HPLC–ICP-MS Analysis of Arsenic in Rice

October 2013 Volume 28 Number 10 www.spectroscopyonline.com

HPLC–ICP-MS Analysis of Arsenic in Rice Grating Selection for Raman Spectroscopy Measuring the Agreement Between Instruments Following Calibration Transfer FT-IR Analysis of Detergent Removal from Biological Samples

magentablackcyanyellow ES326484_Spec1013_CV1.pgs 09.26.2013 20:00 ADV Raman without complexity Once the domain of analytical chemistry and spectroscopy experts, Raman has become an important tool for ﬁ elds as diverse as art restoration, forensics, geology and pharmaceuticals. Thermo Scientific™ DXR™ Raman instruments lead the way in making this technique productive and accessible in more applications with an innovative design that virtually eliminates alignment and calibration, and proven software and databases that convert data to answers quickly and easily. Your focus is on answers. Ours is on better ways to get you there. or compromise

• focus on answers at thermoscientiﬁ c.com/raman

Thermo Scientiﬁ c™ DXR™ Raman Microscope Research grade performance with OMNIC software and patented data optimization

Thermo Scientifi c™ DXR™ SmartRaman Spectrometer Highly automated, push-button operated macro- sampling system © 2013 Thermo Fisher Scientifi c Inc. All rights reserved. All trademarks are All trademarks are c Inc. All rights reserved. © 2013 Thermo Fisher Scientifi c and its subsidiaries. of Thermo Fisher Scientifi the property

magentablackcyanyellow ES328092_SPEC1013_CV2_FP.pgs 09.27.2013 03:45 ADV the REMARKABLY BETTER ATOMIC SPECTROSCOPY revolution on

RUNS AIR

Introducing the NEW Agilent 4100 MP-AES. Say goodbye to fl ammable and expensive gases. Say hello to enhanced productivity. The Agilent 4100 Microwave Plasma-Atomic Emission Spectrometer runs entirely on air. It’s safer and more cost effi cient. It enables superior detection and unattended multi-element analysis. It’s the most signifi cant advancement in atomic spectroscopy in decades. In other words: It’s time to run on air.

JOIN THE REVOLUTION. Learn more about the NEW Agilent 4100 MP-AES. www.agilent.com/chem/RunsOnAir

magentablackcyanyellow ES328055_SPEC1013_003_FP.pgs 09.27.2013 03:43 ADV 4 Spectroscopy 28(10) October 2013 www.spectroscopyonline.com

® Raman MANUSCRIPTS: To discuss possible article topics or obtain manuscript preparation without complexity guidelines, contact the editorial director at: (732) 346-3020, e-mail: [email protected]. Publishers assume no responsibility for safety of artwork, photographs, or manuscripts. Once the domain of spectroscopy experts, Raman Every caution is taken to ensure accuracy, but publishers cannot accept responsibility for the information supplied herein or for any opinion expressed.

has become an important tool for ﬁ elds as diverse SUBSCRIPTIONS: For subscription information: Spectroscopy, P.O. Box 6196, Duluth, MN 55806-6196; (888) 527-7008, 7:00 a.m. to 6:00 p.m. CST. Outside the U.S., +1-218-740-6477. as art restoration, forensics and pharmaceuticals. Delivery of Spectroscopy outside the U.S. is 3–14 days after printing. Single-copy price: U.S., $10.00 + $7.00 postage and handling ($17.00 total); Canada and Mexico, $12.00 + $7.00 ™ ™ Thermo Scientiﬁ c DXR Raman instruments postage and handling ($19.00 total); Other international, $15.00 + $7.00 postage and handling make Raman productive and accessible with a design ($22.00 total). CHANGE OF ADDRESS: Send change of address to Spectroscopy, P.O. Box 6196, Duluth, MN that virtually eliminates alignment and calibration, 55806-6196; provide old mailing label as well as new address; include ZIP or postal code. Allow 4–6 weeks for change. Alternately, go to the following URL for address changes or and software and databases that convert data subscription renewal: https://advanstar.replycentral.com/?PID=581 RETURN ALL UNDELIVERABLE CANADIAN ADDRESSES TO: IMEX Global Solutions, P.O. Box to answers quickly and easily. Your focus is on 25542, London, ON N6C 6B2, CANADA. PUBLICATIONS MAIL AGREEMENT No.40612608. answers. Ours is on better ways to get you there. REPRINT SERVICES: Reprints of all articles in this issue and past issues are available (500 minimum). Call 877-652-5295 ext. 121 or e-mail [email protected]. Outside US, UK, direct dial: 281-419-5725. Ext. 121 or compromise DIRECT LIST RENTAL: Contact Tamara Phillips, (440) 891-2773; e-mail: tphillips@ advanstar.com

• thermoscientiﬁ c.com/raman INTERNATIONAL LICENSING: Maureen Cannon, (440) 891-2742, fax: (440) 891-2650; e-mail: [email protected].

ecycled R Pa % p 0 e r

0 e

- t

2 s

0 a % W P r o e st Consum

©2013 Advanstar Communications Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical including Thermo Scientiﬁ c™ DXR™ Raman Microscope by photocopy, recording, or information storage and retrieval without permission in writing Research grade performance with OMNIC from the publisher. Authorization to photocopy items for internal/educational or personal use, software and patented data optimization or the internal/educational or personal use of specific clients is granted by Advanstar Commu- nications Inc. for libraries and other users registered with the Copyright Clearance Center, 222 Rosewood Dr. Danvers, MA 01923, 978-750-8400 fax 978-646-8700 or visit http://www. copyright.com online. For uses beyond those listed above, please direct your written request to Permission Dept. fax 440-756-5255 or email: [email protected].

Advanstar Communications Inc. provides certain customer contact data (such as customers’ names, addresses, phone numbers, and e-mail addresses) to third parties who wish to promote relevant products, services, and other opportunities that may be of interest to you. If you do not want Advanstar Communications Inc. to make your contact information available to third parties for marketing purposes, simply call toll-free 866-529-2922 between the hours of 7:30 a.m. and 5 p.m. CST and a customer service representative will assist you in removing your name from Advanstar’s lists. Outside the U.S., please phone 218-740-6477.

Spectroscopy does not verify any claims or other information appearing in any of the adver- tisements contained in the publication, and cannot take responsibility for any losses or other damages incurred by readers in reliance of such content.

Spectroscopy welcomes unsolicited articles, manuscripts, photographs, illustrations and other materials but cannot be held responsible for their safekeeping or return.

To subscribe, call toll-free 888-527-7008. Outside the U.S. call 218-740-6477.

Advanstar Communications Inc. (www.advanstar.com) is a leading worldwide media company providing integrated marketing solutions for the Fashion, Life Sciences and Powersports industries. Advanstar serves business professionals and consumers in these industries with its portfolio of 91 events, 67 publications and directories, 150 electronic publications and Web sites, as well as educational and direct marketing products and services. Market leading brands © 2013 Thermo Fisher Scientifi c Inc. All rights reserved. All trademarks are All trademarks are c Inc. All rights reserved. © 2013 Thermo Fisher Scientifi c and its subsidiaries. of Thermo Fisher Scientifi the property and a commitment to delivering innovative, quality products and services enables Advanstar to “Connect Our Customers With Theirs.” Advanstar has approximately 1000 employees and currently operates from multiple offices in North America and Europe.

magentablackcyanyellow ES326235_Spec1013_004.pgs 09.26.2013 04:14 ADV ensitivity S

High-

High- R Speed esolution High-

The All-In-One Platform for FTIR Speed, Sensitivity, AND Resolution Shimadzu’s New IRTracer-100 Provides Exceptional Performance Across a Broad Range of Applications

With over 55 years’ experience producing Shimadzu’s IRTracer-100 features: FTIR spectrophotometers, Shimadzu ■ High Speed: Up to 20 spectra/second acquisitions has cultivated a reputation for quality, ■ High Sensitivity: Best-in-class 60,000:1 S/N ratio delivering maximum performance and value. ■ High Resolution: 0.25cm-1 resolution suitable for That quality continues. Introducing the high-precision gas analysis IRTracer-100, a next-generation FTIR system ■ Easy Maintenance: Automatic Dehumidifier and from Shimadzu that redefines excellence in Advanced Dynamic Alignment features FTIR analysis. ■ Outstanding Reliability: Self-diagnostics and validation routine

Learn more about Shimadzu’s IRTracer-100. Optimized for network applications, new LabSolutions IR Call (800) 477-1227 or visit us online at Software features an intuitive user interface as well as an www.ssi.shimadzu.com/Tracer extensive library of spectra and a high-performance search function. User-friendly macro functions automate routine work Order consumables and accessories on-line at http://store.shimadzu.com for enhanced work efficiency. In addition, numerous optional Shimadzu Scientific Instruments Inc., 7102 Riverwood Dr., Columbia, MD 21046, USA programs are available to address all modern laboratory needs.

magentablackcyanyellow ES328093_SPEC1013_005_FP.pgs 09.27.2013 03:45 ADV 6 Spectroscopy 28(10) October 2013 www.spectroscopyonline.com

Science Is ®

Complicated. PUBLISHING & SALES 485F US Highway One South, Suite 210, Iselin, NJ 08830 (732) 596-0276, Fax: (732) 647-1235

Michael J. Tessalone Science Group Publisher, [email protected] Edward Fantuzzi Publisher, [email protected] Stephanie Shaffer East Coast Sales Manager, [email protected] (508) 481-5885

EDITORIAL Laura Bush Editorial Director, [email protected] Megan Evans Managing Editor, [email protected] Stephen A. Brown Group Technical Editor, [email protected] Cindy Delonas Associate Editor, [email protected] Dan Ward Art Director, [email protected]

Russell Pratt Vice President Sales, [email protected] Anne Young Marketing Manager, [email protected] Tamara Phillips Direct List Rentals, [email protected] Wright’s Media Reprints, [email protected] Maureen Cannon Measurement Permissions, [email protected] Jesse Singer Production Manager, [email protected] Jason McConnell Should Be Easy. Audience Development Manager, [email protected] Gail Mantay Audience Development Assistant Manager, [email protected] Do more with the QuickTrace™ M-7600 CVAA Mercury Analyzer. Joe Loggia SSWGHWHFWLRQ Chief Executive Officer FDSDELOLW\ Tom Florio Chief Executive Officer Fashion Group, Executive Vice-President 5REXVWFRPSDFW Tom Ehardt Executive Vice-President, Chief Administrative Officer & Chief Financial Officer UHOLDEOH Georgiann DeCenzo Executive Vice-President

magentablackcyanyellow ES326234_Spec1013_006.pgs 09.26.2013 04:14 ADV magentablackcyanyellow ES328054_SPEC1013_007_FP.pgs 09.27.2013 03:43 ADV 8 Spectroscopy 28(10) October 2013 www.spectroscopyonline.com

    CONTENTS  

COLUMNS October 2013 Volume 28 Number 10 Molecular Spectroscopy Workbench ...... 12 Considerations of Grating Selection in Optimizing a Raman Spectrograph The physics that determine how gratings and spectrographs work are summarized in simple terms for new users of Raman equipment. Fran Adar

Chemometrics in Spectroscopy ...... 24 Calibration Transfer, Part IV: Measuring the Agreement Between Instruments Following Calibration Transfer The statistical methods used for evaluating the agreement between two or more instruments (or methods) for reported analytical results are discussed, with an emphasis on acceptable analytical accuracy and confidence levels using two standard approaches, standard uncertainty or relative standard uncertainty, and Bland-Altman “limits of agreement.” Jerome Workman, Jr. and Howard Mark

Cover image courtesy of ARTICLE Viktor Lugovskoy. A Rapid FT-IR-Based Method for Monitoring Detergent Removal from Biological Samples ...... 34 A novel infrared-based method that permits fast and impartial analysis of detergent removal from biological samples is presented. Ivona Strug, Sara Gutierrez, Amedeo Cappione III, Mayra Jimenez, Mary Jane Mullen, and ON THE WEB Timothy Nadler FACSS-SCIX PODCAST SERIES

Trends in Infrared Spectroscopic Imaging FEATURE An interview with Rohit Bhargava, winner of the 2013 Craver Award. Speciation Analysis Using HPLC–ICP-MS: Arsenic in Rice ...... 40 Bhargava discusses current trends in In a recent web seminar, Julian Tyson, professor of chemistry at the University of IR spectroscopic imaging, including application-specific instrumentation, Massachusetts, Amherst, explained how to develop a reliable method for arsenic speciation. improvements in data interpretation, Here, he answers questions raised during the seminar. and identifying relationships between structure and spectra.

Improving Raman Probes for DEPARTMENTS Cosmetic and Medical Research An interview with Dr. Paul D.A. Pudney of 11 Unilever Discover, Colworth Laboratory, News Spectrum ...... winner of the 2013 William F. Meggars EAS Conference Preview of Spectroscopy Sessions and New Features Award. Pudney discusses the work that his team published in Applied Spectroscopy by Justin Pennington, 2013 program chair on the development a new Raman probe that improves the measurement of skin. Products & Resources ...... 44

spectroscopyonline.com/podcasts Ad Index ...... 50

Like Spectroscopy on Facebook: www.facebook.com/SpectroscopyMagazine Spectroscopy (ISSN 0887-6703 [print], ISSN 1939-1900 [digital]) is published monthly by Advanstar Communications, Inc., 131 West First Street, Duluth, MN 55802-2065. Spectroscopy is distributed free of charge to users and specifiers of spectroscopic equip- Follow Spectroscopy on Twitter: ment in the United States. Spectroscopy is available on a paid subscription basis to nonqualified readers at the rate of: U.S. and pos- https://twitter.com/spectroscopyMag sessions: 1 year (12 issues), $74.95; 2 years (24 issues), $134.50. Canada/Mexico: 1 year, $95; 2 years, $150. International: 1 year (12 issues), $140; 2 years (24 issues), $250. Periodicals postage paid at Duluth, MN 55806 and at additional mailing offices. POSTMASTER: Send Join the Spectroscopy Group on LinkedIn address changes to Spectroscopy, P.O. Box 6196, Duluth, MN 55806-6196. PUBLICATIONS MAIL AGREEMENT NO. 40612608, Return http://linkd.in/SpecGroup Undeliverable Canadian Addresses to: IMEX Global Solutions, P. O. Box 25542, London, ON N6C 6B2, CANADA. Canadian GST number: R-124213133RT001. Printed in the U.S.A.

magentablackcyanyellow ES326494_Spec1013_008.pgs 09.26.2013 20:01 ADV magentablackcyanyellow ES328072_SPEC1013_009_FP.pgs 09.27.2013 03:44 ADV 10 Spectroscopy 28(10) October 2013 www.spectroscopyonline.com Editorial Advisory Board

Fran Adar Horiba Scientific Robert G. Messerschmidt Rare Light, Inc. Ramon M. Barnes University of Massachusetts Francis M. Mirabella Jr. Mirabella Practical Consulting Solutions, Inc. Matthieu Baudelet University of Central Florida John Monti Montgomery College Paul N. Bourassa Blue Moon Inc. Michael L. Myrick University of South Carolina Michael S. Bradley Thermo Fisher Scientific John W. Olesik The Ohio State University Deborah Bradshaw Consultant Jim Rydzak GlaxoSmithKline Kenneth L. Busch Wyvern Associates Jerome Workman Jr. Unity Scientific

Ashok L. Cholli Polnox Corporation Contributing Editors: David Lankin University of Illinois at Chicago, Fran Adar Horiba Scientific College of Pharmacy David W. Ball Cleveland State University Barbara S. Larsen DuPont Central Research Kenneth L. Busch Wyvern Associates and Development Howard Mark Mark Electronics Volker Thomsen Consultant Ian R. Lewis Kaiser Optical Systems Jerome Workman Jr. Unity Scientific Rachael R. Ogorzalek Loo University of California Los Angeles, David Geffen School of Medicine Spectroscopy’s Editorial Advisory Board is a group of distinguished individuals assembled to help the publication fulfill its editorial mission to promote the effec- Howard Mark Mark Electronics tive use of spectroscopic technology as a practical research and measurement tool. With recognized expertise in a wide range of technique and application areas, board R.D. McDowall McDowall Consulting members perform a range of functions, such as reviewing manuscripts, suggesting authors and topics for coverage, and providing the editor with general direction and Gary McGeorge Bristol-Myers Squibb feedback. We are indebted to these scientists for their contributions to the publication and to the spectroscopy community as a whole. Linda Baine McGown Rensselaer Polytechnic Institute

Silicon Drift Detector SDD OEM Components Experimenter’s Kit

5.9 keV 55Fe 125 eV FWHM Counts 25 mm2 x 500 µm

11.2 µs peaking time 6.4 P/B Ratio: 20000/1 keV

Complete X-Ray Spectrometer Energy (keV) Complete XRF System

® www.amptek.com

magentablackcyanyellow ES326316_Spec1013_010.pgs 09.26.2013 04:19 ADV www.spectroscopyonline.com October 2013 Spectroscopy 28(10) 11 News Spectrum EAS Conference Preview of Spectroscopy • Spectroscopy for Surface Science, Nano Materials, and Sessions and New Features Fiber Analysis • Spectroscopic Applications in Biologics Justin Pennington, 2013 program chair • Applications of NMR Spectroscopy from Small Molecules The Eastern Analytical Symposium (EAS) will be held this year to Large Assemblies from November 18 to 20, 2013, in Somerset, New Jersey. EAS • Spectroscopy at Surfaces and Interfaces is the second largest conference and exposition for laboratory • Spectroscopic Applications in Biomedical Sensing science in the United States dedicated to the needs of • Structural Studies by Magnetic Resonance Spectroscopy analytical chemists and those in the allied sciences. It has • What’s New in NIR Analysis? been a center of innovative science and practical solutions • Advanced Vibration Spectroscopy: Instrumentation and for the analytical community for more than 50 years. Offering Applications high quality technical sessions, short courses, workshops, • Spectroscopic Imaging for Dissolution and Pharmaceutical and seminars, the conference provides professional scientists Development and students with continuing education as they expand • Applications of NMR in the Pharmaceutical Industry their knowledge, broaden their networks, and advance their • Vibrate or Rotate — Techniques for Solving Problems with careers. A large exposition of laboratory instruments, supplies, Spectroscopy. and services related to the science is held concurrently with Short courses, taught by experts, emphasize practical the symposia to meet all the needs of attendees. knowledge of a variety of topics to help analysts keep current The conference covers a broad analytical scope including with best practices and new techniques. Whether the need spectroscopy, chromatography, mass spectrometry, is to learn a new analytical technology, understand new microscopy, and chemometrics, as well as a variety of regulations, explore an entirely new analytical field, or just applications in pharmaceuticals, forensics, conservation brush up a new concept in one’s area of expertise, EAS invites science, and surface analysis. Spectroscopy is covered in attendees to take part in one of the spectroscopy-related many of the broader applied technical and award sessions. courses. In addition, there are several specifically applicable to More information about the conference may be found at spectroscopy, including www.eas.org or on LinkedIn. ◾ Market Profile: Terahertz Spectroscopy At the fringe of the infrared spectrum is the terahertz For laboratory applications, terahertz spectroscopy region, which until recently was not significantly utilized is quickly being adopted by major companies for both for spectroscopy. The properties of terahertz spectroscopy quality analysis and product development. Intel’s make it useful for applications in a variety of industries, with adoption of the technique validates its usefulness in the many major companies already adopting the technique. The semiconductor and electronics industries. A number entrance of new competitors and very strong growth of the of major pharmaceutical manufacturers are now technique portend its growth into a major segment of the using the technique for both development and QA–QC molecular spectroscopy market. of solid dosage forms. There is The terahertz region of the significant exploration of terahertz

electromagnetic spectrum lies 25% Semiconductor, electronics, spectroscopy for clinical and between the far infrared region 6% nanotech - 25% medical applications, although Pharmaceuticals - 23% and the microwave region, and 11% 23% much of this work is currently Academia - 21% includes wavelengths of about 0.1– Hospitals and medical based in academic laboratories. 1.0 mm. Terahertz spectroscopy 14% centers, clinical - 14% Global laboratory demand for 21% Government - 11% is somewhat complementary to Other - 6% terahertz spectroscopy was nearly other spectroscopy techniques, $20 million in 2012, with at as a number of compounds have least six major competitors now unique fingerprints in the terahertz offering commercial instruments. region, including some narcotics, Laboratory terahertz spectroscopy demand by industry for 2012. Although 2013 may be a challenging explosives, and various polymorphic forms of active environment for this market, demand is still likely to see pharmaceutical ingredients (APIs). A variety of common mid single-digit growth before returning to strong double- materials, as well as human tissue, are semitransparent digit growth in 2014. to terahertz radiation, and it is non-ionizing, making it The foregoing data were based on SDi’s market analysis safe for human exposure. A wide variety of commercial and perspectives report entitled Global Assessment terahertz spectrometers are already on the market, including Report, 12th Edition: The Laboratory Life Science and conventional frequency-domain systems, time-domain Analytical Instrument Industry, October 2012. For more systems, imaging systems, and portable instruments. information, visit www.strategic-directions.com.

magentablackcyanyellow ES326242_Spec1013_011.pgs 09.26.2013 04:15 ADV 12 Spectroscopy 28(10) October 2013 www.spectroscopyonline.com

Molecular Spectroscopy Workbench Considerations of Grating Selection in Optimizing a Raman Spectrograph

The performance of a Raman spectrograph for a particular application will depend, among other things, on its sensitivity and spectral resolution. The sensitivity will determine how long it will take to record a spectrum with a given signal-to-noise ratio. In turn, the grating reflectivity will determine the optical throughput of the instrument. The spectral resolution will determine how easy it will be to extract subtle information from a spectrum. The spectral resolution is determined by the focal length of the spectrograph and the groove density of the grating used to disperse the light, and will also affect the apparent sensitivity. Note that in many cases, spectral resolution may be improved at the expense of sensitivity. Because many new users of Raman equipment are not familiar with these grating–spectrograph properties, we thought it would be useful to summarize the physics, in sim- plistic terms, that determine how the instruments work.

Fran Adar

ver the past 20 years Raman spectroscopy has Spectrograph Design Ogained popularity because instrumental in- A spectrograph is designed to accept light with many wave- novations have made it easier to use them for lengths, separate the wavelengths in space, and then ÒdetectÓ problem-solving, providing spectra in 1% of the time each wavelength on a multichannel detector, which today is that it took prior to the Raman revolution following the synonymous with a charge-coupled device (CCD). Figure 1 introduction of the holographic notch filters. In addi- is a schematic of a spectrograph. tion, new graduates of chemistry and materials science Figure 1 shows a typical Raman spectrograph. The col- are taking jobs in industry without any graduate educa- lected Raman light is focused onto an entrance slit. After tion in spectroscopy or the operation of spectroscopic passing the slit, it diverges until it reaches a concave mirror instrumentation. Consequently, the new user is faced whose focal length corresponds to the distance between the with myriad choices in configuring a new instrument mirror and slit; after being reflected by the mirror, the light or in optimizing an existing instrument for a given ex- is Òcollimated.Ó When the light hits the grating, which is an periment. Understanding how the spectrograph core of array of finely spaced lines on a reflective surface, there is a Raman instrument works will aid the novice in pro- constructive and destructive interference, which is wave- ducing quality, defensible results for solving industrial length and angle dependent. Consequently, each wavelength problems or characterizing new materials. is reflected at a different angle (1). As each wavelength is

magentablackcyanyellow ES326246_Spec1013_012.pgs 09.26.2013 04:16 ADV PinAAcle 900 AA Series is a registered trademark of PerkinElmer, Inc. All other trademarks are the property of their respective owners. ® ©2011-2012 PerkinElmer,©2011-2012 Inc. 400260_01. All rights reserved. PerkinElmer

Discover T e PinAAcle Of Performance In AA. Engineered around cut ing-edge f ber-optic technology for enhanced light throughput and superior detection limits, the new PinAAcle™ 900 AA series is the latest innovation from the world leader in atomic absorption. Available in f ame, furnace, or combination models, PinAAcle instruments of er exactly the level of performance you need with the smallest footprint of any combined f ame/graphite furnace AA system on the market. Experience peak performance and unmatched productivity. Step up to the PinAAcle 900 AA series from PerkinElmer.

Learn more by watching the video at www.perkinelmer.com/pinaacleNA1

magentablackcyanyellow ES328071_SPEC1013_013_FP.pgs 09.27.2013 03:44 ADV 14 Spectroscopy 28(10) October 2013 www.spectroscopyonline.com

Of course, the Raman spectrum is only meaningful when the wavelength Collimating values are converted to Raman shift mirror units, also in the software, according to equation 1: Entrance slit = ± = Raman Laser Vibration [1]

Camera where ν is derived from λ according to Grating mirror equation 2:

-1 (cm ) = 1/ (nm) [2]

Typical widths of lines in a Raman Array spectrum are between 1 and 10 cm-1 detector full width at half maximum (FWHM). If parameters are selected that produce Figure 1: Schematic of a dispersive Raman spectrograph. 1 cm-1/pixel, then about 1000 cm-1 can be covered on a CCD that has 1024 pixels in the long direction, which is the spectral dispersion direction. In prin- ciple, the selection of the grating would be straightforward, but as we will see, Grating there are important characteristics that have to be acknowledged in the choice. Just to get oriented to these effects,

i examine the behavior shown in Figure 1 -1 i 3 of the 155 cm line of sulfur that 2 α sinγa = kλ/[2cosα] was recorded with the 633-nm line of i α d f i i a HeNe laser on instruments whose 1 tanu1 = / = tan ( 1’- 2’)= γ γ tan[arcin{kλ1-sin( a+α)} - ( a+α)] focal length varied between 150 mm d where 1 is the distance and 1920 mm. Depending on the goal between pixels at λ1 and λ2 of the measurement, it may or may not Focusing u 1 be important to resolve the compo- element nents in the spectrum; how much will f be necessary to resolve will determine the selection of gratings as will be discussed in the following sections. λ2 λ1

Figure 2: Schematic showing how a pixel position is converted to a wavelength. Grating Characteristics The important grating characteristics then reflected from the camera mir- quantities determining the separation that have to be matched with the de- ror onto the array detector (CCD), it is on the camera of two wavelengths will sired instrument characteristics are focused at a different position on the be the incident angle of the light on • the dispersion, which is a result of array. The wavelength of the light on the grating, the diffracted angles, as the groove density (g/mm) and focal each array pixel can then be calculated determined by these equations, and length, and from the known equations of grating the focal length of the focusing ele- • the reflectivity, which is a compli- physics, as shown in Figure 2 (2). ment (2). When a Raman instrument cated function of the groove density, It is not the purpose of this article is designed, the spectral dispersion at a the groove profile (including the to derive the equations that determine given wavelength is selected, and then blaze angle), the angle at which the the conversion, but only to indicate to the angles and focal length are calcu- grating is used, and the metallic the new user the origin of the “magic” lated to produce the desired result. In coating (3,4). inside the software that enables the addition, there is optical software that spectral “image” on the camera to be enables asymmetrizing the geometry Dispersion converted to a spectrum, that is, a plot so that the images are kept as tight as It is important to recognize that be- of intensity (counts or counts/s) ver- possible on the surface of the detector, cause the relevant x-axis units in a sus Raman shift (cm-1). The physical which, of course, is flat. Raman spectrum are cm-1, the disper-

magentablackcyanyellow ES326248_Spec1013_014.pgs 09.26.2013 04:16 ADV HELPING YOU GUARD AGAINST THE NEXT MELAMINE

Urea. Cyanuric acid. Dicyandiamide. Melamine. Like you, Rob Packer wants to guard against the most common economic adulterants found in milk powder. But it’s the potential new adulterants out there that worry him – and should worry you. Our DairyGuard™ Milk Powder Analyzer screens for known and unknown economic adulterants in milk powder, as well as the nutrition parameters you measure today. With no sample prep, and no scientific background required. DairyGuard. Get prepared for what’s next. www.perkinelmer.com/dairyguard

magentablackcyanyellow ES328074_SPEC1013_015_FP.pgs 09.27.2013 03:44 ADV 16 Spectroscopy 28(10) October 2013 www.spectroscopyonline.com

sion is wavelength dependent, which will be illustrated in Figures 4a and 4b. Figures 4a and 4b, respectively, show dispersion curves for the same gratings in a 300-mm focal length spectrograph, in terms of nm and cm-1. We first examine the dispersion in wavelength units (nm/pixel). The dispersion curve for the 600-g/mm grating over the entire range shown (200–1775 nm) is almost flat, varying between 0.144 and 0.138 nm/pixel. On the other hand the behavior of

Intensity (arbitrary units) the other grating is quite different. For all values of λ > 1080 nm, the dispersion is pinned to 0 nm/pixel. 120 130 140 150 160 170 180 At these long wavelengths, a grating with this value for the groove density Raman shift (cm-1) will not diffract. To put it another way, at about 1080 nm the light exits Figure 3: 155 cm-1 line of crystalline sulfur recorded with instruments whose focal length varied the grating at a grazing angle and between 1920, 640, 460, and 250 mm (from top to bottom). (Courtesy of Sergey Mamedov of the diffraction angle for any longer Horiba Scientific.) wavelengths will be imaginary. In addition, over the usable range of the grating the dispersion is more or less flat between 200 and 600 nm, after (a) 600 1800 which the dispersion decreases at

0.16000 an accelerating rate. But all of these

0.14000 dispersion values are in nm/pixel, -1 0.144 0.144 and what we really want is cm /pixel, 0.12000 0.1420.1420.143 0.143 0.144 0.1440.144 0.144 which is shown in Figure 4b. To

/pix) 0.10000 -1 understand why these curves are so 0.08000 different, we need to calculate cm-1/Å 0.06000 0.048 which has a conversion factor of -1/ 0.0480.048 0.048 0.0470.047 0.045 0.04000 0.0410.038 λ2. Table I illustrates how rapidly this 0.018 Dispersion (cm 0.02000 factor is changing with wavelength. 0.00000 Inspection of Figure 4b indicates 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 -1 -0.02000 that the dispersion in cm /pixel is Wavelength (nm) changing rapidly for both gratings. If one is configuring a system for ~3 -1 (b) 600 1800 cm /pixel in the red (λ ~ 600 nm), a 600-g/mm grating will work quite fine, 100.00 but if there will be a laser emitting near 400 nm, the 1800-g/mm grating would provide better dispersion.

/pix) 10.00

-1 Note that these curves are specific for a particular focal length mono. In Figures 5a and 5b we compare a 300- 1.00 mm focal spectrograph to an 800-mm

Dispersion (cm focal spectrograph. An important point to note is that the long wave-

0.10 length termination is the same in both 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 systems because it is determined by the Wavelength (nm) angle for grazing exit diffraction, an inherent property of each grating, not Figure 4: Dispersion in (a) nm/pixel and (b) cm-1/pixel for an 1800- and 600-g/mm grating in a the focal length of the spectrograph in 300-mm focal length spectrograph. which the grating is being used.

magentablackcyanyellow ES326245_Spec1013_016.pgs 09.26.2013 04:16 ADV INTRODUCING MOXTEK®’S NEW

Tin multi-layer coating of uniform thickness

Ultra-thin polymer Carbon support flm structure

AP5 X-ray windows can be used anywhere an AP3 windows is used. With more solid angle and better transmission, you’ll want to see what this window can do. See what you’re missing, see the light.

• More solid angle • Maximum transmission of low energy x-rays • Higher count rate • High mechanical strength • sacuum Ɵght • Reliable • >ight reũecƟon

Visit www.moxtek.com for more information

452 West 1260 North / Orem, UT 84057 USA Toll Free/ 1.800.758.3110 / Local/ 1.801.225.0930 [email protected] / www.moxtek.com ISO 9001:2008

magentablackcyanyellow ES328057_SPEC1013_017_FP.pgs 09.27.2013 03:43 ADV 18 Spectroscopy 28(10) October 2013 www.spectroscopyonline.com

Table I: Dispersion in cm-1/ Å as a function of wavelength Table II: Dispersion/pixel at 300 nm for λ 200 300 400 500 600 700 800 900 1000 a 300 versus 800 mm focal length spec- nm nm nm nm nm nm nm nm nm trograph equipped with 2400 and 3600 g/mm gratings cm-1/Å 25.00 11.11 6.25 4.00 2.78 2.04 1.56 1.23 1.00 λ = 300 nm 2400 g/mm 3600 g/mm 300 mm FL 4.0 cm-1/pixel 2.5 cm-1/pixel 800 mm FL 1.5 cm-1/pixel 1.0 cm-1/pixel

2400 (a) 3600 Table III: Dispersion/pixel at 200 nm for 10.00 a 300 versus 800 mm focal length spectrograph equipped with 2400 and 3600 g/mm gratings λ = 200 nm 2400 g/mm 3600 g/mm /pix) -1 300 mm FL 9.0 cm-1/pixel 6.0 cm-1/pixel 1.00 800 mm FL 3.0 cm-1/pixel 2.0 cm-1/pixel

Dispersion (cm Reflectivity The grating reflectivity is a complicated function of the groove shape and spac- 0.10 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 ing, and the polarization of the light hit- Wavelength (nm) ting the grating. The article cited earlier (3) describes in detail the physics deter- 2400 (b) 3600 mining the grating behavior. Until the middle of the last century, at about the 10.00 time when lasers became available, gratings were produced by cutting a metal 1.00 surface with a diamond-tipped ruling /pix)

-1 engine. Such equipment produced flat facets on the grating. Within a few years 0.10 of the introduction of the laser it was recognized that it is possible to produce 0.01 gratings by exposing surfaces coated Dispersion (cm with photoresist to interfering laser beams. Major advantages of gratings 0.00 produced with this technology included Wavelength (nm) a virtual elimination of grating ghosts arising from defects in the ruling en- -1 Figure 5: Dispersion in cm /pixel of 2400- and 3600-g/mm gratings mounted in (a) a 300-mm gine, and much higher groove densities. focal length spectrograph and (b) an 800-mm focal length spectrograph. What was perhaps not recognized with the earliest gratings was that the quasi- Aside from the fact that the long Keeping in mind that the sharp- sinusoidal groove profile produced by wavelength cut-offs are the same in the est feature that can be observed has the holographic process exaggerated two systems, it is also important to note a half width of 2 pixels, one can see the ripples in the reflectivity curves that the range of use for these gratings is how a short focal length instrument called Wood’s anomalies. However, the somewhat limited in the visible part of can begin to limit what can be differ- subsequent use of ion etching provided the spectrum. The 2400-g/mm grating entiated when exciting in the UV. If a means to engineer the profile to pro- will diffract out to almost 800 nm, but we go to an even shorter wavelength, duce high reflectivity in the region of the 3600-g/mm grating will not diffract the limiting resolution on a short interest (4). much beyond 500 nm. These gratings focal length system will probably Figure 6 shows the reflectivity curve are, in fact, usually used in the blue and not be adequate for many studies, as for an 1800-g/mm holographic grating UV part of the spectrum where their shown in Table III. that has been optimized for use be- dispersion is so much better than that of For the reader interested in a tween 450 and 850 nm. There are three lower groove density gratings. Table II more detailed, but practical expla- curves on this figure. The green curve enables an easy comparison of the dis- nation of how a spectrograph func- is labeled TM and shows the reflectivity persion per pixel at 300 nm for the two tions, it can be found in The Optics for light polarized perpendicular to the gratings in the two systems. of Spectroscopy (2). grooves of the grating. The red curve is

magentablackcyanyellow ES326247_Spec1013_018.pgs 09.26.2013 04:16 ADV Transforming Metals Prep!

NO VESSEL EASY TO CLEANING USE

CHALLENGING SAMPLES

FASTEST RUN TIMES UltraWAVE

HIGH MIXED THROUGH- BATCHES PUT

BENCHTOP HIGH THROUGHPUT MICROWAVE DIGESTION SYSTEM

The remarkable new UltraWAVE features Milestone’s unique Single Reaction Cham-

ber (SRC) technology in a fully automated system. Unlike closed vessel digestion, the Ul-

traWAVE uses disposable glass vials, and different sample types can be digested si-

multaneously, greatly increasing productivity, and reducing overall cost per digestion.

UltraWAVE - Transforming metals sample prep!

Watch Videos, Download Brochures, Application Notes, or Request A Demo at:

AVAILABLE WEBCASTS: milestonesci.com/ultrawave USP <232>/<233> & FOOD SAMPLE PREP

magentablackcyanyellow ES328069_SPEC1013_019_FP.pgs 09.27.2013 03:44 ADV 20 Spectroscopy 28(10) October 2013 www.spectroscopyonline.com

but an instrument response correction will eliminate all wavelength-dependent 100 sensitivities. 90 80 The holographic 1200-g/mm grat- 70 TE ing presented a surprise when it first 60 50 TM appeared. The groove density was 40 Unpolarized convenient for the desired dispersion, 30 but it was implemented before severe 20

Relative efﬁciency (%) Relative efﬁciency 10 problems in the Wood’s anomalies were 0 observed. The original non-optimized 300 350 400 450 500 550 600 650 700 750 800 850 holographic grating had an extremely Wavelength (nm) sharp anomaly where the reflectiv- Figure 6: Efficiency curve for an 1800-g/mm grating. ity dropped to essentially 0% near 650 nm. Apparently, the groove profile and diffraction angle were such that all the diffracted light grazed off the grating 100 at a particular wavelength. But this was 90 subsequently corrected by optimizing 80 70 the groove profile for wavelengths of 60 TE interest; the optimization had the effect 50 TM 750 nm of softening the anomaly and moving 40 Unpolarized it to other wavelength regions. Figure 7 30 shows the efficiency profiles for several 20 1200-g/mm gratings that have been

Relative efﬁciency (%) Relative efﬁciency 10 0 optimized for different wavelengths. 400 450 500 550 600 650 700 750 800 850 900 950 1000 Visual inspection illustrates how ef- Wavelength (nm) fectively the ion-etching has optimized 100 90 the reflectivity for different wavelength 80 regions. But it should be noted that all 70 60 TE three of these gratings still have the 50 630 nm TM anomaly near 650 nm, which represents 40 30 about a 10% drop in intensity. 20 10

Relative efﬁciency (%) Relative efﬁciency 0 Comparison of Spectra 300 350 400 450 500 550 600 650 700 750 800 Recorded with Different Gratings Wavelength (nm) After all this somewhat theoretical 100 500 nm 90 discussion it would probably be appro- 80 70 priate to illustrate the effects of select- 60 TE ing different gratings for a measure- 50 40 TM ment. Figure 8 shows the spectrum of 30 a slightly fluorescent paper recorded 20 10 with 300-, 600-, 1200-, and 1800-g/

Relative efﬁciency (%) Relative efﬁciency 0 300 400 500 600 700 800 900 1000 1100 1200 mm gratings on the Aramis (460-mm Wavelength (nm) focal length) using the 532-nm laser. The top of Figure 8 shows the spec- Figure 7: Reflectivity curves for three 1200-g/mm gratings optimized for different regions of the trum, as recorded with the different spectrum. The top grating has been optimized for 750 nm, the middle grating for 630 nm, and gratings. The middle of the figure the bottom for 500 nm. shows the same spectrum after the removal of the baseline. The bottom labeled TE and has been measured with ties and what kind of polarization the of the figure shows a smaller region light polarized parallel to the grooves. sampling is presenting, a grating can of the middle figure to better estimate The blue curve is the average of the two be used on the short wavelength side the relative intensities and to visualize and represents unpolarized light. Note of the cross-over between TE and TM. the noise. First, I want to point out that that equipment manufacturers often For this grating the ripples that were I recorded the spectra with acquisition select a grating where the TM curve is mentioned above appear between 300 times meant to compensate for the dif- higher in the region of interest. But, if and 425 nm. In most cases they will not ferences in photon flux per pixel that the user is aware of the grating proper- present a problem to the spectroscopist, follows from the differences in disper-

magentablackcyanyellow ES326260_Spec1013_020.pgs 09.26.2013 04:16 ADV Goodbye, Bradford Assays! Drive your research forward with IR-based quantitation.

With Direct Detect® spectrometry, the ﬁ rst infrared (IR)-based biomolecular quantitation system, there’s no sample prep, messy cuvettes or waste—with one-time standard curves. The Direct Detect® system distinguishes proteins and peptides from sample components, such as lipids, that interfere with classical quantitation methods. Now you can achieve truly accurate results without the pitfalls of colorimetric assays, even for most lysates and complex samples.

Quit Assays Forever — Quantitate Directly. Learn more at: www.emdmillipore.com/DirectDetect

Rat Liver Lysate 0.09

0.07 Lipid Protein 0.08 0.06 0.05 0.07 0.04 0.03 0.06

0.02 Amide Amide Absorbance units Absorbance units I II 0.05 0.01 0.2 0.00 1700 1650 1600 1550 1500 1450 310000 3000 2900 2800 2700 1 Wavenumber cm-1 Wavenumber cm- 0.15 Accurate IR-based protein quantitation in a lipid-rich 0.10 lysate. The most intense

Absorbance units regions of lipid absorbance 0.05 are distinct from the

4000 3500 3000 2500 2000 1500 1000 500 protein’s Amide I signal. Wavenumber cm-1

EMD Millipore is a division of Merck KGaA, Darmstadt, Germany

EMD Millipore and the M logo are trademarks of Merck KGaA, Darmstadt, Germany. Direct Detect is a registered trademark of Merck KGaA, Darmstadt, Germany. © 2013 EMD Millipore Corporation, Billerica, MA. All rights reserved. BS GEN-13-09140 09/2013

magentablackcyanyellow ES328103_SPEC1013_021_FP.pgs 09.27.2013 03:45 ADV 22 Spectroscopy 28(10) October 2013 www.spectroscopyonline.com

Table IV: Approximate change in intensities of a sharp (~1080 cm-1) and somewhat sate for the differences in dispersion. broad band (~1600 cm-1) as a function of grating groove density But what is potentially more interest- 1800 g/mm 1200 g/mm 600 g/mm 300 g/mm ing is the change in relative intensity ~1080 cm-1 1 1.5 2 3 with dispersion. For instance, Table IV shows the changes in intensities of ~1600 cm -1 1 1.75 4 7 the lines at ~1080 and ~1600 cm-1 as a function of groove density. The fact that the band intensities are increasing at different rates is a result of the (a) different bandwidths, and the fact that 15,000 one band actually is composed of over- lapping components; as the dispersion decreases each pixel is integrating a 10,000 greater number of wavenumbers. This can have some rather interest- ing implications for the fluorescence 5000 background. Inspection of the top Intensity (counts/s) of Figure 8 indicates that the background increases by more than a fac- 1000 2000 3000 4000 tor of 10-fold when comparing the Raman shift (cm-1) (b) 5000 1800-g/mm spectrum to that of the 300-g/mm spectrum! In addition, 4000 when there is such a high background it becomes difficult to differentiate between weaker Raman features and 3000 artifacts in the background. The background-subtracted spectra in Figure 8 2000 illustrate this point because a simple

Intensity (counts/s) background subtraction cannot elimi- 1000 nate the random and pattern noise in the background. 0 1000 2000 3000 4000 Comments -1 Raman shift (cm ) I should point out that Figure 8 shows (c) 5000 that complete spectra were recorded for all gratings. That means that even 4000 though a particular grating–laser combination will provide X cm-1 in a

3000 single shot, the full spectra were recorded by scanning the entire range of interest — in this case, 100–4000 2000 cm-1. That means that there is no loss

Intensity (counts/s) in capability when using a high groove 1000 density grating for higher spectral resolution. This figure also shows 0 that the broad background does not 1000 1500 scale with the same factor as a sharp Raman shift (cm-1) band when changing grating (that is, dispersion). We were previously aware Figure 8: Raman spectra of slightly fluorescent paper recorded with 532-nm laser on the that the ratio of a sharp to broad band Aramis (460 mm focal length), using 300-, 600-, 1200-, and 1800-g/mm gratings, adjusting would change when the dispersion of the integration times to scale with the groove density (1, 2, 4, and 6 s). Top, as recorded; the recording instrument is changed, middle, after baseline subtraction, with vertical displacement for clarity; bottom, after baseline but were surprised to see the same ef- subtraction, expanded region. fect on the ratio of sharp band to (very) sion with the various gratings. Because and aperture at which the grating is broad background. What this means is there are other factors effecting the being used), this adjustment in acqui- that when recording spectra of not-so- intensities (such as grating reflectivity sition time does not totally compen- clean samples, the expected sensitivity

magentablackcyanyellow ES326259_Spec1013_022.pgs 09.26.2013 04:16 ADV www.spectroscopyonline.com October 2013 Spectroscopy 28(10) 23

advantage from recording low resolution spectra may be cancelled by the rapid increase in background! While it is always possible to subtract the background levels, the noise does not subtract, as can be seen in the middle and bottom traces in Figure 8. It would be wise to keep this in mind when dealing with samples with interfering backgrounds — it is always best to use conditions that will enable collecting spectra without a background rather than try to deal with it after-the-fact. For those interested in this point, we expect to deal with this more fully sometime in the near future. Raman images: Summary This column has been written to enable the analyst who is new to spectroscopy to understand what the considerations now in are in choosing a grating or focal length for a Raman spectrograph. Maybe you do not understand all of the grating high def physics that determines the grating functionality, but you should at least understand what parameters the choice of grating will optimize. Of course, the selection of laser wavelengths is independent of this, and actually has to happen before an intelligent choice of gratings can be made. I would love to hear from any of you about how useful this has been.

Acknowledgments My thanks to Emmanuel Leroy of Horiba Scientific for the Excel program that provides the capability to plot instrumental dispersions for various grating–spectrograph focal length combinations. The ability to present these plots helps to clarify what is happening. The grating reflectivity plots come from Horiba Scientific grating catalog (4). Pharmaceutical References (1) M. Born and E. Wolf, Principles of Optics – Electromagnetic tablet Theory of Propagation, Interference and Diffraction of Light, 5th Edition (Pergamon Press, Oxford 1975), Chapter VIII. (2) J.M. Lerner and A. Thevenon, The Optics of Spectroscopy, A Tutorial http://www.horiba.com/us/en/scientific/products/ Clear, crisp optics-tutorial/. (3) E.G. Loewen, M. Nivière, and D. Maystre, Appl. Opt. 16(10), Raman images 2711–2721 (1977). Renishaw’s new WiRE 4 (4) Scientific Diffraction Gratings/Custom Gratings – Product Raman software enables Catalog and Capabilities – available by contacting Horiba you to capture and review Scientific. very large Raman datasets Fran Adar is the Principal Raman and produce high def nition Applications Scientist for Horiba Scientific Raman images. These can (Edison, New Jersey). She can be reached by be as large and crisp as e-mail at [email protected] you like, without pixelation.

Apply innovation Improve your images

For more information on this topic, please visit: Renishaw Inc. Hoffman Estates, IL www.spectroscopyonline.com/adar www.renishaw.com/raman

magentablackcyanyellow ES326258_Spec1013_023.pgs 09.26.2013 04:16 ADV 24 Spectroscopy 28(10) October 2013 www.spectroscopyonline.com

Chemometrics in Spectroscopy Calibration Transfer, Part IV: Measuring the Agreement Between Instruments Following Calibration Transfer This is our 100th “Chemometrics in Spectroscopy” column, and when including “Statistics in Spectroscopy,” there are now a total of 138 columns. We began in 1986 and have been working since that time to cover in-depth discussions of both basic and difficult subjects. In this newest series, we have been discussing the subject of multivariate calibration transfer (or calibration transfer) and the determination of acceptable error for spectroscopy. We have covered the basic spectroscopy theory of spectroscopic measurement in reflection, discussed the concepts of measuring and understanding instrument differences, and provided an overview of the mathematics used for transferring calibrations and testing transfer efficacy. In this installment, we discuss the statistical methods used for evaluating the agreement between two or more instruments (or methods) for reported analytical results. The emphasis is on acceptable analytical accuracy and confidence levels using two standard approaches: standard uncertainty or relative standard uncertainty, and Bland-Altman “limits of agreement.”

Jerome Workman, Jr. and Howard Mark

s we have discussed in this series (1–3), calibration We note that the issue of calibration transfer disappears if Atransfer involves several steps. The basic spectra are the instruments are precisely alike. If instruments are the initially measured on at least one instrument (that is, “same” then one sample placed on any of the instruments the parent, primary, or master instrument) and combined will predict precisely the “same” result. Because instruments with the corresponding reference chemical information (that are not alike and also change over time, the use of calibra- is, actual values) for the development of calibration models. tion transfer techniques is often applied to produce the best These models are maintained on the original instrument attempt at model or data transfer. As mentioned in the first over time, are used to make the initial calibration, and are installment of this series (1), there are important issues of at- transferred to other instruments (that is, child, secondary, tempting to match calibrations using optical spectroscopy to or transfer instruments) to enable analysis using the child the reference values. The spectroscopy measures the volume instruments with minimal corrections and intervention. fractions of the various components of a mixture.

magentablackcyanyellow ES326283_Spec1013_024.pgs 09.26.2013 04:18 ADV magentablackcyanyellow ES328070_SPEC1013_025_FP.pgs 09.27.2013 03:44 ADV 26 Spectroscopy 28(10) October 2013 www.spectroscopyonline.com

Table I: Data used for illustration, instruments (or methods A, B, C, and D) more samples on the child (transfer) instrument until the model is basically Sample A1 A2 B1 B2 C1 D1 Number updated based on the child or transfer 1 12.4 12.2 12.1 12.0 14.9 16.1 instrument characteristics. We have previously described the scenario in 2 12.9 12.5 12.5 12.6 15.5 16.8 which a user has multiple products 3 14.0 14.4 13.9 13.7 16.8 18.2 and constituents and must check each 4 16.0 16.3 15.9 15.7 19.2 20.8 constituent for the efficacy of calibra- 5 13.2 13.0 12.9 12.8 15.8 17.2 tion transfer. This is accomplished 6 12.8 12.4 12.7 12.9 15.4 16.6 by measuring 10–20 product samples 7 14.5 14.8 14.9 14.5 17.4 18.9 for each constituent and comparing the average laboratory reference value 8 13.0 13.3 13.4 13.7 15.6 16.9 to the average predicted value for 9 13.6 13.3 13.5 13.4 16.3 17.7 each constituent, and then adjusting 10 12.7 12.4 12.6 12.9 15.2 16.5 each constituent model with a new 11 14.2 14.5 14.6 14.1 17.0 18.5 bias value, resulting in an extremely 12 16.3 16.6 16.4 16.2 19.6 21.2 tedious and unsatisfying procedure. 13 17.8 17.9 17.9 17.4 21.4 23.1 Such transfer of calibration is also accomplished by recalibration on the 14 18.0 18.3 18.5 18.5 21.6 23.4 child instrument or by blending sam- 15 14.5 14.2 13.9 13.5 17.4 18.9 ples measured on multiple instruments 16 17.2 17.0 17.5 17.7 20.6 22.4 into a single calibration. Although the 17 14.4 14.1 14.6 14.8 17.3 18.7 blending approach improves method 18 15.2 15.4 15.7 15.5 18.2 19.8 robustness (or ruggedness) for pre- 19 16.6 16.7 16.5 16.3 19.9 21.6 dicted results across instruments using the same calibration, it is not appli- 20 13.5 13.2 13.1 13.4 16.2 17.6 cable for all applications, for analytes having small net signals, or for achiev- Historically, instrument calibrations Multivariate calibration transfer, ing the optimum accuracy. have been performed using the existing or simply calibration transfer, is a set analytical methods to provide the refer- of software algorithms, and physical How to Tell if Two Instrument ence values. These existing methods materials (or standards) measured on Predictions, or Method Results, have often overwhelmingly used weight multiple instruments, used to move Are Statistically Alike percents for their results. Until this par- calibrations from one instrument to The main question when comparing adigm changes and analysts start using another. All the techniques used to parent to child instrument predictions, the correct units for reporting their re- date involve measuring samples on a reference laboratory method to an sults we have to live with this situation. the parent, primary (calibration) and instrument prediction, or results from We will also have to recognize that the child, secondary (transfer) instru- two completely different reference reference values may be some prescribed ments and then applying a variety of methods, is how to know if the differ- analysis method, the weight fraction of algorithmic approaches for the trans- ences are meaningful or significant materials, the volume percent of compo- fer procedure. The most common ap- and when they are not. There is always sition, or sometimes some phenomeno- proaches involve partial least squares some difference expected, since an logical measurement having no known (PLS) calibration models with bias or imperfect world allows for a certain relation to underlying chemical entities, slope corrections for predicted results, amount of “natural” variation. How- resulting from some arbitrary definition or the application of piecewise direct ever when are those differences consid- developed within a specific industry standardization (PDS) combined with ered statistically significant differences, or application. Amongst ourselves we small adjustments in bias or slope or when are the differences too great to have sometimes termed these reference of predicted values. Many other ap- be acceptable? There are a number of methods “equivalent to throwing the proaches have been published and reference papers and guides to tell us sample at the wall and seeing how long compared, but for many users these how to compute differences, diagnose it sticks!” The current assumption is the are not practicable or have not been their significance, and describe the nonlinearity caused by differences in adopted and made commercially avail- types of errors involved between meth- the spectroscopy and the reported refer- able for various reasons. ods, instruments, and analytical tech- ence values must be compensated for In any specific situation, if the pre- niques of many types. The analytical by using calibration practices. This may scribed method for calibration transfer method can be based on spectroscopy not be as simple as presupposed, but re- does not produce satisfactory results, and multivariate calibration methods, quires further research. the user simply begins to measure other instrumental methods, or even

magentablackcyanyellow ES326277_Spec1013_026.pgs 09.26.2013 04:17 ADV magenta black cyan yellow ES328056_SPEC1013_027_FP.pgs 09.27.2013 03:43 ADV • • • offers leadingedgeNIR Spectroscopy. Metrohm NIRSystems include solutionsforNearInfrared Metrohm extendsitsportfolioto NIRSystems Metrohm www.metrohmusa.com/nirs by Metrohm is servicedandsupported expertise has over40yearsofNIR technology

No Sample 19.0 too Tough... 17.0 15.0

13.0 Instrument (or method) B

11.0 11.0 13.0 15.0 17.0 19.0 Instrument (or method) A

Figure 1: Instrument or method A1 (x-axis) as compared to instrument or method B1 (y-axis), and with data points compared to perfect line of equality.

gravimetric methods. We have included several of the most noted references in the reference section of this column. One classic reference of importance for comparing methods is by Youden and Steiner (4). This reference describes some of the issues we will discuss in this column as well as details regarding collaborative laboratory tests, ranking of laboratories for accuracy, outlier determination, ruggedness tests for methods, and diagnosing the various types of errors in Analyze hard materials with analytical results. no sample preparation Let us begin with a set of measurement data as shown in Table I. This is simulation data that is fairly representative of spectroscopy data following multivariate calibration. These Monitor structural changes data could refer to different methods, such as methods A, at high temperatures B, C, and D; or to different instruments. For our discussion we will designate that the data is from four different instruments A to D for 20 samples. The original data from a Confirm your sampling area calibrated instrument is A1. The results of data transferred to other instruments is represented by B, C, and D. There with digital imaging are duplicate measurements for A as A1 and A2, and for B as B1 and B2, respectively. From these data we will perform an Enjoy years of service and quality results analysis and look for levels of uncertainty and acceptability for the analytical results. Note: C1 and D1 data are used in from the monolithic diamond interface Figure 2 and will be referred to along with A2 and B2 in the next installment of this column.

...with PIKE single reflection Standard Uncertainty and Relative Standard Uncertainty Diamond ATR GladiATR First, we look at the definitions of uncertainty as described by the United States National Institute of Standards and Technology (NIST), a National Metrological Institute (NMI), which is a nonregulatory agency of the United www.piketech.com States Department of Commerce. The lion’s share of NIST’s purpose is to advance measurement science, measurement tel: 608-274-2721 standards, and measurement technology. Their charter is to define measurements from first principles that can be verified world-wide and used as standards for making measurements of any kind related to commerce or technology.

magentablackcyanyellow ES326281_Spec1013_028.pgs 09.26.2013 04:18 ADV www.spectroscopyonline.com October 2013 Spectroscopy 28(10) 29

24.0 PIKE 22.0 ™ PIKE FlexIR™ 20.0 GladiATR

18.0

16.0

14.0 Alternate instrument (or method) results 12.0 12.0 14.0 16.0 18.0 20.0 Instrument (or method) A

Figure 2: Instrument or method A1 (x-axis) as compared to instruments (or methods) C or D (y-axis), showing line of equality for perfect agreement as well as other perfect correlations that are not in analytical agreement. This indicates correlation as an imperfect representation of agreement between methods.

0.7 Mean + 2SD 0.6 0.5 0.4 0.3 0.2 0.1 The GladiATR advances FTIR-ATR technology to Mean 0.0 the next level. Its monolithic diamond sampling -0.1 -0.2 interface and precision optics offer the most energy -0.3 -0.4 efficient design, 4000 – 50 cm-1 working spectral -0.5 -0.6 range, and the highest quality data. Optional Difference (A1 – B1) for each sample Difference Mean - 2SD -0.7 10.0 12.0 14.0 16.0 18.0 20.0 temperature control enables studies of materials at Average (A1 and B1) for each sample temperatures up to 300°C. The built-in camera Figure 3: The Bland-Altman plot indicating the difference plotted against the allows for simultaneous viewing and data collection mean for each sample for instrument (or methods) A1 and B1. (Note: For random, normally distributed data we would expect one sample out of 20 to assuring that what you see is what you sample. be outside the ±2 SD range.) Try this most versatile accessory and simplify your analyses without compromising your results. The NIST definition for uncertainty is quite specific, as explained below (5,6).

Uncertainty Defined FTIR sampling made easier NIST Definitions The standard uncertainty u(y) of a measurement result y is the estimated standard deviation of y.

The relative standard uncertainty ur(y) of a measurement result y is defined by u (y) = u(y)/|y|, where y is not equal r PIKE Technologies to 0. If the probability distribution characterized by the www.piketech.com measurement result y and its standard uncertainty u(y) tel: 608-274-2721 is approximately normal (Gaussian), and u(y) is a reliable estimate of the standard deviation of y, then the interval

magentablackcyanyellow ES326275_Spec1013_029.pgs 09.26.2013 04:17 ADV 30 Spectroscopy 28(10) October 2013 www.spectroscopyonline.com

y – u(y) to y + u(y) is expected to 1 N y =X = ∑ X encompass approximately 68% of i i N k=1 i,k [2] the distribution of values that could where N is the total number of Xi reasonably be attributed to the value (that is, the number of instruments, of the quantity Y of which y is an methods, or models being com- estimate. This implies that it is be- pared). So our estimated analytical lieved with an approximate level of value (y ) is the mean for a number i • confidence of 68% that Y is greater of measurements of that sample (Xi) than or equal to y – u(y), and is less using the analytical method pre- than or equal to y + u(y), which is scribed. And it follows that the stan-

commonly written as Y = y ± u(y). dard uncertainty u(Xi) with reference The use of a concise notation if, for to the measured values (Xi) is equal example, y = 1 234.567 89 U and u(y) to the estimated standard deviation

= 0.000 11 U, where U is the unit of of the mean.

1 y, then Y = (1 234.567 89 ± 0.000 11) N ( 2 U. A more concise form of this ex- u(X )=s (X )= 1 ∑ (X – X )2 [3] i i (n n i,k i ( Ð1)k=1 pression, and one that is commonly used, is Y = 1 234.567 89(11) U, where So to apply this to our data illustration

it is understood that the number in from Table I we use A1 and B1 as X1 parentheses is the numerical value of and X2, therefore

the standard uncertainty referred to

u(y )=u(X )=s (X )= i i i 1 in the corresponding last digits of the ( N 2 quoted result. 1 ∑ (X – X )2 (n n i,k i [4] ( Ð1)k=1 Estimates of Uncertainty

For the data in Table I we simpli- where u(yi) is the estimated standard fied our comparison by selecting uncertainty for a series of measure- only A1 and B1 data noting that one ments on multiple samples where might compare multiple analysis the mean value of the measurements for multiple instruments as a more for each sample is used for compari- powerful test of the estimated stan- son. The equations above are often dard deviation of a measurement used for multiple measurements of result for any method or instrument a single physical constant. For our combination. However, the ap- application using only an A1 and B1 proach shown here is adequate for measurement for each sample, we estimates of uncertainty for typical compute the variance for each of the analysis and calibration transfer ap- 20 samples and pool the results to plications. We note, using the NIST give us our estimate of standard un-

nomenclature, that certainty u(yi) as

( )=u( )=s ( )=

y = f (X , X ,..., X ) [1] u yi Xi Xi 1 2 N 1 N ( 2 = 1 ∑ σ 2 ( i,k) [5] where y is the estimated analytical (n(nÐ1)k=1 value for any sample as a function of a series of measurement quanti- Compiling these results yields the fol-

ties such as X1, X2, . . ., XN; and lowing for the standard uncertainty where each Xi is an independent reported as u(yi) = 0.226 observation (or measurement). We also note that when using this no- Relative Standard Uncertainty

menclature the A1 and B1 measure- This is denoted as ur(yi) = u(yi)/|yi| and ments for each sample are denoted so applying the previously computed

as Xi measurements. We note the results we note that yi is equal to 14.648 value (yi) for each sample measure- and |yi| = 14.648 as the mean of all A1 ment is estimated as the sample and B1 values. Therefore, the u(yi) = mean from N independent measure- 0.226 and so the relative standard un-

ments and is denoted as Xi,k , giving certainty is reported as ur(yi)= u(yi)/|yi| us the relationship as follows. = 0.226/14.648 = 0.0154.

magentablackcyanyellow ES326276_Spec1013_030.pgs 09.26.2013 04:17 ADV www.spectroscopyonline.com October 2013 Spectroscopy 28(10) 31

Confidence Levels Reported: Bland-Altman method rather than The paper we cite is entitled “Sta- The confidence levels would be ex- a possibly more familiar calculation tistical Methods for Assessing Agree- pressed as follows: because Bland-Altman is the standard ment Between Two Methods of Clini- for clinical data in an industry with cal Measurement,” and was written

For _68%: yi ± u(yi) = 14.648 ± 0.226 stringent requirements for analyti- by J. Martin Bland and Douglas G. = 14.42 – 14.87 cal results. There is an entire series Altman (7). This paper describes the of publications by these statisticians errors often made when comparing

For _95%: yi ± 2 . u(yi) = 14.648 ± and this method is still taught and methods. The authors summarize the 0.452 = 14.20 – 15.10 used for clinical analysis criteria. contents of this paper as follows, “In One of the possible reasons there are clinical measurement comparison of So the expression of certainty for this no approved clinical methods using a new measurement technique with example of data would be as y = 14.648 near-infrared (NIR) spectroscopy is an established one is often needed to ± 0.226U or y = 14.648(0.226)U. that it does not stand up to critical see whether they agree sufficiently Thus, for the analyst the final step analysis requirements and the NIR for the new to replace the old. Such is deciding if this is a satisfactory re- groups look at data using unconven- investigations are often analyzed sult across the two instruments. This tional terminology and methods that inappropriately, notably by using concept requires much more discus- do not stand up to serious analytical correlation coefficients. The use of sion, one which we will present in a scrutiny. With 27,000 citations in the correlation is misleading. An alter- future column. literature, and having it included in native approach, based on graphical standard university bioengineering techniques and simple calculations, Bland-Altman curriculum, this would be consid- is described.” So, what can be learned “Limits of Agreement” ered a more standard approach in by using the techniques described in For a second look at the data in Table the wider analytical community. The this paper to compare results from I, we refer to one of the definitive NIR and chemometric communities analytical methods? When methods papers used for method compari- would do well to learn from these are compared following calibration, sons within the highly regulated older and thoroughly tested method- one attempts to assess the degree of clinical sciences. We introduce the ologies and referenced papers. agreement between them. Bland and

Rigaku Corporation and its Global Subsidiaries website: www.Rigaku.com | email: [email protected]

magentablackcyanyellow ES326279_Spec1013_031.pgs 09.26.2013 04:18 ADV 32 Spectroscopy 28(10) October 2013 www.spectroscopyonline.com

Altman discount completely the use the strength of a relationship between method and the second method (or of correlation as a useful parameter to variables — not that the analytical re- instruments A and B in our example) assess analytical agreement. Their ar- sults agree; a change in scale does not for each sample. Such a plot uses the guments are given in this discussion. affect correlation, but drastically af- mean and plus or minus two standard For this method comparison, two fects agreement; correlation depends deviations as the upper and lower single measurements are taken for on the range of the true quantity comparison thresholds. each sample on each instrument as (analyte) in the sample; tests of signif- To assess if the data are in close A1 and A2. The first measurement is icance are mostly irrelevant between enough agreement for our analyti- used for comparison and the second two similar analytical methods; and cal purposes between A1 and B1 we for a repeatability study (which we data in poor agreement analytically compute the bias or mean difference – will cover in a later installment). For can be highly correlated (Figure 2). (d), the standard deviation of the dif- this method of comparison a line Figure 2 shows three analytical sets ferences (s or SD), and the expected of equality plot is made to compare with perfect correlation but poor “limits of agreement.” These are – two methods or two instruments. agreement. expressed as d ± 2s for a 95% confi- The various x, y data points are plot- A Bland-Altman plot shown in dence level. ted against a perfect straight line of Figure 3, which will be extremely The mean difference is computed equality. The authors make the point familiar to clinical analysts, dem- as the average of all the differences that correlation (r) measures the onstrates a good visual comparison between the comparative instru- strength of the relationship between technique to evaluate the agreement ments, for each sample as n two variables, but it does not measure between two methods or instruments. ∑ A –B ( i i) the agreement between them (Figure The x-axis (abscissa) for each sample d = i=1 [6] i n 1). Perfect agreement is indicated by is represented by the average value the data points lying directly on the for each sample obtained from the The standard deviation for this com- line of equality. A perfect correla- comparison results (using two meth- parison for the set of samples is com- tion is indicated if all the points lie ods or two instruments). The y-axis puted as

along any straight line. The authors (ordinate) for each sample is repre- n n ∑ A –B 2 ∑ D2 emphasize that: correlation indicates sented by the difference between one ( i i) i s = i=1 = i=1 [7] i √ 2n √ 2n

And so for our data A1 and A2 in Table I we find the following results: – Systems smart Bias is d = -0.226; standard deviation for all samples is s = -0.015; 95% – confidence level is d + 2s = 0.438 and enough – d – 2s = -0.468. If this number is considered too to help you large for a 95% confidence of the result agreement then these meth- spot a fake ods are not considered equivalent. On the other hand if these limits of agreement are acceptable for the application where they are used, then this is acceptable. In a clinical situation, a physician determines the level of accuracy or agreement required for critical intervention decision making; this would be analo- At EDAX, we understand how you see the world. Our Orbis Micro-XRF Analyzer is gous to a process control supervisor designed to make your life easier. Our SDD option improves trace element sensitivity, or analytical scientist assessing the acceptable level of agreement be- allows 40% faster data collection. Optional chamber viewport lets you feel secure tween comparative methods to use analyzing even your most precious artifacts. Easy. Fast. Flexible. the alternate method or instrument Find out how to power your next insight at edax.com/micro-xrf as a substitute. Conclusion These two similar, classic, and well-accepted methods have been

magentablackcyanyellow ES326285_Spec1013_032.pgs 09.26.2013 04:18 ADV www.spectroscopyonline.com October 2013 Spectroscopy 28(10) 33

applied to analytical results in Jerome Work- Howard Mark commerce and clinical analysis. In- man, Jr. serves on serves on the Editorial cluded below are references for the the Editorial Advisory Advisory Board of reader’s further study of this subject Board of Spectroscopy Spectroscopy and runs of comparing two or more analyti- and is the Executive a consulting service, cal methods or instruments. Other Vice President of En- Mark Electronics references discussing the details of gineering at Unity Sci- (Suffern, New York). comparing analytical methods are entific, LLC, (Brookfield, Connecticut). He can be reached via e-mail: also provided (8–16). He is also an adjunct professor at U.S. [email protected] National University (La Jolla, Califor- For more information on References nia), and Liberty University (Lynch- this topic, please visit: (1) H. Mark and J. Workman, Spectros- burg, Virginia). His e-mail address is www.spectroscopyonline.com copy 28(2), 24–37 (2013). [email protected] (2) H. Mark and J. Workman, Spectros- copy 28(5), 12–25 (2013). (3) H. Mark and J. Workman, Spectros- copy 28(6), 28–35 (2013). (4) W. Youden and E.H. Steiner, Statisti- SPECTROMETERS LASERS TOTAL SOLUTIONS cal Manual of the AOAC, (Association of Official Analytical Chemists, Arling- Your Spectroscopy Partner ton, Virginia, 1984). (5) http://physics.nist.gov/cgi-bin/cuu/ Info/Constants/definitions.html. (6) B.N. Taylor and C.E. Kuyatt, “Guide- lines for Evaluating and Expressing the Uncertainty of NIST Measurement Results” (NIST Technical Note 1297, Innovative September 1994 Edition). (7) J.M. Bland and D.G. Altman, Lancet 1, Enhanced & Revolutionary: i-Raman Plus 307–10 (1986). Now with the most (8) W. Horwitz, Anal. Chem. 54(1), 67A– advanced technology 76A (1982). t %FFQ5&$PPMJOHGPSVQUP.JOVUFTPG (9) P. Hall and B. Selinger, Anal. Chem. *OUFHSBUJPO5JNF 61, 1465–1466 (1989). t )JHIMZ4FOTJUJWF#BDLUIJOOFE$$%%FUFDUPS t 1BUFOUFE$MFBO-B[F¥-BTFS4UBCJMJ[BUJPO (10) D. Rocke and S. Lorenzato, Techno- t )JHI5ISPVHIQVU'4QFDUSPHSBQI metrics 37(2), 176–184 (1995). t 4QFDUSBM3FTPMVUJPOBT'JOFBTDN (11) J.C. Miller and J.N. Miller, Statistics for t 4QFDUSBM3BOHFGSPNDNUPDN Analytical Chemistry, Second Edition t 'JCFS0QUJD1SPCFGPS'MFYJCMF4BNQMJOH t 4BNQMJOH4UBHF $VWFUUF)PMEFS BOE (Ellis Horwood, Upper Saddle River, $IFNPNFUSJD4PGUXBSF"WBJMBCMF New Jersey, 1992), pp. 63–64. (12) W.J. Dixon and F.J. Massey, Jr., Intro- duction to Statistical Analysis, Fourth Edition), W.J. Dixon, Ed. (McGraw-Hill, New York, 1983), pp. 377, 548. (13) D.B. Rohrabacher, Anal. Chem. 63, 139 (1991). (14) H. Mark and J. Workman, Chemomet- rics in Spectroscopy (Elsevier, Aca- demic Press, 2007), Chapters 34–39, 71–73. (15) ASTM E1655 - 05 (2012) Standard Practices for Infrared Multivariate Quantitative Analysis (2012). Call now to stay on 1-855-BW-RAMAN (16) H. Mark and J. Workman, Statistics in the cutting edge! www.bwtek.com Spectroscopy, 2nd Edition (Elsevier, Academic Press, 2003), Chapter 7, pp. 59–69.

magentablackcyanyellow ES326274_Spec1013_033.pgs 09.26.2013 04:17 ADV 34 Spectroscopy 28(10) October 2013 www.spectroscopyonline.com

A Rapid FT-IR-Based Method for Monitoring Detergent Removal from Biological Samples

In life science research, detergents are primarily used in sample preparation to liberate cellular components through membrane disruption and to solubilize lipid-associated proteins. However, detergents can interfere with applications such as enzyme-linked immunosorbent assays, isoelectric focusing, nuclear magnetic resonance spectroscopy, and mass spectrometry, and therefore, often need to be removed before analysis. Here we report a novel infrared-based method that permits fast and impartial analysis of detergent removal from biological samples. Because this method is spectrally based and label-free, determination of protein concentration can be performed simulta- neously. An additional benefit of this technique is the small volume (2 µL) required for analysis, a significant fact given the often precious nature of biological samples.

Ivona Strug, Sara Gutierrez, Amedeo Cappione III, Mayra Jimenez, Mary Jane Mullen, and Timothy Nadler

etergents, water-soluble surface-active agents distin- The extraction of proteins from cells or tissue requires de- Dguished by their amphipathic structure, are used ex- tergents to facilitate dissociation and solubilization. Although tensively in protein biochemistry. Principally, this large detergents are critical to this process, as well as numerous other family of molecules can be used to isolate, solubilize, and preparative methodologies used in protein research, they can stabilize membrane proteins (1–5). Depending on their struc- interfere with many downstream applications, thus limiting tural properties, certain detergents promote solubilization and analytical sensitivity, so they have to be removed before the disaggregation of recombinant proteins during the process of analysis (7–9). The most commonly used detergent removal extraction and purification (1), and others facilitate protein methods include size-based exclusion (dialysis and gel filtra- stabilization and crystallization (6). Detergents are also used tion), hydrophobic adsorption, and ion-exchange chromatog- to reduce background by minimizing nonspecific binding or raphy (10). The choice of an appropriate method is dictated by protein precipitation in a wide range of immunoassays (2,4). inherent properties of the detergents used including hydropho- All detergent molecules contain a hydrophilic head and a bicity, CMC, micelle size, and charge. hydrophobic chain (or tail). This unique composition permits UV absorbance, an array of colorimetric methods, and mass their spontaneous, but ordered aggregation in aqueous media spectrometry (MS) are current options to monitor and op- resulting in the formation of stable micelles. In their pure form, timize the efficiency of detergent removal (8,11,12). Because detergent micelles contain from three to several hundred mole- of the presence of aromatic rings, the concentration of com- cules, and the number of molecules in the micellar form (aggre- pounds, such as Triton X (100 and 114) and NP-40, can be gation number) is an important characteristic of the detergent. estimated by absorbance at 275 and 280 nm (8). However, ap- Consequently, detergent micelles can vary greatly in size and plication of this method is very limited given that detergent’s the molecular weight of the micelle represents another useful absorbance overlaps with that of the protein. Specifically, for parameter describing detergent properties. Micelle formation complex biological samples, such as cell culture or tissue ly- is initiated after a detergent content reaches a certain thresh- sates, it is impossible to establish pre-existing levels of protein old termed the critical micelle concentration (CMC). Generally, and distinguish the absorbance signals produced by detergent detergents must be used at concentrations above their CMC to and protein. Total organic carbon (TOC) analysis, yet another be effective (1–4). method applied to monitoring detergent removal, is also con-

magentablackcyan ES326333_Spec1013_034.pgs 09.26.2013 04:19 ADV www.spectroscopyonline.com October 2013 Spectroscopy 28(10) 35

founded by protein interference and is thus restricted to measuring detergent levels in the flow-through (removed) (a) 3.5 fraction. The use of liquid chromatography–tandem mass spectrometry (LC– 3.0 MS) or matrix-assisted laser desorption– 2.5 ionization (MALDI) has been reported in the case of monitoring the removal 2.0 of Tween 20 and BRIJ-35 (8). SDS con- 1.5 centration is typically measured by a 1.0 colorimetric approach using Stain-All Absorbance (AU) dye (11). Similarly, concentrated sulfuric 0.5 acid and phenol (12) are frequently used 0.0 for concentration estimations of glyco- 3500 3000 2500 2000 1500 1000 500 sidic and bile salt-based detergents like Wavenumber (cm-1) octyl glucoside and CHAPS. In summary, none of the existing platforms or (b) (c) methodologies appears capable of broad- based detergent content measurement in biological samples. Here, we report on the development of a novel Fourier transform infrared (FT-IR)-based method for fast and impartial analysis of detergent concentration in biological samples. The method Figure 1: Sample presentation to FT-IR beam: (a) MIR signature of PTFE membrane; (b) the design × uses a hydrophilic polytetrafluoroeth- of sample spot allowing for retention of analyzed sample in the mid-IR beam (20 magnification); (c) distribution of the sample dried on the spot. Sample: 2 µL of 4 mg/mL cytochrome c in ylene (PTFE) membrane engineered for CytoBuster protein extraction reagent (20× magnification).

magentablackcyanyellow ES326338_Spec1013_035.pgs 09.26.2013 04:20 ADV 36 Spectroscopy 28(10) October 2013 www.spectroscopyonline.com

sample retention and optimal transpar- ency in regions of the IR spectra used for analysis of biological samples. Aque- ous samples are applied directly onto the membrane, dried, and analyzed by FT-IR, with minimal volume require- NP-40 ment (2 µL). Because of the specificity of the absorbance bands, the technique offers simultaneous detection of multiple Digitonin species, such as protein and detergents reported here, without interfering with one another. This dual capability is well CHAPS suited for the optimization of protein preparative processes where the reduc- tion of detergent content is ultimately Sodium required. deoxycholate

Octylglucoside Experimental All measurements were performed using a Direct Detect spectrometer (EMD Mil- SDS lipore), an FT-IR system analyzing the spectral range 500–6000 cm-1. CTAB Detergent Quantitation To verify the application of FT-IR to 3000 2950 2900 2850 2800 monitor detergent removal, 0.5% deoxy- Wavenumber (cm-1) cholic acid sodium salt (EMD Millipore), 1% NP-40 (Sigma), 1× RIPA lysis buffer Figure 2: Overlap of the IR bands originating from stretching vibrations of C-H bonds collected (EMD Millipore) prepared with Dul- for various detergents. becco’s phosphate buffered saline (PBS, EMD Millipore), and CytoBuster protein extraction reagent (EMD Millipore) were used. In each case, 2 µL of sample solu- 100.0 NP-40 10K NP-40 30K tion was applied onto the membrane po- NP-40 100K Sodium deoxycholate 10K sition; this volume was used for all con- Sodium deoxycholate 30K centration estimates. The spectrometer 80.0 Sodium deoxycholate 100K RIPA 10K was initially calibrated with each of the RIPA 30K analyzed detergents and both lysis buf- RIPA 100K CytoBuster 10K fers. Sample concentrations were deter- 60.0 CytoBuster 30K mined in reference to their respective CytoBuster 100K calibration methods.

40.0 Protein Quantitation Bovine gamma globulins (Bayer) at 2 Detergent removed (%) mg/mL in PBS and bovine liver tissue lysates (prepared with CytoBuster and 20.0 RIPA lysis buffers), total protein content at around 3 mg/mL each, were used to demonstrate the application of the 0.0 FT-IR-based method for simultaneous 1 2 3 Spin number monitoring of detergent removal and protein recovery. Protein concentration Figure 3: Efficiency of detergent removal from PBS performed using Amicon Ultra 0.5 centrifugal in the original sample was used as a ref- devices (10, 30, and 100 kDa NMWL). Starting detergent concentrations were as follows: 1% NP- erence point for 100% recovery. Protein 40, 0.5% sodium deoxycholate, 1× RIPA and 1× CytoBuster. Retentates were analyzed following concentrations were determined in ref- each of three successive centrifugal “wash” cycles. The graph plots % detergent removal as a erence to a calibration curve developed function of spin step. using bovine serum albumin (BSA)

magentablackcyanyellow ES326334_Spec1013_036.pgs 09.26.2013 04:20 ADV www.spectroscopyonline.com October 2013 Spectroscopy 28(10) 37

from the National Institute of Standards and are utilized in the reported method traction reagent. The latter two buffers, and Technology (NIST) diluted in PBS (Figure 2). Comprehensive analysis of each containing multiple detergent types, (EMD Millipore). various detergents demonstrated a high are commonly used for the extraction degree of variability in the slope of the and solubilization of cellular proteins. Detergent Removal calibration curves (data not shown), Linear calibration curves for each of Size-based separation via centrifugal dia- strongly suggesting the requirement for the analyzed detergents and lysis buf- filtration was established as the method individualized calibrations. fers were prepared before the analysis. for detergent removal. For this study, To verify efficacy of an FT-IR-based Size-based separation via centrifugal Amicon Ultra (0.5 and 2.0) centrifugal approach to monitoring detergent re- diafiltration using Amicon Ultra 0.5 and filters (EMD Millipore) were used ac- moval, the following solutions were 2.0 (NMWL; 10 kDa, 30 kDa, and 100 cording to the manufacturer’s recom- employed: NP-40 (nonionic detergent), kDa) filters has been used as a detergent mendations. Briefly, 500 µL of investi- sodium deoxycholate (anionic detergent), removal method of choice. Overall, de- gated detergent or buffer solution was RIPA buffer, and CytoBuster protein ex- tergent properties could have been used placed in AU-0.5 or AU-2.0 devices. In the case of the AU-2.0 device, the volume was adjusted to 2 mL by addition of 1.5 mL of PBS. The devices were spun at 14,000g for 5 min (AU-0.5) or 7500g SPECTROMETERS LASERS TOTAL SOLUTIONS for 5–15 min (AU-2.0). The retentate Your Spectroscopy Partner volume from each device was adjusted to 500 µL by the addition of PBS; the remaining detergent concentration was estimated by FT-IR. For each sample, the above procedure was repeated twice for a total of three spin-removal and analysis cycles. Detergent removal was analyzed using three devices per each nominal molecular weight limit (NMWL; 10 kDa, 30 kDa, and 100 kDa). Results and Discussion Sample Presentation to the IR Beam FT-IR analysis of aqueous-based biological samples is achieved by application of the analyte onto a hydrophilic PTFE membrane incorporated into a measurement card. The membrane spectrum contains a strong signal between 1100 and 1300 cm-1, but it is transparent in the IR region above 1300 cm-1 used for the analysis of biological samples (Figure 1a). The hydrophilic spot for sample application (see Figures 1b and 1c) is surrounded Do MoreWITH 1064! by a hydrophobic ring generated by me- Designed for your most challenging biological samples. chanical removal of the hydrophilic surface that retains an analyte permitting precise presentation of the entire sample High-performing and to the IR beam (13,14). user-friendly, The B&W Tek Raman line-up is on FT-IR-Based Monitoring your side. The frst & only name in mobile molecular of Detergent Removal spectroscopy. Because of their complex chemical com- positions, detergents absorb in many different regions of the IR spectrum. The symmetric stretching vibrations of C-H bonds (2840–2870 cm-1) provided www.bwtek.com 1-855-BW-RAMAN an ideal candidate for detergent analysis

magentablackcyanyellow ES326335_Spec1013_037.pgs 09.26.2013 04:20 ADV 38 Spectroscopy 28(10) October 2013 www.spectroscopyonline.com

ciently removed through use of the 100 kDa NMWL format. Interestingly, Cy- toBuster was removed below reliable detection level (0.016×) in three spins while RIPA was resistant to clearance by the two devices harboring smaller NMWL 3500 3000 2500 2000 1500 membranes. Wavenumber (cm-1) Simultaneous FT-IR-Based Monitoring of Detergent Removal and Protein Recovery

3000 2950 2900 2850 2800 00 1750 1700 1650 1600 1550 1500 14 The primary structure of all proteins and peptides is defined by a chain of amino acids linked covalently via Figure 4: Simultaneous quantitation of protein and detergent (4 mg/mL BSA in the presence of peptide (amide) bonds. Amide bonds -1 1% SDS). The top part of the box shows IR signal registered between 3500 and 1400 cm . The absorb electromagnetic radiation in bottom part of the box shows a magnification of the areas of FT-IR spectrum characteristic to the multiple regions of the IR spectrum, protein and the detergent signals. The spectrum of the detergent (no protein) is shown in green. including a strong band at 1600–1690 The FT-IR spectrum of protein in the detergent containing buffer is shown in blue. cm-1 (“Amide I”). The reported FT-IR method utilizes strength of the Amide I band to determine protein concentra-

12.0 tion. Because of spectral separation of NP-40 10K the analysis regions used for detergent NP-40 30K NP-40 100K and protein quantitation, collection of Sodium deoxycholate 10K a single FT-IR spectrum permits si- 10.0 Sodium deoxycholate 30K multaneous analysis of both molecular Sodium deoxycholate 100K RIPA 10K species (Figure 4). RIPA 30K Critical to the process of detergent 8.0 RIPA 100K CytoBuster 10K removal is the need to retain protein, CytoBuster 30K both content and concentration. Inter- CytoBuster 100K estingly, the results obtained during the 6.0 analysis of detergent removal from samples containing protein could not have been predicted based solely on proper- 4.0 ties of the detergent. Figure 5 clearly shows that when mixed with protein, in this example IgG, sodium deoxycholate is as difficult to remove by size-based 2.0 diafiltration as NP-40 or RIPA. Within the analyzed buffer set, only the Cyto- Buster reagent offered effective detergent Percentage of remaining protein/percentage detergent 0.0 0.5 1 2 3 removal (~90% in three spins) without Spin number compromising protein retention. More- over, the lysates required fourfold dilu- Figure 5: Efficiency of detergent removal from IgG (starting concentration 2 mg/mL in PBS) tion (achieved through use of the larger containing solutions performed using Amicon Ultra 0.5 centrifugal devices (10, 30, and 100 AU-2.0 device) to produce noticeable de- kDa NMWL). Starting detergent concentrations were as follows: 1% NP-40, 0.5% sodium tergent removal. With three spin cycles deoxycholate, 1× RIPA, and 1× CytoBuster. Retentates were analyzed following each of three using the AU-2.0 device, CytoBuster successive centrifugal “wash” cycles. The graph plots relative cleanup performance (% protein removal exceeded 90% (Figure 6). How- retained/% detergent removed) as a function of spin step. ever, detergent clearance was accom- panied by a substantial loss of protein in predicting efficiency of the removal instrument detection level (0.031%) in content (~60%), suggesting that for het- by size exclusion (Figure 3). Specifically, less than three centrifugation cycles. erogeneous mixtures containing a wide sodium deoxycholate, characterized by In stark contrast, the NP-40 detergent, range of molecular weight species, such a relatively high CMC (2 mM) and small which is characterized by relatively low as cell lysates, size-exclusion methodolo- micelle size (MW = 1200–1500 Da), was CMC (0.29 mM) and high micelle size gies may not be the optimal approach for effectively cleared to below the reliable (MW = 90 kDa), could only be effi- detergent removal.

magentablackcyanyellow ES326336_Spec1013_038.pgs 09.26.2013 04:20 ADV www.spectroscopyonline.com October 2013 Spectroscopy 28(10) 39

Conclusion FT-IR spectroscopy is a rapid, universal, and highly effective 4.0 method for detection and analysis of biomolecules suspended RIPA 10K RIPA 30K in solution that is based on the unique spectral signatures RIPA 100K of analyzed species. More specifically, the work presented 3.5 CytoBuster 10K CytoBuster 30K here demonstrated the application of FT-IR for simultaneous CytoBuster 100K determination of protein and detergent content in biological 3.0 samples. Most notably, reported quantitation requires single-

time instrument calibration with the detergent of interest 2.5 permitting standardization across both sample types and processing runs. Protein quantitation is performed using 2.0 preloaded, standard NIST BSA calibration. In addition, the reported method can be used in parallel with the majority of known detergent removal techniques. Overall, the results 1.5 suggest that optimization of a sample preparation workflow, including detergent removal, are confounded not only by the 1.0 nature of the starting material (cell or tissue lysate), but also Percentage of remaining protein/percentage remianing detergent by the detection requirements and sensitivities of the down- 0.5 1 2 3 stream analytical tools. Spin number References Figure 6: Efficiency of detergent removal from bovine liver lysates (1) R.M. Garavito and S. Ferguson-Miller, J. Biol. Chem. 276, (starting total protein concentration 3 mg/mL) performed using Amicon 32403–30406 (2001). Ultra 2.0 centrifugal devices (10, 30, and 100 kDa NMWL). 1× RIPA and (2) J.M. Neugenbauer, Methods Enzymol. 182, 239–253 (1990). 1× CytoBuster were used to prepare lysates. Retentates were analyzed (3) C. Bordier, J. Biol. Chem. 256, 1604–1607 (1981). following each of three successive centrifugal “wash” cycles. The graph (4) L.M. Hjelmeland, Methods Enzymol. 182, 253–264 (1990). plots relative cleanup performance (% protein retained/% detergent (5) T. Arnold and D. Linke, BioTechniques 43, 427–440 (2007). removed) as a function of spin number. (6) G.G. Prive, Methods 41, 388–397 (2007). (7) B.S. Antharavally, Curr. Protoc. Protein Sci. 69, 6.12.1–6.12.7 (2012). (8) B.S. Antharavally, K.A. Mallia, M.M. Rosenblatt, A.M. Spectroscopy Salunkhe, J.C. Rogers, P. Haney, and N. Haghdoost, Anal. Biochem. 416, 39–44 (2011). (9) L.M. Hjelmeland, Methods Enzymol. 182, 277–282 (1990). Sampling Solutions (10) A.M. Seddon, P. Curnow, and P.J. Booth, Biochim. Biophys. Acta. 1666, 105–117 (2004). (11) F. Rusconi, E. Valton, R. Nguyen, and E. Dufourc, Anal. Bio- Accessories for analysis of chem. 295, 31–37 (2001). (12) A. Urbani and T. Worne, Anal. Biochem. 336, 117–124 gas, solid, and liquid samples. (2005). Contact us with your (13) D.R. Gagnon, R.M. Pieper, and J.E. Aysta, Patent application requirements. US005764355A (1998). (14) E. Chernokalskaya, V. Joshi, P. Clark, C. Utzat, R. Amara, and Complete list of products T. Rider, U.S. Provisional Patent Application No. 61/475434 available in the new catalog. (2011). Call, or download Ivona Strug, Sara Gutierrez, Amedeo Cappione your free copy on-line. III, Mayra Jimenez, Mary Jane Mullen, and Timothy Nadler are with the Bioscience Division of EMD Millipore Corporation, in Danvers, Massachusetts. Direct correspon- dence about this article to: [email protected] ◾ FTIR, NIR and UV-Vis sampling made easier

www.piketech.com For more information on this topic, please visit our homepage at: www.spectroscopyonline.com [email protected] tel: 608-274-2721

magentablackcyanyellow ES326337_Spec1013_039.pgs 09.26.2013 04:20 ADV 40 Spectroscopy 28(10) October 2013 www.spectroscopyonline.com

Speciation Analysis Using HPLC–ICP-MS: Arsenic in Rice

The contamination of apple juice and rice with potentially harmful arsenic compounds has been highlighted in the news recently, including in a November 2012 article from Consumer Reports (1). For most people, rice is far and away the major source of potentially harmful arsenic compounds in our diet. Relevant agencies have recently issued revised guidelines for human consumption, and there is discussion concerning possible regulation of the arsenic content of foods. Analytical methods capable of distinguishing between inorganic and methylated compounds are necessary to support any such regulation, because arsenic compounds vary widely in toxicity. Recent articles in the analytical chemistry literature illustrate the difficulties of making reliable measurements of the arsenic compounds in rice. Sample preparation is challenging, there is no agreement over the best chromatographic conditions, compound-dependent responses are evident, validation is difficult (if not impossible), and there is ambiguity over the severity of starch matrix effects. In a recent web seminar, Julian Tyson, professor of chemistry at the University of Massachusetts, Amherst, explained how to develop a reliable method for arsenic speciation. Below, he answers questions raised during the web seminar. A recording of the web seminar titled “Speciation Analysis: A Critical Look at Methods Involving HPLC with ICP-MS Detection, with a Focus on Rice” is available for free at http://www.spectroscopyonline.com/ArsenicRiceWebinar2013

What is the source of the arsenic that is present in rice? Cooking in excess water and then discarding the water will Tyson: The arsenic compounds are either in the soil or in decrease the amount of arsenic compound consumed. More the irrigation water. The compounds in the soil already may details appear in a 2009 paper by Raab and colleagues (2). include a significant residue from past agricultural practices, which, in the case of cotton growing in the United States, How much monomethylarsonate (MMA) do you expect to find involved spraying with dimethylarsinate (cacodylic acid), in rice? monomethyl arsenate, and arsenic acid. Tyson: I expect to find just single-digit microgram-per-kilo- gram levels of arsenic as monomethylarsonate. Almost all the Could genetically engineered rice reduce the amount of arsenic reports of arsenic speciation in rice from around the world in rice? show that MMA amounts to no more than a few percent of Tyson: I think that genetic engineering could help. In the total arsenic. Often, it cannot be detected at all. The one fact, Professor Om Parkash here at the University of Mas- possible exception is Korean rice. See Paik and colleagues (3). sachusetts is working on this. You can read more at http://www.bio.umass.edu/plantbio/faculty/parkash.html. What rice certified reference material is available for the arsenic species? Given that inorganic arsenic compounds are polar, would they Tyson: According to the web site of the National Institute of be extracted from the rice (into the water) during cooking? Standards and Technology (NIST), both the rice and wheat Tyson: Yes, all of the arsenic species are extracted by hot flour reference materials are out of stock. Fortunately, as was water. The problem is that in standard rice cooking methods, pointed out by a member of the web seminar audience, the the cooking water is absorbed into the rice and thus eaten. National Metrology Institute of Japan (NMIJ) has issued

magentablackcyan ES326300_Spec1013_040.pgs 09.26.2013 04:18 ADV magentablackcyanyellow ES328823_SPEC1013_041_FP.pgs 09.27.2013 21:31 ADV 42 Spectroscopy 28(10) October 2013 www.spectroscopyonline.com

material CRM 7503-a, Arsenic Com- do not exclude any particles by siev- use the standard nebulizers and spray pounds and Trace Elements in White ing. Heat under pressure. Degrade the chambers that come with our ICP in- Rice. CRM 7503-a is certified for ar- starch with trifluoracetic acid and cen- strument. senite, arsenate, and dimethylarsinate trifuge (filtration not needed). Then se- — which account for 97 of the 98 µg/ lect a high performance liquid chroma- Can you please explain why gradient elu- kg total. Inorganic arsenic accounts tography (HPLC) procedure in which tion is a problem? I am using ion chro- for 86% of the total. This standard is a arsenate is eluted first, so that you are matography (IC). white rice flour (whereas the NIST ma- not waiting for the other components to Tys on: The problem that can arise with terial was a brown rice flour, I believe). be eluted first. Use isocratic elution, but gradient elution is that the change in The Japanese reference material is avail- investigate the effect of increasing the mobile-phase composition changes the able in the United States from Wako temperature (if you can). conditions in the plasma, so that the sig- Chemicals (http://www.wakousa.com), nal for a given flux of arsenic going into with a lead time of about three weeks. Could you describe the interface between the plasma changes, and so you see what Even if you do buy this material, the liquid chromatography (LC) and the appears to be compound-dependent re- however, you should also measure spike inductively coupled plasma (ICP) instru- sponses, but those are probably mostly recoveries. The spikes should be added ments? mobile-phase composition dependent before the sample digestion–extraction. Tyson: Typically, all that is needed to responses. Once you have decided on connect these two instruments is an ap- your gradient, you will almost cer- What would be the best overall method propriate length of PEEK tubing. The tainly need to calibrate the method by for measuring total inorganic arsenic, in- flow rate of most LC separations is com- injecting a mixture of the species you cluding the digestion and other sample patible with that of conventional ICP want to determine and creating separate preparation steps? nebulizers, so the column eluent can be calibration curves for each species. How Tyson: If you are not interested in dif- delivered directly to the ICP nebulizer. many standards to use is your decision. ferentiating between arsenite and arse- We have used miniature spray chambers We usually check that the calibration nate, try a procedure in which peroxide in the past in an effort to minimize ex- is linear and then inject one (or maybe is a component of the extraction re- tracolumn broadening, but the effects two) standards. How long before you agents. Grind to less than 200 µm and are relatively minor and currently we calibrate again is a decision you need to make based on your experience with the stability of your instrument. If you add an arsenic compound as an internal standard, you could use any changes in the response to this compound as a diagnostic for drift in your instrument response. The other issue, as I am sure you are aware, is that with a gradient you need to re-equilibrate the column before injecting the next sample. This can be quite time consuming. We have approached this in the past by using two columns, one of which was being re-equilibrated while the other was in analysis mode. (This approach also requires another pump, of course.)

What are your thoughts on the use of speciated isotope dilution ICP-MS techniques for speciation? Tyson: Speciated isotope dilution ICP- MS is a powerful concept for the inves- tigation of species interconversion in an analytical method. I believe that most of the work done so far has been with mercury, tin, and chromium. I do not think that the technique can work for arsenic, which is monoisotopic in nature, and although isotopes ranging from 67As to 86As are known, none has a half-life

magentablackcyanyellow ES326301_Spec1013_042.pgs 09.26.2013 04:18 ADV www.spectroscopyonline.com October 2013 Spectroscopy 28(10) 43

longer than three months. There was a or electrochemistry was the measure- inger, R. Montoro, J.J. Sloth, R. Rubio, M.J. poster (820-7) at the 2012 Pittcon con- ment technique would not be included. Baxter, J. Feldmann, P. Vermaercke, and ference from Skip Kingston’s group at G. Raber, TrAC Trends Anal. Chem. 30(4), Duquesne describing work in which References 6 41– 651 (2011). arsenite had been labeled with 18O. Fol- (1) Consumer Reports, November 2012 (5) R. Clough, L.R. Drennan-Harris, C.F. Har- lowing LC separation, the species were (http://www.consumerreports.org/cro/ rington, S.J. Hilld, and J.F. Tyson, J. Anal. detected by organic mass spectrometry. magazine/2012/11/arsenic-in-your-food/ At. Spectrom. 27, 1185–1224 (2012). ◾ My take on the literature evidence is index.htm). that under mild acidic or basic condi- (2) A. Raab et al., J. Environ. Monitor. 11, 41 tions (and heating to 95 °C under pres- (2009). For more information on this topic, sure), arsenic species can be extracted (3) M.-K. Paik et al., J. Korean Soc. Appl. Biol. please visit our homepage at: from rice without significant species Chem. 53, 634 (2010). www.spectroscopyonline.com transformation. (4) M.B. de la Calle, H. Emteborg, T.P.J. Lins-

You mentioned that a European pro- ficiency test was conducted to see how accurately arsenic could be measured in a variety of laboratories. What was the outcome of that test in terms of the total arsenic determination? Tyson: In 2009, the European Union Reference Laboratory for Heavy Met- als in Feed and Food organized a pro- ficiency test, IMEP-107, on the determination of total and inorganic arsenic (As) in rice. The goal of the test was to assess laboratories’ ability to determine total and inorganic arsenic in rice. Ap- proximately 98 laboratories reported results for total As and 32 for inorganic As. The outcome for total arsenic determination was about the same as for the speciation analysis: 41% of the 98 participants do not get a satisfactory result. There is a slide in my presentation showing the results. Details of this study are available in the literature (4).

You said that about 70 papers are published every year dealing with arsenic speciation, but the “Atomic Spectrom- etry Update” article only had 40 citations. Why? Tyson: In recent years, an “Atomic Spectrometry Update” review on de- velopments in elemental speciation has been published annually in the Journal of Analytical Atomic Spectroscopy (5). The writers of these articles (of which I am one) are selective in terms of which articles are included in the update. Also, there are quite a lot of articles describing work that does not involve atomic spectrometry as the measurement technique. Work in which a method such as molecular mass spectrometry (possibly in conjunction with gas chromatography)

magentablackcyanyellow ES326299_Spec1013_043.pgs 09.26.2013 04:18 ADV 44 Spectroscopy 28(10) October 2013 www.spectroscopyonline.com Products & resources Spectroscopy and imaging systems UVÐvis spectrophotometer Benchtop spectroscopy and The double-beam imaging systems from Advan- UH5300 UV–vis spectro- test America are designed photometer from Hitachi for use in for pharmaceutical is designed with a 1-nm development and quality con- spectral bandwidth. trol. The systems reportedly The spectrophotometer use terahertz wave technol- reportedly has a seven- ogy to acquire characteristic year factory warranty on spectra based on crystalline its pulsed xenon lamp. structure. According to the According to the com- company, 3-D images provide information about tablet coating thick- pany, standard equipment includes a six-position automatic cell ness and uniformity and enable nondestructive detection of cracks changer. and voids in the tablet core. Advantest America, Inc., Hitachi High Technologies America, Inc., San Jose, CA; www.advantest.com Dallas, TX; www.hitachi-hta.com

FTÐIR Miniature spectrophotometer spectrometer The IRTracer-100 Fourier The Embeda miniature spec- transform infrared spectro- trometer from Ocean Optics,is photometer from Shimadzu designed as a small-footprint is designed for sample analy- optical bench with minimal sis in fields such as pharma- electronics targeted at indus- ceuticals, foods, chemicals, trial process control and OEM and electronics. According to optical measurement systems. the company, the instrument achieves a resolution of According to the company, 0.25 cm-1 and a sample acquisition rate of 20 spectra the instrument includes a 2048-element CCD-array detector and per second. selectable optical bench components, including entrance aper- Shimadzu Scientific Instruments, Columbia, MD; tures, gratings, and order sorting filters. www.ssi.shimadzu.com Ocean Optics, Dunedin, FL; www.oceanoem.com

Microwave ATR accessory sample prepara- attachment tion system An expanded range of attach- The Titan MPS microwave ments for PIKE’s MIRacle single- sample preparation sys- reflection ATR accessory includes tem from PerkinElmer is a specular reflection plate that designed to monitor and reportedly allows most sampling control digestions with plates to be temperature con- contact-free and con- trolled via a dedicated digital unit nection-free sensing via or from a PC with the company’s direct pressure control and direct temperature control monitoring TempPRO software. According to systems. According to the company, the instrument is available in the company, a liquid retainer or flow-through attachment can be added 8- and 16-vessel configurations. to the accessory for working with flowing or difficult samples. PerkinElmer, Waltham, MA; PIKE Technologies, Madison, WI; www.piketech.com www.perkinelmer.com/titanmps

NMR spectrometer UVÐvisÐNIR system Thermo Fisher’s picoSpin The Cary 7000 Universal 80 benchtop NMR spec- Measurement spectro- trometer is designed for photometer from Agilent the collection of routine Technologies is designed high-resolution NMR data to provide measurements of transmission, absolute reflection, on molecules in solution. and scattering without moving the sample. According to the According to the com- company, the system is automated and provides 10 absorbance pany, the instrument has unit capability. a magnet rated at 2 Tesla Agilent Technologies, and can provide resolu- Santa Clara, CA; tion of chemical and structural characteristics not available using www.agilent.com/chem/cary7000UMS lower magnetic field strengths. Thermo Fisher Scientific, San Jose, CA; www.picospin.com

magentablackcyanyellow ES328625_Spec1013_044.pgs 09.27.2013 20:48 ADV Evaluate, Educate, Explore

WiƩ con is tŚe leading conĨerence and eǆposiƟ on Ĩor tŚe latest advances in Laboratory Science. Ʃ ending WiƩ con gives you a uniƋue opportunity to get a hands-on look at cuƫ ng-edge product innovaƟ ons from leading companies. WarƟ cipate in any of the more than March 2-6, 2014 2,000 technical presentaƟ ons to learn about recent discoveries from Chicago, Illinois ǁorld-renoǁned members of the scienƟ Į c community. Improve or www.pittcon.org develop your skills by taking a short course taught by industry experts. &or more informaƟ on on technical sessions, exhibitors and short courses, visit www.piƩ con.org.

magentablackcyanyellow ES328068_SPEC1013_045_FP.pgs 09.27.2013 03:44 ADV 46 Spectroscopy 28(10) October 2013 www.spectroscopyonline.com

Silicon drift detector Detector EDAX’s thermoelectrically Andor’s iDus LDC-DD 416 cooled 50-mm2 silicon drift platform is designed to pro- detector is designed for vide a combination of low dark use in its Orbis micro-XRF noise and high QE. According elemental analyzer system to the company, the detector for high-resolution spectral is suitable for NIR, Raman, acquisition. According to the and photoluminescence and company, the system can be can reduce acquisition times, useful for those who make removing the need for liquid measurements on small frag- nitrogen cooling. ments, coatings, and deposits on thin substrates (such as ink on Andor Technology, paper), biological samples, and trace element analysis using heavy South Windsor, CT; filters to improve sensitivity. www.andor.com EDAX, Mahwah, NJ; www.edax.com

1064-nm NIR Raman spectrometers Raman microscope Portable Raman spectrom- Renishaw’s inVia Raman micro- eters from StellarNet are scope is designed with a 3D designed to use InGaAs imaging capability that allows detectors to capture NIR users to collect and display spectra induced by 1064- Raman data from within trans- nm lasers. According to the parent materials. According to company, the high-resolution the company, the instrument’s model (Raman-HR-TEC-IG) StreamLineHR feature collects achieves 8 cm-1 resolution data from a series of planes with a 1024-element PDA detector, and the standard within materials, processes it, resolution instrument achieves 16 cm-1 resolution and has a and displays it as 3D volume images representing quantities 512-element PDA detector. such as band intensity. StellarNet Inc., Tampa, FL; Renishaw, Hoffman Estates, IL; www.stellarnet.us/PopularConfigurations_ramansystems.htm www.renishaw.com/raman

2014 Sensing

Technology+ CO-LOCATEDCO-LOCATED WITHWITH 2014 Applications Defense+ SPIE DSS Security SPIE DSS Commercial, industrial and scientiﬁc advancements Call for Papers Abstracts due 22 October 2013 — www.spie.org/dss1

Conferences & Courses Location 5–9 May 2014 Baltimore Convention Center Baltimore, Maryland, USA DSS Expo 6–8 May 2014

magentablackcyanyellow ES326318_Spec1013_046.pgs 09.26.2013 04:19 ADV www.spectroscopyonline.com October 2013 Spectroscopy 28(10) 47

Raman spectrometer system for Microspectrometer process control Horiba Scientific’s MicOS A Raman spectrometer microspectrometer system for process con- reportedly combines a trol from tec5USA is microscope head directly based on the company’s coupled to a three-grating MultiSpec spectrometer imaging spectrometer and system. According to the a research-grade scientific company, the system camera. According to the includes a laser light company the instrument is source, a high-resolution compatible over the UV– spectrometer, and a fiber-optic Raman probe. visible and NIR (200 nm to tec5USA, 1600 nm) spectral range. Plainview, NY; Horiba Scientific, Edison, NJ; www.tec5usa.com www.microspectroscopy.com

Mercury analyzer Microwave digestion The QuickTrace M-7600 Milestone’s UltraWAVE micro- mercury analyzer from CETAC wave digestion system uses Technologies is designed the company’s single reaction with TCP–IP communica- chamber technology for use in tions that are preconfigured metals digestions. According to per laboratory preferences. the company, the system uses According to the company, a single pressurized vessel for the analyzer is designed for all samples, allowing for simul- ultratrace to sub-parts-per- taneous digestion of up to 22 million mercury quantitation samples. The system reportedly and is capable of switching between low parts-per-million and can accommodate a maximum parts-per-trillion analysis without hardware or tubing temperature of 300 °C and pres- configuration changes. sure of 199 bar. CETAC Technologies, Omaha, NE; www.cetac.com Milestone, Inc., Shelton, CT; www.milestonesci.com/ultrawave

Simultaneous Fluorescence and Absorbance for Rapid, Reagent-Free Monitoring of Drinking and Waste Water ON DEMAND WEBCAST

EVENT OVERVIEW: KEY LEARNING OBJECTIVES: This web seminar describes the advantages of the simultaneous fuorescence excitation-emission mapping and absorbance spectra (FEEM-Abs) method for analyzing 8IBUJTþVPSFTDFODFFYDJUBUJPOFNJTTJPO mining water, waste water and drinking water. Highlighted features of the method include mapping and simultaneous absorbance the capabilities for rapid (seconds to minutes) and essentially continuous monitoring spectral analysis and why is it so of membrane-based mining water decontamination, biochemical oxygen demand rapid and sensitive to organic chemicals. (BOD) in waste water, and disinfection by-product (DBP) precursors in addition to algal 2) Understand that biochemical oxygen contaminations in drinking water. As key examples it is shown that the conventional 5 day demand (BOD) and disinfection byproduct BOD and DBP formation potential test results correlate strongly with spectral components formation (DBP) potential tests take up to quantifed in seconds to minutes by the FEEM-Abs method. 5 days while fow through a treatment plant can be measured in hours. EVENT TEASER APPEARING IN E-SOLUTIONS MAILINGS: 3) Understand why fuorescence is sensitive to biochemical oxygen demand parameter PRESENTER: Learn how to save weeks of time, effort and money in waste water treatment. using rapid (seconds to minutes) Adam Gilmore reagent-free fuorescence and absorbance 4) Understand why fuorescence is sensitive monitoring for biochemical oxygen demand and to disinfection by-product precursors in trihalomethane formation potential tests in waste drinking water treatment. and drinking water treatment, respectively. 5) How can fuorescence help control WHO SHOULD ATTEND: coagulant and treatment conditions to reduce disinfection by-product formation t8BUFSUSFBUNFOUFOHJOFFSTBOEDIFNJTUTSFTQPOTJCMFGPSSFTFBSDIJOHBOEPSJOUFSFTUFEJO potential. implementing improved technology for waste and drinking water treatment. t3FTFBSDITDJFOUJTUTGSPNFOWJSPONFOUBM HFPDIFNJDBMBOEIZESPMPHZýFMETJOUFSFTUFEJO Sponsored by Presented by transferring technology to industrial water treatment. ® For questions, contact Kristen Farrell at [email protected]

magentablackcyanyellow ES326317_Spec1013_047.pgs 09.26.2013 04:19 ADV 48 Spectroscopy 28(10) October 2013 www.spectroscopyonline.com

STATEMENT OF OWNERSHIP, MANAGEMENT, AND CIRCULATION Hierarchical modeling software (Requester Publications Only) The Unscrambler X hierarchical model (Required by 39 USC 3685) development module and the Unscrambler 1. Publication Title: Spectroscopy X hierarchical model engine from CAMO Hierarchical modeling solutions Software are designed for R&D, production, 2. Publication Number: 0887-6703 Hierarchical Model Hierarchical Model Development Module Engine engineering, and QA–QC environments that 3. Filing Date: 9/30/13 require developing and implementing run- 4. Issue Frequency: Monthly time control models for process monitoring 5. Number of Issues Published Annually: 12 application. According to the company, they 6. Annual Subscription Price (if any): $79.95

A powerful decision-tree approach to multivariate 7. Complete Mailing Address of Known Office of Publication: 131 West First Street, Duluth, are suitable for use by production supervi- modeling of complex data St. Louis County, Minnesota 55802-2065 Make modeling of Join ClassiÀcation and Use Hierarchical Models non-linear data more Prediction models using on-line with instrument sors, quality assurance teams, technical ser- efÀcient and robust multi-level modeling integration Contact Person: Jason McConnell vices, and control system designers. Telephone: (218) 740-6359 CAMO Software, 8. Complete Mailing Address of Headquarters or General Business Office of Publisher: Woodbridge, NJ; 2501 Colorado Avenue, Suite 280, Santa Monica, CA 90404. www.camo.com 9. Full Names and Complete Mailing Addresses of Publisher: Michael J. Tessalone, 485F US Highway 1 South, Ste. 210, Iselin, NJ 08830 Editorial Director: Laura Bush, 485F US Highway 1 South, Ste. 210, Iselin, NJ 08830 Managing Editor: Megan Evans, 485F US Highway 1 South, Ste. 210, Iselin, NJ 08830 10. This publication is owned by: Advanstar Communications Inc., 2501 Colorado Avenue, Suite 280, Santa Monica, CA 90404. The sole shareholder of Advanstar Communications Inc. is: Advanstar, Portable Raman system Inc., whose mailing address is 2501 Colorado Avenue, Suite 280, Santa Monica, The i-Raman EX 1064-nm por- CA 90404. table Raman spectrometer from 11. Advanstar Communications Inc. is a borrower under Credit Agreements dated June 6, 2013, with B&W Tek is designed as an exten- various lenders as named therein from time to time. As of June 6, 2013, the agent for the lenders is: Goldman Sachs Lending Partners LLC, Administrative Agent, 30 Hudson St, 4th Floor, Jersey sion of the company’s i-Raman City, NJ 07302. portable spectrometer. According 12. Does Not Apply to the company, the system is 13. Publication Title: Spectroscopy equipped with a high-sensitivity 14. Issue Date for Circulation Data Below: August 2013 inGaAs array detector with deep 15. Extent and Nature of Circulation TE cooling and a high dynamic range. Average B&W Tek, No. Copies No. Copies of Each Issue Single Issue Newark, DE; During Published www.bwtek.com/products/i-raman-ex/ Preceding Nearest to 12 Months Filing Date A. Total Number of Copies 22,005 21,573 B. Legitimate Paid and/or Requested Distribution 1. Outside County Paid/Requested Mail Subscriptions Stated on PS Form 3541 18,403 16,947 X-ray fluorescence analyzer 2. In County Paid/Requested Mail Subscriptions The Mesa 50 energy dispersive Stated on PS Form 3541 0 0 X-ray fluorescence analyzer 3. Sales Through Dealers and Carriers, from Horiba Scientific is Street Vendors, Counter Sales, and Other designed for screening samples Paid or Requested Distribution Outside USPS 49 49 4. Requested Copies Distributed by containing hazardous elements Other Mail Classes Through the USPS 0 0 such as lead, cadmium, chro- C. Total Paid and/or Requested Circulation mium, mercury, bromine for (Sum of 15b (1), (2), (3), and (4)) 18,452 16,996 RoHS, and chlorine for halogen- D. Non-requested Distribution free As and Sb applications. 1. Outside County Non-requested Copies According to the company, the Stated on PS Form 3541 3,064 3,378 analyzer is available with four 2. In-County Non-requested Copies filters and three spot sizes. Stated on PS Form 3541 0 0 Horiba Scientific, Edison, NJ; www.horiba.com 3. Non-requested Copies Distributed Through the USPS by Other Classes of Mail 0 0 4. Non-requested Copies Distributed Outside the Mail 302 1,097 E. Total Non-requested Distribution (Sum of 15d (1), (2), (3) and (4)) 3,366 4,475 Handheld NIR spectrometer F. Total Distribution (Sum of 15c and e) 21,818 21,471 The handheld microPHAZIR AG G. Copies not Distributed 187 102 analyzer from Thermo Scien- H. Total (Sum of 15f and g) 22,005 21,573 tific is designed to allow manu- I. Percent Paid and/or Requested Circulation 84.57% 79.16% facturers to perform on-site 16. □ Total circulation includes electronic copies. NIR analysis to test ingredients Report circulation on PS Form 3526-X worksheet. and finished feeds to optimize 17. Publication of Statement of Ownership for a Requester Publication is required and will be nutrient formulations, reduce printed in the October issue of this publication. manufacturing costs, and Name and Title of Editor, Publisher, Business Manager, or Owner: Anne Brugman, Audience Development Director improve consistency. According Date: 9/30/2013 to the company, results can be directly exported into spread- I certify that the statements made by me above are correct and complete. sheets, laboratory information management systems, or feed control systems. Thermo Fisher Scientific, San Jose, CA; www.thermoscientific.com/feed

magentablackcyanyellow ES326319_Spec1013_048.pgs 09.26.2013 04:19 ADV Presented by NANOPARTICLE CHARACTERIZATION USING ® ICP-MS FOR FOOD ANALYSIS: WHAT ARE Sponsored by WE LOOKING FOR, AND WHY?

LIVE WEBCAST: Tuesday, September 24, 2013 at 8:00 am PDT • 11:00am EDT • 16:00 BST • 17:00 CEST

EVENT OVERVIEW: WHO SHOULD ATTEND: Two principal techniques for nanoparticle characterization have been de- • The food safety/ veloped in recent years using ICP-MS as a detection system; Field-Flow- testing community Fractionation (FFF) and direct analysis of single particles. • Academic researchers in food Join this webinar to discover the pros and cons of the two techniques or nanoparticle technology and gain valuable insight into new challenges emerging from impending • Nanoparticle manufacturers legislation for nanoparticle characterization in foods. • Toxicologists The increasing use of engineered nanoparticles in various materials and • Membes of legislative food has sparked concern over the potential risks to human and envi- and regulatory bodies ronmental health. Although nanoparticles can also be naturally occur- • Cosmetics analysts ring, the exceptional properties of nanoparticles, such as the superior antibacterial properties of silver nanoparticles, has increased their use • Pharmaceutical analysts in several applications and has urged legislative bodies to define what is understood by nanoparticles and to evaluate means to characterize PRESENTERS: them. Due to the elemental content of nanoparticles, ICP-MS is often an attractive tool for their detection. Shona McSheehy Ducos ICP-MS Product Manager KEY LEARNING OBJECTIVES: Thermo Fisher Scientific 1. Discover what type of new legislation is anticipated for food analysis regarding nanoparticles Erik H. Larsen Senior researcher and 2. Find out what is meant by a number-based vs. group leader mass-based distribution of nanoparticles Technical University 3. Single particle ICP-MS (spICP-MS) – how to do of Denmark National Food Institute it and the pros and cons of the technique 4. Field flow fractionation ICP-MS (FFF-ICP-MS) – MODERATOR: how to do it and the pros and cons of the technique Laura Bush 5. Understand what type of information can be EditoriaL Director gathered from these techniques Spectroscopy 6. Find out what is required for sample preparation to analyse nanoparticles For questions, contact Kristen Farrell at [email protected]

magentablackcyanyellow ES328824_SPEC1013_049_FP.pgs 09.27.2013 21:31 ADV 50 Spectroscopy 28(10) October 2013 www.spectroscopyonline.com ® S

Is your X-RAY ֢ ֣ ֤ ֥ ֦֧֨ ֢֩֡ ֪֥ ֥ ְֱ֦֧֧֭֮֫֬֡֡֯֯֫֬֬֫֬֡Components֡֡ ֢ ֣֡ ֤ ֤֡ ֥ ֤ ֦ ֧ ֣ Supplier֡ ֡֡ ֡ Hitting the Mark?

֡ Innovative Solutions #1 Customer Satisfaction Quality Products www.moxtek.com ֩ ְֱִֵֶַ֧֧֧֦֨֯֡֡֬֩ ִ֣ ֣ ֲֳ֦֦֪֡ ֺֺ 452 W 1260 Nָ ֺ Orem, UT 84057ָָ

!∀∀!# ∃%%& ® 1.800.758.3110 ָֺ∋ָֹֺֺֻּּֽ MOXTEK ASSU┾ Call for Application Notes Spectroscopy is planning to publish the next edition of The Application Notebook in December 2013. As always, the publication will include paid position vendor application notes that describe techniques and applications of all forms of spectroscopy that are of immediate interest THE MOST SENSITIVE to users in industry, academia, and government. HANDHELD RAMAN ANALYZER If your company is interested in participating in this special supplement, contact: ¾Rapid Analysis to Verify 100% of In- coming Raw Material Michael J. Tessalone, Group Publisher • (732) 346-3016 ¾F ULL 21 CFR Part 11 Compliant Edward Fantuzzi, Publisher • (732) 346-3015

or Enwave Optronics, Inc. Stephanie Shaffer, East Coast Sales Manager • (508) 481-5885 ZZZHQZDYHRSWFRP_VDOHV#HQZDYHRSWFRP_7HO---

Ad Index ADVERTISER PG# ADVERTISER PG# ADVERTISER PG# Advantest America, Inc. 42 EMD Millipore Corporation 21 Ocean Optics, Inc. 7 Agilent Technologies 3 Enwave Optronics, Inc. 50 PerkinElmer Corporation 13, 15 Amptek 10 Harrick Scientific Group 9 Photonis 35 Applied Rigaku Technologies 31 Hitachi High Technologies America CV3 PIKE Technologies 28, 29, 39 Avantes BV 50 Horiba Scientific 47, CV4 Pittcon 45 B&W Tek, Inc. 33, 37 Metrohm USA 27 Renishaw, Inc. 23 Camo Software AS 25 Milestone, Inc. 19 Shimadzu Scientific Instruments 5 CEM Corporation 41 Moxtek, Inc. 17, 50 SPIE 46 CETAC Technologies, Inc. 6 New Era Enterprises, Inc. 50 StellarNet, Inc. 43 EDAX, Inc. 32 Nippon Instruments North America 30 Thermo Fisher Scientific CV2, 4, 49

magentablackcyanyellow ES326353_Spec1013_050.pgs 09.26.2013 04:20 ADV Spectroscopy Unleashed The New Way to Work in Your Lab

The New Hitachi UH5300 double-beam spectrophotometer is equipped with the latest IT technology and an environmentally conscious design. System control is achieved wirelessly via an iPad® or wireless-enabled PC, or wired with a PC. Its long-life pulsed Xenon lamp is covered by a market-leading 7-year factory warranty(*). With a wide range of accessories available, the UH5300 is a cost-effective solution bringing a new way of work to your laboratory.

Learn more about Hitachi's spectroscopy products:

www.hitachi-hta.com/spectroscopy

Hitachi—The Reliability Leader in Spectroscopy

Hitachi High Technologies America, Inc. toll free: 800-548-9001 (US & Canada) www.hitachi-hta.com email: [email protected]

magentablackcyanyellow ES328094_SPEC1013_CV3_FP.pgs 09.27.2013 03:45 ADV If you didn’t get a chance to see us at SCIX 2013, contact us for information on our products, posters and presentations.

www.horiba.com/raman email: [email protected]

magentablackcyanyellow ES328073_SPEC1013_CV4_FP.pgs 09.27.2013 03:44 ADV