About the Book
Sampling is an important, but often invisible, part of everyday life in science, technology, industry, society and commerce where decisions are made based on analytical results, which must be based on reliable samples. However, there is a very long and complex pathway from heterogeneous materials in "lots" such as satchels, bags, drums, vessels, truck loads, railroad cars, shiploads, stockpiles (in the kg-ton range) to the miniscule laboratory aliquot (in the g-g range), which is what is actually analysed. Exactly how to acquire a documented, representative analytical result across mass-reduction of up to six orders of magnitude is far from a direct materials handling issue. There are specific principles and rules behind representativity. TOS to the fore!
This book presents the Theory and Practice of Sampling (TOS) starting from level zero in a novel didactic framework without excessive mathematics and statistics. It represents 20 years of teaching experience which has developed into a unique conceptual framework with which the TOS' six principles and four unit operations can be understood in a unifying manner, enabling the reader to start sampling in a correct fashion right away. The book covers sampling from stationary lots, from moving, dynamic lots (process sampling) and has a vital focus on sampling in the analytical laboratory. It contains a wealth of complementing cases, examples and references (most of which are accessible on-line) meant to inspire and motivate the reader to individual skills- building and further self-study. The book has been assessed and reviewed extensively.
Table of Contents:
Foreword xi
1 Theory of Sampling (TOS)-the missing link before analysis 1
1.1 A framework for representative sampling 5
1.2 What comes before analysis-the TOS! 7
1.3 What this book promises... 9
1.4 References 10
2 Theory of Sampling (TOS)-fundamental definitions and concepts 11
2.1 Lot dimensionality 12
2.2 Sampling terminology-the tower of Babel 15
2.3 References 19
3 Heterogeneity-the root of all evil (part 1) 21
3.1 Introduction to the concept of sampling errors (excerpt from DS3077) 22
3.2 Heterogeneity-the basics 25
3.3 Materials, sampling targets and lots 25
3.4 Homogeneity-heterogeneity 28
3.5 Scale 30
3.6 Heterogeneity vs sampling 31
3.7 References 32
4 Heterogeneity-the root of all evil (part 2) 33
4.1 Introduction 33
4.2 Constitutional Heterogeneity (CH) 36
4.3 Distributional Heterogeneity (DH) 37
4.4 Heterogeneity vs practical sampling 41
4.5 "Structured heterogeneity" 46
4.6 The fundamental insight on how to counteract heterogeneity 48
4.7 References 49
5 "Sampling-is not gambling" 51
5.1 Introduction 51
5.2 Enough analogy 52
6 Pierre Gy's key concept of sampling errors 57
6.1 Rational understanding of heterogeneity and appropriate sampling 57
6.2 Although complex, there is hope 62
6.3 How to sample representatively: the TOS 64
6.4 References 66
7 Composite sampling I: the Fundamental Sampling Principle 71
7.1 WHAT TO DO with all this heterogeneity? 71
8 Composite sampling II: lot dimensionality transformation 83
8.1 1-D lots: conveniently elongated lots 83
8.2 Process sampling 85
8.3 Process sampling generalised 86
8.4 Q 87
8.5 Lot dimensionality transformation (LDT) 91
8.6 References 91
9 Sampling quality assessment: the replication experiment 93
9.1 Background 93
9.2 Clarification 95
9.3 Quantifying total empirical variability-the replication experiment 100
9.4 Relative sampling variability 101
9.5 Notes and references 109
10 Sampling quality criteria 111
10.1 Sampling quality criteria 111
10.2 First SQC component-definition of analyte(s) 112
10.3 Second SQC component-delineating the decision unit (DU) 112
10.4 Third SQC component-inference and confidence 113
10.5 Perspectives 115
10.6 Summary 117
10.7 References 117
11 There are standards-and there is the standard 119
11.1 First light 119
11.2 Analysis of sampling standards for solid biofuels 121
11.3 Analysis of grain sampling guide 123
11.4 Sampling for GMO risk assessment 125
11.5 Examples of too glib recommendations 125
11.6 TOS competence is crucial 127
11.7 Que faire? 129
11.8 DS 3077 Horizontal-a new standard for representative sampling. Design, history and acknowledgements 129
11.9 Chapter references 138
12 Spear sampling: a bane at all scales 141
12.1 Introduction 141
12.2 Spear sampling-at all scales 143
12.3 Not always bad-there is hope 147
12.4 Conclusions 148
12.5 References 150
13 Into the laboratory... the TOS still reigns supreme 151
13.1 Representative sampling-a scale invariant endeavour 151
13.2 Size does not matter-only heterogeneity, and how to counteract it 155
13.3 And there is more to be done in the lab ... 156
13.4 Further reading 159
14 Representative mass reduction in the laboratory: riffle splitting galore 161
14.1 Introduction 161
14.2 Riffle splitting 162
14.3 Automation-enter the rotary divider 169
14.4 Benchmark study 171
14.5 The ultimate method/equipment ranking for the laboratory 173
14.6 Conclusions 175
14.7 References 176
15 Introduction to process sampling 179
15.1 Lot dimensionality: ease of practical sampling 179
15.2 Lot dimensionality transformation 183
15.3 Process sampling 184
15.4 1-D lot heterogeneity 187
15.5 Variographic analysis: a first brief 188
15.6 Interpretation of variograms 189
15.7 References 194
16 Process sampling: the importance of correct increment extraction 195
16.1 Moving, or static, 1-D lots: increment cutting must be TOS-correct 195
16.2 "Sooner or later"... 200
16.3 References 201
17 The variographic experiment 203
17.1 The variogram 204
17.2 References 211
18 Experimental validation of a primary sampling system for iron ore pellets 215
18.1 Introduction: status of current ISO standards 215
18.2 Fundamental Sampling Principle and basic requirements for iron ore sampling systems 216
18.3 Principles and general requirements for checking sampling bias 218
18.4 Validation experiment 220
18.5 Experimental results 220
18.6 Discussion 222
18.7 References 223
19 Industrial variographic analysis for continuous sampling system validation 225
19.1 Variographic analysis 225
19.2 Continuous control of sampling systems 225
19.3 9-12.5 mm size fraction of iron ore pellets 226
19.4 Specific surface area of magnetite slurry 228
19.5 Iron grade in magnetite slurry 231
19.6 Conclusions 232
19.7 Acknowledgements 233
19.8 References 233
20 Theory of Sampling (TOS): pro et contra 235
20.1 A powerful case for the TOS in trade and commerce 236
20.2 Cases against the TOS (science, technology, commerce, trade) 240
20.3 Important reading with which to catch the attention of newcomers to the TOS 245
21 Following the TOS will save you a lot of money (pun intended) 247
21.1 Case 1: Always mind analysis 248
21.2 Case 2: Saving a client from a wrong, expensive investment 250
21.3 Case 3: The hidden costs-profit gained by using the TOS 254
21.4 Case 4: The cost of assuming standard normality for serial data 255
21.5 Lessons learned 259
21.6 References 260
22 A tale of two laboratories I: the challenge 263
22.1 Introduction (scientific, technological) 264
22.2 There is analysis... and there is analysis+ 265
22.3 The core issue 267
22.4 The crux of the matter 268
22.5 The complete argument 271
22.6 The meaning of it all 272
22.7 Inside and outside the complacent four walls of the analytical laboratory 275
22.8 "One fine day"... 276
22.9 The really important aspect: costs or gains 277
22.10 What in the world? 278
22.11 References 279
23 A tale of two laboratories II: resolution 281
23.1 Epiphany interpretation I: knowingly closing one's eyes or not? 281
23.2 Epiphany interpretation II: the economic dilemma 283
23.3 Epiphany interpretation III: the moral resolution 285
23.4 Laboratory B's new vision and mission 287
23.5 Can this really lead to increased commercial success? 288
23.6 Acknowledgements 289
23.7 References 289
24 Sampling commitment-and what it takes... 291
24.1 Historical context 291
24.2 Awareness 291
24.3 Minimum competence level 292
24.4 Vade mecum 294
24.5 Trouble with some standards 296
24.6 In practice... 297
24.7 What could be argument(s) against ... 298
24.8 Practice, practice, practice... 299
24.9 The last word 302
24.10 References 302
24.11 Further reading (a first selection) 304
25 Representative sampling and society 307
25.1 Sampling: from the point of view of buyers, consumers, citizens 307
25.2 The way forward: some proposals 311
25.3 Beyond traditional application fields 313
25.4 Conclusions 316
25.5 References 317
26 Epilogue: what's next? 319
About the Author :
Kim H. Esbensen, PhD, Dr (hon), research professor in Geoscience Data Analysis and Sampling at GEUS, the National Geological Surveys of Denmark and Greenland (2010-2015), chemometrics & sampling professor at Aalborg University, Denmark (2001-2015), professor (Process Analytical Technologies) at Telemark Institute of Technology, Norway (1990-2000 and 2010-2015) and professeur associe, Universite du Quebec a Chicoutimi (2013-2016). From 2015 he phased out a more than 35-year academic career for a quest as an independent consultant: www.kheconsult.com. However, as he could not terminate his love for teaching, he is still on a roll as an international visiting-, guest- and affiliate professor around the world.
A geologist/geochemist/data analyst by training, he has been working for three decades at the forefront of chemometrics, but since 2000 he has devoted most of his scientific R&D to the theme of representative sampling of heterogeneous materials, processes and systems (Theory of Sampling, TOS), and PAT (Process Analytical Technology). He is a member of several scientific societies, has published over 250 peer-reviewed papers and is the author of a widely used textbook, Multivariate Data Analysis (35,000 copies), published in its 6th edition in 2018. He was the originator and chairman of the taskforce behind the world's first horizontal (matrix-independent) sampling standard DS 3077 (2013). He is editor of the science magazine TOS forum (https://www.impopen.com/tos-forum) and for the Sampling Column in Spectroscopy Europe/Asia (https://www.spectroscopyeurope.com/sampling).
Esbensen is fond of the right kind of friends and dogs, swinging jazz, fine cuisine, good wine, contemporary art and classical music. His has been collecting science fiction novels for more decades than he is comfortable contemplating, still, as ever, it's all in the future...
