Process Analytical Technology: Spectroscopic Tools and Implementation Strategies for the Chemical and Pharmaceutical Industries

Process Analytical Technology explores the concepts of PAT and its application in the chemical and pharmaceutical industry from the point of view of the analytical chemist. In this new edition all of the original chapters have been updated and revised, and new chapters covering the important topics of sampling, NMR, fluorescence, and acoustic chemometrics have been added. Coverage includes: Implementation of Process Analytical Technologies UV-Visible Spectroscopy for On-line Analysis Infrared Spectroscopy for Process Analytical Applications Process Raman Spectroscopy Process NMR Spectrscopy: Technology and On-line Applications Fluorescent Sensing and Process Analytical Applications Chemometrics in Process Analytical Technology (PAT) On-Line PAT Applications of Spectroscopy in the Pharmaceutical Industry Future Trends for PAT for Increased Process Understanding and Growing Applications in Biomanufacturing NIR Chemical Imaging This volume is an important starting point for anyone wanting to implement PAT and is intended not only to assist a newcomer to the field but also to provide up-to-date information for those who practice process analytical chemistry and PAT. It is relevant for chemists, chemical and process engineers, and analytical chemists working on process development, scale-up and production in the pharmaceutical, fine and specialty chemicals industries, as well as for academic chemistry, chemical engineering, chemometrics and pharmaceutical science research groups focussing on PAT. Review from the First Edition “The book provides an excellent first port of call for anyone seeking material and discussions to understand the area better. It deserves to be found in every library that serves those who are active in the field of Process Analytical Technology.”—Current Engineering Practice

Table of Contents:
Preface to the Second Edition xvii List of Contributors xix List of Abbreviations xxi 1 Overview of Process Analysis and PAT 1 Jason E. Dickens 1.1 Introduction 1 1.1.1 Historical perspective 3 1.1.2 Business drivers 4 1.2 Execution of Process Analysis Projects 5 1.2.1 Wisdoms 5 1.2.2 Team structure 6 1.2.3 Project life cycle 6 1.2.4 Project scoping 9 1.2.5 Common challenges and pitfalls 10 1.3 Process Instrumentation 12 1.3.1 Process instrumentation types 12 1.3.2 Novel process instrumentation 12 1.4 Conclusions 13 1.5 Glossary of Acronyms and Terms 14 References 14 2 Implementation of Process Analytical Technologies 17 Robert Guenard and Gert Thurau 2.1 Introduction to Implementation of Process Analytical Technologies (PAT) in the Industrial Setting 17 2.1.1 Definition of process analytics 18 2.1.2 Differences between process analyzers and laboratory analysis 19 2.1.3 General industrial drivers for PA 19 2.1.4 Types of applications (R&D versus manufacturing) 20 2.1.5 Organizational considerations 20 2.2 Generalized Process Analytics Work Process 23 2.2.1 Project identification and definition 24 2.2.2 Analytical application development 26 2.2.3 Design, specify and procure 26 2.2.4 Implementation in production 28 2.2.5 Routine operation 29 2.2.6 Continuous improvement 30 2.3 Considerations for PAT Implementation in the Pharmaceutical Industry 30 2.3.1 Introduction 30 2.3.2 Business model 30 2.3.3 Technical differences 31 2.3.4 Regulatory Aspects of Process Analytics in the Pharmaceutical Industry –the Concept of Quality by Design 33 2.4 Conclusions 36 References 36 3 Process Sampling: Theory of Sampling – the Missing Link in Process Analytical Technologies (PAT) 37 Kim H. Esbensen and Peter Paasch-Mortensen 3.1 Introduction 37 3.2 Theory of Sampling – Introduction 39 3.2.1 Heterogeneity 41 3.2.2 Constitutional heterogeneity 41 3.2.3 Distributional heterogeneity 42 3.2.4 Structurally correct sampling 45 3.2.5 Incorrect sampling error 45 3.2.6 Increment delimitation error 45 3.2.7 Increment extraction error 46 3.2.8 Increment preparation error 46 3.2.9 Increment weighing error 47 3.2.10 Total sampling error 48 3.2.11 Global estimation error 48 3.3 Mass Reduction as a Specific Sampling Procedure 48 3.4 Fundamental Sampling Principle 51 3.5 Sampling – a Very Practical Issue 51 3.5.1 Sampling unit operations 52 3.5.2 Understanding process sampling: 0-D versus 1-D LOTS 52 3.5.3 Grab sampling – 0-D and 1-D 54 3.5.4 Correct process sampling: increment delimitation/extraction 56 3.5.5 PAT versus correct process sampling – what is required? 58 3.6 Reactors and Vessels – Identical Process Sampling Issues 60 3.6.1 Correct process sampling with existing process technology 62 3.6.2 Upward flux – representative colocated PAT sampling 62 3.6.3 Upstream colocated PAT sampler 64 3.7 Heterogeneity Characterization of 1-D lots: Variography 66 3.7.1 Process sampling modes 67 3.7.2 The experimental variogram 67 3.7.3 Sampling plan simulation and estimation of TSE 71 3.7.4 TSE estimation for 0-D lots – batch sampling 72 3.7.5 Corporate QC benefits of variographic analysis 73 3.8 Data Quality – New Insight from the TOS 75 3.9 Validation in Chemometrics and PAT 76 3.10 Summary 78 References 79 4 UV-visible Spectroscopy for On-line Analysis 81 Marcel A. Liauw, Lewis C. Baylor and Patrick E. O’Rourke 4.1 Introduction 81 4.2 Theory 82 4.2.1 Chemical concentration 82 4.2.2 Color 84 4.2.3 Film thickness 85 4.2.4 Turbidity 85 4.2.5 Plasmons/nanoparticles 85 4.3 Instrumentation 85 4.4 Sample Interface 86 4.4.1 Cuvette/vial 87 4.4.2 Flow cells 87 4.4.3 Insertion probe 87 4.4.4 Reflectance probe 89 4.5 Implementation 89 4.5.1 A complete process analyzer 89 4.5.2 Troubleshooting 89 4.6 Applications 91 4.6.1 Gas and vapor analysis 92 4.6.2 Liquid analysis 92 4.6.3 Solid analysis 96 4.6.4 Other applications 99 4.7 Detailed Application Notes 100 4.7.1 Gas and vapor analysis: toluene 100 4.7.2 Liquid analysis: breakthrough curves 101 4.7.3 Solids analysis: extruded plastic color 101 4.7.4 Film thickness determination: polymer 103 4.8 Conclusion 104 References 104 5 Near-infrared Spectroscopy for Process Analytical Technology: Theory, Technology and Implementation 107 Michael B. Simpson 5.1 Introduction 107 5.2 Theory of Near-infrared Spectroscopy 112 5.3 Analyser Technologies in the Near-infrared 114 5.3.1 Light sources and detectors for near-infrared analyzers 114 5.3.2 The scanning grating monochromator and polychromator diode-array 119 5.3.3 The acousto-optic tunable filter (AOTF) analyzer 123 5.3.4 Fourier transform near-infrared analyzers 127 5.3.5 Emerging technologies in process NIR analyzers 134 5.4 The Sampling Interface 136 5.4.1 Introduction 136 5.4.2 Problem samples: liquids, slurries and solids 142 5.4.3 The use of fiber optics 145 5.5 Practical Examples of Near-infrared Analytical Applications 147 5.5.1 Refinery hydrocarbon streams 148 5.5.2 Polyols, ethoxylated derivatives, ethylene oxide/propylene oxide polyether polyols 149 5.5.3 Oleochemicals, fatty acids, fatty amines and biodiesel 151 5.6 Conclusion 152 References 153 6 Infrared Spectroscopy for Process Analytical Applications 157 John P. Coates 6.1 Introduction 157 6.2 Practical Aspects of IR Spectroscopy 161 6.3 Instrumentation Design and Technology 163 6.4 Process IR Instrumentation 166 6.4.1 Commercially available IR instruments 167 6.4.2 Important IR component technologies 172 6.4.3 New technologies for IR components and instruments 176 6.4.4 Requirements for process infrared analyzers 178 6.4.5 Sample handling for IR process analyzers 185 6.4.6 Issues for consideration in the implementation of process IR 187 6.5 Applications of Process IR Analyzers 189 6.6 Process IR Analyzers: a Review 191 6.7 Trends and Directions 192 References 193 7 Raman Spectroscopy 195 Nancy L. Jestel 7.1 Attractive Features of Raman Spectroscopy 195 7.1.1 Quantitative information 195 7.1.2 Flexible sample forms and sizes used as accessed without damage 196 7.1.3 Flexible sample interfaces 196 7.1.4 Attractive spectral properties and advantageous selection rules 197 7.1.5 High sampling rate 197 7.1.6 Stable and robust equipment 198 7.2 Potential Issues with Raman Spectroscopy 198 7.2.1 High background signals 198 7.2.2 Stability 198 7.2.3 Too much and still too little sensitivity 199 7.2.4 Personnel experience 199 7.2.5 Cost 200 7.3 Fundamentals of Raman Spectroscopy 200 7.4 Raman Instrumentation 203 7.4.1 Safety 203 7.4.2 Laser wavelength selection 204 7.4.3 Laser power and stability 204 7.4.4 Spectrometer 205 7.4.5 Sample interface (probes) 206 7.4.6 Communications 208 7.4.7 Maintenance 209 7.5 Quantitative Raman 209 7.6 Applications 212 7.6.1 Acylation, alkylation, catalytic cracking, and transesterification 213 7.6.2 Bioreactors 213 7.6.3 Blending 214 7.6.4 Calcination 214 7.6.5 Catalysis 215 7.6.6 Chlorination 216 7.6.7 Counterfeit pharmaceuticals 217 7.6.8 Extrusion 218 7.6.9 Forensics 218 7.6.10 Hydrogenation 218 7.6.11 Hydrolysis 219 7.6.12 Medical diagnostics 219 7.6.13 Microwave-assisted organic synthesis 219 7.6.14 Mobile or field uses 220 7.6.15 Natural products 220 7.6.16 Orientation, stress, or strain 221 7.6.17 Ozonolysis 222 7.6.18 Polymerization 222 7.6.19 Polymer curing 224 7.6.20 Polymorphs (crystal forms) 225 7.6.21 Product properties 228 7.6.22 Purification: distillation, filtration, drying 229 7.6.23 Thin films or coatings 229 7.7 Current State of Process Raman Spectroscopy 230 References 231 8 Near-infrared Chemical Imaging for Product and Process Understanding 245 E. Neil Lewis, Joseph W. Schoppelrei, Lisa Makein, Linda H. Kidder and Eunah Lee 8.1 The PAT Initiative 245 8.2 The Role of Near-infrared Chemical Imaging (NIR-CI) in the Pharmaceutical Industry 246 8.2.1 Characterization of solid dosage forms 246 8.2.2 ‘A picture is worth a thousand words’ 247 8.3 Evolution of NIR Imaging Instrumentation 247 8.3.1 Spatially resolved spectroscopy – mapping 247 8.3.2 The infrared focal-plane array 247 8.3.3 Wavelength selection 248 8.3.4 The benefits of NIR spectroscopy 248 8.3.5 NIR imaging instrumentation 249 8.4 Chemical Imaging Principles 251 8.4.1 The hypercube 251 8.4.2 Data analysis 251 8.4.3 Spectral correction 252 8.4.4 Spectral preprocessing 253 8.4.5 Classification 253 8.4.6 Image processing – statistical 255 8.4.7 Image processing – morphology 257 8.5 PAT Applications 257 8.5.1 Content uniformity measurements – ‘self calibrating’ 258 8.5.2 Quality assurance – imaging an intact blister pack 260 8.5.3 Contaminant detection 261 8.5.4 Imaging of coatings – advanced design delivery systems 263 8.6 Processing Case Study: Estimating ‘Abundance’ of Sample Components 267 8.6.1 Experimental 268 8.6.2 Spectral correction and preprocessing 268 8.6.3 Analysis 268 8.6.4 Conclusions 273 8.7 Processing Case Study: Determining Blend Homogeneity Through Statistical Analysis 273 8.7.1 Experimental 273 8.7.2 Observing visual contrast in the image 274 8.7.3 Statistical analysis of the image 274 8.7.4 Blend uniformity measurement 276 8.7.5 Conclusions 276 8.8 Final Thoughts 277 Acknowledgements 278 References 278 9 Acoustic Chemometric Monitoring of Industrial Production Processes 281 Maths Halstensen and Kim H. Esbensen 9.1 What is Acoustic Chemometrics? 281 9.2 How Acoustic Chemometrics Works 282 9.2.1 Acoustic sensors 282 9.2.2 Mounting acoustic sensors (accelerometers) 283 9.2.3 Signal processing 284 9.2.4 Chemometric data analysis 284 9.2.5 Acoustic chemometrics as a PAT tool 284 9.3 Industrial Production Process Monitoring 285 9.3.1 Fluidized bed granulation monitoring 285 9.3.2 Pilot scale studies 286 9.3.3 Monitoring of a start-up sequence of a continuous fluidized bed granulator 291 9.3.4 Process monitoring as an early warning of critical shutdown situations 295 9.3.5 Acoustic chemometrics for fluid flow quantification 296 9.4 Available On-line Acoustic Chemometric Equipment 299 9.5 Discussion 301 9.5.1 Granulator monitoring 301 9.5.2 Process state monitoring 301 9.5.3 Ammonia concentration monitoring 301 9.6 Conclusions 302 References 302 10 Process NMR Spectroscopy: Technology and On-line Applications 303 John C. Edwards and Paul J. Giammatteo 10.1 Introduction 303 10.2 NMR Spectroscopy Overview 305 10.2.1 The NMR phenomenon 305 10.2.2 Time–domain-NMR: utilization of the FID and spin relaxation 309 10.2.3 High-resolution NMR: obtaining a spectrum with resolved chemical shift information 312 10.3 Process NMR Instrumentation 313 10.3.1 Spectrometer and magnet design 313 10.3.2 Sampling and experimental design 316 10.4 Postprocessing Methodologies for NMR Data 317 10.5 Advantages and Limitations of NMR as a Process Analytical Technology 320 10.5.1 Advantages 320 10.5.2 Limitations 321 10.6 On-line and At-line Applications 321 10.6.1 Time–domain NMR 322 10.6.2 High-resolution NMR: chemometric applications 323 10.7 Current Development and Applications 330 10.8 Conclusions 331 References 332 11 Fluorescent Sensing and Process Analytical Applications 337 Jason E. Dickens 11.1 Introduction 337 11.2 Luminescence Fundamentals 338 11.2.1 Luminescence nomenclature 338 11.2.2 Luminescence processes 338 11.2.3 Fluorophore classification 338 11.3 LIF Sensing Fundamentals 341 11.3.1 LIF sensing classification 341 11.3.2 Luminescence spectroscopy 342 11.3.3 LIF signal response function 343 11.4 LIF Sensing Instrumentation 343 11.4.1 LIF photometric instrument specification 345 11.4.2 LIF Instrument selection 347 11.5 Luminescent Detection Risks 347 11.6 Process Analytical Technology Applications 348 11.6.1 Petrochemical, chemical and nuclear field applications 349 11.6.2 Pharmaceutical PAT applications 349 11.7 Conclusions 350 References 351 12 Chemometrics in Process Analytical Technology (PAT) 353 Charles E. Miller 12.1 Introduction 353 12.1.1 What is chemometrics? 353 12.1.2 Some history 354 12.1.3 Some philosophy 355 12.1.4 Chemometrics in analytical chemistry? 355 12.1.5 Chemometrics in process analytical chemistry? 356 12.2 Foundations of Chemometrics 356 12.2.1 Notation 356 12.2.2 Some basic statistics 358 12.2.3 Linear regression 359 12.2.4 Multiple linear regression 361 12.2.5 Principal components analysis (PCA) 362 12.2.6 Design of experiments (DOE) 366 12.3 Chemometric Methods in PAT 368 12.3.1 Data preprocessing 369 12.3.2 Quantitative model building 377 12.3.3 Qualitative model building 389 12.3.4 Exploratory analysis 397 12.4 Overfitting and Model Validation 407 12.4.1 Overfitting and underfitting 407 12.4.2 Test set validation 408 12.4.3 Cross validation 410 12.5 Outliers 413 12.5.1 Introduction to outliers 413 12.5.2 Outlier detection and remediation 413 12.6 Calibration Strategies in PAT 416 12.6.1 The ‘calibration strategy space’ 417 12.6.2 Strategies for direct versus inverse modeling methods 418 12.6.3 Hybrid strategies 419 12.7 Sample and Variable Selection in Chemometrics 420 12.7.1 Sample selection 420 12.7.2 Variable selection 421 12.8 Troubleshooting/Improving an Existing Method 425 12.8.1 Method assessment 425 12.8.2 Model improvement strategies 425 12.9 Calibration Transfer and Instrument Standardization 426 12.9.1 Slope/intercept adjustment 428 12.9.2 Piecewise direct standardization (PDS) 428 12.9.3 Generalized least squares (GLS) weighting 429 12.9.4 Shenk–Westerhaus method 429 12.9.5 Other transfer/standardization methods 429 12.10 Chemometric Model Deployment Issues in PAT 430 12.10.1 Outliers in prediction 430 12.10.2 Deployment software 432 12.10.3 Data systems, and control system integration 432 12.10.4 Method updating 433 12.11 People Issues 433 12.12 The Final Word 434 References 434 13 On-line PAT Applications of Spectroscopy in the Pharmaceutical Industry 439 Brandye Smith-Goettler 13.1 Background 439 13.2 Reaction Monitoring 441 13.3 Crystallization 442 13.4 API Drying 443 13.5 Nanomilling 444 13.6 Hot-melt Extrusion 445 13.7 Granulation 446 13.7.1 Wet granulation 446 13.7.2 Roller compaction 449 13.8 Powder Blending 450 13.8.1 Lubrication 451 13.8.2 Powder flow 451 13.9 Compression 452 13.10 Coating 452 13.11 Biologics 453 13.11.1 Fermentation 453 13.11.2 Freeze-drying 454 13.12 Cleaning Validation 454 13.13 Conclusions 455 References 455 14 NIR spectroscopy in Pharmaceutical Analysis: Off-line and At-line PAT Applications 463 Marcelo Blanco Romía and Manel Alcalá Bernárdez 14.1 Introduction 463 14.1.1 Operational procedures 464 14.1.2 Instrument qualification 466 14.2 Foundation of Qualitative Method Development 466 14.2.1 Pattern recognition methods 467 14.2.2 Construction of spectral libraries 468 14.2.3 Identification and qualification 470 14.3 Foundation of Quantitative Method Development 471 14.3.1 Selection and preparation of samples 472 14.3.2 Preparation and selection of samples 473 14.3.3 Determination of reference values 474 14.3.4 Acquisition of spectra 474 14.3.5 Construction of the calibration model 475 14.3.6 Model validation 476 14.3.7 Prediction of new samples 476 14.4 Method Validation 476 14.5 Calibration Transfer 476 14.6 Pharmaceutical Applications 478 14.6.1 Identification of raw materials 478 14.6.2 Homogeneity 478 14.6.3 Moisture 480 14.6.4 Determination of physical parameters 481 14.6.5 Determination of chemical composition 483 14.7 Conclusions 485 References 486 15 Near-infrared Spectroscopy (NIR) as a PAT Tool in the Chemical Industry: Added Value and Implementation Challenges 493 Ann M. Brearley and Susan J. Foulk 15.1 Introduction 493 15.2 Successful Process Analyzer Implementation 494 15.2.1 A process for successful process analyzer implementation 494 15.2.2 How NIR process analyzers contribute to business value 497 15.2.3 Issues to consider in setting technical requirements for a process analyzer 498 15.2.4 Capabilities and limitations of NIR 499 15.2.5 General challenges in process analyzer implementation 500 15.2.6 Approaches to calibrating an NIR analyzer on-line 502 15.2.7 Special challenges in NIR monitoring of polymer melts 505 15.3 Example Applications 506 15.3.1 Monitoring monomer conversion during emulsion polymerization 506 15.3.2 Monitoring a diethylbenzene isomer separation process 508 15.3.3 Monitoring the composition of copolymers and polymer blends in an extruder 509 15.3.4 Rapid identification of carpet face fiber 512 15.3.5 Monitoring the composition of spinning solution 514 15.3.6 Monitoring end groups and viscosity in polyester melts 516 15.3.7 In-line monitoring of a copolymerization reaction 518 References 520 16 Future Trends for PAT for Increased Process Understanding and Growing Applications in Biomanufacturing 521 Katherine A. Bakeev and Jose C. Menezes 16.1 Introduction 521 16.2 Regulatory Guidance and its Impact on PAT 522 16.3 Going Beyond Process Analyzers Towards Solutions 524 16.3.1 Design of experiments for risk-based analysis 526 16.3.2 Sample and process fingerprinting with PAT tools 527 16.3.3 Design and Control Spaces 528 16.3.4 Chemometrics and process analysis 528 16.4 Emerging Application Areas of PAT 529 16.4.1 Biofuels 529 16.4.2 Biomanufacturing 530 16.5 New and Emerging Sensor and Control Technologies 531 16.5.1 Terahertz spectroscopy 531 16.5.2 Integrated sensing and processing 532 16.5.3 Dielectric spectroscopy 533 16.5.4 Process chromatography 533 16.5.5 Mass spectrometry 534 16.5.6 Microwave resonance 534 16.5.7 Novel sensors 535 16.5.8 Inferential sensors 536 16.6 Advances in Sampling: NeSSI 537 16.7 Challenges Ahead 537 16.7.1 Continuous process validation 538 16.7.2 Data challenges: data handling and fusion 539 16.7.3 Regulatory challenges 539 16.7.4 Enterprise systems for managing data 539 16.8 Conclusion 540 References 540 Index 545

About the Author :
Katherine A. Bakeev is Principal Scientist with GlaxoSmithKline in King of Prussia, PA where she works in the Process Analytics and Chemometrics group supporting chemical development. She has twelve years of industrial experience including work in process analytical chemistry with ISP, and as a product specialist for Foss NIRSystems. She holds a PhD in Polymer Science and Engineering from the University of Massachusetts in Amherst, and a Masters in Technology Management from Stevens Institute of Technology. She has given numerous presentations on the use of near-infrared spectroscopy (NIR), and in 2007 received the Coblentz Society Craver Award for her work in NIR and chemometrics.

Review :
"Overall, this excellent compilation is highly recommended." (Organic Process Research and Development, January 2011)