time, catalyst loading, solvent identity, reactant stoichiometry).
  • Define the feasible range or set of options for each variable.

• Initial Experimental Design:

  • Perform a small set of initial experiments (e.g., 6-16) using a space-filling design (e.g., Latin Hypercube Sampling) or based on historical knowledge to gather baseline data.

• Implement Bayesian Optimization Loop:

  • Model Training: Train a surrogate model (e.g., Gaussian Process) using all available experimental data (initial + subsequent iterations).
  • Suggestion: Use the acquisition function (e.g., Expected Improvement, Upper Confidence Bound) to suggest the next most informative experiment(s) by balancing exploration and exploitation.
  • Execution: Conduct the suggested experiment(s) in the laboratory and record the results.
  • Iteration: Repeat the model-update-suggest-execute cycle until performance objectives are met or the experimental budget is exhausted.

Case Study and Data Presentation

BMS: PMI Prediction and BO in API Synthesis

A team from Bristol Myers Squibb demonstrated the integrated use of these tools for a real clinical candidate [31]. The approach consisted of two phases:

Phase 1: PMI-Based Route Selection A predictive PMI app was used to compare the efficiency of different synthetic routes before any lab work, allowing the team to select the most sustainable option for development.

Phase 2: Reaction Optimization with EDBO+ A specific transformation that had been optimized using traditional One-Factor-at-a-Time (OFAT) methods was re-optimized using the EDBO+ Bayesian optimization platform.

The table below quantitatively compares the performance of the traditional approach versus Bayesian optimization.

Table 1: Performance Comparison of OFAT vs. Bayesian Optimization for a Model API Synthesis

Optimization Method	Number of Experiments	Final Yield (%)	Final Enantiomeric Excess (ee%)	Implied Experimental Efficiency (Yield/Experiment)
One-Factor-at-a-Time (OFAT)	500	70	91	0.14
Bayesian Optimization (EDBO+)	24	80	91	3.33

The results show that Bayesian Optimization achieved a superior yield (80%) in only 24 experiments, a dramatic increase in efficiency compared to the 500 experiments required by OFAT [31].

Machine Learning for Solubility Prediction in Green Processing

Machine learning also aids green chemistry by predicting key physicochemical properties, reducing experimental measurement needs. For instance, developing nanomedicines via supercritical CO₂ processing requires accurate drug solubility data [44].

Table 2: Performance of Machine Learning Models in Predicting Drug Solubility in Supercritical CO₂

Machine Learning Model	R-squared (R²)	Mean Absolute Error (MAE)	Mean Absolute Percentage Error (MAPE)
Random Forest (RF)	0.857	2.90E-06	3.15E-01
Extra Trees (ET)	0.998	1.98E-06	2.27E-01
Gradient Boosting (GB)	0.992	1.10E-06	1.16E-01

In another study, an Artificial Neural Network (ANN) model was developed to simulate the solubility of Oxaprozin in supercritical CO₂, achieving an R² greater than 0.99 for both training and validation, providing a robust tool for process design [44].

The Scientist's Toolkit

Implementing ML and BO requires access to specialized software tools. The following table lists key platforms and their applications.

Table 3: Key Software Tools for Bayesian Optimization and Machine Learning in Chemistry

Tool Name	Type / Key Features	Application in Chemical Optimization	License
EDBO/EDBO+ [31] [43]	Experimental Design via Bayesian Optimization platform	Multi-objective reaction optimization; Open-source and freely available.	Free
Summit [42]	Software framework for chemical reaction optimization	Includes benchmarks and multiple optimization strategies (e.g., TSEMO) for chemistry.	MIT
BoTorch [43]	Python library for Bayesian Optimization	Built on PyTorch; provides advanced tools for developing BO algorithms.	MIT
TensorFlow / PyTorch [41]	Deep Learning frameworks	Building custom neural network models for property prediction (e.g., solubility).	Apache / BSD
Scikit-learn [41]	Machine Learning library in Python	Provides standard ML algorithms (Random Forest, Gradient Boosting) for data modeling.	BSD

Experimental Setup and Reagents

The following materials and resources are essential for executing the wet-lab component of the optimization protocols.

Table 4: Essential Research Reagent Solutions and Materials

Item	Function / Application	Example / Note
Automated Parallel Reactor System	Enables high-throughput execution of multiple reaction conditions in parallel.	Essential for efficiently gathering data for BO iterations.
Analytical Equipment (HPLC, UPLC, GC)	For rapid and accurate quantification of reaction outcomes (yield, conversion, ee).	Critical for generating the high-quality data required for ML model training.
Supercritical CO₂ Apparatus	Used for solubility measurement and nanoparticle formation in green processing.	Includes a high-pressure vessel, CO₂ supply, pump, and temperature control [44].
Common Solvents & Reagents	Solvents (e.g., MeCN, EtOAc, 2-MeTHF), catalysts, reactants.	Prefer solvents with better green credentials (e.g., Cyrene, 2-MeTHF) to improve PMI.
Candidate Catalysts	A diverse library of catalysts for screening and optimization.	Particularly important for catalytic reactions (e.g., K–Sn–H–Y-30-dealuminated zeolite) [14].

In the pursuit of sustainable pharmaceutical manufacturing, the Process Mass Intensity (PMI) has emerged as a key green chemistry metric. PMI measures the total mass of materials used to produce a unit mass of the final product, providing a comprehensive view of process efficiency and environmental impact. Solvents often constitute the largest proportion of mass in Active Pharmaceutical Ingredient (API) synthesis, sometimes accounting for 50-60% of the total materials used and contributing significantly to post-treatment greenhouse gas emissions [45]. Consequently, optimizing solvent selection and recovery presents one of the most impactful opportunities for reducing PMI. This application note provides researchers and drug development professionals with structured protocols and assessment tools to systematically evaluate and implement greener solvent strategies, aligning with the principles of green chemistry and the broader objectives of sustainable process design.

Green Solvent Assessment and Selection Guides

The CHEM21 Solvent Selection Guide

The CHEM21 selection guide represents a consensus-based approach developed through pharmaceutical industry collaboration. It classifies solvents into three main categories—Recommended, Problematic, and Hazardous—based on safety, health, and environmental (SHE) criteria scores that range from 1 (lowest hazard) to 10 (highest hazard) [46].

Table 1: CHEM21 Solvent Classification and SHE Scores

Solvent	BP (°C)	Safety Score	Health Score	Environment Score	Recommended Usage
Water	100	1	1	1	Recommended
Ethanol	78	4	3	3	Recommended
Ethyl Acetate	77	5	3	3	Recommended
Acetone	56	5	3	5	Recommended
Heptane	98	4	2	7	Problematic
DMF	153	2	6	5	Hazardous
DCM	40	1	6	7	Hazardous

The SHE scores are derived from easily available physical properties and GHS (Globally Harmonized System) statements. The safety score primarily considers flash point, with contributions from auto-ignition temperature, resistivity, and peroxide formation ability. The health score derives from the most stringent GHS H3xx statements with a contribution from boiling point. The environment score considers both volatility (boiling point) and GHS H4xx statements [46].

Holistic Multi-Criteria Assessment Framework

Beyond simple classification, a comprehensive solvent assessment should integrate environmental, health, and safety (EHS) considerations with technical performance and economic factors [47]. This is particularly important as some bio-based solvents initially perceived as green may still present significant hazards. For instance, d-limonene, while derived from natural sources, is classified as potentially fatal if swallowed or inhaled and very toxic to aquatic life with long-lasting effects [47]. This underscores the necessity of careful evaluation for all solvent alternatives, regardless of their origin.

Experimental Protocols for Solvent Evaluation and Optimization

Protocol 1: Reaction Optimization and Solvent Effect Analysis

Purpose: To determine optimal solvent conditions that maximize reaction efficiency while maintaining green chemistry principles.

Materials and Equipment:

• Reaction substrates and catalysts
• Candidate green solvents (e.g., from CHEM21 "Recommended" list)
• NMR spectrometer or HPLC system for reaction monitoring
• Temperature-controlled reaction blocks
• Data processing spreadsheet (e.g., VTNA-compatible template)

Procedure:

• Experimental Design: Select a diverse set of 8-12 green solvents spanning various polarities and functional groups.
• Kinetic Profiling: Conduct reactions in each solvent under identical conditions (concentration, temperature, stirring). Monitor conversion over time using appropriate analytical methods (e.g., ¹H NMR) [48].
• Variable Time Normalization Analysis (VTNA): Input concentration-time data into VTNA-compatible spreadsheet to determine reaction orders without complex mathematical derivations [48].
• Linear Solvation Energy Relationship (LSER): Correlate ln(k) with Kamlet-Abboud-Taft solvatochromic parameters (α = hydrogen bond donating ability, β = hydrogen bond accepting ability, π* = dipolarity/polarizability) to understand solvent effects mechanistically [48].
• Greenness-Rate Correlation: Plot ln(k) against solvent greenness scores (from CHEM21 guide) to identify solvents offering optimal balance of performance and sustainability [48].

Data Interpretation: The resulting LSER equation (e.g., ln(k) = -12.1 + 3.1β + 4.2π*) reveals how solvent properties influence reaction rates, enabling prediction of performance in untested solvents [48].

Protocol 2: Computational Solvent Screening Using COSMO-RS

Purpose: To efficiently screen green solvent candidates for solubility and process suitability using computational methods before experimental validation.

Materials and Equipment:

• COSMO-RS software (e.g., COSMOtherm, TURBOMOLE)
• Molecular structures of solute and solvent candidates
• Computing resources for quantum chemical calculations

Procedure:

• Structure Optimization: Generate optimized 3D molecular structures for all compounds using quantum chemical methods (e.g., Density Functional Theory) [49].
• σ-Potential Calculation: Perform COSMO calculations to obtain σ-profiles characterizing surface polarity distributions [49].
• Solubility Prediction: Use COSMO-RS to compute activity coefficients and predict solubility in neat solvents and binary mixtures [49].
• Candidate Ranking: Rank solvents based on predicted solubility and greenness metrics [49].
• Experimental Validation: Select top candidates for experimental verification using shake-flask method with HPLC or UV-Vis quantification [49].

Data Interpretation: Compare computational predictions with experimental results to validate the model. Systems showing computational-experimental agreement can be expanded to related compound systems with increased confidence.

Diagram 1: Green solvent selection workflow for PMI reduction.

Solvent Recovery and Life Cycle Considerations

Effective solvent strategies must consider the complete lifecycle, including recovery and end-of-life treatment. The cumulative energy demand (CED) of solvent production can be calculated to determine whether incineration with energy recovery or purification by distillation offers the greater reduction in environmental impact [50]. Generally, functionalized solvents with longer production routes (e.g., DMF) are best recycled to retain the energy and value invested in their original synthesis, while many hydrocarbons are best incinerated according to simplified LCA approaches [50].

Table 2: Solvent Recovery Energy Considerations

Solvent	Production Energy (MJ/kg)	Distillation Energy (MJ/kg)	Recommended EOL
n-Hexane	90	25	Incineration
Toluene	95	28	Incineration
THF	170	40	Distillation
DMF	130	35	Distillation
Ethanol	50	20	Context-dependent

The Scientist's Toolkit: Research Reagent Solutions

Table 3: Essential Materials for Green Solvent Implementation

Reagent/Material	Function	Application Notes
CHEM21 Solvent Selection Guide	Classification system	Categorizes solvents as Recommended, Problematic, or Hazardous based on SHE criteria [46]
SUSSOL Software	AI-based solvent selection	Uses self-organizing maps to cluster solvents by properties and suggest alternatives [45]
COSMO-RS Software	Computational screening	Predicts solubility and affinity using quantum chemistry without extensive experimentation [49]
Bio-based Solvents (e.g., Ethyl Lactate, d-Limonene)	Renewable alternatives	Derived from biomass; require full EHS assessment despite bio-based origin [51] [47]
Deep Eutectic Solvents (DES)	Tunable solvent systems	Combination of H-bond donor/acceptor; biodegradable, low volatility [51]
Ionic Liquids	Designer solvents	Negligible vapor pressure; tunable properties; assess synthesis environmental impact [51]
Supercritical CO₂	Non-polar extraction	Non-toxic, tunable density/polarity with pressure; requires energy for pressurization [51]
VTNA Spreadsheet	Kinetic analysis	Determines reaction orders without complex mathematical derivations [48]

Systematic solvent selection and recovery represents one of the most effective strategies for reducing PMI in pharmaceutical development. By implementing the structured protocols and assessment frameworks outlined in this application note, researchers can significantly improve process sustainability while maintaining analytical and synthetic efficiency. The integration of computational screening tools, comprehensive lifecycle assessment, and rigorous experimental validation creates a robust methodology for identifying and implementing greener solvent systems that align with the broader objectives of green chemistry and sustainable pharmaceutical manufacturing.

Overcoming Data and Confidentiality Challenges in PMI Assessment

Process Mass Intensity (PMI) has emerged as a crucial mass-based metric for evaluating the environmental performance and resource efficiency of chemical processes, particularly in pharmaceutical development [17]. It is defined as the total mass of materials used to produce a unit mass of the target product, accounting for reactants, reagents, solvents, and catalysts [17]. While Life Cycle Assessment (LCA) represents the gold standard for environmental impact evaluation, its extensive data requirements and complexity often render it impractical for early-stage process development [2] [26]. PMI offers a simplified alternative that aligns with the principles of green chemistry by focusing on mass efficiency and waste reduction [2] [10].

However, the application of PMI faces significant challenges related to data availability and confidentiality constraints, especially when attempting to expand system boundaries beyond gate-to-gate assessments to include upstream value chain impacts [2]. This application note addresses these challenges by providing structured methodologies and tools that enable researchers to conduct meaningful PMI assessments even under data-limited conditions, ensuring that green chemistry principles are effectively integrated into pharmaceutical development workflows.

Current Challenges in PMI Assessment

Data Limitations and System Boundary Definitions

The primary challenge in PMI assessment lies in defining appropriate system boundaries and obtaining sufficient data for meaningful calculations. Traditional gate-to-gate PMI (considering only materials directly used in the manufacturing process) has proven insufficient for accurately predicting environmental impacts, as it ignores significant upstream resource consumption [2] [52]. Recent research demonstrates that expanding system boundaries to cradle-to-gate approaches strengthens correlations with LCA environmental impacts for fifteen of sixteen impact categories [2]. However, this expansion substantially increases data requirements and introduces confidentiality concerns when dealing with proprietary chemistries and supply chain information.

Table 1: Comparison of PMI System Boundaries and Their Limitations

System Boundary	Definition	Data Requirements	Key Limitations
Gate-to-Gate	Includes only materials directly used within the manufacturing facility	Low: Requires only process mass balance	Fails to capture upstream impacts; poor correlation with LCA results [2] [52]
Cradle-to-Gate	Includes all materials from resource extraction to factory gate	High: Requires full supply chain data	Data intensive; confidential supplier information; complex calculations [2] [26]
Value-Chain Adjusted	Selectively includes key upstream materials based on impact significance	Moderate: Prioritizes data collection on high-impact materials	Requires methodology to identify critical pathway elements [2]

Confidentiality Constraints in Pharmaceutical Development

The proprietary nature of pharmaceutical process development creates substantial barriers to comprehensive PMI assessment. Chemical routes, synthetic methodologies, and specific reagent selections constitute valuable intellectual property that companies are reluctant to share, even internally across departments [2]. This confidentiality imperative conflicts with the transparent data sharing needed for accurate value-chain mass intensity calculations. Furthermore, the transition toward a defossilized chemical industry introduces additional complexity, as the environmental implications of material choices are becoming increasingly time-dependent and process-specific [2].

Methodologies for Enhanced PMI Assessment

Value-Chain Mass Intensity (VCMI) Framework

The Value-Chain Mass Intensity framework addresses system boundary limitations by systematically expanding PMI calculations to include upstream material flows. Eichwald et al. (2025) propose categorizing value chain products into seven classes based on the Central Product Classification system, enabling gradual boundary expansion [2]. This approach allows researchers to identify which upstream components contribute most significantly to environmental impacts and prioritize data collection efforts accordingly.

The methodology proceeds through these critical steps:

• Gate-to-Gate PMI Calculation: Establish baseline mass intensity using traditional factory entrance to exit boundaries
• Value Chain Mapping: Identify all major material inputs and classify them according to standardized product categories
• Selective Boundary Expansion: Incrementally include product categories based on their predicted environmental significance
• Correlation Analysis: Validate the expanded PMI against available LCA data or proxies
• Critical Material Identification: Pinpoint the specific input materials that serve as proxies for environmental impacts

Table 2: Experimental Protocol for Systematic VCMI Assessment

Protocol Step	Key Activities	Data Inputs	Output Deliverables
Process Characterization	Document reaction stoichiometry, yields, solvent masses, purification materials	Experimental data; process flow diagrams	Complete mass balance for direct inputs [17]
Value Chain Classification	Categorize all input materials using CPC codes; identify data gaps	Chemical inventories; supplier information	Value chain map with classified material flows [2]
Data Gap Resolution	Apply hierarchy of estimation methods: stoichiometric calculations, similar processes, predictive models	Literature data; analogous processes; predictive algorithms	Complete inventory with uncertainty estimates [53]
Impact Correlation	Calculate Spearman correlation coefficients between PMI/VCMI and LCA impact categories	LCA data; impact assessment results	Correlation matrix identifying key relationships [2]
Validation	Compare results with streamlined LCA tools; benchmark against industry standards	ACS GCI PMI-LCA tool; iGAL metrics [26] [17]	Validated assessment with confidence intervals

Streamlined PMI-LCA Integration

The Streamlined PMI-LCA Tool, developed through collaboration with the ACS Green Chemistry Institute Pharmaceutical Roundtable, represents a practical approach to integrating mass-based metrics with environmental impact assessment [26]. This tool combines the data accessibility of PMI with a cradle-to-gate approach to include the environmental footprint of synthesis raw materials without requiring full LCA expertise [26]. The tool operates through:

• Material-Specific Emission Factors: Utilizing adapted LCA databases with pharmaceutical-relevant emission factors
• Hotspot Identification: Pinpointing process steps with disproportionate environmental impacts
• Scenario Analysis: Enabling comparison of alternative routes under development
• Minimal Data Requirements: Functioning with limited inputs while identifying critical data gaps

The workflow for implementing this integrated approach is detailed below:

Diagram 1: PMI Assessment Workflow

Research Reagent Solutions

Implementing robust PMI assessment under data constraints requires specialized tools and resources. The following table details key solutions available to researchers:

Table 3: Essential Research Reagent Solutions for PMI Assessment

Tool/Resource	Type	Key Functionality	Accessibility
ACS GCI PMI Calculator	Spreadsheet-based tool	Calculates PMI for linear and convergent syntheses; handles solvent recovery calculations [17]	Free download from ACS GCI website
Streamlined PMI-LCA Tool	Integrated assessment tool	Combines PMI with cradle-to-gate LCA factors; identifies environmental hotspots [26] [53]	Under development; available through ACS GCI partnership
iGAL (Innovation Greenness Assessment Lead)	Standardized metrics platform	Provides relative process greenness score; enables benchmarking across processes [17]	Available through IQ Consortium
ecoinvent Database	LCA inventory database	Provides emission factors and resource consumption data for common chemicals [2] [53]	Licensed database; requires subscription
Convergent PMI Calculator	Advanced calculation tool	Handles complex synthesis pathways with multiple branches and convergence points [17]	Free download from ACS GCI website

Experimental Protocol for Confidential PMI Assessment

Data-Limited Value Chain Expansion

This protocol enables researchers to conduct meaningful PMI assessments when complete value chain data is unavailable due to confidentiality or data gap constraints. The methodology is adapted from recent research on mass intensity correlations and systematic boundary expansion [2].

Materials and Data Requirements

• Process mass balance data (reactants, solvents, reagents, catalysts)
• Chemical structures and basic properties of all input materials
• Available supplier information (without confidential details)
• Literature data on analogous processes or materials
• Stoichiometric calculations for theoretical material production

Step-by-Step Procedure

• Establish Baseline Gate-to-Gate PMI

  • Sum the total mass of all materials used in the synthesis (including reaction, workup, and purification)
  • Divide by the mass of product obtained
  • Document using formula: PMI = (Total Mass Input) / (Mass Product) [17]

• Classify Input Materials Using CPC Categories

  • Categorize all input materials according to the seven Central Product Classification categories
  • Identify which categories contain materials with potentially high upstream mass intensities
  • Prioritize data collection on these high-impact categories [2]

• Apply Data Estimation Hierarchy

  • For confidential or unavailable data, apply this estimation hierarchy:
    • First: Use stoichiometric calculations based on chemical reactions
    • Second: Apply similar-class materials from available databases
    • Third: Use predictive models for mass intensity of analogous chemicals
    • Fourth: Apply class-average expansion factors [53]

• Calculate Expanded VCMI

  • Systematically expand system boundaries by including one product category at a time
  • Calculate the incremental increase in mass intensity for each expansion
  • Stop expansion when additional categories contribute less than 5% to total mass intensity [2]

• Validate with Correlation Analysis

  • Compare calculated VCMI values with available LCA data points
  • Calculate Spearman correlation coefficients for different boundary expansions
  • Select the most parsimonious system boundary that maintains strong correlation (>0.7) with environmental impacts of interest [2]

The relationship between system boundary expansion and environmental impact correlation is visualized below:

Diagram 2: System Boundary vs Impact Correlation

Confidentiality-Preserving Calculation Methods

This protocol enables meaningful PMI assessment while protecting proprietary process information through structured calculation methods that maintain confidentiality.

Materials and Specialized Equipment

• Standardized PMI calculation spreadsheet (ACS GCI or equivalent)
• Process flow diagram with confidential information redacted
• Category-based emission factors rather than material-specific data
• Anonymized benchmarking database

Step-by-Step Procedure

• Process Segmentation and Anonymization

  • Divide the synthetic route into discrete steps
  • Replace confidential reagent names with generic functional classifications (e.g., "coupling reagent" instead of specific chemical name)
  • Retain mass quantities while removing identifying structural information

• Category-Based Impact Assessment

  • Assign each input material to a standardized category
  • Use category-average upstream mass intensity factors
  • Apply class-specific expansion multipliers based on published values [2]

• Relative Improvement Calculation

  • Calculate PMI for alternative process routes using consistent methodology
  • Express results as percentage improvement rather than absolute values
  • Use normalized metrics such as iGAL scores for external communication [17]

• Confidential Benchmarking

  • Compare results against anonymized industry benchmarks
  • Utilize the ACS GCI PMI-LCA tool which incorporates aggregated industry data [53]
  • Position performance within quartiles rather than disclosing specific values

Expected Results and Interpretation

Correlation Between PMI and Environmental Impacts

Systematic expansion of system boundaries significantly improves the correlation between mass-based metrics and environmental impacts. Recent research demonstrates that cradle-to-gate mass intensities show stronger correlations with LCA impacts compared to traditional gate-to-gate PMI [2]. The specific strength of correlation varies by environmental impact category, with climate change, resource depletion, and energy consumption typically showing the strongest relationships to mass-based metrics.

The variation in correlation stems from the fact that different environmental impacts are approximated by distinct sets of key input materials that serve as proxies for those impacts [2]. For example:

• Climate Change: Strongly correlated with carbon-intensive inputs like coal and certain hydrocarbons
• Water Consumption: Linked to materials with high water footprint in production
• Resource Depletion: Associated with metals and mineral-based reagents

Case Study: MK-7264 API Development

The application of streamlined PMI assessment in the development of MK-7264 active pharmaceutical ingredient demonstrates the practical utility of these methodologies. Through iterative PMI assessment and process optimization, researchers reduced the PMI from 366 to 88 over the course of process development [26]. This significant improvement was achieved through:

• Solvent Optimization: Reduction and recovery of high-mass intensity solvents
• Route Selection: Choosing synthetic pathways with inherent mass efficiency
• Catalyst Development: Implementing more efficient catalytic systems
• Process Intensification: Combining multiple steps and improving yields

This case exemplifies how systematic PMI assessment, even with data limitations, can drive substantial improvements in process sustainability while maintaining confidentiality of proprietary chemistry.

Troubleshooting and Optimization

Common Data Gap Challenges

• Missing Upstream Data: Apply category-average expansion factors based on chemical class
• Confidential Reagents: Use functional equivalents with known environmental profiles
• Proprietary Intermediates: Estimate based on stoichiometric minimums from reaction equations
• Complex Supply Chains: Focus on high-mass inputs which typically dominate the environmental impact

Methodology Validation

• Cross-Validation: Compare results from multiple estimation approaches
• Sensitivity Analysis: Test how assumptions affect overall conclusions
• Benchmarking: Compare against similar processes in literature or industry databases
• Uncertainty Quantification: Document ranges and confidence intervals for estimates

The methodologies presented in this application note enable researchers to overcome the significant data and confidentiality challenges inherent in comprehensive PMI assessment. By implementing the Value-Chain Mass Intensity framework and streamlined PMI-LCA integration, researchers can obtain meaningful assessments of environmental performance even with limited data availability. The systematic approach to boundary expansion and the use of category-based estimation methods provide practical pathways for incorporating green chemistry principles into pharmaceutical development while protecting intellectual property.

As the chemical industry continues its transition toward a defossilized, circular economy, these methodologies will evolve in importance, enabling researchers to make informed decisions that balance environmental performance with practical constraints. The ongoing development of tools like the web-based PMI-LCA application through the ACS GCI Pharmaceutical Roundtable will further enhance accessibility and standardization across the industry [53].

Beyond Mass: Validating PMI Against Environmental Impact and Other Metrics

How Well Does PMI Correlate with LCA Environmental Impacts?

The drive towards a more sustainable chemical industry necessitates robust methods for evaluating the environmental performance of chemical processes. While Life Cycle Assessment (LCA) is the recommended holistic method for evaluating multiple environmental impacts across a product's entire life cycle, its application is often hampered by the need for extensive life-cycle data, which can be difficult to collect due to lack of measurements or confidentiality, making the process time-consuming and expensive [2]. As a practical solution, the chemical industry, particularly the pharmaceutical sector, has widely adopted simpler "green chemistry metrics," with Process Mass Intensity (PMI) being a prominent example [54] [22].

PMI is a mass-based metric, representing the total mass of materials used to produce a unit mass of a chemical product [22]. Its use assumes that lower mass expenditures lead to lower environmental impacts through less waste production and higher resource efficiency [2]. However, a critical question remains: How well does this simple mass-based metric correlate with the multi-faceted environmental impacts calculated by LCA? This application note addresses this question by summarizing recent research findings, presenting quantitative data, and providing protocols for researchers and drug development professionals to apply these insights.

Key Findings: Correlation Between PMI and LCA

A seminal 2025 study by Eichwald et al. systematically analyzed the correlation between mass intensities and LCA environmental impacts, investigating the effect of expanding system boundaries from gate-to-gate (PMI) to cradle-to-gate (Value-Chain Mass Intensity, or VCMI) [54] [40] [2].

The Critical Role of System Boundaries

• Gate-to-Gate PMI is Insufficient: The study found that the traditional (gate-to-gate) PMI, which only considers materials used within the factory, is not a reliable proxy for environmental impacts [2]. This confirms that a gate-to-gate boundary is too limited.
• Expanding Boundaries Improves Correlation: Expanding the system boundary from gate-to-gate to cradle-to-gate strengthened correlations for fifteen out of sixteen environmental impacts analyzed [54] [2]. Cradle-to-gate accounting includes the mass of natural resources required from the upstream value chain, providing a more complete picture.

Quantitative Correlation Analysis

The research evaluated Spearman correlation coefficients between eight mass intensities (one PMI and seven VCMIs with varying upstream boundaries) and sixteen LCA environmental impact categories [2]. The table below summarizes the findings for key environmental impacts, illustrating how the correlation strength changes with system boundary expansion.

Table 1: Effect of System Boundary on Correlation Between Mass Intensity and LCA Environmental Impacts (Adapted from Eichwald et al., 2025)

Environmental Impact Category	Gate-to-Gate (PMI) Correlation	Cradle-to-Gate (VCMI) Correlation	Key Upstream Driver
Climate Change	Weak	Stronger	Coal, Natural Gas [54] [2]
Water Use	Weak	Stronger	Materials associated with high water consumption [2]
Resource Depletion	Weak	Stronger	Metal Ores, Fossil Resources [2]
Toxicity-Related Impacts	Weak	Varies	Specific chemicals and their synthesis pathways [2]

Limitations and the "One Metric Fits All" Problem

A crucial finding is that a single mass-based metric cannot fully capture the multi-criteria nature of environmental sustainability [54] [2].

• Distinct Material Drivers: Different environmental impacts are driven by distinct sets of "key input materials" in the value chain. For example, coal is a key proxy for climate change, while specific metals might be key for resource depletion [2].
• Time Sensitivity: The reliability of mass intensities as a proxy is time-sensitive. As the industry transitions towards a defossilized economy, the environmental impact of a material like coal may change, making mass-based assessments less reliable over time [54] [2].
• Energy Blind Spot: PMI does not account for energy consumption, which is a key driver of sustainability, particularly in biologics manufacturing [55]. A process with a higher PMI might have lower overall energy consumption per unit produced, making it more sustainable than a lower-PMI process [55].

Experimental Protocols

Protocol for Analyzing PMI-LCA Correlation

This protocol outlines the methodology based on the seminal study by Eichwald et al.

1. Goal and Scope Definition: - Objective: To determine the correlation strength between mass intensity metrics and a set of LCA environmental impact categories. - System Boundaries: Define and compare at least two system boundaries: Gate-to-Gate (PMI) and Cradle-to-Gate (VCMI).

2. Inventory Modeling: - Data Collection: Compile life cycle inventory data for a large set (e.g., >100) of chemical production routes [2]. The ecoinvent database is a suitable source [2]. - Mass Intensity Calculation: - PMI: For each chemical process, calculate PMI as the total mass of input materials (kg) divided by the mass of the product (kg). Inputs include all reagents, solvents, and catalysts used within the process steps [22]. - VCMI: Calculate VCMI by expanding the system boundary to include the cumulative mass of natural resources extracted from the environment to produce the input materials [2]. - LCA Impact Assessment: Calculate a comprehensive set of environmental impact categories (e.g., 16 categories including climate change, water use, resource depletion) for each chemical process using a standard life cycle impact assessment method (e.g., ReCiPe) [2].

3. Correlation Analysis: - Statistical Analysis: Calculate the Spearman rank correlation coefficient between each mass intensity metric (PMI, VCMI) and each LCA environmental impact category [2]. - Interpretation: Analyze how the correlation coefficients change when moving from a gate-to-gate (PMI) to a cradle-to-gate (VCMI) perspective.

Protocol for Conducting a Simplified LCA using a PMI-LCA Tool

The ACS GCI Pharmaceutical Roundtable has developed tools to bridge the gap between simple metrics and full LCA [53] [56].

1. Process Input Definition: - Define Synthesis Steps: Map the synthetic route, including all reaction steps, workup, and purification stages. The tool can handle linear and convergent syntheses [53]. - Input Masses: For each step, input the masses of all raw materials, solvents, catalysts, and process chemicals [22] [53]. - Output Mass: Define the mass of the final Active Pharmaceutical Ingredient (API) or intermediate produced.

2. Tool Execution: - PMI Calculation: The tool automatically calculates the gate-to-gate PMI based on the mass balance [22]. - LCA Estimation: The tool uses built-in life cycle inventory data to estimate environmental impacts, identifying environmental "hotspots" [53].

3. Interpretation and Scenario Analysis: - Compare Routes: Use the tool to compare different synthetic routes for the same API, considering both PMI and LCA impacts [56]. - Sensitivity Analysis: Test the impact of changing key parameters, such as solvent recycling rates or utility sources, on the overall environmental footprint [53].

Visualization of the Relationship Between PMI and LCA

The following diagram illustrates the core conceptual relationship and findings regarding PMI and LCA.

Figure 1: Relationship between Mass Metrics and LCA. Expanding the system boundary from PMI to VCMI strengthens the correlation with LCA, but key upstream materials are the true drivers of environmental impacts.

The Scientist's Toolkit: Key Research Reagents & Solutions

Table 2: Essential Tools and Data for Environmental Assessment in Chemical Research

Tool / Data Type	Function / Description	Example Sources / Notes
PMI Calculator	Benchmarks the mass efficiency of a synthetic process. A simple metric to track green chemistry performance.	ACS GCI Pharmaceutical Roundtable Tool [22]
Convergent PMI Calculator	Enhances the standard PMI calculator to accommodate more complex, convergent syntheses with multiple branches.	ACS GCI Pharmaceutical Roundtable Tool [22]
PMI-LCA Tool	A simplified LCA tool that estimates environmental impacts (hotspots) based on process mass input data. Bridges PMI and full LCA.	ACS GCI Pharmaceutical Roundtable [53] [56]
Life Cycle Inventory (LCI) Database	Provides secondary data on the environmental inputs and outputs of materials and energy processes. Essential for cradle-to-gate modeling.	ecoinvent database [2]
Pharmaceutical-Grade LCI Data	Specialized emission factors that reflect the higher purity and intensive processing of pharmaceutical-grade materials, improving accuracy.	Under development by ACS GCIPR [53]

The correlation between PMI and LCA environmental impacts is not robust when using a traditional gate-to-gate PMI. However, systematically expanding the system boundary to a cradle-to-gate perspective (VCMI) significantly strengthens this correlation for most impact categories. Despite this improvement, PMI and VCMI remain single-dimensional mass metrics and cannot fully represent the multi-criteria environmental profile of a chemical process, as they are blind to factors like energy source and inherent toxicity.

For researchers and drug development professionals, this means:

• Move Beyond Gate-to-Gate: Always consider upstream material impacts. The VCMI approach provides a more accurate assessment.
• Use Metrics Judiciously: Do not rely solely on mass intensity to declare a process "green." It is a useful screening tool and benchmark but not a comprehensive environmental indicator.
• Adopt Simplified LCA Tools: Tools like the ACS GCI PMI-LCA tool represent the future of practical environmental assessment in process development, offering a more reliable pathway to genuine sustainability by integrating mass inputs with life cycle impact data [53] [56]. Future research should focus on refining these simplified LCA methods and developing more accurate, time-sensitive life cycle inventory data for the chemical industry [54] [2].

In the pursuit of sustainable manufacturing, researchers and drug development professionals require reliable, quantitative metrics to guide decision-making. Process Mass Intensity (PMI), defined as the total mass of materials used to produce a unit mass of a product, has emerged as a key green chemistry metric due to its simplicity and minimal data requirements [11]. However, the environmental significance of PMI is profoundly influenced by the chosen system boundary [40].

This application note demonstrates why expanding the system boundary from gate-to-gate to cradle-to-gate significantly enhances PMI's value as a proxy for comprehensive environmental impacts. A cradle-to-gate boundary includes all value chain processes, from raw material extraction (cradle) to the factory gate, whereas gate-to-gate only considers materials directly used within the factory [2]. Evidence from a recent systematic correlation analysis confirms that cradle-to-gate PMI offers a more reliable approximation for a wider range of environmental impacts compared to its gate-to-gate counterpart [40].

Comparative Analysis of Mass Intensity System Boundaries

Definitions and Key Characteristics

The system boundary determines which stages of a product's life cycle are included in the PMI calculation, directly influencing the result and its interpretation.

Table 1: System Boundary Definitions and Scopes

System Boundary	Definition	Included Stages	Excluded Stages
Gate-to-Gate	Considers only the immediate manufacturing process [2].	- Reaction materials- Solvents- Catalysts- Purification materials [11]	- Upstream raw material production- Extraction- Transportation- Use phase- End-of-life
Cradle-to-Gate	Includes all processes from raw material extraction to the factory gate [24] [57].	- Raw material acquisition- Processing- Manufacturing & assembly- Packaging [57]	- Product distribution- Use phase- End-of-life disposal [24]

Quantitative Correlation with LCA Environmental Impacts

A 2025 systematic study evaluated the correlation between mass intensities and sixteen Life Cycle Assessment (LCA) environmental impacts. The findings demonstrate the superior performance of cradle-to-gate boundaries.

Table 2: Correlation Strength of Mass Intensities with LCA Environmental Impacts [40]

Environmental Impact Category	Gate-to-Gate PMI	Cradle-to-Gate VCMI
Climate Change	Weak	Stronger
Human Health	Weak	Stronger
Ecosystem Quality	Weak	Stronger
Resource Depletion	Weak	Stronger
... (12 other impact categories)	Weak	Stronger for 11/12

The expansion of the system boundary from gate-to-gate to cradle-to-gate strengthened correlations for fifteen out of sixteen environmental impacts [40]. This improvement occurs because a cradle-to-gate boundary accounts for key input materials (e.g., coal, metals) that serve as proxies for value-chain processes with significant environmental footprints [40] [2].

Diagram 1: System boundaries define which life cycle stages are included in an assessment. Cradle-to-gate encompasses upstream impacts, providing a more complete picture than gate-to-gate [24] [57].

Experimental Protocol: Calculating Cradle-to-Gate Value-Chain Mass Intensity (VCMI)

Objective

To calculate a cradle-to-gate Value-Chain Mass Intensity (VCMI) for a chemical process by systematically accounting for natural resources consumed across the upstream value chain.

Methodology Workflow

Diagram 2: VCMI calculation requires tracing all inputs back to natural resources [2].

Step-by-Step Procedure

• Goal and Scope Definition

  • Define the functional unit, typically 1 kg of the final chemical product (e.g., an Active Pharmaceutical Ingredient - API) [35].
  • Clearly state that the analysis employs a cradle-to-gate system boundary.

• Gate-to-Gate Inventory Compilation

  • Record the mass (in kg) of all input materials used in the immediate chemical synthesis process. This includes:
    • Starting materials and reagents
    • Solvents
    • Catalysts
    • Any other ancillary materials [11]

• Value Chain Mapping and Classification

  • For each input material identified in Step 2, map its upstream production chain.
  • Classify all value chain products into one of seven standardized Central Product Classification (CPC) classes [2]. This systematic categorization allows for a modular expansion of the system boundary.

• Natural Resource Inventory (NRI) Compilation

  • Trace the mass of each classified value chain product back to the original natural resources required for its production.
  • The NRI should include masses of [2]:
    • Fossil energy carriers (e.g., crude oil, natural gas, coal)
    • Metal ores
    • Mineral resources
    • Biomass
    • Water

• VCMI Calculation

  • Sum the masses of all natural resources compiled in the NRI.
  • VCMI = Total mass of natural resources from the NRI (kg) / Mass of the functional unit (1 kg of product)

Table 3: Key Research Reagent Solutions and Tools

Item	Function/Description	Application Note
LCA Databases (e.g., ecoinvent)	Provide life cycle inventory data for common chemicals and materials, essential for estimating upstream impacts [35].	Critical for compiling the Natural Resource Inventory (NRI). Data gaps for novel chemicals remain a key challenge [35].
Central Product Classification (CPC)	A standardized system for categorizing products. The 2025 study used 7 CPC classes to systematically expand the system boundary [2].	Enables modular and standardized VCMI calculation.
Streamlined PMI-LCA Tools	Combined metrics tools developed by industry consortia like the ACS Green Chemistry Institute Pharmaceutical Roundtable [26].	Designed for rapid assessment in process development, integrating PMI with cradle-to-gate environmental footprint data [26].
Retrosynthetic LCA Workflow	A method to address data gaps for chemicals missing from LCA databases by using retrosynthesis to build life cycle inventory data [35].	An iterative, closed-loop approach that enhances LCA accuracy for complex molecules like APIs [35].

Application in Pharmaceutical Development: A Case Study

The synthesis of the antiviral drug Letermovir exemplifies the practical application of cradle-to-gate assessment. An LCA-guided analysis of the synthesis route identified a Pd-catalyzed Heck cross-coupling and an enantioselective addition as environmental "hotspots" [35]. This nuanced insight, which considered the upstream footprint of catalysts and reagents, would not be apparent from a simple gate-to-gate PMI calculation. The analysis enabled a targeted optimization of the synthesis route to reduce its overall environmental impact [35].

The choice of system boundary is not merely a technicality but a fundamental determinant of a metric's environmental relevance. While gate-to-gate PMI is a useful measure of immediate process efficiency, it fails to account for the significant environmental burdens embedded in the supply chain. The cradle-to-gate PMI, or Value-Chain Mass Intensity (VCMI), provides a more robust proxy for a wider range of LCA environmental impacts by including the mass of natural resources consumed throughout the upstream value chain. For researchers and drug development professionals aiming for genuine sustainability improvements, adopting a cradle-to-gate perspective is a critical step towards more accurate environmental assessment and effective process optimization.

The adoption of green chemistry principles requires robust metrics to quantify the environmental performance of chemical processes. For researchers and drug development professionals, selecting the appropriate metric is crucial for accurate sustainability reporting and process optimization. This application note provides a comparative analysis of four prominent metrics—Process Mass Intensity (PMI), E-Factor, Atom Economy (AE), and Eco-Footprint—detailing their methodologies, applications, and limitations within pharmaceutical and fine chemical contexts.

Metric Definitions and Core Principles

The following table summarizes the fundamental characteristics, formulas, and primary applications of each metric.

Table 1: Core Definitions and Characteristics of Key Green Chemistry Metrics

Metric	Definition	Core Formula	What It Measures	Primary Application Context
Process Mass Intensity (PMI)	Total mass of materials used per unit mass of product [2] [17]	PMI = Total Mass Input (kg) / Mass of Product (kg) [17]	Total resource consumption efficiency (includes reactants, solvents, reagents) [17]	Pharmaceutical process development and optimization [5] [17]
E-Factor	Mass of waste generated per unit mass of product [5] [58]	E-Factor = Total Mass of Waste (kg) / Mass of Product (kg) [5]	Waste generation, aligned with waste prevention principle [5]	Fine chemicals, pharmaceuticals, and bulk chemicals manufacture [5] [58]
Atom Economy (AE)	Molecular weight of desired product vs. total molecular weight of all reactants [5] [11]	AE = (MW of Product / Σ MW of Reactants) × 100% [5]	Intrinsic efficiency of a chemical reaction's design [5]	Early-stage route scouting and reaction design [5]
Eco-Footprint (Carbon Footprint)	Total greenhouse gas emissions throughout a product's life cycle, expressed in CO₂ equivalents (CO₂e) [59]	Carbon Footprint = Σ GHG Emissions (kg CO₂e) [59]	Climate change impact from all life cycle stages (raw materials, production, use, end-of-life) [60] [59]	Comparative environmental impact assessment of products, like inhalers [60] [61] [59]

Experimental Protocols for Metric Determination

Protocol for PMI Determination

Principle: PMI measures the total mass of resources (reactants, reagents, solvents, catalysts) required to produce a specified mass of a chemical product [17]. A cradle-to-gate system boundary, which includes upstream value chain materials, provides a more environmentally representative result than a gate-to-gate boundary [2].

Procedure:

• Define System Boundary: Decide between gate-to-gate (own process only) or cradle-to-gate (includes upstream materials). For environmental approximation, cradle-to-gate is recommended [2].
• Compile Mass Inventory: Record the masses (in kg) of all input materials entering the defined system boundary to produce 1 kg of the final product. This includes:
  • Reaction substrates and reagents
  • Solvents (for reaction and purification)
  • Catalysts and ligands
  • Acids, bases, and other additives
  • Water (optional, depending on reporting standards)
• Calculate PMI: Sum the masses of all inputs and divide by the mass of the isolated, purified product. PMI (kg/kg) = (Σ Mass of Inputs) / (Mass of Product)
• Interpretation: A lower PMI indicates higher mass efficiency. The ideal PMI is 1, representing a process where no materials are lost or wasted [17].

Protocol for E-Factor Determination

Principle: The E-Factor quantifies the waste generated in a process, with the ideal being zero [5]. It is crucial to define the inclusion of solvents and water, leading to simple E-Factor (sEF) and complete E-Factor (cEF) variants [5].

Procedure:

• Define Waste Scope: Specify if calculating sEF (excludes solvents and water) or cEF (includes all solvents and water without recycling) [5].
• Determine Product Mass: Isolate and weigh the final product (in kg).
• Calculate Total Waste Mass: For cEF, sum the masses of all input materials (as in PMI calculation) and subtract the mass of the product. Mass of Waste (kg) = Σ Mass of Inputs - Mass of Product
• Calculate E-Factor: Divide the total waste mass by the product mass. E-Factor (kg/kg) = Mass of Waste / Mass of Product
• Interpretation: A lower E-Factor signifies less waste generation. The cEF provides a more comprehensive assessment, particularly for solvent-intensive pharmaceutical processes [5].

Protocol for Atom Economy Calculation

Principle: Atom Economy is a theoretical calculation assessing the inherent efficiency of a reaction's stoichiometric equation, indicating what fraction of reactant atoms are incorporated into the desired product [5] [11].

Procedure:

• Write the Balanced Equation: Use the stoichiometrically balanced chemical equation for the reaction.
• Sum Molecular Weights: Calculate the sum of the molecular weights (g/mol) for all reactants.
• Identify Product MW: Note the molecular weight of the desired product.
• Calculate Atom Economy: AE (%) = (MW of Desired Product / Σ MW of Reactants) × 100%
• Interpretation: A higher AE percentage indicates a more atom-efficient reaction design. Rearrangement and addition reactions typically have 100% AE, while substitutions and eliminations have lower values [5].

Protocol for Carbon Footprint (Eco-Footprint) Determination

Principle: This metric uses Life Cycle Assessment (LCA) to quantify the total greenhouse gas emissions associated with a product, reported in kilograms of CO₂ equivalents (kg CO₂e) [59].

Procedure:

• Define Goal and Scope: Clearly define the product system, functional unit (e.g., "per dose" of an inhaler), and system boundaries (cradle-to-grave recommended) [59].
• Compile Life Cycle Inventory (LCI): Collect data on all energy and material flows across the product's life cycle (raw material extraction, manufacturing, transportation, use, and end-of-life disposal) [59].
• Apply Emission Factors: Convert all inventory data into GHG emissions using standardized emission factors (e.g., from databases like ecoinvent) [59].
• Calculate Carbon Footprint: Sum the GHG emissions from all life cycle stages to obtain the total kg CO₂e per functional unit.
• Interpretation: A lower carbon footprint indicates a lower contribution to climate change. This metric is essential for capturing environmental impacts that mass-based metrics miss, such as emissions from propellants in inhalers [60] [59].

Comparative Analysis and Data Visualization

Workflow for Green Metric Selection and Application

The following diagram illustrates the decision-making workflow for selecting and applying the appropriate green chemistry metrics based on the development stage and assessment goal.

Quantitative Comparison of Inhaler Technologies

The following table presents a real-world comparative analysis of pressurized metered-dose inhalers (pMDIs) and dry powder inhalers (DPIs), illustrating how these metrics reveal trade-offs between mass efficiency and environmental impact.

Table 2: Comparative Analysis of Inhaler Technologies: pMDI vs. DPI [60] [61] [59]

Metric	Pressurized Metered-Dose Inhaler (pMDI)	Dry Powder Inhaler (DPI)	Comparative Analysis & Key Insight
Carbon Footprint (per dose)	~0.125 kg CO₂e (for HFC-134a propellant) [60] [61]	~0.01 kg CO₂e (average) [61]	DPIs have a 92-99% lower carbon footprint than pMDIs [60] [61]. Mass-based metrics alone cannot capture this dramatic difference.
Process Mass Intensity (PMI)	Expected to be low for the device itself.	Expected to be low for the device itself.	PMI is less relevant for comparing these devices, as the dominant environmental impact (propellant emissions) is not reflected in the mass of inputs.
E-Factor	Would be very high if propellant is counted as waste.	Would be relatively low.	The E-Factor highlights the propellant as waste, but the Carbon Footprint quantifies its severe environmental impact.
Key Driver of Environmental Impact	HFC propellants (potent greenhouse gases) [59].	Patient's inspiratory effort (no propellant) [59].	This case demonstrates the critical limitation of mass-based metrics and the necessity of LCA for a complete environmental picture.

The Scientist's Toolkit: Key Reagents and Materials

This section details critical materials and tools used in the application and advancement of green chemistry metrics.

Table 3: Essential Research Reagents and Tools for Green Metrics Application

Item / Tool	Function / Description	Relevance to Green Metrics
ACS GCI PMI Calculator	A tool to quickly determine PMI values for chemical processes [17].	Standardizes PMI calculation, enables quick comparison and benchmarking of processes during development [17].
Solvent Selection Guides	In-house guides from pharmaceutical companies categorizing solvents as "preferred," "usable," or "undesirable" [5].	Complements mass metrics (PMI, E-Factor) by addressing the environmental and EHS (Environmental, Health, Safety) impact of the largest mass component in pharmaceuticals: solvents [5].
LCA Databases (e.g., ecoinvent)	Databases providing life cycle inventory data for common materials, energy, and transport [2] [59].	Essential for calculating robust Carbon Footprint and Eco-Footprint data, moving beyond simple mass accounting [2].
Low-GWP Propellants (e.g., HFO-1234ze(E))	New propellants with Global Warming Potential (GWP) <1 under development for pMDIs [59].	A research solution directly aimed at addressing the high carbon footprint identified by LCA studies of traditional inhalers [59].

Synergies and Limitations

Each metric serves a distinct purpose, and their combined use provides the most comprehensive picture.

• PMI vs. E-Factor: PMI and E-Factor are fundamentally linked (PMI = E-Factor + 1) [5] [17]. PMI focuses on total resource consumption, making it valuable for cost and efficiency analysis. The E-Factor's strength is its direct focus on waste minimization, aligning with a core principle of green chemistry [5].
• The System Boundary Problem: A significant limitation of PMI and E-Factor is the dependency of their values on the chosen system boundary (e.g., gate-to-gate vs. cradle-to-gate) [2] [5]. Expanding the boundary to include upstream materials (cradle-to-gate) strengthens their correlation with broader environmental impacts but increases data complexity [2].
• Beyond Mass: The most critical limitation of PMI, E-Factor, and AE is that they are mass-based and do not account for the inherent hazardousness or environmental impact of the substances involved [2] [5]. A process with a low E-Factor that generates a small amount of highly toxic waste is not truly "green." This is quantified by the Environmental Quotient (EQ), where EQ = E-Factor × Q, and Q represents the unfriendliness of the waste [5].
• The Role of LCA and Carbon Footprint: As demonstrated in the inhaler case study, LCA and carbon footprint analysis are indispensable for capturing environmental impacts like global warming potential, which are invisible to mass metrics [60] [59]. They address the multi-criteria nature of environmental sustainability, which a single mass-based metric cannot fully capture [2].

For researchers and drug development professionals, the path to robust environmental assessment is clear:

• Use Atom Economy for rapid, theoretical route screening.
• Employ PMI and E-Factor with a cradle-to-gate boundary to drive mass efficiency and waste reduction during process optimization.
• Finally, apply Life Cycle Assessment (Carbon Footprint) to validate true environmental performance and avoid problem-shifting, ensuring that green advances reported on the basis of mass metrics translate into genuine environmental benefits [2] [59].

Process Mass Intensity (PMI) is a pivotal green chemistry metric, defined as the total mass of materials used to produce a unit mass of a desired product. It is calculated as: PMI = Total Mass Input (kg) / Mass of Product (kg) A lower PMI indicates higher resource efficiency and is a primary indicator for waste reduction [62]. This metric has been widely adopted by the chemical and pharmaceutical industries to guide the development of more sustainable manufacturing processes, with notable successes. For instance, Merck reduced the PMI for a cancer treatment antibody-drug conjugate by approximately 75% through a streamlined synthesis [63].

Key Limitations of Mass-Based Metrics

While PMI provides a valuable measure of material efficiency, it fails to capture the full spectrum of environmental impacts. The table below summarizes its core limitations.

Table 1: Core Limitations of Process Mass Intensity (PMI)

Limitation	Description
Lacks Environmental Mechanism	PMI is a mass-based accounting metric that does not reflect any direct interaction with the environment or model specific environmental mechanisms [62].
Neglects Material Origins and Toxicity	The metric does not differentiate between renewable and fossil-based feedstocks, nor does it account for the toxicity, abundance, or hazardous properties of the materials used [62] [26].
Omits Energy Consumption	The use of energy, including its source (e.g., renewable vs. fossil fuels) and associated emissions (e.g., CO₂), is completely neglected in the calculation [62].
Oversimplifies Waste Impact	PMI treats all waste streams equally, ignoring critical factors such as waste properties, treatment methods, and final environmental fate, which determine actual ecological impact [62].
System Boundary Ambiguity	No single, standardized system boundary exists for PMI calculation, leading to inconsistent assessments. A narrow "gate-to-gate" boundary often excludes significant upstream impacts [62].

Current Research and Advanced Analysis

The Critical Role of System Boundaries

Recent research underscores that expanding the system boundary from a narrow gate-to-gate view (PMI) to a broader cradle-to-gate view is crucial. A 2025 systematic study by Eichwald et al. demonstrated that this expansion strengthens the correlation between mass intensity and 15 out of 16 LCA environmental impact categories [62]. The concept of Value-Chain Mass Intensity (VCMI) was introduced, which accounts for the natural resources required to produce the input materials. The study found that the ability of a mass-based metric to approximate an environmental impact category depends on a few key input materials that act as proxies. For example, the input material "coal" implies a combustion process that emits CO₂, making it a key proxy for climate change impact [62].

Correlation with Life Cycle Assessment (LCA)

The following table synthesizes findings from correlation analyses between mass intensities and LCA impacts, highlighting the conditional reliability of PMI.

Table 2: Correlation Analysis Between Mass Intensities and LCA Environmental Impacts

Research Finding	Implication for PMI as a Proxy
A broader cradle-to-gate system boundary strengthens correlation with LCA for most impact categories [62].	A narrow gate-to-gate PMI is an unreliable proxy for overall environmental performance.
Different environmental impacts are approximated by distinct sets of key input materials [62].	A single mass-based metric cannot capture the multi-criteria nature of environmental sustainability.
The reliability of mass-based assessment is highly time-sensitive, especially during the transition to a defossilized economy [62].	A process with a low PMI today might have a different environmental profile tomorrow as the energy grid and supply chains evolve.
A streamlined PMI-LCA tool that combines PMI with cradle-to-gate footprint data can effectively guide process development [26].	Augmenting PMI with LCA data provides a more accurate and practical solution for routine evaluation.

Experimental Protocols for Environmental Metric Evaluation

Protocol 1: Systematic Correlation Analysis Between Mass Intensities and LCA

1. Objective: To quantitatively determine if and how strongly different mass intensities correlate with a comprehensive set of LCA environmental impact categories.

2. Experimental Design:

• Sample Set: Select a large, diverse set of chemical production processes (e.g., n > 100) from a commercial LCA database like ecoinvent [62].
• Variables:
  • Independent Variables: Calculate multiple mass intensities with varying system boundaries (e.g., gate-to-gate PMI, seven cradle-to-gate VCMIs).
  • Dependent Variables: Calculate a wide range of LCA environmental impacts (e.g., 16 categories such as climate change, freshwater eutrophication, land use) for each process.

3. Methodology:

• System Boundary Definition for VCMI: Categorize all value-chain products into distinct classes (e.g., based on Central Product Classification). Systematically expand the VCMI system boundary by stepwise inclusion of these product classes [62].
• Data Analysis: Calculate Spearman correlation coefficients between each mass intensity metric and each LCA impact category. This non-parametric method assesses monotonic relationships without assuming linearity [62].

4. Key Measurements & Analysis:

• Identify which mass intensity system boundaries show statistically significant correlations with which LCA categories.
• Analyze the causal relationships by identifying "key input materials" (e.g., coal, metals) that drive the correlation for specific impact categories [62].

Protocol 2: Application of a Streamlined PMI-LCA Tool

1. Objective: To integrate cradle-to-gate environmental footprint data into PMI calculations for rapid, data-informed decision-making during chemical process development.

2. Experimental Workflow:

3. Procedures:

• Step 1 (Route Definition): Establish the detailed synthetic route and process conditions for the target molecule (e.g., an Active Pharmaceutical Ingredient like MK-7264) [26].
• Step 2 (PMI Calculation): Perform a mass balance to calculate the traditional gate-to-gate PMI.
• Step 3 (Inventory Compilation): Create a comprehensive list of all raw materials used in the synthesis, excluding solvents and reagents accounted for in Step 2.
• Step 4 (LCA Data Mapping): Assign corresponding cradle-to-gate environmental impact factors (e.g., kg CO₂-eq per kg material) to each raw material from a reputable LCA database.
• Step 5 (Footprint Aggregation): Calculate the total cradle-to-gate environmental footprint by summing the contributions from all raw materials.
• Step 6 (Interpretation & Prioritization): Use the combined PMI and footprint results to identify environmental hotspots and prioritize research efforts (e.g., focus on replacing high-footprint reagents) [26].

The Scientist's Toolkit: Research Reagent Solutions

Table 3: Essential Tools and Databases for Advanced Environmental Assessment

Tool / Resource	Function in Environmental Assessment
LCA Databases (e.g., ecoinvent)	Provide life cycle inventory data and impact assessment factors for thousands of chemicals and materials, enabling cradle-to-gate footprint calculation [62] [26].
Streamlined PMI-LCA Tools	Combine the simplicity of PMI with the rigor of LCA by integrating footprint data directly into process development workflows, requiring minimal extra data [26].
Chemical Inventory Screening	Tools to screen raw material inventories against cost and commercial availability criteria (e.g., Sigma-Aldrich catalog, <$100/mol) to define consistent system boundaries [62].
Spearman Correlation Analysis	A statistical method used to evaluate the strength and direction of the monotonic relationship between mass-based metrics and LCA impacts, without assuming linearity [62].
Central Product Classification (CPC)	A standardized system for categorizing products, enabling the systematic expansion of value-chain system boundaries in VCMI calculations [62].

The limitations of PMI are inherent to its fundamental nature as a single, mass-based metric attempting to represent a multi-dimensional problem. While it remains a useful indicator for resource efficiency within a defined scope, it cannot reliably proxy the full spectrum of environmental impacts on its own. Future research and practice must move towards simplified LCA methods and tools that are accessible to chemists and engineers, providing a more direct, accurate, and comprehensive assessment of environmental performance, particularly during the critical transition to a low-carbon, circular economy [62].

In the pursuit of a more sustainable chemical industry, the environmental performance of chemical processes, particularly in pharmaceutical development, must be rigorously evaluated. While Process Mass Intensity (PMI) has emerged as a widely adopted green chemistry metric for its simplicity and minimal data requirements, recent research underscores its limitations as a standalone proxy for comprehensive environmental impact [2]. Life Cycle Assessment (LCA) represents the gold standard for holistic environmental evaluation but faces practical barriers due to its data intensity and complexity [2] [64]. This creates a critical methodological gap in chemical development. The future of assessment lies in the strategic integration of PMI with simplified LCA methods, creating a tiered approach that balances practicality with environmental relevance. This paradigm shift is essential for guiding genuine sustainability improvements in drug development and beyond, moving beyond single-score metrics to embrace the multi-criteria nature of environmental sustainability [2].

Quantitative Comparison of Assessment Metrics

Table 1: Comparative Analysis of Mass Intensity and LCA Approaches

Metric/Method	System Boundary	Key Outputs	Data Requirements	Primary Strengths	Key Limitations
Process Mass Intensity (PMI)	Gate-to-Gate	Mass of input per mass of product (kg/kg) [2]	Low (process mass balance only)	Simple, quick to calculate, useful for internal process tracking [2]	Fails to capture supply chain impacts; ignores material hazard and energy source [2]
Value-Chain Mass Intensity (VCMI)	Cradle-to-Gate	Mass of natural resources per mass of product (kg/kg) [2]	Moderate to High (requires upstream resource data)	Stronger correlation with LCA impacts than gate-to-gate PMI [2]	Still a single-score, mass-based metric; does not differentiate impact types [2]
Simplified LCA	Cradle-to-Gate (Recommended minimum) [64]	Multiple impact scores (e.g., GWP, HH, EQ, NR) [35]	High, but can be streamlined with tools and proxies	Multi-criteria assessment; captures trade-offs; identifies environmental hotspots [2] [35]	Data-intensive; requires expertise and access to LCA databases [2]
Iterative Closed-Loop LCA [35]	Cradle-to-Gate	Detailed impact profiles for complex molecules	Very High (uses retrosynthesis to fill data gaps)	High accuracy for complex syntheses; enables robust route comparison [35]	Time-consuming; not suitable for very early-stage development

Experimental Protocols for Integrated Assessment

Protocol 1: Calculating Value-Chain Mass Intensity (VCMI)

This protocol outlines a systematic method for expanding PMI calculations to a cradle-to-gate boundary, thereby improving its correlation with full LCA impacts [2].

• Define the Product System: Clearly identify the chemical product and the final synthesis step to be assessed.
• Compile Gate-to-Gate Inventory: Quantify all mass inputs (reactants, reagents, solvents, catalysts) and the product output for the immediate chemical process.
• Calculate Baseline PMI: Compute the Process Mass Intensity using the standard formula: PMI = (Total Mass of Inputs to the Process) / (Mass of Product).
• Map the Value Chain: For each input identified in Step 2, trace its production pathway backwards to the extraction of natural resources (the "cradle").
• Classify Upstream Products: Categorize value-chain products into standardized classes (e.g., based on Central Product Classification) [2]. Common classes include bulk petrochemicals, metals, minerals, and agricultural products.
• Calculate VCMI: Sum the masses of all natural resources identified in the value chain mapping. The VCMI is calculated as: VCMI = (Total Mass of Natural Resources from Value Chain) / (Mass of Product).

Protocol 2: Simplified LCA for API Synthesis Route Selection

This protocol provides a streamlined methodology for conducting a cradle-to-gate LCA to compare alternative synthetic routes, based on tools like the PMI-LCA tool from the ACS GCI Pharmaceutical Roundtable and iterative approaches from recent literature [65] [35].

• Goal and Scope Definition:

  • Functional Unit: Define as 1 kg of purified Active Pharmaceutical Ingredient (API) [35].
  • System Boundary: Apply a cradle-to-gate boundary, encompassing raw material extraction, transportation, and all synthesis and purification steps up to the final API [64].
  • Impact Categories: Select a relevant set of indicators. At a minimum, include Global Warming Potential (GWP), and ideally expand to endpoint categories such as Human Health (HH), Ecosystem Quality (EQ), and Natural Resources (NR) [35].

• Life Cycle Inventory (LCI) Compilation:

  • Create Process Flow Diagram: Map all unit processes for each synthetic route.
  • Compile Input/Output Data: For each process step, collect data on mass and energy flows.
  • Address Data Gaps: For chemicals not found in LCA databases (e.g., ecoinvent), employ a tiered approach:
    • Tier A (Preferred): Use proxy data from analogous chemicals or class-averages.
    • Tier B (Advanced): Apply an iterative retrosynthetic approach to model the life cycle inventory of missing chemicals by breaking them down to basic building blocks available in the database [35].

• Life Cycle Impact Assessment (LCIA):

  • Classify and Characterize: Use dedicated LCA software (e.g., Brightway2) and established LCIA methods (e.g., ReCiPe 2016) to calculate impact category results [35].
  • Contribution Analysis: Analyze the results to identify "hotspots" – processes or chemicals that contribute most significantly to the overall environmental impact.

Diagram 1: Simplified LCA workflow for API synthesis. The process handles data gaps via proxies or retrosynthesis.

The Scientist's Toolkit: Key Reagent Solutions

Table 2: Research Reagent Solutions for Sustainable Synthesis

Reagent/Category	Function in Synthesis	Sustainability Consideration	Example & Application Note
Brønsted-Acid Catalysts	Enantioselective catalysis for chiral center formation [35]	Reduces need for heavy metal catalysts; often lower embedded energy.	Used in a novel Mukaiyama-Mannich addition during Letermovir synthesis to avoid metal-based catalysts [35].
Boron-Based Reducing Agents	Reduction of carbonyls and other functional groups.	Lower environmental impact compared to traditional metal hydrides.	Replaced LiAlH4 reduction in an exploratory route for Letermovir, mitigating negative environmental impact [35].
Bio-Derived Phase-Transfer Catalysts (PTC)	Facilitate reactions between reactants in immiscible phases.	Derived from renewable resources (e.g., biomass).	A cinchonidine-derived PTC was used in an enantioselective addition; its LCA inventory considered biomass origin [35].
Palladium Catalysts	Cross-coupling reactions (e.g., Heck, Suzuki).	High environmental impact hotspot; use should be optimized and justified.	Identified as a major hotspot in the published Letermovir route via a Heck coupling [35].

Case Study: Integrated Assessment of Antiviral API Synthesis

A 2025 study on the synthesis of the antiviral drug Letermovir provides a compelling real-world application of integrated PMI and LCA assessment [35]. The research compared the published, award-winning manufacturing route with a novel de novo synthesis.

• PMI Analysis: Both routes were first analyzed using traditional mass-based metrics to understand material efficiency.
• LCA Integration: A comprehensive, iterative LCA was then conducted, following a protocol similar to Protocol 2. This involved a cradle-to-gate assessment with a functional unit of 1 kg of Letermovir, evaluating impacts on GWP, HH, EQ, and NR [35].
• Key Findings and Hotspots:
  • The LCA revealed that the Pd-catalyzed Heck coupling in the published route was a significant environmental hotspot, an insight that would be missed by PMI alone [35].
  • In the de novo route, an enantioselective Mukaiyama–Mannich addition using Brønsted-acid catalysis was identified as the main hotspot, but with a different environmental profile than the Heck coupling [35].
  • The study demonstrated that large solvent volumes for purification were a common issue in both routes, highlighting an area for future green chemistry optimization that is also captured by high PMI values [35].

This case underscores how PMI and LCA are complementary. PMI can flag material-intensive processes, while LCA pinpoints the specific chemistry creating the greatest environmental burden, allowing chemists to make informed, targeted improvements.

Diagram 2: PMI and LCA integration reveals different but complementary process hotspots for informed decision-making.

Conclusion

Process Mass Intensity stands as an indispensable, practical metric for driving efficiency and waste reduction in pharmaceutical development and beyond. Its true power is unlocked when applied with clearly defined, cradle-to-gate system boundaries and when used as part of a suite of tools, including predictive analytics and machine learning for optimization. However, as the industry transitions toward a defossilized, low-carbon economy, PMI's limitation as a pure mass-based metric becomes apparent; it cannot fully capture the multi-criteria nature of environmental sustainability. For the future, the most robust approach will integrate PMI with emerging simplified Life Cycle Assessment methods. This will enable researchers and drug developers to not only minimize mass waste but also make truly informed decisions that reduce overall environmental impact, ensuring that green chemistry advances deliver genuine and comprehensive ecological benefits.

References

• 1. Process Mass Intensity → Term [https://lifestyle.sustainability-directory.com/term/process-mass-intensity/]
• 2. Assessing mass intensity as a green chemistry metric [https://pubs.rsc.org/en/content/articlehtml/2025/gc/d5gc02994e]
• 3. Mass Intensity - an overview [https://www.sciencedirect.com/topics/chemistry/mass-intensity]
• 4. Process Mass Intensity (PMI): A Holistic Analysis of Current ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC11002941/]
• 5. Metrics of green chemistry: Waste minimization [https://www.sciencedirect.com/science/article/abs/pii/S2452223621001255]
• 6. Relationships between step and cumulative PMI and E- ... [https://www.degruyterbrill.com/document/doi/10.1515/gps-2018-0131/html?lang=en&srsltid=AfmBOor1ITWwcYBECd6-G_MklK1BZjes6Rt3kSzWoEJbL0WCdVupf6Aw]
• 7. 13.10: The E-Factor in Green Chemistry [https://chem.libretexts.org/Bookshelves/Environmental_Chemistry/Green_Chemistry_and_the_Ten_Commandments_of_Sustainability_(Manahan)/13%3A_The_Anthrosphere_Industrial_Ecology_and_Green_Chemistry/13.10%3A_The_E-Factor_in_Green_Chemistry]
• 8. Green chemistry metrics [https://en.wikipedia.org/wiki/Green_chemistry_metrics]
• 9. Green Chemistry Metrics for the Fine Chemical ... [https://chemanager-online.com/en/topics/green-chemistry-metrics-for-the-fine-chemical-pharmaceutical-and-related-industries]
• 10. Green Chemistry Metrics with Special Reference to ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC6272361/]
• 11. Green Chemistry Metrics, A Review [https://www.mdpi.com/2227-9717/10/7/1274]
• 12. Enhancing drug development planning through the use of ... [https://www.pmi.org/learning/library/enhancing-drug-development-facilitated-planning-8940]
• 13. project management principles in clinical trial process [https://www.pmi.org/learning/library/project-management-principles-clinical-trial-pharmaceutical-8206]
• 14. Green chemistry metrics: Insights from case studies in fine ... [https://www.sciencedirect.com/science/article/pii/S2352554125001603]
• 15. 12 Principles of Green Chemistry | Sustainability Global [https://sustainabilityglobal.org/12-principles-of-green-chemistry/]
• 16. Principles of green chemistry: building a sustainable future [https://link.springer.com/article/10.1007/s44371-025-00152-9]
• 17. Process Mass Intensity Metric [https://learning.acsgcipr.org/guides-and-metrics/metrics/process-mass-intensity-metric/]
• 18. Twelve Principles of Green Chemistry [https://openbooks.lib.msu.edu/orgchemlabmanual/chapter/twelve-principles-of-green-chemistry/]
• 19. Green Chemistry Using Smarter Metrics: A Life Cycle ... [https://acsgcipr.org/green-chemistry-using-smarter-metrics-a-life-cycle-approach/]
• 20. Green Chemistry and Engineering Metrics [https://www.acs.org/green-chemistry-sustainability/research-innovation/engineerings-metrics.html]
• 21. Fine Chemicals Market | Global Market Analysis Report [https://www.futuremarketinsights.com/reports/fine-chemicals-market]
• 22. Process Mass Intensity Calculator [https://acsgcipr.org/tools/process-mass-intensity/]
• 23. Green chemistry and sustainability metrics in the pharmaceutical manufacturing sector - ScienceDirect [https://www.sciencedirect.com/science/article/abs/pii/S2452223621001188]
• 24. Cradle-to-Gate in LCA – What is it & How does it work? [https://ecochain.com/blog/cradle-to-gate-in-lca/]
• 25. Cradle to Gate vs. Cradle to Grave: What's the Difference? [https://www.earthster.org/post/cradle-to-gate-vs-cradle-to-grave-what-s-the-difference]
• 26. Green and sustainable metrics: Charting the course for green-by-design small molecule API synthesis - ScienceDirect [https://www.sciencedirect.com/science/article/pii/S2666086522000662]
• 27. Green and sustainable chemistry – The case for a systems ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC8654969/]
• 28. Revisiting the Basics of Life Cycle Assessment and ... [https://www.mdpi.com/2071-1050/17/16/7444]
• 29. The PMI Predictor - a Web App Enabling Green-by-Design ... [https://chemrxiv.org/engage/chemrxiv/article-details/60c74011567dfe6429ec3b01]
• 30. PMI Prediction Tool [https://members.acsgcipr.org/pmi-prediction-tool/]
• 31. PMI Prediction and Bayesian Optimization: Two Tools with ... [https://acsgcipr.org/pmi-prediction-and-bayesian-optimization-two-tools-with-one-goal-towards-the-development-of-greener-by-design-synthesis-of-apis/]
• 32. Green Chemistry Research Tools, Guides & Best Practices [https://www.acs.org/green-chemistry-sustainability/research-innovation/research-tools.html]
• 33. API manufacturing: The cost of decarbonization [https://www.mckinsey.com/industries/life-sciences/our-insights/decarbonizing-api-manufacturing-unpacking-the-cost-and-regulatory-requirements]
• 34. Recent advances in green chemistry approaches for ... [https://www.sciencedirect.com/science/article/abs/pii/S2950357424000295]
• 35. Integrated Life Cycle Assessment Guides Sustainability in ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12593353/]
• 36. ACS GCI Pharmaceutical Roundtable Announces 2025 ... [https://chemistryforsustainability.org/community-news/acs-gci-pharmaceutical-roundtable-announces-2025-industry-award-winners]
• 37. Data Science and Modeling for Green Chemistry [https://www.acs.org/funding/awards/data-science-and-modeling-for-green-chemistry.html]
• 38. New Approach Methodologies: NAMs [https://www.pmda.go.jp/english/review-services/0022.html]
• 39. A green approach to 2-arylquinolines via palladium- ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12584958/]
• 40. Assessing mass intensity as a green chemistry metric [https://pubs.rsc.org/en/content/articlelanding/2025/gc/d5gc02994e]
• 41. Applications of machine learning in drug discovery and ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC6552674/]
• 42. Bayesian Optimization for Chemical Synthesis in the Era of ... [https://www.mdpi.com/2227-9717/13/9/2687]
• 43. Bayesian optimisation for chemical problems - RSC Publishing [https://pubs.rsc.org/en/content/articlehtml/2024/dd/d3dd00234a]
• 44. Development of machine learning model and analysis study of ... [https://arabjchem.org/development-of-machine-learning-model-and-analysis-study-of-drug-solubility-in-supercritical-solvent-for-green-technology-development/]
• 45. SUSSOL—Using Artificial Intelligence for Greener Solvent ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC7411708/]
• 46. The CHEM21 Solvent Selection Guide [https://learning.acsgcipr.org/guides-and-metrics/solvent-selection-guides/the-chem21-solvent-selection-guide/]
• 47. A Preselection Method for Screening Green Solvents in the ... [https://www.mdpi.com/2071-1050/17/8/3285]
• 48. Reaction Optimization for Greener Chemistry with a ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC9738638/]
• 49. New Screening Protocol for Effective Green Solvents ... [https://www.mdpi.com/1420-3049/27/10/3323]
• 50. Tools and techniques for solvent selection: green solvent ... [https://sustainablechemicalprocesses.springeropen.com/articles/10.1186/s40508-016-0051-z]
• 51. Green solvents and their role in modern sample ... [https://link.springer.com/article/10.1007/s44371-025-00325-6]
• 52. Why we might be misusing process mass intensity (PMI) ... [https://pubs.rsc.org/en/content/articlelanding/2020/gc/c9gc01537j]
• 53. PMI-LCA Tool Development Challenge [https://acsgcipr.org/funding-awards/pmi-lca-tool-development-challenge/]
• 54. Assessing mass intensity as a green chemistry metric [https://www.sciencedirect.com/org/science/article/pii/S1463926225006594]
• 55. Comparison of process mass intensity (PMI) of continuous ... [https://www.sciencedirect.com/science/article/pii/S1871678422000619]
• 56. Green Chemistry Using Smarter Metrics: A Life Cycle ... [https://www.acs.org/acs-webinars/library/pmi-lca-tool.html]
• 57. Cradle‑to‑Gate (LCA) [https://www.climatepartner.com/en/knowledge/glossary/cradle-to-gate]
• 58. The E factor 25 years on: the rise of green chemistry and sustainability - Green Chemistry (RSC Publishing) [https://pubs.rsc.org/en/content/articlelanding/2017/gc/c6gc02157c]
• 59. The environmental impact of inhaled therapy - ERS Publications [https://publications.ersnet.org/content/erj/60/1/2102106]
• 60. Switching to the Dry Powder Inhaler: Disease Control with a ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12623521/]
• 61. Trends in inhaler use and associated carbon footprint [https://pmc.ncbi.nlm.nih.gov/articles/PMC12406821/]
• 62. Assessing mass intensity as a green chemistry metric [https://pubs.rsc.org/it-it/content/articlehtml/2025/gc/d5gc02994e]
• 63. Celebrating Breakthroughs in Sustainable Manufacturing [https://www.acs.org/acs-webinars/library/green-pharma-breakthroughs-2025.html]
• 64. A proposal of twelve principles for LCA of chemicals [https://pubs.rsc.org/en/content/articlehtml/2025/gc/d4gc04844j?page=search]
• 65. PMI Life Cycle Assessment [https://acsgcipr.org/tools/pmi-life-cycle-assessment/]