Integration of Quality by Design (QbD) with green chemistry in pharmaceutical Analysis: A sustainable Approach

The pharmaceutical industry operates within one of the most rigorously regulated analytical environments in the world. Analytical methods underpin every stage of the drug product lifecycle—from early-stage raw material characterisation through process monitoring, finished product release testing, stability studies, and post-market surveillance. The obligation to assure product quality whilst minimising the environmental and occupational health footprint of analytical operations has grown steadily more pressing, driven by regulatory evolution, institutional sustainability commitments, and the practical economics of large-scale pharmaceutical manufacturing ^[1]

Traditional approaches to analytical method development were empirical and univariate: one variable was changed at a time (OFAT), and the resulting method was validated according to ICH Q2(R1) criteria for specificity, linearity, accuracy, precision, detection and quantitation limits, range, and robustness.^[2] Such methods provided limited understanding of variable interactions, method boundaries, or real-world robustness, and generated no systematic pressure toward greener reagent choices or reduced solvent consumption.

QbD emerged as a transformative paradigm with publication of ICH Q8(R2) in 2009, complemented by Q9 (risk management), Q10 (quality systems), and most significantly ICH Q14 (Analytical Procedure Development, 2022)—the first ICH guideline explicitly addressing analytical method development under a QbD approach^.[3,4,5] Q14 establishes the Analytical Target Profile (ATP), endorses the MODR concept, and supports submission of enhanced method development data, enabling post-approval within-MODR adjustments without prior regulatory approval—a significant operational incentive for industry adoption.^[3] Concurrently, Green Analytical Chemistry (GAC) matured from aspirational principles into a toolbox of quantitative metrics^[6] The 12 Principles of Green Chemistry, first articulated by Anastas and Warner in 1998,^[7] were adapted to analytical chemistry by Namiesnik and colleagues, who established the conceptual vocabulary of waste minimisation, direct analysis, and solvent reduction in analytical practice.^[6] The convergence of QbD and Green Chemistry constitutes a practical, operationalizable strategy for developing pharmaceutical analytical methods that are simultaneously scientifically sound, regulatory-compliant, and environmentally responsible. The DoE framework central to QbD enables multi-objective optimisation incorporating green metrics such as AGREE^[8] and AES^[9] as response variables alongside chromatographic resolution and sensitivity, identifying operating conditions that satisfy all goals simultaneously ^{[10, 11]}

This review aims to provide a comprehensive, critical synthesis of the literature on the integration of QbD with Green Chemistry in pharmaceutical analysis. It establishes theoretical foundations of each paradigm independently, examines the tools and metrics through which integration is achieved, surveys case study evidence across diverse drug classes and analytical techniques, and addresses regulatory perspectives, challenges, and future directions.

FOUNDATIONS OF QUALITY BY DESIGN IN PHARMACEUTICAL ANALYSIS

Historical Development and Regulatory Context

The intellectual origins of QbD in pharmaceutical science trace to Joseph M. Juran's quality planning concepts, emphasising the importance of designing quality into products from the outset rather than testing it in post-hoc. The pharmaceutical adaptation was catalysed by FDA Process Analytical Technology (PAT) initiatives in the early 2000s, which sought to modernise pharmaceutical manufacturing through real-time monitoring and scientific principles^.[12]

The ICH Q8(R2) guideline introduced the Quality Target Product Profile (QTPP) and design space concept, establishing the philosophical template for QbD in pharmaceutical development^[1] The subsequent Q9(R1) and Q10 guidelines provided risk management and quality system frameworks within which QbD operates^.[5] The landmark ICH Q14 guideline, finalised in 2022, explicitly establishes the ATP as the analytical counterpart to the QTPP and endorses the MODR as the analytical equivalent of the pharmaceutical design space^[3]

The ICH Q2(R2) revision (2022), harmonised with Q14, updated the classical validation framework to accommodate QbD-developed methods, including performance characterisation across the MODR rather than at a single operating point^[4] The FDA and EMA have both issued reflection papers supporting QbD approaches and indicating acceptance of enhanced analytical method submissions that define a MODR, enabling post-approval method adjustments without prior regulatory approval^[3,4]

KEY ELEMENTS OF THE QBD FRAMEWORK FOR ANALYTICAL METHODS

Analytical Target Profile (ATP)

The ATP, defined in ICH Q14, is a prospective statement of the performance characteristics an analytical procedure must achieve to be fit for its intended purpose ^[3] It encompasses specificity, accuracy, precision (repeatability and intermediate precision), linearity, range, detection and quantitation limits, and robustness criteria, all defined at the outset of method development. The ATP drives every subsequent development decision, ensuring that method optimisation is goal-directed rather than arbitrary. Sustainability criteria—minimum acceptable AGREE score, maximum permissible organic solvent volume per analysis, maximum waste generation—can and should be incorporated into the ATP when developing an integrated QbD-Green method ^[8,10]

Risk Assessment Tools

Risk assessment tools are applied early in the QbD workflow to identify Critical Method Attributes (CMAs) and Critical Method Parameters (CMPs). The Ishikawa (fishbone) diagram provides a visual cause-and-effect mapping of method parameters onto performance attributes^[13] Failure Mode and Effects Analysis (FMEA) assigns risk priority numbers (RPNs) based on the severity, probability, and detectability of potential failure modes^[5,13] Outputs of risk assessment directly inform factor selection for subsequent DoE studies, ensuring experimental effort is focused on parameters with genuine potential to affect method performance.

Design of Experiments (DoE)

DoE represents the most powerful and distinctive tool in the QbD toolkit. Unlike OFAT approaches, DoE examines multiple factors simultaneously in a structured, statistically sound experimental design, enabling estimation of main effects, interaction effects, and curvature effects with a fraction of the experimental effort required by full factorial exploration^[13] Commonly employed designs in pharmaceutical analytical QbD include: Plackett-Burman (PB) designs for screening (identifying significant factors from a large candidate set using n+1 experiments for n factors); Box-Behnken Designs (BBD) for response surface modelling without requiring extreme factor combinations; Central Composite Designs (CCD) for full quadratic response surface modelling; and D-optimal designs for constrained or irregular experimental regions ^[13,14] Experimental data from DoE studies are fitted to mathematical response surface models by multivariate regression. Overlay of multiple response surface models—for resolution, peak tailing, analysis time, and green metric score—produces a graphical MODR that simultaneously satisfies all ATP requirements ^[10,11]

 Method Operable Design Region (MODR)

The MODR is the multidimensional factor space within which method performance is demonstrated to meet all ATP requirements with high statistical confidence. Its definition is the culmination of the QbD development process ^[3]. For methods operating within an approved MODR, minor adjustments to method parameters do not require prior regulatory approval, provided adjustments remain within the MODR—a powerful commercial and operational advantage ^[3,1] Monte Carlo simulation and interval analysis are used to evaluate MODR robustness and probability of compliance ^{[15, 16]}

Method Validation Within the QbD Framework

Method validation under QbD, as articulated in ICH Q2(R2) and contextualised by Q14, characterises performance across the MODR rather than at a single operating point^[4,3] This is conceptually aligned with the "Accuracy Profile" approach to validation, which provides a probabilistic assessment of the proportion of analytical results expected to fall within predefined acceptance criteria across the design space^[2] .Validation thus provides statistical confidence that the method meets ATP requirements not just under nominal conditions but throughout the claimed MODR.

FOUNDATIONS OF GREEN CHEMISTRY AND GREEN ANALYTICAL CHEMISTRY

The 12 Principles of Green Chemistry

Green Chemistry was formally defined through the 12 Principles articulated by Paul T. Anastas and John C. Warner in their 1998 monograph "Green Chemistry: Theory and Practice.”^[7]Their application to analytical chemistry has been the subject of sustained scholarly attention since the early 2000s, culminating in the establishment of Green Analytical Chemistry (GAC) as a recognised sub-discipline^.[6]

The 12 Principles most directly applicable to analytical method development are: (1) Prevention—prevent waste rather than treat it; (2) Atom Economy—maximise incorporation of starting materials into products, translating in analysis to minimal sample and reagent volumes; (3) Less Hazardous Synthesis—use and generate substances with little toxicity; (5) Safer Solvents and Auxiliaries—avoid auxiliary substances where possible, and when used make them innocuous; (6) Energy Efficiency—minimise energy requirements; (10) Design for Degradation—breakdown into innocuous degradation products; and (11) Real-time Analysis—enable in-process monitoring to avoid formation of hazardous substances. ^[7,6]

Evolution Towards Green Analytical Chemistry (GAC)

The formal articulation of GAC as a discipline is attributed primarily to Janusz Namiesnik and colleagues at Gdansk University of Technology, whose series of review papers from the early 2000s established the conceptual framework and vocabulary of the field.^[6] Namiesnik defined GAC as the development and application of analytical procedures that reduce hazards to human health and environment whilst achieving equivalent or superior analytical performance.

Key GAC strategies applicable to pharmaceutical analysis include: minimisation of sample preparation steps and volumes; replacement of organic solvents with water-based or less hazardous alternatives; use of direct analysis techniques; miniaturisation of analytical platforms; and use of renewable or recyclable reagents and solvents^.[6] These strategies align closely with the efficiency-maximisation goals of QbD, creating natural synergies the integrated framework exploits.

Green Metric Assessment Tools

A critical enabler of the quantitative integration of green chemistry with QbD has been the development of numerical green metric tools. These assign quantitative scores to analytical procedures based on environmental, health, and safety (EHS) profiles, enabling objective comparison and incorporation as DoE response variables^{. [8,9,17]}

National Environmental Method Index (NEMI)

NEMI, proposed in 2003, evaluates procedures against four binary criteria: use of persistent, bioaccumulative, or toxic (PBT) chemicals; hazardous chemical use; hazardous waste generation; and extreme pH (below 2 or above 12). Methods satisfying all four criteria are classified as "green.”^[17] Whilst NEMI provides a simple, easily communicable result, its binary nature limits discriminatory power and ability to drive incremental improvements.

Analytical Eco-Scale (AES)

The AES, developed by Van Aken, Schepdael, Sinnaeve, and Deconinck in 2011, uses a penalty-point system starting from 100, subtracting points for hazardous reagents, energy consumption, waste generation, and occupational hazards^.[9] Scores above 75 denote "excellent green analysis," 50–75 "acceptable green analysis," and below 50 "inadequate green analysis." The continuous, quantitative AES score is directly amenable to use as a DoE response variable.

Green Analytical Procedure Index (GAPI)

GAPI, developed by Plotka-Wasylka in 2016, uses a pictographic "traffic light" system based on five pentagons representing: sample collection and storage; sample preparation; analytical technique; reagents and solvents; and waste management.^[18] Each pentagon is subdivided into assessment fields scored green, yellow, or red for low, moderate, or high concern respectively. A revised version published in 2018 added additional assessment criteria covering the full analytical procedure lifecycle^.[17]

Analytical Greenness Metric (AGREE)

AGREE, proposed by Nowak, Kochana, and Pachla-Wisniewska in 2021, represents the most comprehensive and quantitatively rigorous of currently available green metric tools^.[8] It evaluates procedures against all 12 principles of GAC, assigning a numerical score from 0 (least green) to 1 (fully green) with a colour-coded circular pictogram. AGREE weights each principle according to relative importance and provides a single composite score fully comparable across different analytical procedures and techniques, making it optimal for DoE integration.

Blue Applicability Grade Index (BAGI)

BAGI, introduced by Nowak and Plotka-Wasylka in 2021 as a complement to AGREE, evaluates practical applicability and economic feasibility of analytical procedures^.[15] BAGI considers sample throughput, cost per analysis, instrument availability, and skill requirements, providing a "blue" applicability score. The combination of AGREE and BAGI scores enables comprehensive multi-criteria optimisation within a QbD framework, balancing sustainability with operational practicality.

Table 1: Comparison of Green Metric Assessment Tools Used in Pharmaceutical Analytical QbD

INTEGRATION OF QBD AND GREEN CHEMISTRY: A UNIFIED FRAMEWORK

Conceptual Basis for Integration

The conceptual alignment between QbD and Green Chemistry arises from their shared emphasis on proactive, knowledge-led design rather than reactive empirical trial-and-error^.[1,7] Both paradigms reject the notion that quality can be assured purely through post-hoc testing, and both mandate systematic exploration of design space to identify optimal operating conditions satisfying multiple performance criteria simultaneously.

In practical terms, integration is achieved by incorporating green metric scores—AGREE,^[8]AES,^[9] or GAPI^[10,11]] Conventionally, QbD method optimisation considers responses such as chromatographic resolution (Rs), peak symmetry (As), retention time, and sensitivity. The QbD-Green integration adds AGREE score, AES penalty points, organic solvent consumption volume, waste generation, and energy consumption as additional simultaneous optimisation targets. Desirability function analysis converts all responses to a common 0–1 scale and computes a composite desirability that is maximised to identify the globally optimal operating point.^[14,13]

Systematic Workflow for Integrated QbD-Green Analytical Method Development

A systematic workflow for integrated QbD-Green Chemistry method development, synthesised from current literature and regulatory guidance, proceeds through six stages:^[3,4,10,13]

• Stage 1 – ATP Definition: Establish required performance criteria including specificity, precision, accuracy, and linearity. Explicitly include sustainability criteria in the ATP: minimum acceptable AGREE score, maximum permissible organic solvent consumption per analysis, and acceptable waste generation limits.^[3,8]]
• Stage 2 – CMP Identification and Screening: Generate candidate method parameters via literature review and expert consultation. Apply FMEA and Ishikawa diagrams to prioritise parameters. Apply a Plackett-Burman or fractional factorial screening DoE to identify parameters with significant effects on both analytical performance and green metrics.^[13,5]
• Stage 3 – Response Surface Optimisation: Apply BBD or CCD to systematically explore critical parameters across all analytical performance and green sustainability responses. Fit polynomial models and validate via ANOVA, R², lack-of-fit test, and residual analysis.^[14,13]
• Stage 4 – MODR Construction: Use overlay contour plots and desirability function optimisation to define the MODR satisfying all analytical performance and sustainability targets simultaneously. Evaluate MODR robustness with Monte Carlo simulation.^[3,11]
• Stage 5 – Validation of the Optimised Green Method: Validate per ICH Q2(R2) and Q14 at the optimal point and MODR boundaries. Calculate and document AGREE, AES, GAPI, and BAGI scores. Compare against benchmark methods to quantify sustainability improvements.^[4,3,8,15]
• Stage 6 – Documentation and Regulatory Submission: Prepare documentation including ATP, risk assessment outputs, DoE plan and results, response surface models, MODR definition, and green metric scores. Submit under the enhanced approach supported by ICH Q14.^[3]

Selection of Experimental Designs for Integrated Optimisation

The choice of experimental design within the integrated QbD-Green framework is governed by the number of critical factors, the degree of model complexity required, practical experimental constraints, and the nature of the response surface.^[13,14] For complex multi-response optimisation incorporating green metrics, CCD and BBD are preferred due to their ability to fit second-order response surface models capturing the curvature of responses such as chromatographic resolution and AGREE score in the vicinity of the optimum.^[14]

Mixture designs are particularly relevant to green mobile phase optimisation, where mobile phase component fractions must sum to unity, creating a constrained experimental region requiring dedicated design strategies.^[19] Simplex-lattice and simplex-centroid designs enable simultaneous optimisation of mobile phase composition across ternary or quaternary solvent mixtures whilst incorporating green metric evaluation of alternative solvent systems.

APPLICATIONS IN PHARMACEUTICAL ANALYSIS: CASE STUDIES AND EVIDENCE

High-Performance Liquid Chromatography (HPLC)

HPLC remains the most widely employed technique in pharmaceutical analysis, and consequently the largest body of literature on QbD-Green Chemistry integration pertains to HPLC method development.^[14,20] The critical role of mobile phase composition and flow rate in determining both chromatographic performance and solvent consumption makes HPLC particularly amenable to multi-objective QbD-Green optimisation.

Orlandini et al. published landmark studies demonstrating simultaneous optimisation of analytical performance and greenness in HPLC methods for pharmaceutical compounds. Employing Box-Behnken designs with acetonitrile/methanol/buffer mobile phase variables and column temperature as factors, these authors incorporated AES penalty points alongside chromatographic responses in desirability function optimisation.^[10,11] The approach reduced acetonitrile consumption substantially compared to reference pharmacopoeial methods whilst maintaining complete baseline resolution of all analytes, establishing a methodological template widely adopted in subsequent literature.

Kumar et al. applied a QbD framework with Plackett-Burman screening followed by Box-Behnken optimisation to develop a green RP-HPLC method for simultaneous estimation of amlodipine and Olmesartan in combined antihypertensive tablets.^[21] Risk assessment using an Ishikawa diagram identified mobile phase acetonitrile content, buffer concentration, and pH as critical method parameters. DoE optimisation achieved an AES score of 76 (excellent green) at the optimal operating point, with robustness demonstrated across the established MODR.

Chavez-Blanco et al. demonstrated DoE-optimised HPLC methods for simultaneous determination of metformin and glipizide in combined antidiabetic formulations.^[22] A central composite design optimised mobile phase pH, organic modifier fraction, and flow rate, with AES and analysis time as co-optimised responses. The resulting method employed an ammonium formate/methanol mobile phase, replacing a conventional acetonitrile-phosphate buffer system, achieving a 40% reduction in organic solvent use alongside a 25% reduction in analysis time without compromise of analytical performance.

Sahu et al. provided comprehensive methodological guidance on the application of experimental designs in HPLC method development and validation within pharmaceutical QbD frameworks, documenting design selection criteria, model validation approaches, and MODR construction strategies across multiple drug classes.^[14]

Ultra-Performance Liquid Chromatography (UPLC)

UPLC, employing sub-2-µm particle columns at higher operating pressures, inherently offers significant greenness advantages through dramatically reduced analysis times and mobile phase volumes. Integration of QbD with UPLC development amplifies these advantages by enabling systematic identification of operating conditions that maximise these efficiencies.^[23]

Patel et al. demonstrated QbD-based development of a UPLC method for the simultaneous determination of multiple antihypertensive drugs.^[23] A central composite design was used to optimise gradient conditions, flow rate, and column temperature. The final method consumed substantially less mobile phase compared to an equivalent HPLC method—representing over 85% reduction in solvent consumption. AGREE score assessment yielded a value categorised as "good" green practice, and the authors demonstrated incorporation of this score as a DoE response alongside peak resolution, peak width, and retention time—an important methodological contribution to the integrated QbD-Green literature.

Sahu et al. reported QbD-optimised UPLC methods for antiretroviral drug combinations.^[20] Box-Behnken designs with acetonitrile percentage, aqueous phase pH, and flow rate as factors, with desirability function analysis incorporating AGREE score and chromatographic resolution, identified optimal mobile phases achieving complete resolution in under 3 minutes with minimal solvent consumption—exemplifying the practical efficiency gains of the integrated approach.

Micellar Liquid Chromatography (MLC) and Micellar Electrokinetic Capillary Chromatography (MEKC)

Micellar liquid chromatography (MLC), employing surfactant solutions above their critical micelle concentration as the mobile phase, is inherently one of the greener alternatives to conventional reversed-phase HPLC, as surfactants replace organic solvents as the primary modulator of retention, enabling predominantly aqueous operation.^[19]

Garcia-Alvarez-Coque et al. have extensively documented QbD-based MLC method development for pharmaceutical applications, demonstrating that Box-Behnken designs effectively characterise the response surface of chromatographic resolution with respect to sodium dodecyl sulphate (SDS) concentration, organic modifier content, and pH, whilst AES scores consistently exceeded 80 for optimal mobile phase compositions—well above the "excellent green analysis" threshold.^[19]

Shakya et al. applied QbD to optimise an MEKC method for simultaneous determination of four B-group vitamins in pharmaceutical preparations.^[16] A central composite design with SDS concentration, borate buffer concentration, and applied voltage as factors yielded a method with AGREE score categorised as excellent, achieving complete separation of all four vitamins within 8 minutes. The use of an aqueous SDS electrolyte system generated no organic solvent waste, and the short analysis time minimised energy consumption—both key AGREE criteria.

UV-Visible Spectrophotometry

UV-Visible spectrophotometry, as a simpler and more widely accessible analytical technique, presents distinct opportunities and constraints for QbD-Green integration. The absence of chromatographic separation means primary greenness drivers are reagent toxicity in derivative spectrophotometric methods, sample volume, and waste generation.^[24]

Wróbel et al. developed a QbD-optimised derivative spectrophotometric method for quantification of sulfamethoxazole in pharmaceutical preparations.^[24] A Plackett-Burman design screened ten reaction variables, identifying reagent concentration, reaction temperature, and reaction time as critical parameters. Box-Behnken optimisation of these three factors yielded conditions minimising reagent consumption whilst maximising colour development. Waste generation was reduced by 60% compared to the conventional method, with AES score achieving acceptable green classification—demonstrating that even relatively simple techniques benefit substantially from QbD-Green integration.

Near-Infrared (NIR) and Raman Spectroscopy

NIR spectroscopy represents perhaps the most inherently green analytical technique applicable to pharmaceutical analysis, enabling non-destructive, reagent-free, rapid analysis of solid and liquid samples with minimal or no sample preparation. Integration of QbD with NIR method development is manifested primarily in systematic calibration model development.^[25]

De Beer et al. demonstrated the application of QbD principles to development of NIR reflectance methods for blend uniformity monitoring in solid dosage form manufacturing.^[25] The systematic design of calibration standards using a mixture design, combined with partial least squares (PLS) regression model development and cross-validation, produced methods with exceptional predictive performance across ATP-defined concentration ranges. NIR methods require no solvents, generate no liquid waste, and can be applied non-destructively, earning maximum AGREE scores for all environmental criteria—a compelling combination of analytical quality and sustainability.

Raman spectroscopy, similarly non-destructive and reagent-free, has been subject to comparable QbD-based method development.^[24] Application of DoE to optimisation of instrumental parameters (laser power, exposure time, accumulation number) alongside sampling conditions enables systematic MODR characterisation within a QbD framework, whilst the intrinsic greenness of the technique is captured in AGREE and AES assessments that invariably yield excellent scores.

Electroanalytical Methods

Electroanalytical techniques, particularly differential pulse voltammetry (DPV) and square wave voltammetry (SWV), have received growing attention in pharmaceutical analysis as green alternatives to chromatographic methods for determination of electroactive compounds. QbD-based development involves optimisation of electrode material, electrolyte composition, pulse parameters, and scan rate using DoE approaches.^[20]

Application of Box-Behnken designs to optimise DPV conditions for pharmaceutical compounds, incorporating AGREE score as a co-optimised response alongside sensitivity and peak resolution, has demonstrated the excellent inherent greenness of voltammetric techniques.^[20] The use of aqueous electrolyte systems generates no organic solvent waste, and the rapid analysis time minimises energy consumption—both key AGREE criteria—enabling AGREE scores above 0.85 in optimal aqueous electrolyte DPV systems. The absence of complex sample preparation further reduces the environmental footprint of electroanalytical pharmaceutical methods.

Table 2: Summary of Representative QbD-Green Chemistry Integrated Pharmaceutical Analytical Methods