Raman Spectroscopy of Biological Pigments Solid Mixed Matrices in MALDI/TOF-MS Solutions for Innovation

Vol. 28 No. 1 February/March 2016

Informing European spectroscopists for over 40 years Raman spectroscopy of biological pigments Solid mixed matrices in MALDI/TOF-MS Solutions for Innovation

UltraCOOL Probe · For maximum sensitivity · 800 and 600MHz - 5mm C-H and H-X type · 13C : > x5 higher sensitivity · 1H: > x4 higher sensitivity

JEOL - NMR Probes Advanced Technology and Performance

The SuperCOOL probe · Low cost nitrogen cooled probe · 400MHz 5mm tunable type · 13C : x3 higher sensitivity · 1H: > x3 higher sensitivity

www.jeol.com

AID_probes_a4v1.indd 1 14/04/2014 10:20 EDITORIAL Connecting Global Competence F As you will have noticed from this issue’s cover, we are making a 3 colourful start to 2016. In the first article on “The analytical niche for Raman spectroscopy in biological pigment research”, Daniel U2 Thomas and Cushla McGoverin suggest that Raman spectroscopy may have a particularly valuable role in pigment biology TU research. Pigments are almost universal in biology and are the basis of much of what we find attractive in flowers, birds and sea life, such as the fan corals on the cover. The authors show how NE3 RE4 Raman spectroscopy can be used to quickly confirm the presence of a pigment as well as providing more detailed knowledge TW about unknown pigments. The bio theme moves to mass spectrometry in “Solid mixed matrices and their advantages in matrix-assisted laser desorption/ O4 ionisation time-of-flight mass spectrometry” by Marek Šebela. Getting the most from various matrices for use in matrix-assisted GL RK laser desorption/ionisation (MALDI) has always been a bit of an art, and the introduction of mixed matrices increases the number of possible combinations but may improve reproducibility and so 03 simplify analysis in the end. The author describes mixed matrices for a range of samples including proteins, peptides, oligo - B saccharides, oligonucleotides, lipids, polymers and even intact 2 microbial cells! In the Tony Davies Column, vast amounts of data and how you AL2 handle them are investigated by Tony, and Shane Ellis, Benjamin Balluff and Ron Heeren from the Maastricht MultiModal Molecular Imaging Institute. “Spectroscopic data handling at petabyte scale” shows how one institute is dealing with truly huge amounts of data, both in its collection and in its distribution to Everything scientists for interpretation and analysis. At the same time, the related to biotechnology institute has been able to incorporate best practice around the in one hall FAIR Data Stewardship of scientific information. Peter Jenks looks back to the BERM 14 conference on biological and environmental reference materials in the Quality Matters Elementary Column. The next conference in the series returns to Europe: Berlin in June 2018. to your success. In the Sampling Column, Kim Esbensen and Claas Wagner continue our education about representative sampling. In The world’s largest network of trade fairs for laboratory technology, instrumental analysis and “Sampling quality assessment: the replication experiment”, they biotechnology features the complete range provide an overview of the issue of replication, which may not of products and services for all of your laboratory be as straightforward as might be expected at first. needs—in industry and research. The scientific We also have a Product Focus on Atomic Spectroscopy, many highlight is the analytica conference, where the international elite discusses the latest findings New Products, even before the tsunami that one can expect in biochemistry and laboratory medicine. from Pittcon next month, news of literature and nearly three Monacofiere, Tel. +39 02 4070 8301, pages of upcoming conferences, courses and exhibitions. [email protected]

May 10–13, 2016 Messe München 25th International Trade Fair for Laboratory Technology, Analysis, Biotechnology and analytica conference www.analytica.de

www.spectroscopyeurope.com

ana16_Biotech_70x297_SPECTROSCOPY_E-ITA.indd 1 18.12.15 15:39 Vol. 28 No. 1 February 2016 ISSN 0966-0941

Informing European spectroscopists for over 40 years Raman spectroscopy of biological pigments Solid mixed matrices in MALDI/TOF-MS CONTENTS

3 Editorial 6 The analytical niche for Raman spectroscopy in biological pigment research Daniel Thomas and Cushla McGoverin believe Daniel B. Thomas and Cushla M. McGoverin that Raman spectroscopy has a niche for the investigation of biological pigments in many organisms including these sea fan corals. Read about this on page 6. 10 Solid mixed matrices and their advantages in matrix-assisted laser desorption/ionisation time-of-flight Publisher mass spectrometry Ian Michael E-mail: [email protected] Marek Šebela

Article Editor John Chalmers 15 Tony Davies Column VSConsulting, Stokesley, UK. E-mail: [email protected] Spectroscopic data handling at petabyte scale

Advertising Sales UK and Ireland 18 Quality Matters Ian Michael BERM 14 retrospective: autumn in Maryland, USA IM Publications, 6 Charlton Mill, Charlton, Chichester, West Sussex PO18 0HY, United Kingdom. Tel: +44-1243-811334, Fax: +44-1243-811711, 20 Sampling Column E-mail: [email protected]. Sampling quality assessment: the replication experiment Americas Joe Tomaszewski John Wiley & Sons Inc. 26 Product Focus Tel: +1-908-514-0776 Atomic Spectroscopy E: [email protected] Europe and 27 New Products the Rest of World Stephen Parkes and Charlotte Redfern John Wiley & Sons Ltd, The Atrium, Southern Gate, 31 Literature Chichester, West Sussex PO19 8SQ, UK, Tel: +44-1243-770367, Fax: +44-1243-770432, E-mail: [email protected] and [email protected] 32 Diary Paid subscriptions Advertisers index Journals Subscriptions Department, John Wiley & 35 Sons, Ltd, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ, UK. Spectroscopy Europe is a controlled circulation journal, published seven times a year and avail- Tel +44-1243 779777, Fax +44-1243 843232, able free-of-charge to qualifying individuals in Europe. Others can subscribe at the rate of €152 e-mail: [email protected] (Europe), £108 (UK), $208 (ROW, postage included) for the seven issues published in 2016. All paid subscription enquiries should be addressed to: Spectroscopy Europe, John Wiley & Sons Ltd, Spectroscopy Europe is a joint publication of Journals Administration Department, The Atrium, Southern Gate, Chichester, West Sussex PO19 John Wiley & Sons Ltd, The Atrium, Southern 8SQ, UK. Gate, Chichester, West Sussex PO19 8SQ, UK, and IM Publications LLP, 6 Charlton Mill, Charlton, Copyright 2016 John Wiley & Sons Ltd and IM Publications LLP Chichester, West Sussex PO18 0HY, UK. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, Vol. 28 No. 1 or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or February/March 2016 otherwise, without the prior permission of the publisher in writing. Printed in the UK by Headley Brothers Ltd, Ashford, Kent.

4 SPECTROSCOPYEUROPE Shimadzu_SpectroEurope_0216:Layout 1 15.12.15 16:33 Seite 1

IRTracer-100 – highly improved sensitivity meets expandability and easy maintenance

It’s a matter of specs

The IRTracer-100 provides quick and easy analysis in a variety of fields. This FTIR spectrophotometer for near, mid and far infrared range is perfect for pharmaceutical, chemical, food and environmental applications as well as for microanalysis (crime, forensics). The IRTracer-100 monitors high speed reactions within a second.

• Outstanding sensitivity • LabSolutions IR software quickly and easily obtains high quality data unifies data handling with other laboratory hardware, e.g. LC and GC from Shimadzu • High-speed scanning of 20 spectra in a second in rapid scan mode • High performance library and search software containing 12,000 spectra and a unique contaminant • Expandability search program through software and hardware options

The Shimadzu IRTracer-100 New Levels of Performance and Quality Created by Excellent Sensitivity, Speed and Resolution

www.shimadzu.eu VOL. 28 NO. 1 (2016) ARTICLE

The analytical niche for Raman spectroscopy in biological pigment research

Daniel B. Thomasa and Cushla M. McGoverinb aInstitute of Natural and Mathematical Sciences, Massey University, Auckland, New Zealand. E-mail: [email protected] bThe Dodd Walls Centre for Photonic and Quantum Technologies, Department of Physics, University of Auckland, Auckland, New Zealand

Introduction ists. Ultraviolet/visible spectroscopy HPLC and UV/vis. The aim of this short Pigments have important roles in the (UV/vis) and high-performance liquid review article is to show how Raman physiologies and ecologies of a diverse chromatography (HPLC) are staples of spectroscopy is being used in biological range of organisms, from unicellu - biochrome research, where the former is pigment research, beginning with studies lar algae to fungi, plants and animals non-destructive and the latter is routinely of carotenoid pigmentation. (see, for example, Figure 1). Biological used for chemical identification. Raman pigmentation is used for communication spectroscopy combines both of these Carotenoids and camouflage, as well as maintain- advantages into a single technique, but The characteristic colours of canaries, ing health, for producing and circulat - is relatively under-utilised in pigment marigolds, carrots and butterflies are all ing respiratory gases, and many other biology research. Raman spectroscopy examples of the yellow through orange functions. Pigmentation research often provides information about functional to red colours conferred by carotenoids. encourages researchers with a back- groups in a molecule and can be used Carotenoids contain 40 carbon atoms ground in biology to become familiar to identify biological pigments. Raman and are divided into two groups on the with analytical instrumentation more spectroscopy has the potential for wide basis of whether oxygen is present or commonly used by dedicated chem - use in pigmentation research alongside not, xanthophylls and carotenes, respec-

Figure 1. Recent studies with Raman spectroscopy have focussed on anthocyanins and carotenoids as well as many other classes of biological pigment. Carotenoid pigments confer the yellow, orange and red hues to the feathers of many bird species, and anthocyanin pigments are responsible for red, pink, blue and purple colours in many flowers. Image credit: Cushla McGoverin and Daniel Thomas.

6 SPECTROSCOPYEUROPE www.spectroscopyeurope.com VOL. 28 NO. 1 (2016) ARTICLE

tively. Carotenoids are composed of a Unsubstituted linear hydes (polyenals), pigment compounds conjugated double bond chain which polyacetylenes for which the structure was first predicted may or may not be terminated by ring Bright biochromes are also common in with Raman spectroscopy.5 As with carot- structures. These shared structural traits marine environments, e.g. the red skele- enoid molecules, the conjugated carbon result in three carotenoid characteris - ton of Corallium sea fan corals. Sea fans backbones of psittacofulvins give rise to tic peaks within Raman spectra: C=C and many other octocorals (i.e. corals strong C=C (e.g. 1500–1520 cm–1) and stretching at 1500–1535 cm–1, C–C with eight-fold symmetry) can also have C–C (e.g. 1100–1120 cm–1) peaks in stretching at 1145–1165 cm–1 and C– yellow, pink and purple skeletons, and Raman spectra. The backbones of psitta- –1 CH3 deformation at 1000–1010 cm . these exclusively-marine animals have cofulvins are not methylated as they are

The structural differences between a long history of being harvested for in carotenoids though, and the C–CH 3 carotenoids in the terminating rings, use in jewellery and other ornaments. deformation mode (e.g. 1000 cm–1) that polyene side chains and polyene length Despite the centuries of tradable value, helps identify a carotenoid is absent in alter the Raman shift of these three the pigmentary basis for octocoral skel- spectra from parrot feathers (see Figure dominant peaks, and introduce smaller etons has remained largely unknown. 2). In contrast, spectra from octocor- peaks which are carotenoid specific. It Relatively recent studies of pigmented als have a minor peak around 1004– –1 is therefore possible using these peaks octocoral skeletons with Raman spec- 1020 cm attributed to C–CH3 rocking. and their positions thereof to use Raman troscopy sought an end to the mystery Although a fully resolved structure for the spectroscopy for the identification of and recovered spectra with major peaks octocoral pigment is yet to be reported, –1 carotenoids. This approach was recently around 1100–1120 cm and 1500– the presence of a C–CH3 rocking peak used to non-destructively characterise 1520 cm–1, but the spectra did not have in the former is strong evidence that the most abundant carotenoid in the a strong peak at 1000 cm–1. These spec- octocorals and parrot feathers have feathers of 36 bird species.1 However, tra suggested that octocoral pigments distinct pigments. Raman spectroscopy these spectra were all collected from a may be “psittacofulvins”, a group of has shown that octocoral and parrot similar matrix, keratin of the bird feather, pigments that had previously only been pigments have similar architecture, an and therefore matrix effects did not reported from parrot feathers.3,4 unsubstituted linear polyacetylene back- introduce a significant source of spectral Parrots are unusual among birds in bone,4 but that “psittacofulvins” are still a variance that would perturb carotenoid that the red, orange and yellow colours parrot-specific plumage pigment. characterisation when comparing the of their plumage are not the result of feather spectra. Raman spectra of carot- carotenoid pigmentation. Instead, parrots Tetrapyrroles enoids are environment dependent; achieve brightly coloured plumage with Birds achieve vivid colouration with a therefore, spectra from biological tissues a group of polyunsaturated linear alde- suite of chemically-distinct pigments, should not be compared to extracted carotenoid spectra for the purposes of identifying carotenoids. For exam - ple, mango tissue (Mangifera indica) contains a single form of carotenoid, b-carotene; the C=C stretching mode peak in the Raman spectrum of mango tissue occurs at 1529 cm–1, however, in the Raman spectrum recorded from the extracted β-carotene this peak is at 1515 cm –1.2 The differences between extract and matrix-bound carotenoid spectra is, in part, a consequence of shape changes induced by the matrix. Changes in the planarity of the polyene chain result in peak shifting, particularly the C=C stretching band. In addition, the matrix may provide environments of different hydrophobicity, which will affect the extent to which the highly Figure 2. Raman spectrum from a Purple-breasted Cotinga ( ) feather pigmented hydrophobic carotenoids interact with Cotinga cotinga with carotenoids (upper spectrum) and from a Dusky Parrot (Pionus fuscus) feather pigmented the surrounding matrix, e.g. the red with polyenal “psittacofulvin” pigments (lower spectrum). The peak around 1000 cm–1 in the carotenoid lycopene is present in toma- upper spectrum identifies C–CH3 deformation and represents a clear difference between the two toes as microcrystalline aggregates due spectra. We can therefore conclude that the conjugated backbone of the polyenal “psittacofulvin” to the aqueous nature of the matrix. pigment is not methylated. www.spectroscopyeurope.com SPECTROSCOPYEUROPE 7 VOL. 28 NO. 1 (2016) ARTICLE

including the tetrapyrrole pigments that Moa are extinct ostrich-like birds that requires no specialised sample prepara- colour the surfaces of eggshells. Brown inhabited New Zealand until the 15 th tion and analyses are rapid (i.e. speci- eggshells are coloured with protopor- century. Eggshell fragments from upland men handling time is short). Known phyrin IX (e.g. chicken eggs), the precur- moa (Megalapteryx didinus) are olive- pigments in biological tissues can be sor macrocycle for haeme, and the green or faint blue and suggest that quickly identified (e.g. carotenoids in classic “eggshell blue” and blue-green biological pigments can be preserved feathers), and there is the potential for shell colours are conferred with biliver- in eggshells over extended time scales. new pigments to be discovered. For din, a linear tetrapyrrole pigment broken Ancient eggshell fragments with pigmen- example, an unusual yellow feather down from haeme. Blue-green and tation are rare and destructive analyses pigment was recently identified in the brown pigments help to camouflage are not always viable. Instead, faint blue feathers of some penguin species.8 The eggs and can aid parents in distinguish- moa eggshell fragments were analysed solubility, light absorption and fluores- ing their own eggs from an egg laid by with non-destructive Raman spectros- cence properties of the penguin pigment a brood parasite (e.g. cuckoo). While copy, confirming the presence of biliver- were revealed to be substantially differ- most pigmented eggshells are brown, din.7 Beyond presence or absence data, ent from any known plumage pigment. blue-green or a mixture thereof, some information about pigment concentration More information about the chemical comparatively rare eggshells are pink, could provide valuable insight into the identity of the penguin pigment was orange or bright green.6 The bright green ecology of these extinct birds. Pigment sought with Raman spectroscopy. eggshells of an Elegant Crested Tinamou concentrations could potentially be Functional groups in the penguin (Eudromia elegans) were recently stud- gained from Raman spectral peak ratios pigment identified with Raman spec- ied with Raman spectroscopy.7 Seven calibrated with concentration data from troscopy were consistent with a nitro - peaks in Raman spectra from the HPLC. The strongest peak in a Raman gen-bearing heterocyclic ring.9 A distinct Elegant Crested Tinamou eggshells both spectrum of a pigmented eggshell is peak at 1466 cm–1 was attributed to the identified a tetrapyrrole pigment and attributed to the symmetric stretching of stretching of bonds in a lactam ring (i.e. distinguished biliverdin from protopor- the carbonate anion in calcium carbon- cyclic amide) and peaks at 1285 cm–1 phyrin (1619, 1588, 1467, 1295, 1248, ate.7 The relative intensities of carbon- and 1351 cm–1 were attributed to C–N 1174 and 970 cm–1). The two most ate and pigment peaks are expected to stretching and C–C stretching. Nitrogen intense peaks from the pigment were change with concentration, providing a heterocycles occur in several types of at 1295 cm–1 and 1248 cm–1 which non-destructive method for quantifying pigment including pterins, porphyrins and identified the C=C and C–N stretching pigment levels. linear tetrapyrroles. The yellow penguin of the pyrrole subunits. Hence, Raman pigment and the tetrapyrrole pigment spectroscopy showed that biliverdin is Unresolved pigments bilirubin have broadly similar Raman responsible for a wide blue to green Colourful animals held in museum spectra, except that the penguin pigment colour gamut in eggshells. Moreover, the collections are excellent resources for has a peak at 1577 cm–1 with the high- ability to identify biliverdin in eggshells studying evolutionary patterns in biologi- est relative intensity.9 The Raman spec- without destroying the specimen is a cal pigmentation. Raman spectroscopy is trum from the penguin pigment does useful application for studying ancient ideally suited for surveying the pigments not otherwise match any other published remains (i.e. zooarchaeology). of museum specimens as the technique spectrum. The penguin pigment has proven challenging to extract from the “Bright biochromes are also common in marine feather for mass spectrometric analyses environments...” and structural elucidation. Resolving the structure of the yellow penguin pigment will provide deep insight into the function and evolutionary origins of the unusual yellow pigment.

Anthocyanins Raman spectroscopy is widely used to confirm the presence of a known compound in a particular sample. However, when the sensitivity and chemical specificity of standard Raman spectroscopy are insufficient for an analytical problem, the surface enhancement effect may be needed. In surface-enhanced Raman spectroscopy (SERS) specific molecular vibrations are enhanced when

8 SPECTROSCOPYEUROPE www.spectroscopyeurope.com VOL. 28 NO. 1 (2016) ARTICLE

carotenoids in feathers using Raman spec- University of Chicago Press, Chicago, IL, USA the molecule is adsorbed to, or in prox- troscopy”, Anal. Methods 6, 1301 (2014) . (2014). doi: http://dx.doi.org/10.7208/ imity of, a coinage metal surface with doi: http://dx.doi.org/10.1039/C3AY41870G chicago/9780226057811.001.0001 nanometric roughness (e.g. silver or 2. V.E. de Oliveira, H.V. Castro, H.G.M. Edwards 7. D.B. Thomas, M.E. Hauber, D. Hanley, G.I.N. and L.F.C. de Oliveira, “Carotenes and carot- Waterhouse, S. Fraser and K.C. Gordon, gold nanoparticles, silver or gold nano- enoids in natural biological samples: a “Analysing avian eggshell pigments with patterned surfaces). Surface-enhanced Raman spectroscopic analysis”, J. Raman Raman spectroscopy”, J. Exp. Biol. 218, 2670 Raman spectroscopy has recently been Spectrosc. 41, 642 (2010). doi: http:// (2015). doi: http://dx.doi.org/10.1242/ dx.doi.org/10.1002/jrs.2493 jeb.124917 used to characterise plant sources of 3. L.F. Maia, B.G. Fleury, B.G. Lages, J.P. Barbosa, 8. K.J. McGraw, M.B. Toomey, P.M. Nolan, N.I. anthocyanins (red, purple and blue A.C. Pinto, H.V. Castro, V.E. de Oliveira, Morehouse, M. Massaro and P. Jouventin, plant pigments).10 Anthocyanins have a H.G.M. Edwards and L.F.C. de Oliveira, “A description of unique fluorescent yellow “Identification of reddish pigments in octo- pigments in penguin feathers”, Pigment Cell long history of use as dyes particularly corals by Raman spectroscopy”, J. Raman Res. 20, 301 (2007). doi: http://dx.doi. for textiles. SERS spectra were recorded Spectrosc. 42, 653 (2011). doi: http://dx.doi. org/10.1111/j.1600-0749.2007.00386.x from plants pigmented with anthocya- org/10.1002/jrs.2758 9. D.B. Thomas, C.M. McGoverin, K.J. McGraw, 4. R.F. Fernandes, L.F. Maia, M.R.C. Couri, L.A.S. H.F. James and O. Madden, “Vibrational nins (bilberry, elderberry, purple corn, Costa and L.F.C. de Oliveira, “Identification spectroscopic analyses of unique yellow sumac and hollyhock) and extracts from of reddish pigments in octocorals by Raman feather pigments (spheniscins) in penguins”, aged and unaged dyed textiles. Spectral spectroscopy”, Spectrochim. Acta A 134, J. Roy. Soc. Interface 10, 20121065 434 (2015). doi: http://dx.doi.org/10.1016/j. (2013). doi: http://dx.doi.org/10.1098/ peaks characteristic to anthocyanins were saa.2014.06.022 rsif.2012.1065 observed in the spectra recorded from 5. M. Veronelli, G. Zerbi and R. Stradi, “In situ 10. C. Zaffino, S. Bruni, B. Russo, R. Pilu, C. each plant species sample. resonance Raman spectra of carotenoids Lago and G.M. Colonna, “Identification of in bird’s feathers”, J. Raman Spectrosc. 26, anthocyanins in plant sources and textiles Moreover, pigments extracted from 683 (1995). doi: http://dx.doi.org/10.1002/ by surface-enhanced Raman spectros - wool fibres (as little as 0.5 mg) produced jrs.1250260815 copy (SERS)”, J. Raman Spectrosc. in press SERS spectra with anthocyanin-identify- 6. M.E. Hauber and J. Bates, The Book of (2015). doi: http://dx.doi.org/10.1002/ 10 Eggs: A Lifesize Guide to the Eggs of jrs.4786 ing peaks. However, even the sensi- Six Hundred of the World’s Bird Species . tive SERS spectra were insufficient for identifying specific anthocyanins or plant sources in the wool spectra. This experiment indicated that it is possible to identify the presence of anthocyanins in textiles even after the colour has faded; EXPERTS in a result that has implications for the anal- FLUORESCENCE SPECTROSCOPY ysis of these dyes in historical textiles.

Conclusion FLS980 Complete Luminescence Laboratory Raman spectroscopy is used in biological Research-grade fluorescence pigment research as a method for quickly spectrometer confirming the presence of a known Single photon counting, biochrome. Researchers have also Fluorescence lifetimes (TCSPC) and used the technique to identify particu- Phosphorescence lifetimes (MCS) lar pigments beyond the generalised >25,000:1 Water Raman SNR pigment class, to investigate the chem- Multiple sources and detectors ical identities of unknown pigments, Upgrades include - NIR coverage, and although not fully explored here, microscopes, plate reader module, polarisation, integrating sphere and to measure pigment concentrations. more Raman spectroscopy is ideal for studying pigments in situ (i.e. without extrac- FS5 Photon Counting Spectrofluorometer tion from biological tissues) as there Fluorescence and phosphorescence is no specialised sample preparation measurements capability, including required. We feel that Raman spectros- lifetimes, up to 1650 nm copy could be more widely used in biol- >6,000:1 Water Raman SNR, the ogy research, and we hope that this brief most sensitive bench-top fluorometer available review encourages pigment researchers to explore the value of this technique for Transmission detector for UV-VIS absorption spectroscopy their own study systems. UV-VIS-NIR range References 1. D.B. Thomas, K.J. McGraw, H.F. James and O. Madden, “Non-destructive descriptions of [email protected] | www.edinst.com www.spectroscopyeurope.com SPECTROSCOPYEUROPE 9 VOL. 28 NO. 1 (2016) ARTICLE

Solid mixed matrices and their advantages in matrix- assisted laser desorption/ ionisation time-of-flight mass spectrometry

Marek Šebela Department of Protein Biochemistry and Proteomics, Centre of the Region Haná for Biotechnological and Agricultural Research, Faculty of Science, Palacký University, Olomouc, Czech Republic. E-mail: [email protected]

Introduction mass spectrometers were originally finding of the possibility to ionise and This review article describes the use of designed for small molecular masses not detect proteins with molecular masses solid mixed matrices in matrix-assisted exceeding 1 kDa.2 above 10 kDa.6 In contrast, Tanaka and laser desorption/ionisation (MALDI) Laser desorption/ionisation for biomol- his coworkers relied on an ultrafine cobalt time-of-flight (TOF) mass spectrometry ecules was independently introduced powder with glycerol for laser desorp- (MS). These are prepared by combining by Michael Karas and Franz Hillenkamp tion and ionisation and to their satisfac- common matrix compounds and applied (Germany) for solids in a crystalline envi- tion, they could measure proteins up to particularly to analyse peptides, proteins, ronment3 and Koichi Tanaka (Japan) in about 35 kDa.4 Nowadays, this method is oligosaccharides, oligonucleotides, the liquid phase.4 The use of lasers to rarely used compared with the approach lipids, synthetic polymers or intact cells. generate ions dates back to the 1960s by Karas and Hillenkamp. Nevertheless, The biggest advantage of this approach when a majority of the initial experiments Koichi Tanaka was awarded the Nobel is that it typically yields homogeneous was done with inorganic compounds. Prize for Chemistry in 2002 together sample spots on the target plate and The primary role for the laser was to with John Bennett Fenn, who in parallel provides highly reproducible measure- produce a very rapid heating, and the discovered electrospray ionisation (ESI), ments accompanied by increased sensi- energy transfer was considered to another soft ionisation technique, which tivity and resolution. occur via the metallic substrate.5 Karas is also well suited to studying biological Since its discovery towards the end and Hillenkamp, who actually coined macromolecules.2 of the 1980s, MALDI, which is usually the term MALDI, first recognised that a associated with a TOF mass analyser in small organic compound, which strongly The role of matrix in commercial instruments, has become a absorbed at the laser wavelength they MALDI process common technology in biological MS.1 used and exhibited a soft ionisation, Competition between the two soft ioni- MALDI-TOF MS has found its application could enhance the ionising laser desorp- sation techniques saw ESI-MS pushed not only in the analysis of peptides and tion process. During analyses with amino forward in its development as it offered proteins, oligonucleotides, oligosaccha- acids, the highly absorbing tryptophan the advantage of being able to be rides, technical polymers and small polar functioned well as a matrix for alanine, coupled on-line with high-performance compounds, but also viruses or even a non-absorbing amino acid, when liquid chromatography. In the 1990s, the intact microbial cells. Initially there was they were present together in the form acceptance of MALDI MS as an analyti- an opinion that it would never be possi- of a 1 : 1 mixture.3 Later on, Karas and cal tool was raised by introducing both ble to analyse large biomolecules using Hillenkamp worked with nicotinic acid delayed extraction of ions (resulting mass spectrometry. It was assumed that, as a matrix with excitation at 266 nm. in high resolution spectra) and post- as first it is necessary to vaporise and Having been aware of results of Koichi source decay fragmentation (allowing then ionise the sample to perform MS, Tanaka and his coworkers,4 they success- one to obtain structural information on large molecules would be decomposed fully turned their increased effort with a selected ions).5 The process of MALDI is long before their vaporisation. Moreover, new post-acceleration detector into a achieved in two steps.7 First, the analyte

10 SPECTROSCOPYEUROPE www.spectroscopyeurope.com VOL. 28 NO. 1 (2016) ARTICLE

mentation, especially in the case of proteins and peptides carrying post-trans- lational modifications. A small number of compounds have been studied with respect to their propensity to induce fragmentation. They are thereby classified as “hot” (assisting the formation of ions with a high internal energy) or “cold”. For example, Karas et al. found that post- source decay (= metastable fragmentation) decreases in the order CHCA or SA > DHB mixed with 2-hydroxy- 5-methoxybenzoic acid (so-called “super DHB”) > 3-hydroxypicolinic acid (HPA) for protonated glycoproteins.12 However, it has been shown that the assignment of “hot” and “cold” matrix may reverse as the analyte or laser energy changes.13 Figure 1. A scheme of the principle of MALDI (adapted from a previous publication7). Mixed matrices for proteins, peptides, to be investigated is typically dissolved in duced cinnamic acid derivatives, such as oligosaccharides and a solvent, which contains an appropriate sinapinic, ferulic and caffeic acids (SA, oligonucleotides matrix compound. On the target plate, a FA, CA, respecively),9 which performed Various approaches have been explored “solid-solution” deposit of analyte-doped excellently for the analysis of proteins to enhance matrix selectivity, such as the matrix crystals is formed on evaporation using ultraviolet lasers with maximum use of additives, but they do not directly of the solvent. As a result, molecules of wavelengths at 355 nm (Nd : YAG) or affect the ionisation and rather influ - the analyte are embedded in an excess 337 nm (nitrogen laser).5 After a few ence co-crystallisation or exclusion of of matrix molecules.7 In the second step, years, the same researchers discov - salt impurities. A big potential resides in which takes place under vacuum condi- ered that α-cyano-4-hydroxycinnamic combining more compounds into binary tions in the ion source of the spectrome- acid (CHCA) was efficient for MALDI- or ternary matrix mixtures. The above- ter, the co-crystals are ablated in portions TOF MS analysis of peptides and glyco- mentioned “super DHB” is a mixture by intense laser pulses (see Figure 1). peptides in the molecular mass range of DHB and 2-hydroxy-5-methoxyben- The exact mechanism of the formation of 500–5000.10 2,5-Dihydroxybenzoic zoic acid in a weight ratio of 9 : 1. This of ions is not completely understood. acid (DHB) has become another matrix binary matrix provides higher yields and Definitely, the matrix is indispensa - suitable for peptides and glycopeptides, signal-to-noise ratios for analyte ions in ble in this process. The rapid heating which was first examined at that time.11 the high mass range: typically it is appli- by laser irradiation causes the accumu- Currently, there are many compounds cable to intact proteins. The reason is lation of energy through excitation of available which can be applied for this elucidated by the formation of a disor- the matrix molecules. A dense plume purpose according to the specific needs, der in the DHB crystal lattice resulting of material containing both the matrix but only a few of them are really good. in “softer” desorption.14 Another combi- and matrix–analyte clusters expands Anyway, matrix choice and optimisa - nation based on DHB involves CHCA into the vacuum of the ion source.8 The tion of the sample preparation proto - as the second matrix component (in a most widely accepted mechanism of col are the most important steps in weight ratio of 1 : 1).15 This mixed matrix primary ionisation involves generation MALDI experiments.7 The final selection proved to be well suited for peptides and of charged species within a cluster abla- is always related to the chemical nature intact proteins (an improved signal inten- tion process or gas-phase proton transfer of the analyte and the respective laser sity and resolution was found especially in the expanding plume from photo- wavelength. The most effective MALDI for glycoproteins). The crystallisation of ionised matrix molecules.7,8 The ions in matrix must simultaneously meet a samples with CHCA/DHB results in a the gas phase are then accelerated by number of general requirements:7 strong homogeneous pattern of CHCA-derived an electrostatic field and move towards absorbance of the laser light at a given crystals in the central part of the sample the analyser. wavelength, ability to sublime, vacuum spot with small DHB crystals at the Interestingly, nearly all of the matrix stability, promotion of analyte ionisation, perimeter. When applied, CHCA/DHB compounds that are routinely used today solubility in analyte-compatible solvents provides both significantly improved were discovered in the early days of and lack of reactivity. The choice of matrix spot-to-spot reproducibility of mass MALDI.5 In 1989, Beavis and Chait intro- is also important for the control of frag- spectra and reduced noise compared www.spectroscopyeurope.com SPECTROSCOPYEUROPE 11 VOL. 28 NO. 1 (2016) ARTICLE

with the use of either component indi- mercaptobenzothiazole proved to be an Mixed matrices for lipids, vidually, which simplifies fully automated excellent first layer for DHB and 6-aza- low-molecular-weight acquisition. An interesting binary matrix 2-thiothymine (ATT; by the way this is compounds, polymers and for measuring phosphopeptides was perfect for investigations on acidic oligo- intact microbial cells discovered by combining HPA with CHCA saccharides in the negative-ion mode) Dithranol (1,8-dihydroxy-9,10-dihydro- in an optimised weight ratio of 1 : 1 (see or could be used alone.17 Combining anthracen-9-one) is recommended for Figure 2). As a result, phosphopeptide DHB with the basic aminopyrazine in MALDI-TOF MS experiments with lipids.7 signals can be acquired reproducibly in a weight ratio of 3 : 1 allowed an excel- A ternary matrix containing dithranol, either the positive- or negative-ion mode lent data acquisition not only for neutral DHB and CHCA (in hexane/isopropa- at a lower laser power than with DHB (a monosaccharides and oligosaccharides nol/dimethylsulfoxide; mixed in an equi- common matrix for phosphopeptides) (including the content of a real plant molar ratio with the addition of sodium or CHCA alone. In addition, the analy- extract) but also maltodextrins up to 40 iodide) has recently been employed sis time is shortened by avoiding search- glucose units.18 In contrast, the presence to analyse lipid impurities in biodiesel, ing for the so-called “sweet” spots as of salts is undesirable when performing which is composed mainly of fatty acid is necessary when working with DHB. MALDI-TOF MS with oligonucleotides methyl esters.20 The biggest advantage Concurrently, neutral losses of phos - because of the formation of adducts of the ternary matrix mixture resided phate group (–80 Da) and phosphoric between the phosphate backbone in the formation of smaller and more acid (–98 Da) are reduced. The mixed of DNA and counter ions (Na +, K+). homogeneous co-crystals with samples matrix is also applicable to intact phos- Nowadays, oligonucleotides are largely (compared with the use of individ - phoproteins (as has been demonstrated utilised as gene probes in molecular ual matrices), which had a positive with molecules up to 30 kDa).16 biology. When the quality evaluation of impact on reproducibility and sensitiv- Oligosaccharides represent a chal- synthetic preparations by MALDI-TOF MS ity. Moreover, semi-quantitative informa- lenge for MALDI-TOF MS as there are is necessary, HPA, ATT or 2′,4′,6′-trihy- tion could be obtained using a calibration typically many problems with low sensi- droxyacetophenone are suitable matri- with standards. Similarly, a binary matrix tivity and strong background noise. ces.7 Diammonium hydrogen citrate is consisting of CHCA and DHB in a weight Neutral oligosaccharides are detected used as a common additive. A major ratio of 1 : 1 in 70% methanol/0.1% in the positive-ion mode as sodium- or obstacle is the relatively poor mass reso- TFA (trifluoroacetic acid) containing potassium-charged pseudomolecular lution, which increases in parallel with 1% piperidine was found efficient for ions for which DHB usually represents the increasing molecular mass accom- MALDI imaging MS of phospholipids.21 the matrix of choice. Oligosaccharides panied by the decrease in signal inten- The binary mixture was deposited onto from human milk have been studied sity and sensitivity. To cope with these a sliced rat brain tissue and, impor - with numerous matrices, their binary troubles, the use of a mixture of HPA tantly, it crystallised homogeneously, mixtures and single-matrix mixtures with with pyrazinecarboxylic acid (4 : 1, v/v, for which resulted in increased reproduc- various additives.17 DHB with 2,4-dinitro- saturated solutions) has been described, ibility and signal intensities as well as a benzoic acid and a diluted NaCl solution which provides better spectral parame- higher number of signals. The matrix was was found optimal for standard analyses ters as well as higher reproducibility and superior to conventional matrices in its with neutral compounds. 5-Chloro-2- tolerance to impurities.19 performance and functioned well in both positive- and negative-ion modes: more than 100 glycerophospholipids and sphingophospholipids could be identified by MS and tandem MS. 21 As a certain disadvantage, a strong background at masses below 500 Da appeared which fortunately did not disturb the use of CHCA/DHB for phospholipids. The problem of background signals at low masses was addressed by Guo and He22 in 2007: a binary matrix combining CHCA and basic 9-aminoacridine (the compounds differ from each other in their affinity to protons) significantly reduced the number of background matrix peaks in both positive- and negative-mode detection of small molecules. Figure 2. MALDI-TOF mass spectrum of a synthetic phosphopeptide carrying phosphorylation at In addition, better signal-to-background a serine residue. The spectrum was acquired using HPA/CHCA mixed matrix.16 ratios were observed for negatively

12 SPECTROSCOPYEUROPE www.spectroscopyeurope.com VOL. 28 NO. 1 (2016) ARTICLE

ent properties required of the matrix by varying the identity and/or concentration of components. As a consequence, the sample preparation step provides more homogeneous crystallisation on the target plate, which ensures higher reproducibility of measurements and is typically accompanied by both higher sensitivity and better resolution of signals.

Acknowledgements This work was supported by grant no. L01204 from the Ministry of Education, Youth and Sports, Czech Republic (the National Program of Sustainability I).

References 1. A. Croxatto, G. Prod’hom and G. Greub, “Applications of MALDI-TOF mass spectrometry in clinical diagnostic microbiol - ogy”, FEMS Microbiol. Rev. 36, 380 (2012). doi: http://dx.doi.org/10.1111/j.1574- Figure 3. MALDI-TOF mass spectra of intact fungal cells. Each spectrum, acquired using FA/SA 25 6976.2011.00298.x mixed matrix, belongs to a different plant pathogen species of the same genus. 2. J. Franzen, “ Mass spectrometry in Bremen, a tribute to Dr Ludolf Jenckel”, in A History of European Mass Spectrometry, Ed by K.R. Jennings. IM Publications, Chichester, p. 152 charged inositol phosphates in complex vidually and then also in the form of vari- (2012). plant extracts. Various binary matrices ous binary mixtures such as CHCA/FA, 3. M. Karas, D. Bachmann and F. Hillenkamp, (e.g. DHB + CHCA, DHB + SA, CHCA + SA CHCA/SA and FA/SA (different mixing “Influence of the wavelength in high-irra- diance ultraviolet laser desorption mass and others based on 2,6-dihydroxyben- ratios and solvents). The mixture of FA spectrometry of organic molecules“, Anal. zoic acid as an isomer of DHB) or ternary and SA in a weight ratio of 1 : 3 (specifi- Chem. 57, 2935 (1985). doi: http://dx.doi. mixtures such as DHB + CHCA + SA or cally 5 : 15 mg mL–1) in acetonitrile/2.5% org/10.1021/ac00291a042 4. K. Tanaka, H. Waki, Y. Ido, S. Akita, Y. Yoshida 2,6-dihydroxybenzoic acid + CHCA + SA (v/v) TFA, 7 : 3, v/v, was finally proven to and T. Yoshida, “Protein and polymer anal- allowed one to achieve an improved be optimum for working with fungi and yses up to m/z 100,000 by laser ioniza- sensitivity in the analysis of polyethyl- oomycetes.25 It allows a reproducible tion time-of-flight mass spectrometry“, Rapid Commun. Mass Spectrom. 2, 151 ene glycols with molecular masses rang- acquisition of signals (because of the (1988). doi: http://dx.doi.org/10.1002/ ing from 1 kDa to 10 kDa (the reason homogeneous crystallisation pattern) rcm.1290020802 probably resides in more homogeneous with relatively high signal-to-noise ratios, 5. M. Karas, “From field desorption to MALDI 23 and to the resurgence of time-of-flight mass matrix-sample co-crystal patterns). which are distributed evenly over a large spectrometry” in A History of European FA is a commonly used matrix for mass region (see Figure 3). Also a binary Mass Spectrometry, Ed by K.R. Jennings. IM intact cell/spore MALDI-TOF MS of fungi. mixture CA/SA or ternary mixture with Publications, Chichester, pp. 70–95 (2012). 6. M. Karas and F. Hillenkamp, “Laser desorp- In respect of the number of ion species CA, FA and SA represent good choices tion ionization of proteins with molecular and signal intensities, FA was shown to for working with fungal cells (unpub - masses exceeding 10,000 daltons“, Anal. be superior, for example, to CHCA, SA, lished results). Chem. 60, 2299 (1988). doi: http://dx.doi. org/10.1021/ac00171a028 2-hydroxy-5-methoxybenzoic acid or 7. E. de Hoffmann and V. Stroobant, “Ion 2′,4′,6′-trihydroxyacetophenone when Conclusions sources”, Mass Spectrometry Principles and Fusarium species were analysed despite The preparation of samples for MALDI- Applications. John Wiley & Sons, Chichester, pp. 33–39 (2007). involving additives or applying binary TOF MS is an exciting experimen - 8. M. Karas and R. Krüger, “Ion formation in matrix combinations such as DHB/2- tal chemistry. The choice of the best MALDI: The cluster ionization mechanism“, hydroxy-5-methoxybenzoic acid and matrix, solvents and sample prepara- Chem. Rev. 103, 427 (2003). doi: http:// 24 dx.doi.org/10.1021/cr010376a CHCA/DHB. One major drawback of FA tion technique are crucial steps on the 9. R.C. Beavis, B.T. Chait and H.M. Fales, is the inhomogeneous look of its co-crys- way to achieving reliable results. To over- “Cinnamic acid derivatives as matrices for tals with the analyte resembling snow- come certain imperfections in the use ultraviolet laser desorption mass spectrometry of proteins“, Rapid Commun. Mass covered branches. In our laboratory, we of common matrix compounds, there Spectrom. 3, 432 (1989). doi: http://dx.doi. have a long and extensive experience is a possibility of applying additives or org/10.1002/rcm.1290031207 in the use of binary matrices for intact making binary or ternary matrix systems. 10. R.C. Beavis, T. Chaudhary and B.T. Chait, “α-Cyano-4-hydroxycinnamic acid as a matrix fungi. To find the optimal matrix, CA, Combining more compounds into a final for matrix-assisted laser desorption mass CHCA, DHB plus FA were evaluated indi- mixture allows optimisation of the differ- spectrometry“, Org. Mass Spectrom. 27, www.spectroscopyeurope.com SPECTROSCOPYEUROPE 13 VOL. 28 NO. 1 (2016) ARTICLE

156 (1992). doi: http://dx.doi.org/10.1002/ peptides in matrix-assisted laser desorp- (2011). doi: http://dx.doi.org/10.1021/ oms.1210270217 tion/ionization mass spectrometry”, Rapid ef201257g 11. K. Strupat, M. Karas and F. Hillenkamp, Commun. Mass Spectrom . 23, 2264 21. S.R. Shanta, L.H. Zhou, Y.S. Park, Y.H. Kim, Y. “2,5-Dihydroxybenzoic acid: a new matrix for (2009). doi: http://dx.doi.org/10.1002/ Kim and K.P. Kim, “Binary matrix for MALDI laser desorption—ionization mass spectrom- rcm.4139 imaging mass spectrometry of phospholip- etry“, Int. J. Mass Spectrom. Ion Proc. 111, 89 17. A. Pfenninger, M. Karas, B. Finke, B. Stahl ids in both ion modes” Anal. Chem. 83, (1991). doi: http://dx.doi.org/10.1016/0168- and G. Sawatzki, “Matrix optimization for 1252 (2011). doi: http://dx.doi.org/10.1021/ 1176(91)85050-V matrix-assisted laser desorption/ioniza- ac1029659 12. M. Karas, U. Bahr, K. Strupat, F. Hillenkamp, tion mass spectrometry of oligosaccharides 22. Z. Guo and L. He, “A binary matrix for back- A. Tsarbopoulos and B.N. Pramanik, “Matrix from human milk”, J. Mass Spectrom. 34, ground suppression in MALDI-MS of small dependence of metastable fragmentation 98 (1999). doi: http://dx.doi.org/10.1002/ molecules”, Anal. Bioanal. Chem. 387, 1939 of glycoproteins in MALDI TOF mass spec- (SICI)1096-9888(199902)34:2<98::AID- (2007). doi: http://dx.doi.org/10.1007/ trometry“, Anal. Chem. 67, 675 (1995). doi: JMS767>3.0.CO;2-N s00216-006-1100-3 http://dx.doi.org/10.1021/ac00099a029 18. M.A. Hashir, G. Stecher and G.K. Bonn, 23. J. Hong, T. Kim and J. Kim, “Tertiary matrices 13. G. Luo, I. Marginean and A. Vertes, “Internal “Identification of low molecular weight carbo- for the analysis of polyethylene glycols using energy of ions generated by matrix-assisted hydrates employing new binary mixtures for MALDI-TOF MS”, Mass Spectrom. Lett. 5, laser desorption/ionization“, Anal. Chem. matrix-assisted laser desorption/ionisation 49 (2014). doi: http://dx.doi.org/10.5478/ 74, 6185 (2002). doi: http://dx.doi. mass spectrometry”, Rapid Commun. Mass MSL.2014.5.2.49 org/10.1021/ac020339z Spectrom. 22, 2185 (2008). doi: http:// 24. J. Kemptner, M. Marchetti-Deschmann, R. 14. M. Karas, H. Ehring, E. Nordhoff, B. Stahl, dx.doi.org/10.1002/rcm.3602 Mach, I.S. Druzhinina, C.P. Kubicek and G. K. Strupat, F. Hillenkamp, M. Grehl and B. 19. L. Zhou, H. Deng, Q. Deng and S. Zhao, “A Allmaier, “Evaluation of matrix-assisted laser Krebs, “Matrix-assisted laser desorption/ mixed matrix of 3-hydroxypicolinic acid and desorption/ionization (MALDI) prepara- ionization mass spectrometry with additives pyrazinecarboxylic acid for matrix-assisted tion techniques for surface characterization to 2,5-dihydroxybenzoic acid“, Org. Mass laser desorption/ionization time-of-flight of intact Fusarium spores by MALDI linear Spectrom. 28, 1476 (1993). doi: http:// mass spectrometry of oligodeoxynucleo- time-of-flight mass spectrometry”, Rapid dx.doi.org/10.1002/oms.1210281219 tides“, Rapid Commun. Mass Spectrom. 18, Commun. Mass Spectrom. 23, 877 (2009). 15. S. Laugesen and P. Roepstorff, “Combination 787 (2004). doi: http://dx.doi.org/10.1002/ doi: http://dx.doi.org/10.1002/rcm.3949 of two matrices results in improved perfor- rcm.1404 25. J. Chalupová, M. Sedlárˇová, M. Helmel, mance of MALDI MS for peptide mass 20. C.R. McAlpin, K.J. Voorhees, T.L. Alleman P. Rˇehulka, M. Marchetti-Deschmann, G. mapping and protein analysis”, J. Am. and R.L. McCormick, “Ternary matrix for the Allmaier and M. Šebela, “MALDI-based intact Soc. Mass Spectrom. 14, 992 (2003). matrix-assisted laser desorption ionization spore mass spectrometry of downy and doi: http://dx.doi.org/10.1016/S1044- time-of-flight mass spectrometry (MALDI– powdery mildews”, ˇ J. Mass Spectrom. 47, 0305(03)00262-9 TOF MS) analysis of non-fuel lipid compo- 978 (2012). doi: http://dx.doi.org/10.1002/ 16. L.H. Zhou, G.Y. Kang and K.P. Kim, “A binary nents in biodiesel”, Energy Fuels 25, 5407 jms.3046 matrix for improved detection of phospho- A History of European Mass Spectrometry Edited by Keith R. Jennings

This book aims to give an insight into how some of the more important developments in European mass spectrometry came about, from the advent of the first commercial instruments to the present day. The various accounts, several of which contain personal reminiscences, both provide a human background to these developments and convey the excitement of being part of the European mass spectrometry community during this period.

■■ The foundations of mass spectrometry in Europe—the first fifty years by Keith R. Jennings ■■ Fundamentals by Nico M.M. Nibbering ■■ GC/MS and LC/MS by Andries Bruins ■■ From field desorption to MALDI and to the resurgence of time-of-flight mass spectrometry by Michael Karas ■■ Mass spectrometry in Manchester by Bob Bateman ■■ Mass spectrometry in Bremen, a tribute to Dr Ludolf Jenckel by Jochen Franzen ■■ A short story about the life of Curt Brunnée by Michael C. ten Noever de Brauw ■■ The European history of peptide and protein mass spectrometry by Peter Roepstorff ■■ Applications to small biomolecules and developments in Central and Eastern Europe by Károly Vékey ■■ Industrial and environmental applications by Jim Scrivens ■■ Scientific societies and meetings in Europe by Alison E. Ashcroft

ISBN: 978-1-906715-04-5 n viii + 285 pp. Price: £30 For more information, visit: www.impublications.com/mshistory/

14 SPECTROSCOPYEUROPE www.spectroscopyeurope.com VOL. 28 NO. 1 (2016) TONY DAVIES COLUMN

Spectroscopic data handling at petabyte scale

Antony N. Davies,a,b Shane R. Ellis,c Benjamin Balluffc and Ron M. Heerenc aStrategic Research Group – Measurement and Analytical Science, Akzo Nobel Chemicals b.V., Deventer, the Netherlands bSERC, Sustainable Environment Research Centre, Faculty of Computing, Engineering and Science, University of South Wales, UK cMaastricht MultiModal Molecular Imaging Institute M4I, Universiteitssingel 50, 6229 ER Maastricht, The Netherlands

For many analytical spectroscopists, data ing approaches for nanomedicine and processing, a significant challenge within handling challenges arise every few years biomedical research. Peter Peters’ team itself. An example of the state-of-the-art when the space on the USB stick used use techniques such as high-resolu - development work being undertaken to move data between the spectrom- tion cryo-electron microscopy to investi- with commercial companies is the beta eter and the office computer becomes gate complex protein structures in cells. testing of a new parallel imaging MS/ full. With the founding of the Maastricht This far-sighted strategic decision by MS nanoTOF II from Physical Electronics MultiModal Molecular Imaging Institute Maastricht University has attracted signif- (PHI) (Figure 1) where the TOF-SIMS M4I at the Brightlands Maastricht Health icant funding to the location and has spectrum (MS1) and the MS/MS spec- Campus and the associated appoint- allowed an unrivalled capability to be trum (MS2) are acquired in parallel. This ment of two new Professors, the inter- established which is still growing. high-information-volume methodology ests of Professor Ron Heeren with the allows researchers to directly compare Division of Imaging Mass Spectrometry Size of the data storm spectra, images or depth profiles from and Professor Peter Peters and the Working together with some of the major MS1 and MS2 of the same three-dimen- Division of Nanoscopy, a perfect storm instrument vendors in our field they are sional volume containing hundreds of of data has been created. At the largest now generating data at the rate of 100s of thousands or more pixels. molecular imaging centre in Europe, Ron GBytes/day. These large amounts of data On another new instrument, the Heeren’s group study high-resolution need to be rapidly and securely stored Bruker rapifleX MALDI Tissuetyper™ molecular imaging of biological systems in a location which is also designed to time-of-flight (TOF) which offers acqui- and polymers through the development be able to serve this data back to the sition rates up to 50 times faster than and application of state-of-the-art mass individual researchers when they need other MALDI imaging systems, they have spectrometry based molecular imag- to begin the task of data analysis and generated results from experiments that were performed on brain sections with pixel sizes ranging from 10 × 10 µm2 to 50 × 50 µm2. The data generated in both positive- and negative-ion modes yielded information-rich and comple- mentary lipid spectra revealing the spatial changes of the lipidome composition throughout the mouse brain. The speed of the instrument allowed an entire mouse brain to be imaged consec- utively in both positive- and negative-ion mode in ~35 minutes.1 These high acquisition speeds allow work on new classes of matrices that are unstable under high vacuum for MALDI-MSI studies, but, of course, this means large amounts of data are now acquired much faster, thus plac- Figure 1. The new Physical Electronics NanoToF II tandem SIMS system in operation at M4I in ing further demands on IT infrastructure. Maastricht. A typical experiment from this instrument www.spectroscopyeurope.com SPECTROSCOPYEUROPE 15 VOL. 28 NO. 1 (2016) TONY DAVIES COLUMN

connected via Gigabit–Ethernet connec- (a) (b) tions to the instruments, data analy - sis clients and university network. In order to reduce the data transfer rates between the storage and the data analysis units, the mass spectrometry imaging (MSI) data is processed and reduced on the fly during acquisition. The latter can lead to a 100- to 1000- fold reduction, depending on the type of data, enabling acceptable response ~7 mm times for the analysis by the researcher. Also MSI data can benefit tremen - dously from parallelised processing, as 100% an MSI dataset is a collection of indi - vidual mass spectra where each spectrum can be treated separately. Hence, 0% commercial as well as in-house developed software make use of multi-core + + + Figure 2. (a) Positive-ion images of [PC(40:6)+K] , [PC(38:6)+K] and [PC(36:1)+K] observed processing systems. At the Maastricht at m/z 972, 844 and 826 and shown in red, blue and green, respectively, acquired with a University there are currently two nodes 20 × 20 μm raster. This image contained 181,723 pixels. (b) Enlarged region showing the comple- of 64 cores and each with 512 GB RAM mentary distributions of these ions in the cerebellum. The corresponding H&E-stained section is shown on the right. Reproduced from Reference 1 with permission; © 2015 John Wiley & Sons memory available. As a partner in the Ltd. Dutch Life Science Grid, it is possible to upscale to greater computational power using clusters of other participat- (see Figure 2) yields data at around are cutting and allows the differentia - ing centres. 10–100 GB per tissue. In this particular tion of tumorous and healthy tissue. This Another important pillar for data case the raw data stream is made up critical information, based on a series of analysis, successful interpretation and of over 181,000 individual mass spec- collected mass spectra, helps ensure all generation of relevant results, is an tra measured at a resolution of 20 µm. tumorous tissue is removed and mini- IT-infrastructure for the integration of the Of course such advancements open the mises the need for follow-up surgery. data with other data. In the context of way to analysis of large tissue cohorts projects that run in collaboration with the for clinical studies. In this respect TBs of Weathering the Academic Hospital of Maastricht (AZM), raw data are expected which must be storm—data handling this can be clinical data or other types treated carefully along with the confiden- infrastructure of data that has been obtained by other tial, associated patient data. In order to efficiently master the in-house techniques from the same sample/ Another area of large data production data tsunami and to provide the research- patient (e.g. genomic data, MRI scans and analysis in this group at Maastricht ers with the opportunity of actually inter- etc.) Other data can also be meta-data is from the team working on developing preting the data volumes and converting related to the experiment such as instru- the medical applications of the Waters them into knowledge (with the associ- ment settings during data acquisition or iKnife Rapid Evaporative Ionisation Mass ated publications of course!) the follow- the sample preparation protocol. This Spectrometry (REIMS) systems and ing infrastructure has been put in place IT-infrastructure of storage and integra- their associated databases. This system (see Figure 3). tion also enables to fulfill the require - allows for molecular analysis of surgi - The demands on that IT-infrastructure ments of the FAIR data criteria. cally removed tissue in real-time during in data handling are two-fold: on one the cutting process by collecting the side huge amounts of data have to be FAIR data smoke produced and introducing it into stored somewhere (storage space), as The M4I is, with other Dutch-based a mass spectrometer (in this case a Xevo the data produced surpasses standard research groups, a member of the Dutch system from Waters). It relies heavily PC storage possibilities. And on the other Techcentre for Life Sciences (DTL) who on the generation and access to tissue hand, this amount of data has to be are promoters of the FAIR Data approach and disease-specific databases that are moved in a short amount of time from (http://www.dtls.nl/fair-data/). Long- compared to the molecular profile of the and to the storage (network speed). term readers of this column will have no tissue in contact with the surgical knife. At the M4I, a petabyte centralised difficulty in recognising and welcoming It thus provides real-time feedback to storage system from Hitachi Data the ideals behind the FAIR data approach. surgeons as to the type of tissue they Systems has been installed which is As they describe it data should be:

16 SPECTROSCOPYEUROPE www.spectroscopyeurope.com VOL. 28 NO. 1 (2016) TONY DAVIES COLUMN

Dutch Life Science Grid raw data Hitachi Data Systems Petabyte Storage

processed imaging data

clinical data DATA IN DATA OUT Gigabit-Ethernet Multi-core processing Data systems experimental meta‐data Inform ‐ation

etc. ….. Knowledge

Researchers process data

Figure 3. Very rough outline of the data generation to publication pathway at M4I.

■■ Findable—easy to find by both To be Findable: To be Re-usable: humans and computer systems and F1. (meta)data are assigned a glob- R1. meta(data) have a plurality of based on mandatory description of ally unique and eternally persistent accurate and relevant attributes the metadata that allow the discov- identifier R1.1. (meta)data are released ery of interesting datasets; F2. data are described with rich with a clear and accessible data ■■ Accessible—stored for the long term metadata usage license such that they can be easily accessed F3. (meta)data are registered or R1.2. (meta)data are associated and/or downloaded with well- indexed in a searchable resource with their provenance defined license and access condi- F4. metadata specify the data iden- R1.3. (meta)data meet domain- tions (Open Access when possible), tifier relevant community standards whether at the level of metadata or at To be Accessible: the level of the actual data content; A1. (meta)data are retrievable by Conclusions ■■ Interoperable—ready to be combined their identifier using a standardised In conclusion it is very pleasing to see with other datasets by humans as communications protocol not only significant investment into well as computer systems; A1.1. the protocol is open, free advanced spectroscopic techniques ■■ Reusable—ready to be used for future and universally implementable being made in Europe during econom- research and to be processed further A1.2. the protocol allows for an ically difficult cycles, but also that the using computational methods. authentication and authorisation longer-term future of the spectroscopic As such the DTL is working with simi- procedure, where necessary data is also at the forefront of the minds larly interested international bodies on A2. metadata are eternally accessible, of those fortunate enough to be receiv- the FAIR Data Stewardship of scientific even when the data are no longer ing this support and a key enabler of information (https://www.force11.org/ available their strategy and, we are confident, of group/fairgroup/fairprinciples). To be Interoperable: their future success. These lay down exactly what steps an I1. (meta)data use a formal, acces- organisation needs to take in order to sible, shared and broadly applicable References meet the ideals of the FAIR data approach. language for knowledge representa- 1. N. Ogrinc Potocˇnik, T. Porta, M. Becker, R.M. Heeren and S.R. Ellis, “Use of advantageous, This is still work in progress but is very tion. volatile matrices enabled by next-generation well aligned as general principles for I2. (meta)data use vocabularies that high-speed matrix-assisted laser desorption/ sensible Big Data archiving not only in the follow FAIR principles ionization time-of-flight imaging employing a scanning laser beam”, Rapid Commun. bio-spectroscopy fields but for all of us I3. (meta)data include qualified refer- Mass Spectrom. 29, 2195–203 (2015). doi: regardless of our specific areas of interest. ences to other (meta)data http://dx.doi.org/10.1002/rcm.7379 www.spectroscopyeurope.com SPECTROSCOPYEUROPE 17 VOL. 28 NO. 1 (2016) QUALITY MATTERS

BERM 14 retrospective: autumn in Maryland, USA

Peter Jenks The Jenks Partnership, Newhaven House, Junction Road, Alderbury, Salisbury, Wiltshire SP5 3AZ, UK. E-mail: [email protected]

The 14th Biological and Environmental ■■ 78 invited and contributed oral pres- Reference Material symposium entations in 14 sessions (BERM14) was held at the Gaylord ■■ 108 posters National Resort and Conference Centre, ■■ Almost four full days of presentations, National Harbour, Maryland, USA, from with parallel sessions in the morning 11 to 15 October 2015. It was, based and afternoon, ran each side of 21 on the feedback received, a resounding plenary and keynote speakers who success, both scientifically and socially. helped set the tone for the following The weather was perfect and the loca- oral presentations. tion, well, impressive as was the local ■■ In total 99 oral presentations organisation. ■■ Two poster sessions allowed 108 Before looking at the BERM 14 Meeting posters to be presented to the dele- highlights it is important to put BERM gates. into context: it is the original symposium ■■ Twenty-three organisations, most of looking at biological and environmental them producers of CRMs and or PT reference materials. The first meeting, supported the event by sponsoring a then known as “BRM” was organised by range of opportunities. Dr Stephen Wise, from NIST and the meeting Dr Wayne Wolf and held in Philadelphia, Scientific Chair, together with his wife proudly Over time the biological aspect of USA, in September 1983: 26 years ago! displaying the plaque given to them to mark BERM has grown, and this time was no It was a delight to see that Wayne was more than 30 years involvement with the different. Indeed with six main sessions part of BERM 14, supporting AOAC with BERM series of symposia. looking at all aspects of metrology and their exhibition booth. Another BERM reference materials in biological systems stalwart, Dr Stephen Wise, was present day and with plenty of dining and social there was much to challenge a traditional in his capacity of Scientific Chair. Steve options in the local area. Feedback from analytical chemist. told me that he had attended the early attendees was universally positive and Notable sessions included: BRM meetings as a (then) young NIST the organisers can be congratulated for Reference Materials for Biosimilars, Scientist. Steve’s enduring support for a job very well done indeed! Pharmaceuticals and Bioanalysis, BERM was marked by the presentation of Back in 1983, that really is 32 years kicked off by a challenging Keynote talk an Award Shield by Hendrik Emons, on ago, just 25 people shared 16 presenta- by Bary Cherney from Amgen. behalf of everyone who attended BERM tions. Indeed, back then the idea of the Symposia over the years. application of sound chemical metrol- Since then, BERM meetings have ogy principles to biological matrices was alternated between Europe and North somewhat novel. For BERM 14 the ever America, with a few variations in that increasing level of interest in reference BERM 13 followed BERM 12 in Europe. materials fuelled interest: key features So after a gap of nine years it was high from BERM 14 include: time BERM returned to the USA. The last ■■ 281 attendees: a new record meeting held in the USA, BERM 10, took ■■ NMIs, government agencies, place between 30 April and 5 May 2006 commercial providers, non-prof- and attracted close to 200 delegates. its, accreditation bodies, users of The venue for BERM 14 was chosen to reference materials BERM delegates winding down in the roof ensure there would be no space restric- ■■ Scientists from 27 countries top bar after a gruelling day in scientific tions, no need to go outside during the attended sessions.

18 SPECTROSCOPYEUROPE www.spectroscopyeurope.com QUALITY MATTERS See us at PITTCON Booth 3318

THE SCIENCE OF COMPLIANCE

The BERM 14 programme demanded parallel sessions: the photo shows a speaker from ANAB talking about the development of ISO 17034.

CRM Developments for Clinical how do you know for certain what you Analyses which included two fascinat- are measuring? ing talks by Hongmei Li and Liuxing Feng The meeting was brought to a close from the National Institute of Metrology by a thoughtful and final plenary session in China looking at the application of by Dr Derek Craston, who holds the “hyphenated” analytical techniques such position of Government Chemist to the as HPLC-ICP-MS and IDMS to some chal- United Kingdom Government, effec- lenging bio molecules. tively the “Supreme Court” for metrology Commutability of Clinical CRMs, disputes between the UK Government lead by Ingrid Zegers from IRMM, looked and UK Business! With the weight of an area which is going to get more and such responsibility Derek has some very more important as the demand for and clear ideas about the need for and use The World Leader in UV, use of CRMs in clinical and biological of CRMs! Visible and NIR Certified metrology increases. So, BERM 14 is over: BERM 15 is due Reference Materials CRM Developments for Food & to take place during June in 2018 and Dietary Supplements got off to a will be in Berlin, Germany, hosted by ISO/IEC 17025 Calibration wonderfully informative and amus - BAM and supported by IRMM. Dr Ulrich ing start with a plenary talk by Mark Panne, a long-time supporter of BERM, NIST Traceable Blumenthal from the American Botanical will be the Symposium Chair. ISO Guide 34 Reference Council in which he highlighted the Material Producer many challenges associated with bring- ing good analytical metrology to a rapidly Lifetime Guarantee evolving and creatively marketed sector of the health care industry. Fast Recalibration Service Reference Materials and Microbiology was kicked off by Raymond Cypess from ATCC who showed just how CRMs and best practices are essential to

PMS 660 close the reproducibility gap that so often PMS Cool Gray 7 0659 appears in biological measurements. [email protected] Confidence in Identification for www.starna.com

Preparation of Biological Reference REFERENCE MATERIALS 4001 And so it is almost over: just time to look PMS 660 +44(0) 20 8501 5550 PMS Cool Gray 7 Materials asked, in a number of differ- forward to BERM 15, which will be in Berlin ent ways, the fundamental question: in 2018. www.spectroscopyeurope.com SPECTROSCOPYEUROPE 19 VOL. 28 NO. 1 (2016) SAMPLING COLUMN

Sampling quality assessment: the replication experiment

Kim H. Esbensena and Claas Wagnerb aKHE Consulting, www.kheconsult.com. E-mail: [email protected] bSampling Consultant. E-mail: [email protected]

This column gives an overview of an the variance of the repeated outcome, be There are in fact many scenarios that issue that has not received proper atten- it small or large, will furnish a measure differ from a nicely bracketed DOE situ- tion for decades, the issue of “replica- of the “total experimental uncertainty”, ation. Indeed most data sets do not tion”. This issue turns out to be complex which will be larger than the strict analyt- originate exclusively from within the and there has been a lot of confusion in ical repeatability. In routine operations complacent four walls of an analytical the literature. Three answers to what is in the analytical laboratory, variability also laboratory. What will be described below often stated in response to the funda- reflects effects from other uncertainty constitutes the opposing end of a full mental question: “what is replicated contributions stemming, for example, spectrum of possibilities in which the exactly?” are i) replicate samples, ii) repli- from small-scale sampling of reactants researcher/data analyst must also recog- cate measurements or iii) replicate anal- involved, which may not necessarily nise significant sampling, handling and ysis (replicate analytical results). Upon represent completely “homogeneous other errors in addition to the effective reflection it is clear that these three stocks”. Added uncertainty contributions TAE. The total sampling error (TSE) will answers are not identical. The often only may also occur from resetting the experi- include all sampling and mass-reduction implied understanding for all three cases mental setup—to what precision can one error effects, all incurred before analysis. is that a beneficial averaging is carried “reset” temperature, pressure, concentra- It is self-evident that these errors must out with the connotation that important tion levels of co-factor chemical species also be included in realistic analytical insight can be gained by “replication”. By after having turned the setup off and error assessments; TAE alone will not replicating the specific process behind cleaned all the experimental equipment? give a relevant, valid estimate of the total replicated samples, measurements and Still, such uncertainty contributions are effective effects influencing the analyti- results, some measure of variability is usually considered acceptable parts of cal results. We are forced to be able obtained; but a measure of what? There the total analytical error (TAE). Often all to furnish a valid estimate of the total are many vague prerequisites and impre- of the above turn out to be of small, or sampling-handling-analysis uncertainty cise assumptions involved, which need vanishing, effect because of the regular estimate (GEE: = TSE + TAE). careful analysis. For starters, i) addresses conditions surrounding a controlled DOE The description below is supposed the pre-laboratory realm, while ii) and iii) situation. to deal comprehensively with the many play out their role in the analytical labora- Stepping back one step, however, one different manifestations surrounding the tory—but even here: are replicate analy- might find it equally relevant to repeat replication issue, such that most realistic sis the same as replicate measurements? the experiment by another technician, scenarios are covered. At the heart-of-the- researcher and/or in another laboratory, matter is a key question: what is meant by Background enter the well-known analytical concept “replicate samples”? This issue will appear From the discipline of design of experi- of reproducibility. There may be more, more complex than may seem the case ments (DOE) comes a strict conceptual smaller or larger effects in this widened at first sight and will receive careful atten- understanding and terminology because context, and careful empirical total effect tion w.r.t. definitions and terminology. It of the controlled surrounding conditions. estimations must always be carried out will also transpire that this issue is inti- In the situation of chemical synthe - in order to arrive at a valid estimate of the mately related to validation in data analysis influenced by several experimental augmented, effective TAE. sis, chemometrics and statistics. factors, temperature, pressure, concentration of co-factors for example, it is easy Behold the whole lot-to-analysis Clarification to understand what a replicate experi- pathway Upon reflection it will be appreciated ment means: one is to repeat the exper- Below we address more external issues, that “replication” can concern the follow- imental run(s) under identical conditions not always on the traditional agenda for ing alternatives in the lot-to-aliquot path- for all controllable factors, taking care to replication, in fact quite often left out, or way from primary sampling to analytical randomise all other factors, in which case forgotten. result:

20 SPECTROSCOPYEUROPE www.spectroscopyeurope.com extract the vial and insert it in the instrument repeatedly allowing a realistic temperature variation to influence TAE because this is a more realistic repetition of the general measurement process in any laboratory than simply leaving the test portion in the instrument. This is a first foray into what is known as “Taguchi thinking1”, which opens up the possibility to focus on possible influencing factors which are not embedded in the experimental design; this could well be of interest in some cases. One important dictum of Taguchi was: Do not necessarily look only for optimal results (which may have large variability) but to results where the response variability is low over a large span of the experimental domain (even if less optimal). This is what Toyota has been practicing for years. Certain scepticism regarding the merits of this approach has been voiced but here we will let the reader decide.

Opening up the relevance of this type of perturbation of the analytical process, to another analyst it may appear equally reasonable to include some, or all, of the “sample preparation” procedures in the replication as well, which should then also be repeated 10 times (point 4 and/or 5 above).

But having broadened the horizon this far, it is an unavoidable logical step to follow up with still further realistic perturbations, which means to also include the tertiary, secondary and in the full measure of things, even also primary sampling errors in the replication concept. Why? Because these are potential uncertainty contributions that of necessity will be in play for any-and-all analytical aliquots ever subjected to measurement! VOL. 28Following NO. 1 (2016) the full impact of the Theory of Sampling (TOS) and its detailed treatment of the phenomenon of heterogeneity, it is in fact clear that the only complete “sampling-and-analysis” scenario that is guaranteed to include all SAMPLINGuncertainty contributions must start withCOLUMN replication of the primary sampling (“replication from the top”). Any less comprehensive replication scenario is bound to be incomplete.

1. Replication of the primary sampling process 2. Replication starting with the second- Replication – where? ary sampling stage (i.e. first mass reduction) in the compound 3. Replication starting with the tertiary sampling process (i.e. lab. mass LOT sampling_analysis LOT reduction) process 1) 4. Replication starting with aliquot preparation (e.g. powder compactifica- tion) 5. Replication starting with aliquot 2) S1 instrument presentation (e.g. surface conditioning) 6. Replication of the analysis (measure- S2 3) ment operation) only (TAE) The last option is the situation corre- ……. ……. sponding to “replicate measurement” in 4) the most restricted case. But does this S3 mean that the analytical aliquot (the vial) stays in the analytical instrument all the time while the analyst simply “presses Fig.Figure 1. Replication 1. Replication can can be be performed performed at at many many stages stages in thein thefull lot-to-aliquotfull lot-to-aliquot pathway, pathway, but which is the most realistic situation pertaining to the general operations not only in the analytical the button” say 10 times? Possibly; in laboratory? It turns out that all replication must meaningfully start “from the top”. which case this furthers a strict estimate 1 Taguchi approach: http://en.wikipedia.org/wiki/Taguchi_methods of TAE only. However, it seems equally relevant to extract the vial and insert it in the instrument repeatedly, allowing a equally reasonable to include some, or comprehensive replication scenario is possible temperature variation to influ- all, of the “sample preparation” proce- bound to be incomplete. ence on TAE because this is a more real- dures in the replication as well, because Repeating the primary sampling, istic repetition of the general work and these part-operations cannot necessar- again say 10 times (preferentially more measurement process in any laboratory ily be performed in completely identical when needed), means that each of 10 than simply leaving the test portion in the fashion. This effect should then also be individually sampled primary samples is instrument. This is a first foray into what repeated, say 10 times (stages 4 and/or being subjected to an identical proto - is known as “Taguchi thinking”,1 which 5 above) in order to acquire a measure col that governs all the ensuing sub- opens up a focus on potentially influenc- of its variance contribution. sampling (mass-reduction), sample ing factors which are not embedded in But having broadened the horizon handling and preparation stages and the experimental design explicitly. Clearly this far, it is a logical step to follow procedures in the laboratory. From the this kind of external thinking is relevant in up with still further realistic perturba - logic of this full representativity path - many situations and should therefore be tions of the measurement process , way, “from lot-to-analytical aliquot”, this included in the replication approach. One which broadly means including also is the only procedure incorporating the important dictum of Taguchi’s is: do not the tertiary, secondary and in the full complete ensemble of uncertainties necessarily look only for optimal results measure of things, even also primary and errors encountered of whatever (which may have large variability), but to sampling errors in the replication nature (sampling, handling, preparation, results where the response variability is concept. Why? Because these are de presentation). The point is that for each low over a large span of the experimen- facto uncertainty contributions that replicated primary sample, all potential tal domain (even if less optimal). This is will have been in play for any-and-all errors will be manifested differently ten a clever way of gaining more information analytical aliquot, ever subjected to individual times giving rise to an accu- about the process involved, be this a measurement! Following the full impact mulated variance which is the most real- production or manufacturing process, or of the Theory of Sampling (TOS) and istic estimate of the total measurement the analytical process itself. Certain scep- its detailed treatment of the phenome- uncertainty (MU).2 In particular this ticisms have been voiced regarding the non of heterogeneity, it is clear that the estimate is bound to include the full merits of this approach, but we will let only complete “sampling-and-analysis” sampling error effects (TSE), which will the reader decide on this matter. scenario that is guaranteed to include often dominate. Opening up for the relevance of this all uncertainty contributions must start In clear contrast, starting at any other type of perturbation of the analytical with replication of the primary sampling of the levels in the list above, stages 2–6 process, to another analyst it may appear (“replication from the top”). Any less will guarantee an incomplete, inferior www.spectroscopyeurope.com SPECTROSCOPYEUROPE 21 VOL. 28 NO. 1 (2016) SAMPLING COLUMN

TSE + TAE estimation, which is structurally “The analyst is not supposed to deal able to understand what was intended, destined to be too low, i.e. unrealistic. with matters outside the laboratory (e.g. nor what was indeed carried out, Should one nevertheless feel sampling)” because of incomplete descriptions in compelled to “shortcut” the full repli - “This department is only charged with the “Method” sections of scientific publication procedure by not starting “from the task of reducing the primary sample cations and technical reports. The issue the top”, one is mandated to describe to manageable proportions, as per codi- is therefore far from trivial, indeed grave the rationale behind this choice and fied laboratory’s instructions” errors are continuously being commit- to provide a full report of what was in “Sampling is automated and carried ted. But rather than address the obvious fact done, lest the user of the analytical out by process analytical technology first question: who is responsible, the results has no way of knowing what was (PAT) sensors; there is no sampling issue way forward shall here be constructive. implicated in the umbrella term ”replica- involved here” The focus shall be on ways and means tion”. “Users” and decision makers, acting “I am not responsible for sampling, I to put an effective end to the confusion on the analytical data, do not like to be only analyse/model the data” surrounding the replication issue, and kept in the dark. ... and similar excuses for not seeing indeed put it to good use instead. Undocumented or unexplained appli- the complete measurement uncer - cation of the term “replicate experi - tainty context. All too often the prob - Quantifying total empirical ments” (or “repeated experiments”) has lem belongs to “somebody else”, with variability—the replication been the source of a significant amount the unavoidable result that the prob - experiment of unnecessary confusion in the past. lem does not receive further attention. Above was outlined how a realistic esti- Many times s2(TAE) has simply been Therefore this stand (“not our respon- mate of the total TSE + TAE, a replication misconstrued to imply s2(TSE + TAE), a sibility”) is always potentially in danger experiment (RE) must always start “from grave error, for which someone or some- of being perpetuated and if so “replicate the top”. This is where replication starts, body (or some ill-considered, incomplete analysis” will still take its point of depar- be this primary sampling in nature, in the protocol) is responsible. But we are here ture at stage 3 (maybe stage 2), but field, sampling in the industrial plant, or not interested in pointing fingers at any never from stage 1, the primary sampling it can be sampling of any target desig- entity (private or legal); it suffices to stop stage. This is not an acceptable situa - nated as the primary lot (examples continuing such practice. tion. There are many occasions in which follow below). The above scenario illustrates an authors, reviewers and even editors have Figure 2 shows the scenario in which unfortunate responsibility compartmen- missed cracking down with the neces- an avid sampler is facing a large lot with talisation, which is sometimes found in sary firmness on such demonstrable the objective of establishing a realistic scientific, industrial, publishing or regula- ambiguities regarding “replication”, with estimate of the average lot concentration tory contexts: the certain result that the reader is not for one (or more) analytes. It is abun-

Figure 2. A primary sampler approaching a significantly heterogeneous lot with a grab sampling RE approach, but deployed with two very different coverage footprints. The left side realises the RE on an irrationally narrow footprint in relation to the full geometrical scale of the lot. The right side attempts to take account of the (hidden) lot heterogeneity by employing a wider footprint as a basis for the RE. These alternative scenarios will result in different relative sampling variability estimates because of the different lot heterogeneities covered. (N.B. neither of these primary sampling procedures succeeds to sample the interior of the lot, so both are not honouring the fundamental sampling principle (FSP).

22 SPECTROSCOPYEUROPE www.spectroscopyeurope.com VOL. 28 NO. 1 (2016) SAMPLING COLUMN

dantly clear that a single grab sample the primary sampling is to be replicated. as manifested by a minimum 10 times stands no chance of ever being able to The RE shall be carried out by a fixed replication of the sampling process being do this job because of the intrinsic distri- procedure that specifies precisely how assessed. RSV therefore measures the butional heterogeneity of the lot. It does the following sub-sampling, mass reduc- total empirical sampling variance influ- not matter whether the lot is small, inter- tion and analysis are to be carried out. It enced by the specific heterogeneity of mediate or large, the point being that is essential that both primary sampling the lot material, as expressed by the this intrinsic heterogeneity is unknown as well as all sub-sampling and mass- current sampling procedure. This is a at the moment of routine sampling. The reduction stages and sample prepara- crucial understanding. There can be no sampler therefore has no other option tion is replicated in a completely identical more relevant summary statistic of the than to act as if it is significantly large. fashion in order not to introduce artificial effect of repeating the full lot-to-aliquot There is no problem assuming this variability in the assessment. pathway procedures (10 or more times) rational stance, the TOS furnishes all Note that when these stipulations are than a RE-based RSV. necessary governing principles and prac- followed it is possible to conduct a RE for In the last decade there has been a tical procedures and equipment assess- any sampling procedure, for example a major discussion in the international ment possibilities so as always to be able grab sampling vs a composite sampling sampling community as to the useful- to deal with significant lot heterogeneity, procedure. ness of a singular, canonical RSV thresh- e.g. Esbensen & Julius (2009).3 It has been found convenient to old; opinions have been diverse. In the By deploying a RE, Figure 2 (right), the employ a standard statistic to the results last few years a consensus has emerged, sampler now has access to a first esti- from a RE. The relative coefficient of vari- however, that indicates a general accept- mate of the effective variability of the ation, CVrel is an informative measure of ance threshold of 20%. RSVs higher sampling procedure, but with TOS it is the relative magnitude of the standard than 20% signify a too-high sampling also clear that there is a grave breach deviation (STD) in relation to the average variability, with the consequence that of the fundamental sampling principle (Xavr) of the replicated analytical results, the sampling procedure tested must be (FSP). expressed as a %: improved so as better to counteract the éù inherent heterogeneity effects in the lot êúSTD CVrel = ´=100 RSV (1) material. Should one elect to accept a Relative sampling êúX variability ëûavr RSV higher than 20% this shall have to It has been found useful to employ a RSV is called the relative sampling be justified and made public to ensure general measure of the sampling variabil- variability (or relative sampling stand- full transparency for all stakeholders. ity as expressed by a RE, enter the RSV: ard deviation). RSV encompasses all The usefulness of the RSV measure the relative sampling variability. sampling and analytical errors combined cannot be underestimated. For whatever The variability of any number of replications can be quantified by extracting and analysing the analytical results from a number of replicate primary samples. RSV threshold(s) These specifically shall have the aim to cover the entire spatial geometry of the RSV: 20% lot as best possible, i.e. spanning the geometrical volume of the primary lot RSV: 33% in an optimal fashion (given the circum- stances), and calculating the resulting RSV: 50% empirical variability based of the resulting analytical results aS. Often a rela- RSV: 85% tively small number of primary samples may suffice for a first survey, though RSV: 120% never less than 10. It is essential that the sampling operations are fully realistic replications of the standard routines, i.e. they shall not be extracted at the Xavr. same general location, Figure 2 (left), which would only result in a local char- Figure 3. Schematic illustration of replication experiment thresholds RSV, e.g. 20%, 33%, 50%, 85% and 120%. Very large relative standard deviations (higher than approximately 85%), when acterisation not at all able to relate to interpreted as representing a standard normal distribution, apparently give rise to negative the effects of the full lot heterogeneity. concentration values. This has no physical meaning, however, and need not cause any untoward What is meant here is that the successive worry; these are but model fitting artefacts, of no practical consequence. The essential information primary sampling events shall take place for the sampler is manifest already when RSV transgresses >20%, i.e. when the sampling proce- at other, equally likely locations where dure is operationally too variable and must be improved upon (TOS). www.spectroscopyeurope.com SPECTROSCOPYEUROPE 23 VOL. 28 NO. 1 (2016) SAMPLING COLUMN

lot material, sampled by whatever procedure, the specific lot/procedure combination can be very quickly assessed. There Singular sampling events can are no untoward practicalities involved always be replicated, e.g. 10 times which might militate against performing a RE assessment; indeed anybody can perform RE assessment on any sampling procedure, or for any sampling equipment etc. It should never be possible to argue for, or against, a specific sampling procedure without a transparent quantitative assessment. RE numbers speak for themselves. The “difficult” issue of sampling is put on a fully understand- able, and very simple operational basis— the RE. Based on an extensive practical experience over 50 years from many applied sectors and fields within science, technology and industry, there are very Figure 4. Examples of replication experiments (RE) that are easily set up. On the left is a many cases on record in which the 20% dynamic process sampling situation, at the right sampling from a stationary lot. Both sampling threshold is exceeded (not infrequently scenarios can be assigned an objective RSV quality index. In order that no misunderstanding may by significant deviations); but there are occur, it is only necessary to perform a proper, calibrating RE once, as part of surveying and char- also an important number of cases in acterising the intrinsic heterogeneity of a specific lot material. which the existing procedure is vindi- cated. A few illustrative examples are given below. But first: what information appropriate RSV, voluntarily described thus be acceptable as is. But in all other is residing in a simple RSV level? and reported in full. cases, TOS-modifications must be imple- Figure 3 illustrates how STD is While it is acceptable to level criticism mented, no exceptions. expressed as a fraction of the general against the suggested threshold (20%), One can therefore view RSV as a flex- level quantified by Xavr. In this illustra- this also entails the obligation to perform ible and relevant sampling procedure tion the white distribution has a STD empirical due diligence in the form of quality index, scaled with the inher- which is exactly 20% of Xavr. Also indi- a RE. Recent industrial, scientific and ent heterogeneity encountered. RSV cated are cases where the empirical technological history is flush with exam- is particularly useful for initial charac- STD forms, e.g. 33%, 50%, 85%... The ples of major surprises brought about terisation of sampling from stationary issue clearly is, at what %-level is one by such simple replication experiments lots, while it is much more customary no longer comfortable with the quantifi- and their attendant RSV. It is either the to use a dynamic, process sampling cation resolution, e.g. for RSV = 50% the intrinsic material heterogeneity which is augmented approach, called vario- signal-to-noise ratio is 1 : 1 only, likely underestimated or, at other times, the graphics when sampling from dynamic not an acceptable situation under any sampling procedure turned out to be lots. RSV and variographics are closely accounts. much less universal than assumed. related approaches fundamentally quan- The canonical RSV threshold, 20%, The purpose of a RE is often to tifying the same heterogeneity; the serves as a general indication only in the assess the validity of an already exist- latter approach is much more powerful, case where nothing is known a priori ing sampling procedure. In practice, the however, due to the fact of its more elab- as to the heterogeneity of the mate - RE can only perform and test a current orate experimental design which allows rial involved. Materials and material sampling procedure as it interacts with a full decomposition of GEE, see, for exam- classes certainly exist that may merit a specific lot material. Should a RSV for this ple, References 5–8. Variographic hetero- higher, or a lower, threshold, for which exploratory survey exceed the canonical, geneity characterisation of dynamic lots the proposed general RSV value can, of or case-specific, threshold, the need for is the subject of a later column in this course, no longer serve. For such cases, complete fulfilment of the TOS has been series. a material-dependent quantification can documented and is therefore mandated, All examples described above pertain be developed, dependent upon the no exceptions allowed. There may be to issues related to sampling and other sampler’s own competence and dili - good reasons to start validation by testing error contributions before analysis. It is gence. The mandate in the sampling an existing sampling procedure—there is noteworthy that some analytical proce- standard DS 30774 is clear: all analyti- always the possibility it may turn out to dures can have significantly large TAE, cal results shall be accompanied by an fall below the pertinent threshold, and e.g. of the order of 10–20% or more,

24 SPECTROSCOPYEUROPE www.spectroscopyeurope.com VOL. 28 NO. 1 (2016) SAMPLING COLUMN

which is then already factored into the empirical RSV level. The principle issues from the few examples given here can be generalised to many other material and lot types. The GEE = TSE + TAE issues are identical for all lot systems. The following examples illustrate how a specific sampling equipment can be assessed with respect to several different materials (with specific heterogeneities), which may result in both pass and fail. RE is a general facility that can in fact be deployed at all stages in the lot-to- aliquot pathway, i.e. also a stages later than the primary sampling stage. If the objective were to assess and compare the two splitters in Figure 6 specifically, the RE may well be initiated at this sub- sampling stage directly (in such a case it is of course still critical to add the sampling error effects from the preced- Figure 5. Upper left: primary process sampler assessed for three different materials, one of ing stages in the final evaluation). which does not pass the test of the dedicated RE (RSV = 78%). Lower right: a complex primary sampler being subjected to a RE with the distinctly worrisome result of RSV = 158%. N.B. illustra- The replication experiment (RE) is a tive examples only, no specific sampler is endorsed, nor renounced. Samplers are sketched only powerful and highly versatile sampling/ in order to illustrate how RE may be used for quantitative assessment. analysis quality assessment facility that can be deployed with great flexibility. It is necessary to be fully specific as to what is meant by “replication” in the situation at hand, i.e. at what stage in the lot-to-analysis pathway is replication to commence. We shall have occasion to employ replication experiments many times in these columns.

Notes and references 1. Taguchi approach: http://en.wikipedia.org/ wiki/Taguchi_methods 2. Issues related to the concept of Measurement Uncertainty (MU), which too often in practice only covers the parts of the analysis process that can be brought under direct laboratory control (while in its full defi- nition purports to cover the entire sampling- handling-analysis pathway, are treated in: K.H. Esbensen and C. Wagner, “Theory of Sampling (TOS) versus Measurement Uncertainty (MU) – a call for integration”, Figure 6. Two laboratory equipment (splitters) subjected to RE assessment, showing highly satis- Trends Anal. Chem. 57, 93–106 (2014). doi: factory quantitative results. N.B. illustrative examples only, no specific sampler is endorsed, nor http://dx.doi.org/10.1016/j.trac.2014.02.007 renounced. Samplers are sketched only in order to illustrate how RE may be used for quantitative 3. K.H. Esbensen and L.P. Julius, assessment. “Representative sampling, data quality, validation – a necessary trinity in chemometrics”, in Comprehensive Chemometrics, of Sampling) – A call for a regulatory para- – III. General Considerations on sampling Ed by S. Brown, R. Tauler and R. Walczak. digm shift”, Int. J. Pharm. 499, 156–174 heterogeneous foods”, Trends Anal. Chem. Elsevier, Oxford, Vol. 4, pp. 1–20 (2009). (2016). doi: http://dx.doi.org/10.1016/j. 32, 178–184 (2012). doi: http//dx.doi. 4. K.H. Esbensen (chairman taskforce F-205 ijpharm.2015.12.038 org/10.1016/j.trac.2011.12.002 2010-2013), DS 3077. Representative 6. K.H. Esbensen, C. Paoletti and P. Minkkinen, 8. P. Minkkinen, K.H. Esbensen and C. Paoletti, Sampling—Horizontal Standard. Danish “Representative sampling of large kernel lots – “Representative sampling of large kernel Standards (2013). http://www.ds.dk I. Theory of Sampling and variographic analysis”, lots – II. Application to soybean sampling 5. K.H. Esbensen, A.D. Román-Ospino, A. Trends Anal. Chem. 32, 154–164 (2012). doi: for GMO control”, Trends Anal. Chem. Sanchez and R.J. Romañach, “Adequacy and http://dx.doi.org/10.1016/j.trac.2011.09.008 32, 165–177 (2012). doi: http://dx.doi. veriﬁability of pharmaceutical mixtures and 7. K.H. Esbensen, C. Paoletti and P. Minkkinen, org/10.1016/j.trac.2011.12.001 dose units by variographic analysis (Theory “Representative sampling of large kernel lots www.spectroscopyeurope.com SPECTROSCOPYEUROPE 25 VOL. 28 NO. 1 (2016) PRODUCT FOCUS

Product Focus on Atomic Spectroscopy

Spectroscopy Europe Product Focuses highlight currently available instrumentation in a particular area of spectroscopy. This Product Focus is on Atomic Spectroscopy, and some companies have provided information on their key products, their applications and features. See our media information (www.spectroscopyeurope.com/advertisers/media-packs) for details of future Product Focuses.

Tel: +44(0)131 555 6662 Tel: 203-925-4602 Adrok Ltd. [email protected] PerkinElmer, Inc. [email protected] www.adrokgroup.com www.perkinelmer.com

PRODUCT: Adrok ADR Test Chamber PRODUCT: NexION® 350 ICP-MS APPLICATIONS: Atomic Dielectric Resonance (ADR) for identifying APPLICATIONS: Environmental monitoring sub-surface geological features n Applications in the oil, gas and water n Food quality/safety testing n USP industries. 232/233/2232 (elemental impurities in KEY FEATURES: Adrok pro- pharmaceutical and nutraceuticals) n vides information on what lies Speciation analyses n Biomonitoring n beneath the earth’s surface Nanomaterials characterisation with no need to drill cores KEY FEATURES: World-class data acquisition rates for complete sam- n Low energy wave forms ple characterisation n Unrivaled stability provides increased uptime and penetrate and return data unparalleled productivity n Simple operation with new workflow-based from the interior of materials, rather than gleaning information from the Syngistix™ cross platform software n Real-time single nanoparticle surfaces of solids n Technology differentiates materials based on several acquisition with optional Nano Application Module parameters, with dielectric permittivity measurement being one of the major functions used for sub-surface mapping PRODUCT NAME: Optima® 8x00 ICP-OES APPLICATIONS: Environmental monitoring Inorganic Tel: +1-540-585-3030 n Food quality/safety testing & nutritional [email protected] elemental analysis n Pharmaceutical & Ventures inorganicventures.com nutraceutical testing n Geochemical/mining n Used oils/lubricants analysis PRODUCT: AACU1 KEY FEATURES: Flat Plate plasma technology lowers operating and APPLICATIONS: ICP-MS n ICP-OES n Atomic Absorption Spectroscopy maintenance costs n Dual View for high and low concentrations in the KEY FEATURES: ISO 9001 n Traceable to NIST n TCT Packaging for same run n PlasmaCam™ viewing camera aids in method develop- longer shelf life ment and productivity n Simple operation with new workflow-based Syngistix™ cross platform software PRODUCT: AAFE1 APPLICATIONS: ICP-MS n ICP-OES n Atomic Absorption Spectroscopy KEY FEATURES: ISO 9001 n Traceable to NIST n TCT Packaging for PRODUCT NAME: PinAAcle™ 900 AA Spectrometers longer shelf life APPLICATIONS: Environmental monitoring n Food quality/safety testing n Biomonitoring LOT-Quantum Design Tel: +44 01372 378822 KEY FEATURES: Flame only, furnace only [email protected] or space-saving designs featuring both n Ltd www.lot-qd.co.uk Choice of deuterium or longitudinal Zeeman background correction n TubeView™ color PRODUCT: Dual AR Coated Back-Illuminated CCD Cameras furnace camera simplifes autosampler operation n Simple operation APPLICATIONS: Raman spectroscopy n UV/NIR photoluminescence n with new workflow-based Syngistix™ cross platform software Multi-track spectroscopy n Fluorescence experiments n Plasma studies KEY FEATURES: Dual AR coating for increased broadband response n Back-illuminated CCD for high quantum effciency n TE cooling for low PRODUCT NAME: PinAAcle 500 Flame AA Spectrometer dark current n Deep depleted sensor for NIR fringe suppression n USB APPLICATIONS: Geochemical/mining 2.0 connectivity n Nutritional elemental analysis n Environmental monitoring PRODUCT: High Resolution Shamrock Spectrographs for KEY FEATURES: Completely corro- Atomic Spectroscopy sion resistant, minimising maintenance APPLICATIONS: Raman spectroscopy n Photoluminescence n Multi- and reducing operating costs n Robust track spectroscopy n Fluorescence experiments n Plasma studies and reliable without compromising on KEY FEATURES: 0.1 nm spectral resolution n Multi-track capability for performance n Flexible operation through easy-to-use Syngistix Touch™ optical fbre input n Double detector output n Image astigmatism correc- or more comprehensive Syngistix™ for AA software n Lowest cost-per- tion optics n Motorised grating turret element analysis using a FAST Flame-enhanced system

26 SPECTROSCOPYEUROPE www.spectroscopyeurope.com VOL. 28 NO. 1 (2016) PRODUCT FOCUS

Shimadzu Tel: +49 (0) 203-76870 Tel: +1-603-886-8400 [email protected] Teledyne Leeman Labs [email protected] Europa GmbH www.shimadzu.eu www.teledyneleemanlabs.com

PRODUCT: Atomic Absorption Spectrophotometer AA-7000 PRODUCT: Hydra II Series Mercury Analysers APPLICATIONS: Environmental n Metallurgy n APPLICATIONS: Mercury analysis n Cold vapour atomic absorption Food and agriculture n Pharmaceuticals n (CVAA) n Cold vapour atomic fluorescence (CVAF)n Thermal decompo- Petrochemistry sition n Amalgamation KEY FEATURES: Dual atomiser double-beam optics n Highest sensitivity n Dual background correction functions n Advanced safety technology n Evolving system confguration with PRODUCT: Inductively Coupled Plasma (ICP-OES and smallest footprint ICP-AES) APPLICATIONS: Elemental analysis n Atomic spectroscopy PRODUCT: ICPE-9800 series KEY FEATURES: Advanced solid-stated array detector (L-PAD) n Full APPLICATIONS: Environmental n wavelength coverage, low stray light optics n Full frame imaging n Radial Food n Water n Pharmaceuticals n axial and dual view confgurations Petrochemistry n Quality Control KEY FEATURES: High-speed simultaneous analysis of multiple elements n Low operating costs n Minimum argon consump- tion n Fast rinse time n Reduced memory effects increase analytical throughput NEW PRODUCTS

diamond ATR element, Hastelloy C-276 the DPR-210, the new probe offers a INFRARED construction and energised PTFE seals, selection of interchangeable ATR mate- allowing it to function over wide ranges rials, such as ZnSe, ZnS, AMTIR-1 and 23-mm diamond ATR of temperatures and pressures, and with germanium. probe aggressive chemical systems, including Axiom Analytical Axiom Analytical has introduced the strong acids and bases. The use of metal- link.spectroscopyeurope.com/28-01-101 DMD-373 diamond ATR probe, which is lic internal lightguides provides perfor- similar to the existing DMD-370 but in a mance stability and long term reliability. smaller (23 mm) diameter, making it suit- Axiom Analytical MASS SPEC able for a wide range of laboratory and link.spectroscopyeurope.com/28-01-100 on-line process applications. The mid-IR Single quad GC-MS transmission of the new probe exceeds Process ATR probe Shimadzu has released the 10%, providing high performance with Axiom Analytical’s new DPR-212 ATR GCMS-QP2020, a high-end single FT-IR spectrometers with room-temper- probe offers similar advantages to their quadrupole gas chromatography mass ature detectors. In common with the DPR-210 ATR probe but in a straight, spectrometer. A newly developed, large- DMD-370, the new probe features a 30-cm long configuration designed to capacity, differential exhaust turbo provide optical transmission of over molecular pump enables hydrogen and 20%. The earlier DPR-210 includes a nitrogen as well as helium to be used right-angle joint which allows it to be inserted vertically into a sample container when mounted in a traditional spectrometer sample compartment. In contrast, the straight DPR-212 is ideal for use with instruments such as the Bruker Alpha or Thermo-Fisher iS Series which allow the probe to point straight down out of the sample compartment. When used with Axiom Analytical’s DMD-373 diamond ATR a more traditional instrument, it can be The GCMS-QP2020 from Shimadzu is a probe is suitable for a wide range of labora- inserted into a sample vessel through a high-end single quadrupole GC mass spec- tory and on-line applications. suitable sealing fitting. In common with trometer. www.spectroscopyeurope.com SPECTROSCOPYEUROPE 27 VOL. 28 NO. 1 (2016) NEW PRODUCTS

as carrier gas. It also allows switching software. Stop-flow or continuous flow between ionisation modes without stop- NMR monitoring is straightforward to set up. ping the mass spectrometer. In the GCMS Magritek Insight software, the Smart SIM method Reaction monitoring with link.spectroscopyeurope.com/28-01-104 creation function automatically creates benchtop NMR programs to obtain data only during the Magritek have released new solutions elution time of the target component, that enable straightforward reaction PHOTONICS thereby achieving high sensitivity. The monitoring on the Spinsolve benchtop system allows simultaneous analysis of NMR spectrometer. One of the largest Femtosecond fibre laser more than twice as many components growth areas for benchtop NMR is reac- TOptica have introduced FemtoFiber ultra as previous models. In addition, meas- tion monitoring where it is being broadly NIR, the third generation of their femto- urements that used to be split into multi- applied in chemistry and pharmaceuti- second fibre laser. It provides pulses of ple methods can now be integrated into cal sciences. NMR has a range of advan- <150 fs at a repetition rate of 80 MHz one. The GCMS-QP2020 is available in tages for reaction monitoring. Spectra and with >500 mW output power at a a number of configurations to suit the wavelength of 780 nm. The laser has components targeted and the phase of a compact design and does not need the samples. A variety of databases are water cooling. available including environmental anal- TOptica ysis, foods analysis, volatile compound link.spectroscopyeurope.com/28-01-105 analysis, biological sample analysis and forensic analysis. Terahertz systems Shimadzu TOptica’s new time-domain terahertz link.spectroscopyeurope.com/28-01-102 platform, TeraFlash, and the CW-terahertz spectrometer, TeraScan, both have a peak dynamic range >90 dB. The Mass spectrometer for TeraFlash generates THz spectra with a bioreactor quantification bandwidth of >5 THz, allowing layer thick- Magritek have released new solutions that The Hiden HPR-40 DSA mass spec- enable reaction monitoring on the Spinsolve trometer systems are designed for moni- benchtop NMR spectrometer. toring dissolved gaseous content in aqueous solutions. All systems feature a media interface with semi-perme are chemically specific, enabling meas- able membrane to enrich the transition urements to be sensitive to molecu - of gases and vapours and simultane - lar structure and reaction kinetics. The ously inhibit the transfer of water vapour. signal has a linear response to concen- Applications include diverse areas of tration and it is generally not sensitive microbiological study including fermen- to the matrix. Results are representative tation culture analysis, biofuel research, of the complete sample and the meas- methane generation, soil core sampling, urement is non-destructive. To enable seawater and freshwater evaluation. The the Spinsolve spectrometer for reac- The TeraFlash, TOptica’s new time-domain systems enable real-time monitoring of tion monitoring Magritek have created terahertz platform. dissolved gaseous species in bioreac- solutions with either glass flow cells or tor operation and in evolved fermenter PTFE flow tubing that fit through the ness measurements down to 20 µm. off gas, including continuous measure- completely clear bore of the instrument. The TeraScan provides frequency reso- ment of CO2, O2 and N2 organic vapours. The glass flow cell option offers a large lution better than 5 MHz, which enables Interface types include slim immersion sample volume in the sampling region of precise spectroscopic measurements of probes for static media and circular flow the spectrometer and a narrow internal trace gases. through membrane carriers for flowing diameter outside of that zone. This gives TOptica media, with choice of membrane and the maximum signal-to-noise output with link.spectroscopyeurope.com/28-01-106 membrane area to optimise sensitivity minimal fluid in the flow system. The for specific gaseous species. For photo- hardware kit is completed with a peristal- Ultra-compact CW laser responsive studies, cuvette-style vessels tic pump, appropriate tubing and a stand modules are available with integral agitation and to provide access below the spectrome- Integrated Optics introduced an upgrade capacities ranging from 2 mL to 500 mL. ter. Complementing the hardware is full to its MatchBox series of ultra-compact Hiden Analytical experiment scripting capability and pump CW laser sources for spectroscopy, sort- link.spectroscopyeurope.com/28-01-103 control available in the latest instrument ing and illumination. The MatchBox2

28 SPECTROSCOPYEUROPE www.spectroscopyeurope.com VOL. 28 NO. 1 (2016) NEW PRODUCTS

standard S20 photocathodes, while lowering dark counts as much as 10× with near-symmetrical pulse height distribution. Currently the spectral ranges cover UV, blue and green wavelengths. The new photocathodes are offered as an option to a number of Photonis photon counting products, and the Hi-QE option can also be specified for the Imaging Photon Camera. The Eagle Raman-S OEM spectrometer plat- The MatchBox series of ultra-compact CW Photonis form from Ibsen Photonics. laser sources from Integrated Optics. link.spectroscopyeurope.com/28-01-108 high efficiency, and optical resolution as series features improved ruggedness, Single photon counter good as 4 cm–1 with a wavelength range extended operational temperature range, for time-correlated of 800–1100 nm. It can be equipped new serial UART interface, laser control applications with either a –10°C cooled BT-CCD or a software and a smaller footprint. The In time-correlated single photon count- –60°C LDC-DD cooled camera. series of OEM laser sources offers more ing (TCSPC), single photons are not Ibsen Photonics than 25 different wavelengths often with only counted but the time of detection link.spectroscopyeurope.com/28-01-110 both SLM (single-longitudinal mode) or is also determined based on a refer - non-SLM spectral options. Four options ence signal. Here, a laser pulse gener- Raman-AFM system of output types are available: free-space, ally serves as a reference. This method Renishaw have added Bruker’s MM, SM and PM fibre. All types of lasers is a statistical counting method. TCSPC Dimension Icon atomic force microscope are offered in the same matchbox-size is used in particular in fluorescence life- to the range of instruments supported by enclosure, unified +5 V DC power input time measurements. This method is their inVia confocal Raman microscope. and serial control interface. All lasers often compared to a stop watch: a laser The combined instrument has a flexible with minor exceptions feature more than pulse excites a sample (time start); arm linking the two instruments, which 500 : 1 polarisation contrast or >20 dB just a few pico- or nanoseconds later, a couples light between them with mirrors. polarisation extinction ratio from a polar- “fluorescence photon” is released (time The arm uses Renishaw’s StreamLineHR isation-maintaining fibre. SLM lasers typi- stop). This time is recorded in a histo - high-resolution mapping technology that cally feature around 50 dB side-mode gram. After many start-stop passes, a can map areas up to 500 µm × 500 µm, suppression ratio (SMSR) with supe - conclusive histogram is created that with position encoders ensuring 100-nm rior centre wavelength stability of few displays the intensity of the fluorescence repeatability. Bruker’s PeakForce QNM picometres over the temperature range depending on time. The COUNT T is complements StreamLineHR by provid- from 15°C to 35°C. The lasers of the equipped with an avalanche photodiode ing even higher resolution nano-mechan- MatchBox2 series are self-contained, (active area of 150 µm) and features a ical information. however, a number of accessories are high detection efficiency of >80% and Renishaw offered to make test and installation a temporal resolution of up to 350 ps. In link.spectroscopyeurope.com/28-01-111 more convenient, especially in scientific addition to fluorescence lifetime meas- setups. urement (FLIM), the timing module is Integrated Optics used in time-resolved fluorescence and SAMPLE PREP link.spectroscopyeurope.com/28-01-107 single-molecule spectroscopy, as well as LIDAR applications. Dioxin and PCB cleanup High quantum efficiency Laser Components The GO-xHT clean-up system from Miura photocathode link.spectroscopyeurope.com/28-01-109 Institute of Environmental Science is avail- Photonis Netherlands has released able in Europe through Shimadzu. The a new photocathode designed to GO-xHT is targetted at dioxins and PCBs provide a combination of extremely RAMAN analyses of food and feedstuffs. It offers low dark counts, fast response time high throughput whilst saving solvents and a high quantum efficiency. Hi-QE OEM Raman and eliminating cross contamination: 1 mL Photocathodes are engineered to spectrometer extracts use less than 100 mL of solvent provide heightened spectral sensitivity The Eagle Raman-S spectrometer plat- per sample. Further, this consists of 85 mL in specific wavelength ranges for use in form from Ibsen Photonics is designed of hexane and 2 mL of toluene and no photon-starved applications. The Hi-QE for OEM integration into weak intensity dichloromethane. Three GO-xHT versions Photocathode increases quantum effi- Raman solutions at 785 nm or 830 nm are available with two, four or six columns ciency (QE) 50% when compared to excitation. It has low stray light level and in parallel. The system can be purchased www.spectroscopyeurope.com SPECTROSCOPYEUROPE 29 VOL. 28 NO. 1 (2016) NEW PRODUCTS

on/off variations from affecting readings. This provides long-term stability and precision up to 3× better than before. If speed is prioritised over precision, the new spectrometers can reduce measurement times achieving analyses of most samples within a few minutes. Operating software has been redesigned and the new TurboQuant II software analyses practically any unkown liquid, powder or Shimadzu are distributing the GO-xHT clean- The EDX Experiment Design and Execution solid sample. up system for dioxins and PCBs. Module from Symbion Systems. Spectro Analytical Instruments link.spectroscopyeurope.com/28-01-115 on its own or packaged with Shimadzu’s It allows spectra to be collected using GCMS-TQ8040 triple quadrupole mass Symbion LX, DX or RX and transferred Very large area CCD for spectrometer, the combination of which directly to Symbion QT chemometrics direct X-ray detection fulfills the new EU 589/2014 regulation without the need for duplicate manual Andor Technology has introduced the new on analysing dioxins and PCBs. data entry. During collection, the user iKon-XL “open-front” (‘SO’) 16-Megapixel Shimadzu can add any number of variables, create CCD camera platform for vacuum ultra- link.spectroscopyeurope.com/28-01-112 a spreadsheet of component values and violet (VUV) and soft X-ray direct detec- annotate a memo block for each spec- tion. This new camera platform is based trum. All of this information is transferred on back-illuminated CCD231-84 and SOFTWARE intact to Symbion QT to facilitate the CCD230-80 sensor variants from e2v building of any number of multivariate delivering up to 95% QE, with deep Online chemometrics calibrations. On deployment, EDX allows depletion options for enhanced sensitiv- MKS Data Analytics has released SIMCA- the user to add multiple trends based on ity in the 4–10 keV region, down to 2 e– online 14, which now has batch evolu- calculations such as peak height, peak read noise and up to 350,000 e– pixel tion OPLS that provides improved ratio or integral as well as chemometric well depth. Its maintenance-free deep-TE interpretation of process monitoring predictions. cooling down to –75°C minimises sensor and the evolution level process model. Symbion Systems dark current and enables the use of long An enhanced Control Advisor works link.spectroscopyeurope.com/28-01-114 exposure times for weak signal detec- with multiple phases for batch evolution, without the need for liquid nitrogen tion models, while supporting continu- or cryo-coolers. The iKon-XL “SO” is suited ous processes, to provide forecasts and X-RAY for large field-of-view laboratory-based or advice on how to improve the outcome synchrotron-based X-ray diffraction, X-ray of the process. An improved interface New line of ED-XRF plasma imaging and spectroscopy or X-ray includes login using Windows creden- spectrometers microscopy experiments. New “Extended tials through Active Directory, summaries Spectro Analytical Instruments have intro- Dynamic Range” technology allows simul- of all configurations into one view, and duced a new line of Spectro Xepos spec- taneous access to the lowest noise and groupings of continuous data by size and trometers. They include developments in maximum well depth within one scan colour for easier and faster visual interpre- adaptive excitation, and tube and detec- and is complemented by up to 18-bit tation. A dedicated simplified workflow tor technologies that improve sensitivity digitisation capability. Flexible connec- assists in updating models and keeping (often by 10× or more). The X-ray tubes tivity is standard through either USB 3.0 track of configurations and project files. in the new instruments remain powered or a long distance direct fibre-optic inter- MKS Data Analytics Solutions on between measurements to prevent face. Images can be read out at a range of link.spectroscopyeurope.com/28-01-113 readout speeds via either single or quad ports, the latter benefiting from the design Automation of attention given to electronic stability and data collection and image quadrant balancing. The iKon-XL chemometric model “SO” comes with a standard DN160CF/8” development CF/CF-203 rotatable flange and a metal- Symbion Systems has introduced the to-metal knife-edge sealing interface EDX Experiment Design and Execution for seamless integration with ultrahigh Module. The new module automates the vacuum chambers. process of spectroscopic data collection Spectro have introduced a new line of their Andor Technology and chemometric model development. Xepos ED-XRF spectrometers. link.spectroscopyeurope.com/28-01-116

30 SPECTROSCOPYEUROPE www.spectroscopyeurope.com VOL. 28 NO. 1 (2016) LITERATURE

Clinical applications Shimadzu has published an applica - tion handbook titled Clinical. Its near 140 pages cover 47 real life applications including Vitamin D, steroids, immuno- suppresants, catecholamines and amino acids analysis, as well as relevant technologies and solutions including chromatography, mass spectrometry, spectroscopy and life science instruments. The book can be downloaded for free from www. shimadzu.eu/clinical. Shimadzu link.spectroscopyeurope.com/28-01-120 Overcoming matrix effects eter are influenced. Such effects, which with XRF include absorption and enhancement, Steel analysis Spectro have published a white paper when taken collectively, are referred to Rigaku has published a new application explaining why overcoming matrix effects generally as matrix effects. For quality report describing the analysis of low- associated with X-ray fluorescence (XRF) control applications, when the sample alloy steel using a benchtop wavelength analysis is critical to achieving consistent matrix is known or can be matched, a dispersive X-ray fluorescence (WDXRF) high-accuracy results. Mitigating Matrix variety of standards-based XRF calculation spectrometer. Rigaku Application Note Effects with Advanced Spectra-Handling procedures are available to compensate # XRF 1042 details the quantitative anal- Functionality When Using XRF for High- for undesirable matrix effects. ysis of elements in low-alloy steels, with Accuracy Elemental Analysis is available However, creating the right basis complete information regarding sample to download. A great advantage of energy for consistently high-accuracy results preparation, method calibration and dispersive X-ray fluorescence (ED-XRF) requires additional spectra handling repeatability. Alloy steel is amalgamated analysis for rapid screening analysis is its functionality to determine the correct net with various elements to improve its ability to measure samples directly with intensities of the measured spectra. The mechanical properties. Steels comprised a minimum of preparation. Realising new white paper explains why this addi- of up to 8% alloying elements are called this benefit, however, requires eliminat- tional functionality is a critical aspect of low-alloy steels. Low-alloy steels are typi- ing potential errors that can result when overcoming matrix effects and ensuring cally designed to achieve better hard- atoms in the sample matrix influence those consistently high-accuracy results. enability. Their mechanical properties the fluorescence of others and thus the Spectro are determined by the concentrations intensities measured by the spectrom- link.spectroscopyeurope.com/28-01-122 of the different elements added to the steel, some of which are at very low levels. The concentrations of elements in molten steel are adjusted during the Silicon Drift Detectors steel making process, typically in electric 5.9 XRF Experimenter’s Kit keV OEM Components furnaces. Rapid analysis of the elemen- 55Fe tal composition is therefore essential. As eV FWHM

Counts 125 part of the control process, analyses of 25 mm2 x 500 µm 11.2 µs peaking time 6.4 slag and raw materials such as quicklime P/B Ratio: 20000/1 keV and ferroalloys are also required. X-ray Energy (keV) fluorescence spectrometers are routinely employed due to their rapid analysis Complete X-Ray Spectrometer FAST SDD® XRF System capabilities and their ability to measure both bulk metal and powders. For the Count Rate = >1,000,000 CPS analysis detailed in the report, certified Resolution Peaking standard reference materials of low-alloy Time steel provided by NIST and JSS (Japan 125 eV FWHM 4 µs Steel Standard) were used to establish 130 eV FWHM 1 µs the calibration, and measurements were ® 140 eV FWHM 0.2 µs performed using the Supermini200 with 160 eV FWHM 0.05 µs a 200 W Pd target X-ray tube. Rigaku www.amptek.com link.spectroscopyeurope.com/28-01-121 www.spectroscopyeurope.com SPECTROSCOPYEUROPE 31 VOL. 28 NO. 1 (2016) DIARY

17–21 April, Baltimore, Maryland, USA. 10–11 May, Monheim, Germany. Young Conferences Photonics Innovations for Advanced Investigators in Lipid Science Meeting. 2016 Technologies and Applications.  steph-  www.dgfett.de/meetings/aktuell/ [email protected],  http://spie.org/x94875. duesseldorf2016/#ort. xml. 13–17 March, San Diego, California, USA. th 251st American Chemical Society National 13–14 May, Tokyo, Japan. 16 Symposium Meeting.  [email protected],  www. 18–20 April, Liverpool, UK. Third on Molecular Spectroscopy.  http://regu- chemistry.org. International Glow Discharge Spectroscopy lus.mtrl1.info.hiroshima-cu.ac.jp/~molspec/e- Symposium (IGDSS2016). Peter Robinson, index.html. 14–18 March, Baltimore, Maryland, USA.  [email protected],  www.ew-gds. American Physical Society March Meeting. com. 14–17 May, Zlatni Pyasatsi, Bulgaria. Developments and Applications of Solid  www.aps.org/meetings/march. 18–21 April, Moama, New South Wales, State NMR to Materials Science, Chemistry 17 March, York, UK. RSC NMR Discussion Australia. 17 th Australian Near Infrared and Engineering Conference.  www.zing- Group (NRMDG) Spring Meeting: In situ Spectroscopy Group (ANISG) / New conferences.com/conferences/solid-state- Monitoring by NMR: What is it all About? Zealand NIRS Society (NZNIRSS) nmr.  www.nmrdg.org.uk. Conference.  www.anisg.com.au/confer- 18–20 May, Lisbon, Portugal. CEM 2016.  ence-2016. 20–22 March, Long Beach, California, USA. www.cem.uk.com. 19 April, Wotten-under-Edge, UK. Inside OSA Optics and Photonics Congress: 19–20 May, Berlin, Germany. 4 th Raman UK Seminar.  www.renishaw.com/ High-Brightness Sources and Light-Driven International Conference on Advanced Interactions.  www.osa.org/en-us/osa_ en/inside-raman-uk-seminar--18268. Applied Raman Spectroscopy (RamanFest org/highbrightnessopc. 20–25 April, Baltimore, Maryland, USA. 2016).  http://ramanfest.org. 21–23 March, Cambridge, UK. Faraday SPIE Next Generation Spectroscopic 23–25 May, Ormylia, Chalkidiki, Greece. Discussion: Advanced Vibrational Technologies IX.  http://spie.org/SIC/ 12 th Biennial International Conference Spectroscopy For Biomedical Applications. conferencedetails/next-generation-spectro- of the Infrared and Raman Users Group  www.rsc.org/events/detail/16963/ scopic-technologies. (IRUG12).  [email protected], advanced-vibrational-spectroscopy-for- th  http://www.irug.org. biomedical-applications-faraday-discussion. 22–26 April, Xi’an, China. 11 International Conference Series on Laser-Light and 25–26 May, London, UK. Analytical Lab 28 March–1 April, Phoenix, Arizona, USA. Interactions with Particles (LIP 2016). Dr Informatics 2016.  http://massinformat- Materials Research Society Spring Meeting. Jiajie Wang,  [email protected],  ics.co.uk/ali/  www.mrs.org/spring2016. http://lip2016.csp.escience.cn. 25–28 May, Antalya, Turkey. 3rd International 31 March–2 April, Eger, Hungary. European 24–27 April, Gothenburg, Sweden. Conference on New Trends in Symposium on Atomic Spectrometry (ESAS European Conference on Nonlinear Chemometrics and Applications (NTCA 2016). Ms Beatrix Schenker, c/o Hungarian Optical Spectroscopy (ECONOS 2016). 2016). Dr Remziye Güzel,  chemometrics@ Chemical Society, H-1015 Budapest, Hattyú u. Stina Hemdal,  econos2016@chalmers. ankara.edu.tr,  http://ntca.ankara.edu.tr. 16. II/8. Hungary,  [email protected], se,  www.chalmers.se/en/conference/  www.esas2016.mke.org.hu. econos2016. 6–9 June, San Francisco, California, USA. BIO International Convention (BIO 2016).  1–6 April, San Diego, California, USA. 25–28 April, Fort Lauderdale, Florida, USA. [email protected],  http://convention.bio. Experimental Biology 2016.  eb@faseb. OSA Optics and Photonics Congress: org/2016. org,  http://experimentalbiology.org. Biomedical Optics.  www.osa.org/en-us/ meetings/optics_and_photonics_congresses/ 5–9 June, Halifax, Nova Scotia, Canada. th 3–7 April, Colchester, UK. The 49th biomedical_optics. 99 Canadian Chemistry Conference and Annual International Meeting of the ESR Exhibition (CSC 2016).  www.csc2016.ca. Spectroscopy Group of the Royal Society 2–3 May, Seattle, Washington, USA. Center 5–10 June, San Jose, California, USA. of Chemistry. Dr Dima Svistunenko,  for Process Analysis & Control (CPAC) Conference on Lasers and Electro-Optics [email protected],  www.esr-group.org/ Spring Meeting. Mel Koch,  kochm@ (CLEO).  www.cleoconference.org. conferences/2016-conference-essex. uw.edu,  http://cpac.apl.washington.edu/ event/2016+CPAC+Spring+Meeting. 5–9 June, San Antonio, Texas, USA. 64th 4–7 April, Brussels, UK. SPIE Photonics ASMS Conference on Mass Spectrometry.  2–6 May, Lille, France. European Materials Europe 2016. http://spie.org/conferences-  [email protected],  www.asms.org. and-exhibitions/photonics-europe. Research Society (E-MRS) 2016 Spring Meeting.  www.european-mrs.com/ th rd 5–8 June, Loen, Norway. 8 Nordic 6–7 April, Prague, Czech Republic. 3 meetings/2016-spring-meeting. Conference on Plasma Spectrochemistry. FoodIntegrity Conference.  foodinteg-  [email protected],  www. [email protected].  www.foodintegrity2016. 8–11 May, Bagnols-sur-Cèze, France. EMAS th nordicplasma.com. eu. 2016—12 Regional Workshop on Electron Probe Microanalysis of Materials Today— 6–9 June, St Augustine, Florida, USA. 17 th 10–15 April, Pittsburgh, Pennsylvania, USA. Practical Aspects.  luc.vantdack@uant- International Workshop on Physical th 57 Experimental Nuclear Magnetic werpen.be,  www.microbeamanalysis.eu/ Characterization of Pharmaceutical Solids Resonance Conference.  www.enc-confer- events. (IWPCPS-17).  www.assainternational. ence.org. com/workshops/iwpcps-17. 9–13 May, Mendoz, Argentina. International 10–15 April, Oxford, UK. IBBI 2016—9th Meeting on Photodynamics and Related 6–10 June, Barcelona, Spain. XVI Conference on Isolated Biomolecules and Aspects.  http://photodynamics9.wix.com/ Chemometrics in Analytical Chemistry.  Biomolecular Interactions.  www.ibbi- phd9#!program/cfcg. [email protected],  www.cacbarcelona. conference.org. com. 9–13 May, Poznan, Poland. The 24th Annual 16–19 April, Salt Lake City, Utah, USA. World Forum on Advanced Materials.  6–10 June, Berlin, Germany. 7th International American Physical Society April Meeting. [email protected],  www. Conference on Spectroscopic Ellipsometry  www.aps.org/meetings/april. polychar24.divisia.pl. (ICSE-7).  http://www.icse-7.de.

32 SPECTROSCOPYEUROPE www.spectroscopyeurope.com VOL. 28 NO. 1 (2016) DIARY

6–9 June, Fort Myers, Florida, USA. 14th 3–6 July, Chamonix Mont-Blanc, France. 25–28 July, Heidelberg, Germany. OSA Pharmaceutical Powder X-ray Diffraction 6th International Conference of the Optics and Photonics Congress: Imaging Symposium (PPXRD-14).  www.icdd.com/ International Association for Spectral and Applied Optics.  www.osa.org/en-us/ ppxrd. Imaging (IASIM 2016).  http://iasim16. meetings/optics_and_photonics_congresses/ sciencesconf.org. imaging_and_applied_optics. 7–10 June, Karlsruhe, Germany. 13th International Conference on the 4–6 July, Liverpool, UK. 18th Biennial 30 July–5 August, Chambersburg, Applications of Magnetic Resonance in National Atomic Spectroscopy Symposium Pennsylvania, USA. 18th International Diffuse Food Science (FOODMR 2016). Leo Nick, (18th BNASS).  www.rsc.org/events/ Reflectance Conference (IDRC 2016).  Conference Bureau, Forschungs-Gesellschaft detail/19910. www.idrc-chambersburg.org. Verfahrens-Technik e.V. Theodor-Heuss-Allee th th 25, D-60486 Frankfurt am Main, Germany, 10–14 July, Stony Brook, New York, USA. 14 1–5 August, Rosemont, Illinois, USA. 65  [email protected],  https://mrfood2016. International Conference on Surface X-ray Annual Denver X-ray Conference (2016 gvt.org. and Neutron Scattering (SXNS14). Gretchen DXC).  www.dxcicdd.com. Cisco,  [email protected],  www.bnl.gov/ th 9–15 June, Novosibirsk, Russia. 12 th sxns14. 14–19 August, Fortaleza, Brazil. 25 International GeoRaman Conference.  International Conference on Raman [email protected],  http://geora- 16–17 July, Biddeford, Maine, USA. Gordon Spectroscopy (ICORS 2016).  www. man2016.igm.nsc.ru. Research Seminar on Vibrational icors2016.org. Spectroscopy.  www.grc.org/programs. 12–15 June, Todi, Italy. 6th CMA4CH aspx?id=16690. 14–19 August, West Dover, Vermont, USA. Mediterraneum Meeting.  http://www. Gordon Research Conference on Molecular cma4ch.org/index2.html. 17–22 July, Santa Fe, New Mexico, USA. OSA Structure Elucidation.  www.grc.org/ Topical Meeting: International Conference programs.aspx?id=17262. 19–24 June, Andover, New Hampshire, on Ultrafast Phenomena.  www.osa.org/ USA. Gordon Research Conference on en-us/meetings/topical_meetings/interna- 20–26 August, Toronto, Canada. 21st Multiphoton Processes.  www.grc.org/ tional_conference_on_ultrafast_phenomen. International Mass Spectrometry programs.aspx?id=11703. Conference.  [email protected],  17–22 July, Biddeford, Maine, USA. Gordon www.imsc2016.ca. 19–24 June, Ludwigsburg, Germany. Research Conference on Vibrational Spectroscopies in Novel Superconductors Spectroscopy.  www.grc.org/programs. 21–25 August, Philadelphia, Pennsylvania, (SNS 2016).  www.fkf.mpg.de/SNS2016. aspx?id=12221. USA. American Chemical Society National Fall Meeting & Exposition.  www.acs.org. 19–24 June, Torun, Poland. 23rd International 17–22 July, Breckenridge, Colorado, USA. 58th Conference on Spectral Line Shapes.  Annual Rocky Mountain Conference on 21–24 August, Los Angeles, California, USA. th [email protected],  http://icsls23.fizyka. Magnetic Resonance.  www.rockychem. 8 Workshop on Hyperspectral Image and umk.pl. com. Signal Processing: Evolution in Remote Sensing (WHISPERS).  www.ieee-whis- 19–24 June, Gothenburg, Sweden. European 17–23 July, Sanya, Hainan Island, China. pers.com. Conference on X-Ray Spectrometry 24th Annual International Conference on th (EXRS2016). EXRS2016, Sweden MEETX Composites/Nano Engineering (ICCE- 24–26 August, Reims, France. 14 Biennial AB, SE-412 94 Gothenburg, Sweden,  24). Professor David Hui, Department of HITRAN Database Conference combined th [email protected],  www.exrs2016.se. Mechanical Engineering, University of New with the 13 Atmospheric Spectroscopy Orleans, New Orleans, LA 70148, USA,  Applications (ASA) Meeting. Maud Rotger, 20–24 June, Champaign-Urbana, Illinois, [email protected],  www.icce-nano.org.  [email protected],  www.univ- USA. 71st International Symposium on reims.fr/site/evenement/asa-hitran/home- Molecular Spectroscopy. Birgit D. McCall, 17–21 July, Istanbul, Turkey. World Polymer accueil,18642,32042.html. International Symposium on Molecular Congress (MACRO 2016). Prof. Yusuf Spectroscopy, University of Illinois, 306B Yagci,  [email protected],  http:// 28 August–1 September, San Diego, California, Noyes Laboratory, 505 South Mathews macro2016.org/default.asp. USA. SPIE Optics + Photonics.  http:// Avenue, Urbana, IL 61801, USA.  birgit@ spie.org/x30582.xml. 18–22 July, Mauritius. International isms.illinois.edu,  http://isms.illinois.edu. Conference on Pure and Applied Chemistry 28 August–2 September, Irkutsk, Russia. Asia- 26–29 June, Inuyama, Japan. 7th International (ICPAC 2016).  [email protected],  Pacific EPR/ESR Symposium (APES 2016). Workshop on Plasma Spectroscopy, IPS http://sites.uom.ac.mu/icpac2016. Dr Dmitriy Polovyanenko,  apes2016@ 2016.  [email protected]. nioch.nsc.ru,  www.apes2016.org. 19–21 July, Hamburg, Germany. International ac.jp,  www.ips2016inuyama.com. Symposium on Environmental Analytical 30 August–3 September, Prague, Czech th 26–30 June, Montreal, Canada. SPEC 2016. Chemistry (ISEAC39) “Environmental Republic. 24 International Conference on  www.spec2016.com. and Food Monitoring”. Prof. Dr Jose High Resolution Molecular Spectroscopy Broekaert, University of Hamburg, Institut (PRAHA2016). Professor Štepán Urban, 27 June–1 July, St Petersburg, Russia. Laser für Anorganische und Angewandte Chemie, University of Chemistry and Technology, Optics 2016.  conference2016@laserop- Lehrstuhl für Analytische Chemie heterogener Faculty of Chemical Engineering, Technická tics.ru,  www.laseroptics.ru. Systeme, Martin-Luther-King Platz 6, 20146 5, CZ-16628 Praha 6, Czech Republic,  Hamburg, Germany.  jose.broekaert@ [email protected],  www.chem.uni- 29 June–1 July, Groningen, The Netherlands. chemie.uni8hamburg.de,  www.iaeac.com. wuppertal.de/conference. 8th International Conference on Coherent Multidimensional Spectroscopy (CMDS 24–28 July, Columbus, Ohio, USA. 31 August–2 September, Edinburgh, UK. Ultra 2016).  www.cmds2016.org. Microscopy & Microanalysis 2016.  www. Fast Imaging of Photochemical Dynamics. microscopy.org/MandM/2016.  www.rsc.org/events/detail/19765. 3–7 July, Aarhus, Denmark. EUROMAR 2016. Prof. Thomas Vosegaard, Interdisciplinary 25–27 July, Glasgow, UK. BSPR 2016 BSR 4–8 September, Torino, Italy. 10th Conference Nanoscience Center and Department of 2016 — Proteomic Approaches to Health of the European Federation of EPR Groups Chemistr, Aarhus University, Denmark,  tv@ and Disease. Karl Burgess,  karl.burgess@ (EFEPR).  [email protected],  www. chem.au.dk,  http://euromar2016.org. glasgow.ac.uk. efepr2016.unito.it. READER SERVICE SPECTROSCOPY europe

Use this page to register for a free subscription to Spectroscopy Europe You can also register online at the address below.

Online version at http://www.spectroscopyeurope.com/register

Your Details

Title (Professor/Dr etc.): City:

First Name: State/Region:

Last Name: Postal Code:

Job Title: Country:

Organisation Name: E-mail:

Address:

Please send me/continue to send me a free copy of Spectroscopy Europe. Signed: ______

Field of Work: Techniques: Area of employment: (please tick ALL that apply) (please tick ALL that apply) (please tick ONE only) Atomic Absorption Agriculture Industry Atomic Emission Analytical Chemistry Independent Lab Chemometrics Biotechnology University/Institute Computers/Automation Chemicals Hospital ICP/MS Electronics/Semiconductors Government Infrared Energy and Fuels Research Inst/Foundation Laser Spectroscopy Environmental Other______Luminescence/Fluorescence Food Mass Spectrometry Job function: (please tick ONE only) Genomics and other -omics Microscopy and Imaging Analyst Instrumentation MRI Quality Control/Assurance Life Science Near Infrared Research Scientist Materials Science NMR, ESR, EPR Lab Manager Medical Sciences Photonics and Optics University Professor Metals and Minerals Raman Engineering/Design Nanotechnology Separation Science Manufacturing/Processing Pharmaceuticals Surface Analysis Teaching Polymers and Peptides UV/Vis Marketing/Sales Security and Forensics X-Ray Diffraction Other______Water X-Ray Spectrometry None

Do you recommend, specify or authorise the purchase of spectroscopic instrumentation or services: Yes No

Your Personal Data (we will never pass your personal details to any third party) John Wiley & Sons Ltd and IM Publications LLP will use the information you have provided to fulfil your request. In addition, we would like to: 1. Contact you by post with information and offers from relevant companies (your details are never disclosed to them). Tick here if you do NOT want to receive this: 2. Contact you by e-mail with information and offers from relevant companies (your e-mail address and details are never disclosed to them). Tick if you WANT to receive this: 3. Your registration to this website includes a regular e-mail newsletter alerting you to the latest content. Tick if you do NOT want to receive this:

Now fax this page to +44-1243-811711, or return by mail to: Spectroscopy Europe, 6 Charlton Mill, Charlton, Chichester, West Sussex PO18 0HY, UK. VOL. 28 NO. 1 (2016) DIARY

11–16 September, Antwerp, Belgium. 5th 8–9 November, Berlin, Germany. 6th Annual International Conference on Vibrational Lab Automation & Robotics (ELA 2016).  Courses Optical Activity. Prof. Dr Christian https://selectbiosciences.com/conferences/ 2016 Johannessen, University of Antwerp, index.aspx?conf=LABAR2016. Groenenborgerlaan 171, B-2020 Antwerp, 29 February–4 March, Gembloux, Belgium. Belgium,  [email protected],  http:// 14–16 November, Somerset, New Jersey, Annual Spectroscopy and Chemometrics www.voa5.org. USA. Eastern Analytical Symposium and training. Juan Antonio Fernández Pierna,  Exposition (EAS 2016).  www.eas.org.  11–14 September, Dresden, Germany. [email protected], www.cra. th wallonie.be/en/events/training-in-vibrational- 20 European Symposium on Polymer 22–25 November, Taipei, Taiwan. 17 th Spectroscopy (ESOPS 20).  www. spectroscopy-and-chemometrics. International Symposium on Luminescence ipfdd.de/en/events/conferences-and- workshops/20th-european-symposium-on- Spectrometry (ISLS 2016).  www.isls2016. 5–8 April, Matera, Italy. European Fourier polymer-spectroscopy-esops. com.tw. Transform Mass Spectrometry Workshop.  www.eftms2016.it. 12–16 September, Chamonix-Mont Blanc, 30 November–3 December, Kagoshima, France. 9th International Conference on Japan. ANS2016: The 5 th Asian NIR 13–16 June, Todi, Italy. Multivariate Analysis Laser Induced Breakdown Spectroscopy Symposium. Prof. Satoru Tsuchikawa, Course, School for Novices.  http://www. (LIBS 2016). Conference office, Université Graduate School of Bioagricultural Sciences, cma4ch.org/frames-cou.html. Claude Bernard Lyon 1, Cellule Congrès de Nagoya University, Japan,  office@ans2016. 13–17 June, Warwick, UK. Interpretation l’UCBL, LIBS 2016, 43, bd du 11 Novembre org,  http://ans2016.org/index.html. 1918, 69622 Villeurbanne Cedex, France. of Infrared and Raman Spectra. James A.  [email protected],  www. de Haseth, IR Courses, Inc., 165 Sunnybrook libs2016-france.org/en. 2017 Drive, Athens, Georgia 30605-3347, USA,   th [email protected], www.ircourses. 13–15 September, Eastbourne, UK. 37 5–10 March, Chicago, Illinois, USA. The org/schedUK.html. Annual Meeting of the British Mass Pittsburgh Conference on Analytical Spectrometry Society (BMSS 2016).  Chemistry and Applied Spectroscopy 11–15 July, Brunswick, Maine, USA. Infrared www.bmss.org.uk/meetings.shtml. (Pittcon 2017).  http://www.pittcon.org. Spectroscopy I: Interpretation of Infrared and Raman Spectra.  www.ircourses.org. 18–23 September, Santa Fe, New Mexico, 5–9 June, Copenhagen, Denmark. ICNIRS USA. 2016 SciX Conference (formerly 2017.  [email protected],  FACSS): Annual National Meeting of the Society for Applied Spectroscopy (SAS) / www.icnirs2017.com. The 43rd Annual North American Meeting Exhibitions 7–10 July, Victoria, Canada. 9th International of the Federation of Analytical Chemistry Symposium on Two-Dimensional 2016 and Spectroscopy Societies.  facss@facss. Correlation Spectroscopy (2DCOS-9).  org,  www.facss.org. 20–23 March, Dubai, United Arab Emirates. www.icavs.org/icavs-9. ARABLAB 2016.  www.arablab.com. 25–30 September, Copenhagen, Denmark. 11–16 July, Victoria, Canada. 9th International 1st International Conference on Infrared, 12–14 April, Moscow, Russia. Analitika 2016. Millimeter, and Terahertz Waves (IRMMW- Conference on Advanced Vibrational  www.analitikaexpo.com. THz 2016).  www.irmmw-thz2016.org. Spectroscopy (ICAVS-9).  www.icavs.org. 25–28 April, Dalian, China. AnalytiX—Annual 26–28 September, Ulm, Germany. 13th 8–13 September, Santa Fe, New Mexico, USA. Conference and Expo of AnalytiX 2016.  Confocal Raman Imaging Symposium. Dr 2017 SciX Conference (formerly FACSS): www.bitcongress.com/Analytix2016. Karin Hollricher, WITec GmbH, Lise-Meitner-Str. Annual National Meeting of the Society 6, 89081 Ulm, Germany,  Karin.Hollrichr@ th for Applied Spectroscopy (SAS) / The 44 10–13 May, Munich, Germany. Analytica witec.de,  http://www.witec.de. Annual North American Meeting of the 2016—25th International Trade Fair 12–13 October, Coventry, UK. Photonex Federation of Analytical Chemistry and for Laboratory Technology, Analysis, 2016 Exhibition Roadshow & Conference. Spectroscopy Societies.  [email protected], Biotechnology and Analytica Conference.  www.photonex.org/index.php.  www.facss.org.  http://www.analytica.de. ADVERTISERS INDEX

Amptek...... 31 Analytica...... 3 Edinburgh Instruments...... 9 IM Publications...... 14 Jeol...... 2 Shimadzu...... 5 Specac...... 36 Starna Scientific...... 19 www.spectroscopyeurope.com SPECTROSCOPYEUROPE 35

As simple as‘spot’, click, read, repeat.

ecac Meet Sp con 2016 Pitt 33 › Booth 33 S 2016 ›AC 35 Booth 10 Lab 2016 Arab 4 › Stand 54

The NEW Specac Pearl® liquids analyser ● The smartest, easiest-to-use FTIR liquid sampling system available. ● Faster and more reliable than any traditional liquid cell - saving time and money. ● Accurate, repeatable pathlengths with wedging option, eliminating troublesome fringing. ● Different pathlengths and window materials, changed in seconds. ● Handles viscous materials with ease.

Pearl®- simpler, easier, faster

THE WORLD’S LEADING SPECTROSCOPIC ACCESSORY MANUFACTURER

UK: 01689 892902 | www.specac.com | US: +1 800 447 2558