How Microscope Objectives Affect the Raman Spectra of Crystals

NotebookApplication SpecialSection

September 2017 Volume 32 Number 9 www.spectroscopyonline.com

What do you need to digest today?

How Microscope Objectives Affect Environmental & Food, Plant & Inorganic Regulatory Animalthe Tissue Raman SpectraCompounds of Crystals

Metallurgical Oils & Plastics What’s PharmaceuticalsNew in the New USP <1058>? In Situ Raman Spectroscopy for Monitoring Reaction Kinetics

2017 Emerging Leader in Molecular Spectroscopy Award Winner Russ Algar iPrep can digest it all

MARS 6 has always been your go to system for digesting routine samples like foods, plants, and soils and now with iPrep you can digest your toughest samples like Kevlar® and Torray Filters or mixed samples like oils and plastics in a single run. At CEM we are dedicated to getting you better digestions so that you achieve better analyses.

The MARS 6 with iPrep vessels and iWave simultaneously processed the four different oil and plastic samples. Each sample was run in duplicate for a total of eight samples. The system automatically adjusts the power instantly to compensate for the varied sample types.

Precise power application provides superior temperature control

Bunker Oil Max Size: 0.25 grams Result: Clear

Before After

From routine to extreme, MARS 6 can digest it all.

www.cem.com/mars6 NotebookApplication SpecialSection

September 2017 Volume 32 Number 9 www.spectroscopyonline.com

How Microscope Objectives Affect the Raman Spectra of Crystals What’s New in the New USP <1058>? In Situ Raman Spectroscopy for Monitoring Reaction Kinetics

2017 Emerging Leader in Molecular Spectroscopy Award Winner Russ Algar

State-of-the-Art X-Ray Solutions t'"454%%¥ t4%% t4J1*/ New in-house manufacturing = The highest performing detectors available

The best detector for optimal results

240

Si-PIN 220 FAST SDD® Detector 25 mm2 Si-PIN >2,000,000 CPS and 122 eV FWHM Resolution 200 13 mm2 Options: 180 Si-PIN t NN2DPMMJNBUFEUPNN2 6 mm2 160 t NN2DPMMJNBUFEUPNN2 SDD 25 mm2 t 8JOEPXT#F NJM N PS$4FSJFT 4J N 140 FAST SDD® 3 4 2 2 Typical Resolution (eV FWHM @ 5.9 keV) Typical 25 mm and 70 mm t 50QBDLBHFGJUTBMM"NQUFLDPOGJHVSBUJPOT 120 0 5 10 15 20 25 30 Peaking Time (μs)

Confi gurations to meet your needs

93BOE19%JHJUBM1VMTF1SPDFTTPS 9$PNQMFUF93BZ4QFDUSPNFUFS 93'&YQFSJNFOUFST,JU

OEM Confi gurations

"TBNQMFPGEFUFDUPSTXJUIQSFBNQMJöFST %JHJUBM1VMTF1SPDFTTPSTBOE1PXFS4VQQMJFT BOEIFBUTJOLT

® For XRF www.amptek.com 4 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com

LED Sources for Spectroscopy ®

With User-Tailorable Spectrum

MANUSCRIPTS: To discuss possible article topics or obtain manuscript preparation guidelines, contact the editorial director at: (732) 346-3020, e-mail: Laura.Bush@ ubm.com. Publishers assume no responsibility for safety of artwork, photographs, or manuscripts. Every caution is taken to ensure accuracy, but publishers cannot accept responsibility for the information supplied herein or for any opinion expressed. 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. 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 postage and handling ($19.00 total); Other international, $15.00 + $7.00 postage and handling ($22.00 total). CHANGE OF ADDRESS: Send change of address to Spectroscopy, P.O. Box 6196, ƕUV-VIS-NIR Fiber-Coupled LED Duluth, MN 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 ƕStrong Blue & UV Component address changes or subscription renewal: http://ubmsubs.ubm.com/?pubid=SPEC ƕHigh Power & High Stability RETURN ALL UNDELIVERABLE CANADIAN ADDRESSES TO: IMEX Global Solutions, P.O. Box 25542, London, ON N6C 6B2, CANADA. PUBLICATIONS MAIL AGREEMENT No.40612608. ƕSolid State, Long Lifetime 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 bkolb@wrightsmedia. com. Outside US, UK, direct dial: 281-419-5725. Ext. 121 www.mightex.com C.A.S.T. DATA AND LIST INFORMATION: Contact Melissa Stillwell, (218) 740-6831; e-mail: [email protected] Mightex [email protected] 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

© 2017 UBM. All rights reserved. No part of this publication may be reproduced or transmit- ted in any form or by any means, electronic or mechanical including by photocopy, recording, ‘Setting the Standard’ or information storage and retrieval without permission in writing from the publisher. Authori- zation to photocopy items for internal/educational or personal use, or the internal/educational or personal use of specific clients is granted by UBM 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].

UBM Americas 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 UBM Ameri- cas 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 UBM Americas lists. Outside Starna Cells the U.S., please phone 218-740-6477. Spectroscopy does not verify any claims or other information appearing in any of the adver- Quality others strive to match tisements contained in the publication, and cannot take responsibility for any losses or other damages incurred by readers in reliance of such content.

By joining optical surfaces Spectroscopy welcomes unsolicited articles, manuscripts, photographs, illustrations and other materials but cannot be held responsible for their safekeeping or return. by heat fusion alone, Starna revolutionised cell quality and To subscribe, call toll-free 888-527-7008. Outside the U.S. call 218-740-6477. performance. Starna Cells Inc. UBM Americas (www.ubmamericas.com) is a leading worldwide media company providing Vast, competitively priced integrated marketing solutions for the Fashion, Life Sciences and Powersports industries. UBM range for every application (800) 228-4482 Americas serves business professionals and consumers in these industries with its portfolio [email protected] of 91 events, 67 publications and directories, 150 electronic publications and Web sites, as Custom design and well as educational and direct marketing products and services. Market leading brands and a manufacture for anything else! www.starnacells.com commitment to delivering innovative, quality products and services enables UBM Americas to “Connect Our Customers With Theirs. UBM Americas has approximately 1000 employees and currently operates from multiple offices in North America and Europe. That cooled CCD is impressive! Have Lookok aatt the aberration-free speccttra!

You FERGIE’s footprint is smaller than the page you’re Heard? reading now!

High-performance spectroscopy can be easy!

How? Simply plug in the new Princeton Instruments MICROSCOPY FERGIE™ and start your work right away… No calibrating. No aligning. No guessing. PHYSICS

Best of all, true novices and seasoned spectroscopists alike can BIOLOGY easily configure and operate FERGIE for popular applications such as Raman, absorption, microspectroscopy, and fluorescence. CHEMISTRY scientists, educators, Join the many and PHARMA engineers worldwide who are embracing this revolution in spectroscopy! INDUSTRY

“I can honestly say FERGIE is changing the way we do spectroscopy in our lab!” – Dr. Mark Waterland, Massey University, New Zealand

Inst ton rum e e c n in t r s P P To learn more about the innovative FERGIE, as well as our new

S I m c a Scholars Grant Program, please visit: www.princetoninstruments.com/FERGIE h r o g l o ar r s Grant P

Global Sales & Support www.princetoninstruments.com | [email protected] tel: +1 609.587.9797

6 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com

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

Michael J. Tessalone Vice President/Group Publisher [email protected] Stephanie Shaffer Publisher [email protected] Edward Fantuzzi Associate Publisher [email protected] Michael Kushner Senior Director, Digital Media [email protected] Laura Bush Editorial Director [email protected] Megan L’Heureux Managing Editor Meg.L’[email protected] Stephen A. Brown Group Technical Editor [email protected] Cindy Delonas Associate Editor [email protected] Kristen Moore Webcast Operations Manager [email protected] Vania Oliveira Project Manager [email protected] Sabina Advani Digital Production Manager [email protected] Wavelength Tunable Optical Isolators for use Kaylynn Chiarello-Ebner with laser diodes and low to medium power Managing Editor, Special Projects [email protected] applications between 532-980nm. Dan Ward Art Director Effectively eliminate optical feedback. [email protected] Anne Lavigne Marketing Manager [email protected] Melissa Stillwell C.A.S.T. Data and List Information [email protected] Wright’s Media Reprints [email protected] Maureen Cannon Permissions [email protected] Jesse Singer Production Manager [email protected] Wendy Bong High Quality | Great Performance Audience Development Manager [email protected]

www.eotech.com | [email protected] Ross Burns . 231-935-4044 Audience Development Assistant Manager Electro-Optics Technology, Inc [email protected]

TM A-TEEM Molecular Fingerprinting

There’s more to colored molecules than meets the eye. Many applications require quantification of individual organic compounds in complex mixtures. Traditional methods for these measurements are HPLC, GC-MS, UV-Vis…But now measurements that took a half hour to a day can be done in minutes with Excitation-Emission Matrix (EEM).

HORIBA has combined an ultrafast CCD that’s up to 4,000 times faster than traditional PMT-based TM ® fluorometers, with a new, patented A-TEEM technology in our Aqualog that uses the absorbance, transmittance and EEM data to fingerprint molecules with high specificity and ultrahigh- sensitivity at a 6 million nm/min emission scan rate! A-TEEM can easily and effectively identify, quantify and understand dynamics of molecules in mixtures, under a variety of conditions... in the blink of an eye.

See it for yourself at www.aqualog.com. But don’t blink, or you’ll miss it! horiba.com/scientific email: [email protected] 8 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com

v 32 n 9 CONTENTS s 2017 ® COLUMNS September 2017 Volume 32 Number 9 Molecular Spectroscopy Workbench ...... 14 The Effect of Microscope Objectives on the Raman Spectra of Crystals David Tuschel The Raman spectra of a particular face of a single crystal can be significantly different if acquired with different microscope objectives. This article explains the underlying physics of changes in relative intensity and even peak position of certain Raman bands depending on the microscope objective used to acquire the spectrum.

Focus on Quality ...... 24 What’s New in the New USP <1058>? R.D. McDowall The new version of United States Pharmacopeia general chapter <1058> “Analytical Instrument Qualification” became effective August 1, 2017. What does this mean for you?

IR Spectral Interpretation Workshop ...... 31 The Carbonyl Group, Part I: Introduction Brian C. Smith An introduction to the IR spectroscopy of the carbonyl group, exploring why the peak is intense and showing how to apply that knowledge to the analysis of the spectra of ketones Cover image courtesy of Frank L Junior/Shutterstock. SPECIAL FEATURE The 2017 Emerging Leader in Molecular Spectroscopy Award ...... 37 Megan L’Heureux ON THE WEB Russ Algar, the winner of Spectroscopy’s 2017 Emerging Leader in Molecular Spectroscopy Award, is developing fluorescence assays and measurement systems based on quantum QUIZ: INTERPRETING SPECTRA dots. One day, these methods may enable smartphone-based medical diagnostics. Take the latest quiz! Are your spectral interpretation skills up to par? Find out by taking the latest quiz from our “IR Spectral Interpretation Workshop” column. PEER-REVIEWED ARTICLE

See the quiz on page 34 of this issue or at: In Situ Raman Spectroscopy Monitoring of the Reaction spectroscopyonline.com/ of Sulfur Trioxide with Polyethylene Fibers in Chlorinated Solvents ...... 42 ir-spectral-interpretation-workshop-o Xiaoyun Chen, Jasson Patton, Bryan Barton, Jui-Ching Lin, Michael Behr, and Zenon Lysenko WEB SEMINARS The apparent reaction kinetics between SO3 and polyethylene are investigated in various halogenated solvents using in situ Raman spectroscopy with an immersion Raman probe, Single Particle Mode or Hyphenated demonstrating the power of in situ Raman spectroscopy to monitor hazardous reactions. ICP-MS? A Discussion of Nanoparticle Analysis in Complex Matrices Dr. Susana Cuello Nuñez, LGC Limited, and Steve Wilbur, Agilent Technologies THE APPLICATION NOTEBOOK ...... 51 A-TEEM™ Molecular Fingerprinting: A New and Exciting Spectroscopy Technique Dr. Adam Gilmore, Horiba Scientific

spectroscopyonline.com/SpecWebSeminars DEPARTMENTS News Spectrum...... 12 Ad Index ...... 50 Like Spectroscopy on Facebook: Products & Resources ...... 48 Call for Application Notes ...... 63 www.facebook.com/SpectroscopyMagazine Follow Spectroscopy on Twitter: Spectroscopy (ISSN 0887-6703 [print], ISSN 1939-1900 [digital]) is published monthly by UBM LLC 131 West First Street, Duluth, https://twitter.com/spectroscopyMag MN 55802-2065. Spectroscopy is distributed free of charge to users and specifiers of spectroscopic equipment in the United States. Spectroscopy is available on a paid subscription basis to nonqualified readers at the rate of: U.S. and possessions: 1 year (12 issues), Join the Spectroscopy Group on LinkedIn $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), http://linkd.in/SpecGroup $250. Periodicals postage paid at Duluth, MN 55806 and at additional mailing offices. POSTMASTER: Send address changes to Spec- troscopy, P.O. Box 6196, Duluth, MN 55806-6196. PUBLICATIONS MAIL AGREEMENT NO. 40612608, Return 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. Inorganic Ventures gives you a little more control.

The key to accurate testing is control. It’s also the key to managing your testing program. Inorganic Ventures off ers more options in Certiﬁ ed Reference Materials (CRMs) to help you get the right tools to control testing accuracy and your budget.

Introducing new 30mL bottles for even better control. Now available for 1,000 and 10,000 ppm stock products.*

Beneﬁ ts: • No hazardous shipping fees • Less waste • Storage and contamination concerns minimized • Packaged with TCT

*Some exclusions may apply.

inorganicventures.com | 1.800.669.6799

300 Technology Drive | Christiansburg, VA 24073 USA International Distribution Center | Santander, Spain

NEW CUSTOMERS CAN SAVE 20%! Use promo code IVGP2020 on your ﬁ rst order.* *Exclusions may apply. Cannot be used with any other discount. 10 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com Editorial Advisory Board

Fran Adar Horiba Scientific Howard Mark Mark Electronics

Matthew J. Baker University of Strathclyde R.D. McDowall McDowall Consulting

Ramon M. Barnes University of Massachusetts Gary McGeorge Bristol-Myers Squibb

Matthieu Baudelet University of Central Florida Linda Baine McGown Rensselaer Polytechnic Institute

Rohit Bhargava University of Illinois at Urbana-Champaign Francis M. Mirabella Jr. Mirabella Practical Consulting Solutions, Inc. Illuminate Paul N. Bourassa Blue Moon Inc. Ellen V. Miseo Michael L. Myrick University of South Carolina Michael S. Bradley Thermo Fisher Scientific John W. Olesik The Ohio State University Deborah Bradshaw Consultant Steven Ray State University of New York at Buffalo Lora L. Brehm The Dow Chemical Company Jim Rydzak Specere Consulting George Chan Lawrence Berkeley National Laboratory Jerome Workman Jr. Unity Scientific David Lankin University of Illinois at Chicago, College of Pharmacy Lu Yang National Research Council Canada

Barbara S. Larsen DuPont Central Research and Development Spectroscopy’s Editorial Advisory Board is a group of distinguished individuals Bernhard Lendl Vienna University of Technology (TU Wien) assembled to help the publication fulfill its editorial mission to promote the effective use of spectroscopic technology as a practical research and measurement tool. With recognized expertise in a wide range of technique and application areas, board Ian R. Lewis Kaiser Optical Systems members perform a range of functions, such as reviewing manuscripts, suggesting authors and topics for coverage, and providing the editor with general direction and Rachael R. Ogorzalek Loo University of California Los Angeles, David feedback. We are indebted to these scientists for their contributions to the publication and to the spectroscopy community as a whole. Geffen School of Medicine

handle very large and/or heavy samples. Accepting samples up to 400 mm diameter, 50 mm thick and 30 kg mass, this system is ideal for analyzing

elemental analysis of large samples. With a variable measurement spot (30 mm to 0.5 mm diameter, with 5-step automatic selection) and mapping capability with multi-point measurements to check for sample uniformity,

control processes. t Analyze Be – U t Elemental range: ppm to % t Thickness range: sub Å to mm t Measurement spot t 30 mm to 0.5 mm diameter t 5-step automatic selection t Mapping capability t Allows multipoint measurements t Sample view camera (option)

Rigaku Corporation and its Global Subsidiaries website: www.Rigaku.com | email: [email protected] Copyright © 2017 PerkinElmer, Inc. 400372_01 All rights reserved. PerkinElmer® is a registered trademark of PerkinElmer, Inc. All other trademarks are the property of their respective owners. EXPAND YOUR RANGE EXTEND YOUR RESOURCES EXTEND YOUR Avio 500ICP-OES For moreinformation,visitperkinelmer.com/avio500 It’s everythingyouwantinanICP-OESsystem. High throughput.Lowcostofownership.Superiorperformance. comes togethertoexpandtherangeofwhatyoucanaccomplish. and highthroughputenabledbyDualViewtechnology,itall consumption ofanyICP,simultaneousbackgroundcorrection, dealing withthemostdifficultsamples.Andlowest argon superior resolution,yourlabcanaccomplishmore,evenwhen investment yourworkdemands.Withhighsensitivityand need withthehigh-qualityperformanceandfasterreturnon The Avio®500ICP-OEScombinestheproductivityyou with lowcostofownership The NewAvio500ICP-OES-Highthroughput 12 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com News Spectrum New York Section of the Society to Professor Richard Van Duyne of Northwestern for Applied Spectroscopy Announces University Speakers for Fall 2017 Meetings Speakers: Professor Richard Van Duyne, “Nanoscale The New York Section of the Society for Applied Spectroscopy Chemical Imaging with Tip-Enhanced Raman (NYSAS) has announced its initial lineup of speakers for its Spectroscopy” fall 2017 meetings. Meetings are held monthly, and are open Amanda J. Haes of the University of Iowa, to all; membership in SAS or the New York chapter is not “Translating SERS into a Robust Detection required. Attendance costs $15–20 and $5 for students to Platform for Uranium in Complex Matrices” cover the cost of a light dinner, and participants are asked Christy L. Haynes of the University of Minnesota, to register in advance to assist with food planning. Meetings “Polymer-Enabled Analytical SERS Sensing” usually start with a networking session, followed by dinner Nicholas Winograd of Pennsylvania State and the presentation. In November, in addition to the regular University, “Imaging Mass Spectrometry on the monthly meeting, the NYSAS will present an award at the Nanoscale with Cluster Ion Beams” Eastern Analytical Symposium. Additional details can be Location: Eastern Analytical Symposium, Crowne Plaza found on the chapter’s website, www.nysas.org. Princeton, Plainsboro, New Jersey

September meeting: December meeting:

Date: Wednesday, September 27, 2017 Date: Friday, December 1, 2017 Time: 5:30–8:30 p.m. Time: 5:30–8:30 p.m.

Speaker: Professor Gene Hall of Rutgers University, “The Speaker: Eric Breitung, Senior Research Scientist at the Raman and IR Spectra of Companion Pets’ Metropolitan Museum of Art, “Materials Testing Nutritional Supplements” at the Met: What Display, Shipping, and Storage Location: Rutgers University Materials Should or Shouldn’t Be Near the Art” Tentative location: Chemistry Building, room 260, Location: Metropolitan Museum of Art, 1000 Fifth Avenue, Busch Campus New York, New York ◾

October meeting: IR QUIZ TIME Date: Wednesday, October 25, 2017 Time: 5:30–8:30 p.m.

Using what you have learned from the July installment Speaker: Emil Ciurczak, consultant, topic TBA of “IR Spectral Interpretation Workshop and previous col- Location: Fairleigh Dickinson University umns, do your best to assign the peaks in this IR spectrum of a gas, determine the functional groups present, November regular meeting: and determine the chemical structure of the molecule that gave rise to this spectrum. Ignore the peaks with an X through them. Date: Thursday, November 9, 2017 Time: 5:30–8:30 p.m. To see the answer, please turn to page 34. Speaker: David Hopkins, NIR consultant, “Why Use the 632

Norris Regression - the Derivative Quotient Math 2971 1.5 in Regression?” 2947

Location: Horiba Scientific, 3880 Park Ave, Edison, 1.0 1778

New Jersey 1249 3328 2886 1140 Absorbance 1460 1376 1166 0.5 1706 3582 November special event: 2103 What: NYSAS at the Eastern Analytical Symposium 0.0

4000 3500 3000 2500 2000 1500 1000 500 Wavenumber (cm-1) Date: Monday, November 13, 2017 Time: 9:00–11:20 a.m. www.spectroscopyonline.com/ir-spectral- Event: Award session at the Eastern Analytical interpretation-workshop-quiz-12 Symposium, presenting the Gold Medal Award SUPERCHARGE Your Spectrometer...

Monolayers and More

The PIKE VeeMAXTM III is a variable-angle accessory that can be configured for specular reflection or ATR experiments. Diamond-turned mirrors deliver highest energy throughput. Heating, spectroelectro- chemical and flow cell options are available. When equipped with a polarizer, it provides sensitivity necessary for the measurement of thin films and monolayers. An automated version allows for complex measurements to be performed fast and reliably. A must- have in any research laboratory.

...with PIKE Accessories

Turn up the power to analyze even your toughest samples with a PIKE accessory in your FTIR sample compartment. Gain greater energy throughput, achieve higher spectral quality and attain faster results. Our accessories fit most spectrometers and many have heating and See the Invisible automation options. Talk to us today about your applications.

PIKE Technologies’ line of long-path gas cells offers quality construction, configur- ability, and ease of use. Pathlengths range from 1 to 20 meters. Heating up to 200 oC is optional. All gas cells feature diamond- turned, gold-coated mirrors for highest performance, precision and chemical resistance. Enclosed transfer optics may be purged to eliminate atmospheric interferences in the spectrum. www.piketech.com . [email protected] . (608) 274-2721 14 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com

Molecular Spectroscopy Workbench The Effect of Microscope Objectives on the Raman Spectra of Crystals

The Raman spectra of a particular face of a single crystal can be significantly different if acquired with different microscope objectives. The purpose of this installment of “Molecular Spectroscopy Workbench” is to inform and educate users of micro-Raman instrumentation of the effect of the microscope objective on the Raman spectra of crystals. Furthermore, we explain the underlying physics of changes in relative intensity and even peak position of certain Raman bands depending on the microscope objective used to acquire the spectrum. Changes in peak position are attributed to phonon directional dispersion sampled through wide-angle microscope objectives with different numerical apertures.

David Tuschel

aser illumination and light collection of Raman spec- illumination and collect the Raman light. This phenomenon trometers can involve mirrors, simple lenses, fiber-optic is most evident in the Raman spectra acquired with different L probes, and microscope objectives. The foremost consid- microscope objectives. The purpose of this installment of “Mo- eration pertaining to these optical elements is how effectively lecular Spectroscopy Workbench” is to inform and educate they deliver the laser illumination and collect the Raman users of micro-Raman instrumentation about the effect of the scattered light, and how strong the Raman signal is for a given microscope objective on the Raman spectra of crystals. laser power density and spectral acquisition time. Regard- One can find detailed theoretical and experimental treat- ing the light-collection optics in particular, the important ments on the effect of microscope objectives and wide-field characteristics include the solid angle of collection and optical collection optics that were published in the early days of micro- throughput. A spectroscopist’s expectation is that the overall Raman spectroscopy (1–3). An excellent source of information signal strength may vary depending on the collection optic on this topic is the chapter written by George Turrell titled used but that the Raman spectrum—that is, the relative inten- “Raman Sampling” in the book Practical Raman Spectroscopy sities of the bands—will not. That expectation is certainly rea- published in 1988 (4). Turrell’s very instructive chapter deals sonable for Raman spectra acquired from liquids, gases, amor- with laser excitation focusing and Raman light collection using phous materials, glasses, and polycrystalline solids. However, wide-angle microscope objectives. Of course, one generally that expectation is not always valid when performing Raman thinks of the microscope objective’s magnification as one of its spectroscopy of single crystals or grains. One can observe most important and relevant characteristics. One should also variations of the peak positions and the relative intensities of be mindful of the objective’s numerical aperture (NA) and its Raman bands depending on the optic used to deliver the laser direct relationship to Raman light collection efficiency. The www.spectroscopyonline.com September 2017 Spectroscopy 32(9) 15 numerical aperture is defined by the following expression:

A NA = n sinθ [1] 1

E where n is the refractive index of the medium through which the light is E passing and θ is the angle between the ray along the optical axis at the A E 1 center of the lens and the ray from the E perimeter of the lens to the sample. E Therefore, a microscope objective with E E,A Intensity (arbitrary units) 1 a high NA will have a greater solid angle of light collection. However, remember that the NA does not just affect the light collection efficiency. 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 950 Raman shift (cm–1) One can envision a variation of the solid angle of the laser beam incident on the sample depending upon the Figure 1: Raman spectra of the X-face of a single crystal of LiNbO3 with the incident laser light × × × NA of the objective. At the very high- polarized along the Y-axis. Spectra were acquired using 10 (brown), 50 LWD (blue), 50 × × est magnification and NA, the laser (green), 100 LWD (navy), and 100 (red) microscope objectives. The symbols A1 and E are the beam comes to a focus at a very short symmetry species to which the bands have been assigned. working distance over a wide angle. This means that the laser beam pass- that from a 25-mm-diameter lens with angles of incidence, whereas the long- ing through a 100× objective with a a 40-mm focal length. The microscope focal-length lens more closely approxi- 0.9 NA focused on the sample will not objective profile can be thought of as a mates a collimated narrow beam of have the same illumination profile as wide-angle cone consisting of varying parallel rays orthogonal to the sample

Redeﬁ ning formulation analysis

Introducing the RA802 Pharmaceutical Analyser Formulate tablets more efficiently with the RA802 Pharmaceutical Analyser; a compact benchtop Raman imaging system, designed exclusively for the pharmaceutical industry.

• Efficiently analyze samples with rough, uneven or • Track the surface live while acquiring surface or curved surfaces. subsurface Raman data. • Reveal detailed chemical and physical information. • Analyze multiple tablets without the need for user • No sample preparation needed; look at solid or liquid intervention. samples. For more information visit www.renishaw.com/RA802

Renishaw IncWest Dundee, ),www.renishaw.com 16 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com

by crystallographers or X, Y, and Z by optical physicists. The crystallographic face is defined by the crystallographic A 1 axis normal to that face; that is, the A E 1 face and axis are orthogonal to each A 1 other. For example, one is looking at the Z-face of a crystal if the line of eye- sight is parallel to the Z-axis. In a cubic crystal, all three axes have the same E refractive index. A uniaxial crystal A E 1 E has two axes with the same refractive 195 index, called the ordinary axes, and Intensity (arbitrary units) E E the third axis, called the extraordinary axis, has a different refractive index from the ordinary axes. All three of 150 200 250 300 350 400 450 500 550 600 650 700 the crystallographic axes of a biaxial Raman shift (cm–1) crystal have different refractive indices. Uniaxial and biaxial single crystals

Figure 2: Raman spectra of the X-face of a single crystal of LiNbO3 with the incident laser light are ideal for demonstrating the con- polarized along the Z-axis. Spectra were acquired using 10× (brown), 50× LWD (blue), 50× sequences of using wide angle micro- × × (green), 100 LWD (navy), and 100 (red) microscope objectives. The symbols A1 and E are the scope objectives on crystalline materi- symmetry species to which the bands have been assigned. als. LiNbO3 is a uniaxial crystal whose ordinary and extraordinary refractive

indices at 530 nm are no = 2.3247 and ne = 2.2355, respectively (5). A correlation exists between the directionally dependent dielectric properties and 155 the corresponding Raman tensors of single crystals. Moreover, the Raman 872 spectrum acquired from a single crystal depends on the crystal class to 582 which the crystal belongs, the crystal’s 432 237 orientation relative to the direction and 264 332 polarization of the incident light, and the collection angle and polarization of the Raman scattered light. Intensity (arbitrary units) LiNbO3 belongs to the C3v crystal class for which the Raman tensors are 50100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000 a c d c –1 0 0 0 0 0 Raman shift (cm ) A (Z) =(0 a 0 ;( E(X) =(c 0 0 ;( E(Y) =(0 –c d ( 1 0 0 b d 0 0 0 d 0 [2]

Figure 3: Raman spectra of the Z-face of a single crystal of LiNbO3 with the incident laser light where the letter in parentheses next polarized along the X-axis. Spectra were acquired using 10× (brown), 50× LWD (blue), 50× to the symmetry species indicates the (green), 100× LWD (navy), and 100× (red) microscope objectives. crystallographic direction of the lattice vibrational mode polarization. surface. Furthermore, if we define the these effects through spectra of single- The irreducible representations of the direction of illumination as the Z-axis, crystal LiNbO3 and KTiOPO4 acquired LiNbO3 vibrational modes are one can expect that the laser light po- from objectives with different magnifi- Γ = A IR,R +9E IR,R + A larization initially in the XY-plane will cations and numerical apertures. vib 4 1( ) ( ) 5 2 [3] now have some component along the

Z-axis at the wider angles of incidence Raman Spectroscopy The 4A1 and 9E modes are infrared of a microscope objective. The effect of Single-Crystal LiNbO3 and Raman active, whereas the 5A2 of the microscope objective’s NA on In performing Raman spectroscopy modes are silent. Raman spectra illumination and light collection has of single crystals, it is important to acquired from the X-face of a single significant consequences for the prac- understand the nomenclature of crys- crystal of uniaxial LiNbO3 are shown tice of micro-Raman spectroscopy as tallography. The three crystallographic in Figure 1. Spectroscopic selection applied to crystals. We demonstrate axes are generally labeled a, b, and c rules predict that these Raman bands www.spectroscopyonline.com September 2017 Spectroscopy 32(9) 17

Table I: Microscope objectives and their numerical apertures

Numerical Working 155 Magnification Aperture Distance (mm)

10× 0.25 10.6

50× LWD 0.50 10.6 153 582 237 432 50× 0.75 0.38 332 264 578 100× LWD 0.75 4.7 Intensity (arbitrary units) 100× 0.90 0.21

150 200 250 300 350 400 450 500 550 600 650 will belong to the symmetry species Raman shift (cm–1) A1(TO) and E(TO), where the symbols TO and LO stand for the transverse and longitudinal optical modes, re- Figure 4: Raman spectra of the Z-face of a single crystal of LiNbO3 with the incident laser light spectively. The spectra were acquired polarized along the X-axis. Spectra were acquired using 10× (brown), 50× LWD (blue), 50× using 10× (NA 0.25), 50× LWD (NA (green), 100× LWD (navy), and 100× (red) microscope objectives. 0.50), 50× (NA 0.75), 100× LWD (NA 0.75), and 100× (NA 0.90) microscope used in the collection of these or any tive used for laser delivery and light objectives. The laser beam was inci- other spectra presented in this work. collection. This is the type of response, dent upon the X-face and polarized The five spectra are plotted normal- independent of microscope objective, parallel to the Y-axis, one of the or- ized and appear to be nearly identical, which many users of Raman instru- dinary axes. No Raman analyzer was independent of the microscope objec- mentation may incorrectly expect of 18 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com all crystallographic Raman measurements. However, rotate 195 cm-1 band has been assigned to the E(LO) symmetry the incident polarization by 90° such that the incident laser species and is not expected in this configuration (6–9). Why polarization is now parallel to the Z-axis, the extraordinary would a Raman band forbidden in this experimental con- axis, and one observes entirely different Raman spectra. figuration appear in the spectra and why would its relative

The b Raman tensor element of the A1 mode now domi- intensity vary depending on the microscope objective used? nates the spectra shown in Figure 2. The answer to this question can be found in Table I, which The Z-axis polarized spectra are significantly differ- shows the numerical apertures of the objectives used. ent from those obtained with the incident light polarized The numerical aperture is a measure of the solid angle parallel to the Y-axis. Most of the bands appearing in the of light collection of the microscope objective. A large NA Y-polarized spectra of Figure 1 are attributed to the E sym- value corresponds to a greater solid angle. In this case, the metry species, whereas the most pronounced bands in the NA has implications for the laser beam profile incident on

Z-polarized spectra of Figure 2 belong to the A1 symme- the sample. Ideally, one would like to fill the back aperture try species. These differences arise because of the Raman with the laser beam such that it comes to a diffraction polarization selection rules and the interplay of incident limited spot in the field of view. Not all the rays envisioned and scattered light direction, polarization, and the orienta- travelling down the barrel of the microscope objective will tion of the crystal. Selection rules predict that these X-face reach the laser spot at the same angle of incidence. The Raman bands excited with laser light polarized along the objective with the greatest NA will cause the rays passing Z-axis in the backscattering configuration will belong to the through the lens near the perimeter to refract at the larg- symmetry species A1(TO) and E(TO). Most of the Raman est angle. Therefore, one can envision a distribution of bands appear to be of near identical relative strength, in- angles of incidence on the sample that is dependent on the dependent of the microscope objective used to acquire the numerical aperture. The field of view may be normal to spectrum. However, notice that the intensity of a single band the X-face, but that does not mean that all of the rays from at 195 cm-1 depends on the microscope objective used to the objective will be parallel to the X-axis. Some will be deliver the laser beam and collect the Raman scattered light. incident on the sample at wide angles commensurate with The 195 cm-1 band is absent in the spectrum acquired using the numerical aperture. Thus, we see with increasing NA the 10× objective and increases progressively with the 50× the emergence of the 195 cm-1 band that has been assigned LWD, 50×, 100× LWD, and 100× objectives. Moreover, the to the E(LO) symmetry species and should not be observed under this 180° backscattering experimental configuration. These types of differences in Raman spectra that are dependent on the objective used can easily go unnoticed by September 19-22, 2017 one not skilled in the practice of Raman spectroscopy. The purpose of this installment is to make users aware of this Boston Park Plaza Hotel, Boston, MA phenomenon and to explain its origin. Rotation of the crystal by 90° about the Y-axis allows th the laser beam to now be parallel to the extraordinary axis Symposium on the (Z-axis) and incident upon the Z-face. The perpendicular 14 X- and Y-axes are the ordinary axes with identical refrac- Practical Applications of tive indices. Spectra were acquired using 10× (NA 0.25), Mass Spectrometry 50× LWD (NA 0.50), 50× (NA 0.75), 100× LWD (NA 0.75), in the Biotechnology Industry and 100× (NA 0.90) microscope objectives and are shown in Figure 3. Selection rules predict that these Z-face Raman bands excited with laser light polarized along the X-axis in the backscattering configuration will belong to the sym-

metry species A1(LO) and E(TO). The spectra have been -1 normalized to the A1(LO) band at 872 cm , thereby reveal- ing the changes in relative intensities of all the other bands as a function of the microscope objective used. The contrast with the spectra in Figure 1 is quite stark where there were no differences among the spectra acquired with different objectives. Here, we see that all of the Raman bands, to a greater or lesser degree, vary with intensity depending on the objective used. All of the Raman bands in Figure 3 follow the same pattern. The strength of a given band increases Scan here or visit relative to that of the 872 cm-1 band progressively with the www.casss.org for 10×, 50× LWD, 50×, 100× LWD, and 100× objectives. program updates. SHARING SCIENCE SOLUTIONS It is important to note that all the bands do not increase in signal strength by the same proportion relative to the www.spectroscopyonline.com September 2017 Spectroscopy 32(9) 19

872 cm-1 band. For example, the bands at 155 and 582 cm-1 are both weaker than the 872 cm-1 band in the spectrum acquired with the 10× objective. 765 However, in the spectrum acquired using the 100× objective the signal 374 strength of the 155 cm-1 band now significantly exceeds that of the 872 cm-1 216 702 band whereas the signal strength of 327 the 582 cm-1 band is still less than that 268 of the 872 cm-1 band. There are other 637 ways of viewing and analyzing this same phenomenon; we could have Intensity (arbitrary units) plotted the spectra normalized to any of the other bands to see the relative 100 200 300 400 500 600 700 800 900 1000 1100 signal strength variations dependent Raman shift (cm–1) on the microscope objective. The variations in relative signal strengths are because of a change in the distribu- Figure 5: Raman spectra of the Z-face of a single crystal of KTiOPO4 with the incident laser light tion of angles of incidence as we use polarized along the Y-axis. Spectra were acquired using 10× (brown), 50× LWD (blue), 50× objectives with increasing numerical (green), 100× LWD (navy), and 100× (red) microscope objectives. aperture. The 10× objective with NA of 0.25 comes the closest of those used at oblique angles rather than just those respect to the light-collection axis ap- here to deliver a laser beam parallel to closely parallel to the crystallographic pear unexpectedly in the backscatter- the crystallographic axis. As the NA of axis. Consequently, Raman bands that ing configuration. the objective is increased, one can envi- would be expected were the laser beam Not only do we observe a change sion more rays incident on the sample to be incident at an oblique angle with in the relative intensities, some bands

2018 SPRING MEETING & EXHIBIT April 2–6, 2018 | Phoenix, Arizona CALL FOR PAPERS Abstract Submission Opens: September 29, 2017 | Abstract Submission Deadline: October 31, 2017

Ź Characterization, Modeling and Theory Meeting Chairs

Ź Edward Botchwey Electronic and Photonic Materials Georgia Institute of Technology/Emory University

Ź Energy Materials and Technologies Catherine Dubourdieu Helmholtz Zentrum Berlin Ź Manufacturing Quanxi Jia University at Buffalo, The State University of New York/ Ź Nanomaterials Los Alamos National Laboratory Shane Kennett Ź Soft Materials and Biomaterials Exponent Failure Analysis Associates Cheolmin Park Yonsei University

www.mrs.org/spring2018 ® 20 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com

phenomenon is known as phonon directional dispersion. A detailed theoretical treatment can be found in the 375 book by Sherwood (10). A plot of experimentally obtained 216 phonon directional dispersion of the Raman bands of LiNbO3 is shown 327 in the publication by Maimounatou 268 400 and colleagues (9). We see the greatest 293 342 dispersion for the E(TO) bands that 328 appear at 153 and 578 cm-1 when the exciting beam is parallel to the Z-axis. Intensity (arbitrary units) The 153 and 578 cm-1 bands convert to

E(LO) and A1(TO) modes, respectively as the crystal is rotated 90° such that 220 240 260 280 300 320 340 360 380 400 420 the incident beam is finally parallel Raman shift (cm–1) to the X-axis. The phonon directional dispersion shown in Figure 4 of the Figure 6: Raman spectra of the Z-face of a single crystal of KTiOPO4 with the incident laser light Maimounatou publication is consis- polarized along the Y-axis. Spectra were acquired using 10× (brown), 50× LWD (blue), 50× tent with the change in Raman peak (green), 100× LWD (navy), and 100× (red) microscope objectives. position as microscope objectives with higher numerical aperture and greater manifest a shift in peak position as an ing radiation is propagating in a plane off-axis illumination is used. objective with higher numerical aper- containing the extraordinary axis, ture is used. The spectra of Figure 3 are either the XZ- or YZ-planes. Now the Raman Spectroscopy shown in an expanded scale in Figure short-range atomic forces that lead of Single-Crystal KTiOPO4 4 to more clearly see the relative inten- to the extraordinary refractive index Unlike the uniaxial crystal LiNbO3, sity changes and, in some cases, peak and different vibrational force con- KTiOPO4 is a biaxial crystal whose shifts. Specifically, note that the 153 stants of the unique axis play a role. three refractive indices at 530 nm -1 and 578 cm bands in the spectrum If these forces are greater than the are nx = 1.7787, ny = 1.7924, and nz = × acquired using the 10 objective shift long-range forces that lead to the split- 1.8873, respectively (11). Like LiNbO3, -1 to 155 and 582 cm , respectively in ting of longitudinal and transverse KTiOPO4 is a ferroelectric crystal at the spectrum acquired with the 100× phonons, then a phonon of mixed room temperature with high nonlinear objective. The bands at 237, 264, 332, longitudinal and transverse charac- susceptibilities and electrooptic coef- and 432 cm-1 are fixed, showing no ter will be launched by the exciting ficients. Both crystals are commonly shift when using different microscope radiation. Thus, instead of a splitting used for frequency doubling. The objectives. What accounts for the shift of the Raman bands because of the 532-nm laser is produced by guiding in peak position of some bands but not LO and TO phonons, one will ob- the 1064-nm beam from a Nd:YAG others? The answer is phonon direc- serve a single Raman band of mixed laser through a crystal such as LiNbO3 tional dispersion. LO and TO character dependent on or KTiOPO4 oriented appropriately Consider the case of laser light the direction of propagation in the to achieve phase matching. Here we propagating along a general direc- crystal. Imagine a laser beam initially continue our examination of the ef- tion in a crystal and not parallel to a directed along the Z-axis of a single fect of microscope objective numeri- crystallographic axis. Such light can crystal of LiNbO3. If Raman spectra cal aperture on the Raman spectra of excite lattice vibrational waves that were acquired as the crystal is rotated KTiOPO4. resolve into longitudinal and trans- about the Y-axis in small rotational KTiOPO4 belongs to the C2v crystal verse waves. The frequency of the increments until the laser beam was class for which the Raman tensors are longitudinal phonon is higher than finally directed along the X-axis, one a 0 0 0 d 0 0 0 e 0 0 0 that of the transverse phonon. If our would observe that some peaks shift A = (0 b 0 ;( A =(d 0 0 ;( B =(0 0 0 ( ; B =(0 0 f ( [4] 1 0 0 c 2 0 0 0 1 e 0 0 2 0 f 0 phonon is propagating in a general to higher wavenumber as the laser il- direction, we may expect to see a split- lumination rotates from the Z-axis to The irreducible representations of the ting and two Raman bands associated the X-axis. Those lattice vibrational vibrational modes are with the LO and TO phonons. That modes most affected by short range Γ = A IR,R + A R + B IR,R + B IR,R behavior can be expected if the plane atomic forces will be of mixed LO and vib 47 1( ) 48 2( ) 47 1 ( ) 47 2 ( ) [5] contains only the two ordinary axes TO character and their energies will of the uniaxial crystal. However, the depend on the direction of propaga- The 47A1, 48A2, 47B1, and 47B2 modes situation is quite different if the excit- tion through the crystal lattice. This are Raman active, leading to a Raman www.spectroscopyonline.com September 2017 Spectroscopy 32(9) 21

765 768

701

517 547 637 834 Intensity (arbitrary units)

500 520 540 560 580 600 620 640 660 680 700 720 740 760 780 800 820 840 Raman shift (cm–1)

Figure 7: Raman spectra of the Z-face of a single crystal of KTiOPO4 with the incident laser light polarized along the Y-axis. Spectra were acquired using 10× (brown), 50× LWD (blue), 50× (green), 100× LWD (navy), and 100× (red) microscope objectives.

763 768

517 547

701 592 633 Intensity (arbitrary units)

500 520 540 560 580 600 620 640 660 680 700 720 740 760 780 800 820 840 Raman shift (cm–1)

Figure 8: Raman spectra of the Z-face of a single crystal of KTiOPO4 with the incident laser light polarized along the X-axis. Spectra were acquired using 10× (brown), 50× LWD (blue), 50× (green), 100× LWD (navy), and 100× (red) microscope objectives. spectrum much more complicated are plotted normalized to the 765 cm-1 than that of LiNbO3. Nevertheless, we band. It is clear that many of the see the same phenomena and patterns bands manifest the same intensity as a function of microscope objective relative to that of the 765 cm-1 band as we did with our spectra of LiNbO3. independent of the microscope ob- Raman spectra obtained of the Z- jective used. However, note how the face with the laser beam guided along strengths of the bands at 268, 327, the Z-axis with the light polarized 637, and 702 cm-1 vary depending on parallel to the Y-axis are shown in the objective used. In particular, the Figure 5. The spectra were acquired 702 cm-1 band is absent in the spec- using 10× (NA 0.25), 50× LWD (NA trum acquired with the 10× objective. 0.50), 50× (NA 0.75), 100× LWD It appears first in the spectrum ac- (NA 0.75), and 100× (NA 0.90) mi- quired using the 50× LWD objective croscope objectives. All of the spectra and grows progressively more intense 22 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com

as the numerical aperture increases through the 50×, 100× LWD, and 100× objectives. The behavior is much like that NE -1 W of the LiNbO3 band at 195 cm in Figure 2. 863 -1 R Expansion of the scale in the spectral region below 425 cm EFERENCES shown in Figure 6 allows one to more clearly see the changes in relative strength of the bands at 268 and 327 cm-1 when compared to those invariant at 342, 375, and 400 cm-1. The latter are of identical strength independent of the microscope ISO/IEC 17025 & ISO Guide 34 Accredited &HUWLĆHG objective used to acquire the spectrum. We observe the same Simple UV/VIS/NIR Validation progression in these spectra of KTiOPO as we did for LiNbO , Permanently sealed cells for repeat use 5HIHUHQFH 4 3 Absorbance, Stray Light, depending on the microscope objective used. The changes in Wavelength, Resolution Materials relative band strength can be accounted for and explained with NIST Traceable respect to the increasing numerical aperture. As the numerical aperture increases, so does the distribution of angles of incidence of the exciting laser beam. We observed the phenomenon of phonon directional

dispersion in our spectra of the Z-face of LiNbO3 and so we might expect to observe the same effect in our spectra

of KTiOPO4, especially with the exciting beam nominally parallel to the Z-axis. Recall that nz = 1.8873 and is substan- tially greater than nx = 1.7787 and ny = 1.7924. Therefore, the short range and anisotropic atomic forces leading to this birefringence should lead us to expect phonon directional 6WDUQD&HOOV,QF dispersion in KTiOPO4. Indeed, one can see in Figure 6 that -1 × PO Box 1919 Atascadero, CA 93423 the band at 328 cm obtained with the 50 LWD objective -1 × Phone: (800) 228-4482 USA or (805) 466-8855 outside USA shifts to 327 cm in the spectrum acquired using the 100 VDOHV#VWDUQDFHOOVFRPZZZVWDUQDFHOOVFRP objective. An expanded view of these same spectra in the region from 500 to 850 cm-1 is shown in Figure 7. The band at -1 768 cm has been assigned to an A1(LO) mode (12). Here we see that the 517, 547, 701, and 834 cm-1 bands remain fixed, independent of the objective used to acquire the spectra. However, the intense band at 768 cm-1 acquired using the 10× objective shifts progressively to a lower wavenumber with increasing numerical aperture until reaching a value of 765 cm-1 in the spectrum acquired with the 100× objective. The progression of phonon directional dispersion to a lower wavenumber with increasing numerical aperture observed 2017 EASTERN ANALYTICAL SYMPOSIUM & EXPOSITION for KTiOPO4 is the opposite of that observed for LiNbO3 where the bands shifted to a higher wavenumber. We can CROWNE PLAZA attribute the different responses to the different anisotropic NEW LOCATION PRINCETON – CONFERENCE CENTER short-range atomic forces of KTiOPO4 and LiNbO3. The NOVEMBER 13-15, 2017 PLAINSBORO, NJ Z-axis or extraordinary refractive index (ne = 2.2355) of LiNbO3 is less than that of the ordinary refractive index Î Beautiful, modern facility with Î Three-day Technical Program (n = 2.3247). Conversely, the Z-axis refractive index (n = Exposition, Technical Program, covering the broad interests of our o z and Short Courses all under attendees from academia and industry 1.8873) of KTiOPO is greater than either n = 1.7787 or n = one roof and only 30 minutes from while celebrating the analytical 4 x y Somerset, NJ excellence of our EAS Award winners 1.7924. Consequently, the phonon energies shift in a direc- Î New Exposition arrangement Î Short Courses emphasizing a wide for attendees and exhibitors to range of topics for problem solving in tion commensurate with the positive or negative values of better interact and discuss analytical analytical laboratories with interactive instrumentation, products, services, discussions and case studies the birefringence. and supplies Î Professional development The polarization of the incident beam also plays an im- Î Deadline for POSTER abstract Workshops & Employment submission is September 1, 2017 Bureau for career success and portant role. The spectra of the Z-face shown in Figures Seminars specifically for educators, high school & undergraduate 5–7 were all obtained with the incident laser light polarized students, and much more parallel to the crystallographic Y-axis. Spectra of the Z-face eas.org with incident laser light polarized parallel to the X-axis are shown in Figure 8. The spectra appear similar but are not identical to those shown in Figure 7. In particular, note

that the phonon directional dispersion of the A1(LO) band is greater as it progressively shifts from 768 cm-1 acquired www.spectroscopyonline.com September 2017 Spectroscopy 32(9) 23 using the 10× objective to lower (2) C. Bremard, J. Laureyns, J.-C. Merlin, (10) P.M.A. Sherwood, Vibrational Spec- wavenumber with increasing numeri- and G. Turrell, J. Raman Spec. 18, troscopy of Solids (Cambridge Univer- cal aperture until reaching a value of 305–313(1987). sity Press, London, 1972), pp. 83–115. 763 cm-1 in the spectrum acquired (3) C. Bremard, P. Dhamelincourt, J. Lau- (11) A.M. Prokhorov and Y.S. Kuz’minov, with the 100× objective. The phonon reyns, and G. Turrell, Appl. Spec. 39, Ferroelectric Crystals for Laser Radia- directional dispersion of the A1(LO) 1036–1039(1985). tion Control (Adam Hilger, Bristol, mode is even greater for light polar- (4) G. Turrell, in Practical Raman Spec- 1990), p. 307. ized parallel to the X-axis than it is troscopy, D.J. Gardiner and P.R. Graves, (12) M. Rüsing, C. Eigner, P. Mackwitz, G. for light polarized along the Y-axis. Eds. (Springer-Verlag, Berlin, 1989), Berth, C. Silberhorn, and A. Zrenner, J. Therefore, one needs to be cognizant pp. 13–54. Appl. Phys. 119, 044103 (2016). of the polarization of the incident (5) A.M. Prokhorov and Y.S. Kuz’minov, laser light along with the numerical Physics and Chemistry of Crystalline aperture of the microscope objective Lithium Niobate (Adam Hilger, Bristol, David Tuschel is a when acquiring Raman spectra with a 1990), p. 199. Raman applications man- microscope based instrument. (6) S. Margueron, A. Bartasyte, A.M. ager at Horiba Scientific, in Glazer, E. Simon, J. Hlinka, and I. Edison, New Jersey, where Conclusion Gregora, J. Appl. Phys. 111, 104105 he works with Fran Adar. The numerical aperture is a measure of (2012). David is sharing author- ship of this column with the solid angle of light collection of the (7) M.D. Fontana and P. Bourson, Appl. Fran. He can be reached microscope objective. A microscope Phys. Rev. 2, 040602 (2015). at: SpectroscopyEdit@ objective with a high NA will have a (8) S. Sanna, S. Neufeld, M. Rüsing, G. UBM.com greater solid angle of light collection. Berth, A. Zrenner, and W.G. Schmidt, The NA also has implications for the Phys. Rev. B 91, 224302 (2015). laser beam profile incident on the sam- (9) B. Maimounatou, B. Mohamadou, and For more information on ple. The objective with the greatest NA R. Erasmus, Phys. Status Solidi B 253, this topic, please visit: will cause the rays passing through the 573–582(2016). www.spectroscopyonline.com lens near the perimeter to refract at the largest angle. Therefore, one can envision a distribution of angles of incidence that is dependent on the numerical aperture. Consequently, Raman bands that would be expected were the laser beam to be incident at an oblique angle with respect to the light collection axis appear unexpectedly in the backscattering configuration. We observe a change in the relative intensities and some bands manifest a shift in peak position as an objective with higher numerical aperture is used. This phenomenon can be attributed to phonon directional dispersion. The field of view may be normal to a particular crystal face, but that does not mean that all of the rays from the objective will be parallel to that crystallographic axis. Therefore, one needs to be cognizant of the numerical aperture of the microscope objective when acquiring Raman spectra of single crystals or grains with a microscope based instrument.

References (1) G. Turrell, J. Raman Spec. 15, 103– 108(1984). 24 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com

Focus on Quality What’s New in the New USP <1058>?

United States Pharmacopeia general chapter <1058> “Analytical Instrument Qualification” has been updated and became effective August 1, 2017. So, what has changed in the new version?

R.D. McDowall

n the regulated world of good manufacturing practice Why Is Instrument Qualification Important? (GMP) we have regulations that define what should be The simplest answer to this question is that qualification is I done, but leave the interpretation to the individual organi- important so that you know that the instrument is functioning zation on how to do it. However, when we come to the regu- correctly and that you can trust the results it produces when it lated analytical or quality control (QC) laboratory we also have is used to analyze samples. the pharmacopoeias, such as the European Pharmacopoeia However, there is a more important reason in today’s world (EP), Japanese Pharmacopoeia (JP), and United States Pharma- of data integrity—integrated instrument qualification and copeia (USP), to provide further information to help interpret computer validation is an essential component of a data integ- the regulations. These tomes can have monographs for active rity model. The complete four-layer model can be viewed in pharmaceutical ingredients, finished products, and general my recent book, Validation of Chromatography Data Systems chapters that provide requirements for how to apply various (3), but the analytical portion of the model is shown in Figure analytical techniques, such as spectroscopy. 1 and is described in more detail in an earlier “Focus on Qual- ity” column (4). In the Beginning . . . The four layers of the model are Of the major pharmacopoeias, only USP has a general chapter • Foundation: management leadership, policies and proce- on analytical instrument qualification (AIQ) (1). This chapter dures, culture, and ethos came about with a 2003 conference organized by the American • Right instrument and system for the job: instrument qualifi- Association of Pharmaceutical Scientists (AAPS) on analytical cation and computer validation instrument validation. The first decision of the conference was • Right analytical method for the job: development and valida- that the name was wrong and it should be analytical instrument tion of analytical procedures qualification (AIQ). The conference resulted in a white paper • Right analysis for the right reportable result: analysis from (2) that after review and revision became USP general chapter sampling to reporting the result <1058> on AIQ effective in 2008. This chapter described a data The model works from the foundation up with each layer quality triangle, general principles of instrument qualification, providing input to the next. As shown in Figure 1, AIQ and and a general risk classification of analytical equipment, instru- computerized system validation (CSV) come after the foun- ments, and systems. General chapter <1058> did not specify any dation layer, illustrating that if the instrument is not quali- operating parameters or acceptance limits because those can be fied and any software is not validated, the two layers above found in the specific general chapters for analytical techniques. (method validation and sample analysis) will not be effective This column is written so that you will understand the and can compromise data integrity. Interestingly, AIQ and changes that come with the new version and the impact that CSV are missing from many of the data integrity guidance they will have on the way that you qualify and validate instru- documents but you can see the importance in the overall data ments and the associated software, respectively. integrity framework of Figure 1. www.spectroscopyonline.com September 2017 Spectroscopy 32(9) 25

Why Is There a New Revision of USP <1058>? Level 3: The 2008 version of USP <1058> had Right analysis for the right reportable result several issues that I presented during a Data acquired and transformed that are complete, consistent and accurate session co-organized by Paul Smith at an AAPS conference in 2010 and published Level 2: Right analytical procedure for the right job in this column (5). There were three main Validated–veriﬁed under actual conditions of use problems with the first version: • Problem 1: The true role of the supplier Level 1: was missing. The supplier is respon- Right instrument for the right job sible for the instrument specification, Qualiﬁcation or validation (or both) for intended purpose detailed design, and manufacture of the instrument but this responsibility is Foundation: Right culture and ethos for data integrity (DI) not mentioned in <1058>. The reason Management leadership, DI policies and procedures, staff DI training is that the section on design qualification (DQ) mentions that a user can use Figure 1: The four layers of the data integrity model for laboratories within a pharmaceutical the supplier’s specification.Micro However,wav ae Digestionquality system. | Adapted Clean with Chemistry permission from | reference Mercury 4. Analysis user needs to understand the conditions under which the specification was mea- is dumped on the supplier: “The manu- in the summer of 2013, which was circu- sured and how relevant it is to a labora- facturer should perform DQ, validate lated to a few individuals in industry and tory’s use of an instrument. this software, and provide users with a suppliers for review before submission to HELPIN G • Problem 2: Users are responsible for summary of validation. At the user site, the USP. CHE M ISTS DQ. USP <1058> places great emphasis holistic qualification, which involves the Proposed drafts of the new version on the fact that in the design qualifica- entire instrument and software system, is were published for public comment in tion stage is the responsibility of the more efficient than modular validation of Pharmacopeial Forum in 2015 (9) and supplier, “Design qualification (DQ) is the software alone.” 2016 (10) and comments were incor- most suitably performed by the instru- In the days of data integrity, this ap- porated in the updated versions. The ment developerAt or manufacturer.” Milestone This proach, Wis completelye H euntenable.lp C Thehemists. approved USP <1058> final version was Microwave Digestion is wrong. Only users can define their United States Food and Drug Adminis- published in January 2017 in the First instrument30 Y needsears. and must 5 do0 so toPatents. tration (FDA) 20,000guidance on software Gl valio- bSupplemental Users. to USP 40 (11). The chapter define their intended use of the instru- dation (7), quoted by <1058>, was written became effective on August 1, 2017. Clean Chemistry ment and comply with GMP regula- for medical device software, which is not There was an erratum published in Feb- tions (§211.63) (6). configured unlike much of the laboratory ruary, but the only change was reference Mercury Analysis • Problem 3: Poor software validation software used today. of the operational qualification (OQ) guidance. The software qualification testing to the intended use definition and validation description in USP Revision of USP <1058> (user requirements specification [URS]). <1058> is the poorest part of this gen- To try and rectify some of these issues, eral chapter as software is pervasive the revision process of USP <1058> What Has Changed in USP <1058>? throughout Group B instrumentsVisit and Ourstarted Updated in 2012 with the Website publication of First let us look at the overall scope of Group C systems. a “Stimulus to the Revision” process ar- changes between the old and new ver- Although the approach to handling ticle published in Pharmacopeial Forum sions of USP <1058> as shown in Table I. milestonesci.com embedded software in Group B instru- written Chris Burgess and myself (8). ments where the8 firmware66.995.5100 is implicitly This article • milestonesci.comproposed two items: Missing in Action or indirectly validated during the instru- • An integrated approach to analytical The following items were omitted from ment qualification is fine, there are omis- instrument qualification and com- the new version of <1058>: sions. Users need to be aware that both puterized system validation (AIQ- • Differences between qualification and calculations and user defined programs CSV), and validation. This was omitted because must be verified to comply with GMP • More granularity for Group B in- qualification and validation activities requirements in 211.68(b) (6). Note that struments and Group C computer- are integrated in the new version, so the qualification of firmware, which is a ized systems to ensure that calcula- why describe the differences? You need simple and practical approach, is now in- tions and user defined programs to control the instrument and any soft- consistent with Good Automated Manu- were captured in the first group and ware and if you can demonstrate this facturing Practice (GAMP) 5, which has an appropriate amount of validation through the 4Qs process described in dropped Category 2 (firmware). was performed by the users for the the new <1058>, why bother with what Software for Group C systems is the second group. the activity is called? weakest area in the whole chapter <1058>. We used the feedback from that arti- • Table I in the old version of <1058> The responsibility for software validation cle to draft a new version of USP <1058> describes the timing, applicability, and 24 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com

Focus on Quality What’s New in the New USP <1058>?INDUSTRIES MILESTONE

United States Pharmacopeia general chapter <1058> “Analytical Instrument Qualification” has been updated and became effective August 1, 2017. So, what has changed in the new version? HELPING ACADEMIA CANNABIS CLINICAL COSMETICS ENERGY ENVIRONMENTAL C HEMIST S TESTING R.D. McDowall

n the regulated world of good manufacturing practice Why Is Instrument Qualification Important? (GMP) we have regulations that define what should be The simplest answer to this question is that qualification is Discover how Milestone I done, but leave the interpretation to the individual organi- important so that you know that the instrument is functioning zation on how FOODto do it. & However, FEED whenMET weA comeLS to theNUT reguRAC- EUTICAcorrectlyL PH andARM thatA/USP you can trustPOLYME the resultsRS it producesSPECI AwhenLTY it lated analytical or quality control (QC) laboratory we also have is used CtoOMPLI analyzeAN samples.CE CHEMICALS can help your laboratory. the pharmacopoeias, such as the European Pharmacopoeia However, there is a more important reason in today’s world (EP), Japanese Pharmacopoeia (JP), and United States Pharma- of data integrity—integrated instrument qualification and copeia (USP)PROD, to provideU furtherCTS information to help interpret computer validation is an essential component of a data integ- the regulations. These tomes can have monographs for active rity model. The complete four-layer model can be viewed in pharmaceutical ingredients, finished products, and general my recent book, Validation of Chromatography Data Systems chapters that provide requirements for how to apply various (3), but the analytical portion of the model is shown in Figure analytical techniques, such as spectroscopy. 1 and is described in more detail in an earlier “Focus on Qual- ity” column (4). In the Beginning . . . The four layers of the model are milestonesci.com Of the major pharmacopoeias,DIGESTION onlyEXT USPRAC hasTION a general chapterMERC URY• Foundation:CLE managementAN SYNTHESISleadership, policies andASHING proce- ANALYSIS CHEMISTRY on analytical instrument qualification (AIQ) (1). This chapter dures, culture, and ethos came about with a 2003 conference organized by the American • Right instrument and system for the job: instrument qualifi- Association of Pharmaceutical Scientists (AAPS) on analytical cation and computer validation instrument validation. The first decision of the conference was • Right analytical method for the job: development and valida- that the name was wrong and it should be analytical instrument tion of analytical procedures qualification (AIQ). The conference resulted in a white paper • Right analysis for the right reportable result: analysis from (2) that after review and revision became USP general chapter sampling to reporting the result <1058> on AIQ effective in 2008. This chapter described a data The model works from the foundation up with each layer quality triangle, general principles of instrument qualification, providing input to the next. As shown in Figure 1, AIQ and and a general risk classification of analytical equipment, instru- computerized system validation (CSV) come after the foun- ments, and systems. General chapter <1058> did not specify any dation layer, illustrating that if the instrument is not quali- operating parameters or acceptance limits because those can be fied and any software is not validated, the two layers above found in the specific general chapters for analytical techniques. (method validation and sample analysis) will not be effective This column is written so that you will understand the and can compromise data integrity. Interestingly, AIQ and changes that come with the new version and the impact that CSV are missing from many of the data integrity guidance they will have on the way that you qualify and validate instru- documents but you can see the importance in the overall data ments and the associated software, respectively. integrity framework of Figure 1. www.spectroscopyonline.com September 2017 Spectroscopy 32(9) 25

AAPS conference in 2010 and published Level 2: Right analytical procedure for the right job in this column (5). There were three main Validated–veriﬁed under actual conditions of use problems with the first version: • Problem 1: The true role of the supplier Level 1: was missing. The supplier is respon- Right instrument for the right job sible for the instrument specification, Qualiﬁcation or validation (or both) for intended purpose detailed design, and manufacture of

the instrument but this responsibility is Foundation: Right culture and ethos for data integrity (DI) not mentioned in <1058>. The reason Management leadership, DI policies and procedures, staff DI training is that the section on design qualification (DQ) mentions that a user can use Figure 1: The four layers of the data integrity model for laboratories within a pharmaceutical the supplier’s specification. However, a quality system. Adapted with permission from reference 4. user needs to understand the conditions under which the specification was mea- is dumped on the supplier: “The manu- in the summer of 2013, which was circu- sured and how relevant it is to a labora- facturer should perform DQ, validate lated to a few individuals in industry and tory’s use of an instrument. this software, and provide users with a suppliers for review before submission to • Problem 2: Users are responsible for summary of validation. At the user site, the USP. DQ. USP <1058> places great emphasis holistic qualification, which involves the Proposed drafts of the new version on the fact that in the design qualifica- entire instrument and software system, is were published for public comment in tion stage is the responsibility of the more efficient than modular validation of Pharmacopeial Forum in 2015 (9) and supplier, “Design qualification (DQ) is the software alone.” 2016 (10) and comments were incor- most suitably performed by the instru- In the days of data integrity, this ap- porated in the updated versions. The ment developer or manufacturer.” This proach is completely untenable. The approved USP <1058> final version was is wrong. Only users can define their United States Food and Drug Adminis- published in January 2017 in the First instrument needs and must do so to tration (FDA) guidance on software vali- Supplement to USP 40 (11). The chapter define their intended use of the instru- dation (7), quoted by <1058>, was written became effective on August 1, 2017. ment and comply with GMP regula- for medical device software, which is not There was an erratum published in Feb- tions (§211.63) (6). configured unlike much of the laboratory ruary, but the only change was reference • Problem 3: Poor software validation software used today. of the operational qualification (OQ) guidance. The software qualification testing to the intended use definition and validation description in USP Revision of USP <1058> (user requirements specification [URS]). <1058> is the poorest part of this gen- To try and rectify some of these issues, eral chapter as software is pervasive the revision process of USP <1058> What Has Changed in USP <1058>? throughout Group B instruments and started in 2012 with the publication of First let us look at the overall scope of Group C systems. a “Stimulus to the Revision” process ar- changes between the old and new ver- Although the approach to handling ticle published in Pharmacopeial Forum sions of USP <1058> as shown in Table I. embedded software in Group B instru- written Chris Burgess and myself (8). ments where the firmware is implicitly This article proposed two items: Missing in Action or indirectly validated during the instru- • An integrated approach to analytical The following items were omitted from ment qualification is fine, there are omis- instrument qualification and com- the new version of <1058>: sions. Users need to be aware that both puterized system validation (AIQ- • Differences between qualification and calculations and user defined programs CSV), and validation. This was omitted because must be verified to comply with GMP • More granularity for Group B in- qualification and validation activities requirements in 211.68(b) (6). Note that struments and Group C computer- are integrated in the new version, so the qualification of firmware, which is a ized systems to ensure that calcula- why describe the differences? You need simple and practical approach, is now in- tions and user defined programs to control the instrument and any soft- consistent with Good Automated Manu- were captured in the first group and ware and if you can demonstrate this facturing Practice (GAMP) 5, which has an appropriate amount of validation through the 4Qs process described in dropped Category 2 (firmware). was performed by the users for the the new <1058>, why bother with what Software for Group C systems is the second group. the activity is called? weakest area in the whole chapter <1058>. We used the feedback from that arti- • Table I in the old version of <1058> The responsibility for software validation cle to draft a new version of USP <1058> describes the timing, applicability, and 26 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com

activities of each phase of AIQ and it alone software has nothing to do with • Examples of instruments in the three has been dropped from the new ver- AIQ, not surprisingly this section has categories: Because the instrument sion. Rather than give a fixed and rigid been omitted from the new version. The classification depends on the intended approach to AIQ, as the table did, there GAMP 5 (12) and the accompanying use there is no need to give a long list of is more flexibility in the new version good practice guide for laboratory com- instruments or systems in Groups A, of <1058> and omitting this table rein- puterized systems (13) or my chroma- B, and C. It is the intended use of the forces the new approach. tography data system (CDS) validation instrument that defines the group and • Standalone software: Because stand- book (3) will be adequate for this task. providing a list produces anomalies. For

Table I: Comparison of the old and new versions of USP <1058> on analytical instrument qualification Section USP <1058> 2008 Version USP <1058> 2017 Version • Expanded introduction • Can merge activities—for example, IQ and OQ Introduction • Description of Groups A, B, and C moved earlier in chapter • Classification of an instrument depends on the intended use Validation versus • Outline of the differences between qualification the two terms Components of • Data quality triangle unchanged data quality • Essentially the same in the two versions Design Qualification • Emphasis on supplier to perform this • Users must define functional and operations specifications and task intended use • Little if any involvement by the user • Expected to be minimal for commercially available instruments • Users demonstrate fitness for use • Supplier robust design, development, and testing documentation • Change of use triggers review and update of specifications Installation Qualification • IQ needed for pre-owned • Extension of the section to include software installation and IT instruments involvement for interface to a network • Risk assessment for nonqualified instruments to determine if IQ AIQ Process should be performed or not Operational Qualification • Can be merged with IQ • New section on software functions • New section on software configuration or customization • Configure software before OQ testing • Users must review supplier qualification materials • OQ tests refer to instrument specific general chapters Performance Qualification • Expended section on practices for PQ, change control, and periodic review • Timing, applicability, and activities Table 1 for each phase of AIQ • Expansion of section on manufacturers to include suppliers, Roles and service agents, and consultants responsibilities • Requirement for a technical agreement between user and supplier Software • Standalone software • Expanded introduction validation • Firmware now includes control of calculations and user defined programs • Instrument control software expended section Change control • Slimmer and more concise approach to managing change AIQ • Essentially the same in the two versions documentation Instrument • Description of groups A, B, and C categories • Examples of each group Glossary • Definition of seven terms www.spectroscopyonline.com September 2017 Spectroscopy 32(9) 27

example, in the old version of <1058> ment meets them and that data quality <1058> uses the same 4Qs model as the a dissolution bath is listed in Group C and integrity are maintained (11). The 2008 version. Yes . . . but there are some when it should be in Group B if there manufacturer section now includes sup- significant differences. Look at Figure is only firmware and the instrument is pliers, service agents, and consultants 2, which presents the 4Qs model in the calibrated for use. To avoid these argu- to reflect the real world of instrument form of a V model rather that a linear ments the list of examples has been qualification. One new responsibility is flow. This figure was also published in a dropped from the new version. for the supplier or manufacturer to de- “Questions of Quality” column authored velop meaningful specifications for users by Paul Smith and myself (14). However, Additions and Changes to compare with their needs. Incumbent Figure 2 has now been updated to reflect to USP <1058> on both users and suppliers is the need the changes in the new version of USP Of greater interest to readers will be to understand and state, respectively, the <1058>. Look at the green-shaded boxes the changes and additions to the new conditions under which specifications to see the main changes: general chapter, again these can be seen are measured to ensure that laboratory in Table I. Below I will discuss the fol- requirements can be met. We will discuss Design Qualification lowing three areas that reflect the main this further under the 4Qs model in the Design qualification now has two phases changes to the general chapter: next section. associated with it. • Roles and responsibilities Finally, there is a requirement for a • The first phase is for the users to define • Changes to DQ, installation qualifica- technical agreement between users and the intended use of the instrument in a tion (IQ), and OQ phases and how this suppliers for the support and mainte- user requirements specification: Users impacts your approaches to AIQ nance of any Group B instrument and must define functional and opera- • Software validation Group C system. The agreement may tion specifications and intended use take the form of a contract that both par- (11). Although the new <1058> notes Roles and Responsibilities ties need to understand the contents of that this definition is expected to be The USP <1058> update to the “Roles and the responsibilities of each. minimal for commercially available and Responsibilities” section makes users instruments it does not mean slavishly ultimately responsible for specifying their An Updated 4Qs Model copying supplier specifications—espe- needs, ensuring that a selected instru- At first sight, the new version of USP cially if you do not know how any of

Your Mobile Spectroscopy Partner

The all new i-Raman® Pro ST provides easy identification of materials through a variety of packaging and barrier layers! Learn More About the i-Raman Pro ST +1-302-368-7824 www.bwtek.com/ProSeeThrough [email protected] 28 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com

as shown in Figure 2. Executing an OQ

Use outside existing qualification limits or without a corresponding URS or design major instrument upgrade document is planning to fail any qualification. This is one of the major changes in Risk assessment Regular or the new version of USP <1058>. move or major repair • Note that when the new <1058> talks OQ Laboratory user Design Performance verifies Operational Instrument requirements qualification qualification about minimal specifications for com- URS qualification retirement specification (DQ) requirements (OQ) (PQ) mercial instruments, it does not include minimal specifications for software

Installation used to control them for Group C qualiﬁcation (IQ) systems. Here you not only need to

Initial Ongoing Retirement consider the control of the instrument, qualiﬁcation requaliﬁcation and removal but also the acquisition, transforma- Figure 2: Modified 4Qs model for analytical instrument qualification. Adapted with permission tion, storage, and reporting of data and from reference 14. results that includes how data integrity and quality is assured. This activity is the parameters have been measured. that the users’ current requirements not expected to be minimal—especially The output from this process is a URS are in the system and any omissions in today’s regulatory environment. or your intended use definition. are mitigated, where appropriate. • There is also the possibility, for Group • The second phase is the qualification These two sections are where most B instruments, of merging the labora- of the instrument design. This means laboratories get it wrong for reasons such tory user requirements and the DQ: the that you confirm that the selected as we know what we want (therefore why decision to purchase being made on the instrument meets your design speci- bother to document it) or we believed URS (with the proviso that the contents fication or intended use. If looking the supplier’s literature. This is where are adequate). outside of the analytical laboratory, most qualifications fail because there is • Risk management is implicit in the medical device manufacturers call this no specification upon which to base the <1058> classification of the instrument process design verification—ensuring testing in the OQ phase of the process groups and the subgroups of B and C instruments and systems but more needs to be done in the specification and configuration of the software. For example, access controls, data acquisition, and transformation are key areas for managing data integrity risks. • What is also shown in Figure 2 is that if there is any change of use during the operation of the instrument or system, it must trigger a review of the current specifications with an update of them, if appropriate.

Installation Qualification In the new version of <1058>, installation qualification now includes • The installation of software and the involvement of the IT function to interface an instrument to a network • The requirement for conducting an IQ for nonqualified instruments is replaced with a requirement to gather available information and conduct a risk assessment to determine if an IQ should be conducted. In many cases, if an instrument has been installed and maintained by a supplier with records of these activities, but has not been formally qualified, the risk assessment may determine that no IQ should be www.spectroscopyonline.com September 2017 Spectroscopy 32(9) 29

performed, placing increased empha- (and document the settings) before an examples have been removed in the new sis on the OQ phase to demonstrate OQ is conducted, otherwise you’ll be re- version of <1058> and replaced with the fitness for intended use. peating some tests. In practice, however, need to determine the group based on in- • The introduction to <1058> mentions there may be a differentiation of duties tended use, a formal risk assessment now that activities can be merged—for ex- because a supplier may only perform a needs to be performed and documented. ample, IQ and OQ (11). I would see sup- basic qualification of the unconfigured A risk assessment, based on Figure 3 to pliers taking advantage of this option to software leaving the laboratory to con- classify instruments based on their in- have a single IQ and OQ document for figure the application and then conduct tended use has been published by Burgess ease of working, because both phases further verification of the whole system. and myself (15) and is based on the up- are typically conducted by the same in- This point is critical because data in- dated classification used in the new ver- dividual. The combined protocol must tegrity gaps are usually closed through sion of USP <1058> (11). As can be seen be preapproved by the laboratory and configuration of the application. How- from Figure 2, the risk assessment should then post-execution reviewed by them. ever, unless the application is known or be conducted at the start of the process in However, the post-execution review a copy is already installed and config- the DQ phase of work because the out- needs to be conducted while the engi- ured, it is unlikely that the OQ will be come of the risk assessment can influence neer is still on site so that any correc- performed on a configured version be- the extent of work in the OQ phase. tions can be done before the individual cause the laboratory may not know the Rather than classify an item as either leaves the site. process to be automated (for example, Group B or Group C, there is now more hybrid operation, electronic operation, granularity for both groups with three Operational Qualification or incorporate spreadsheet calculations suboptions in each of these two groups. Operational qualification has also been into the software). This increased granularity allows labo- extended to include ratories more flexibility in qualification • A new section on software functions Software Validation Changes and validation approaches, but also fills and the differences between software The major changes to this General the holes from the first version of <1058>. configuration and customization for Chapter occur in the section on software Group B instruments now just Group C systems. validation. They are shown diagrammati- require either qualification of the in- • It is important to configure software cally in Figure 2. Because the instrument strument and either verification of

Collect more light. Keep more light. Detect more light.

This may sound simple, but it’s the driving force behind all we do, because it makes for good spectroscopy. Starting with our own patented high transmissivity VPH gratings, we design low f-number spectrometers and systems that maximize effi ciency at every step.

Benchtop quality data at a compact system price >10x greater sensitivity & measurement speed Superior stray light suppression for high SNR Trace-level limit of detection: Raman, NIR & FL High thermal stability & reproducibility for OEMs

RAMAN | VIS | NIR | FLUORESCENCE

Trade in anywasatchphotonics.com spectrometer - get a 25% discount! [email protected] us for applications expertise | +1 919-544-7785 & testing [email protected] | +1 919-544-7785 30 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com

latory Requirements, Second Edition

USP <1058> (Royal Society of Chemistry, Cambridge, 2017 version USP <1058> 2008 version UK, 2017).

Group A (4) R.D. McDowall, Spectroscopy 31(4), 15–25 (2016). Qualiﬁcation only (5) R.D. McDowall, Spectroscopy 25(11),

Analytical Qualiﬁcation and 24–31 (2010). instrument Group B verify calculations qualiﬁcation (6) Code of Federal Regulations (CFR), 21

Qualification and control user CFR 211, “Current Good Manufacturing programs Practice for Finished Pharmaceutical Qualification and GAMP category nonconfigurable 3 software Products” (Food and Drug Administra- software tion, Sliver Spring, Maryland, 2008). Qualification and GAMP category Group C configurable 4 software software (7) US Food and Drug Administration, Guid-

Qualiﬁcation and GAMP category ance for Industry General Principles of conﬁgurable and 4 software with custom software 5 module Software Validation (FDA, Rockville, Maryland, 2002). Figure 3: Software validation and verification options with the new USP <1058>. Adapted with (8) C. Burgess and R.D. McDowall, Pharma- permission from reference 3. copeial Forum 38(1) (2012). any embedded calculations if used or traceability matrix will be mandatory. (9) “USP <1058> Analytical Instrument specification, build, and test of any user Qualification in Process Revision,” Phar- defined programs. Summary macopeial Forum 41(3) (2015). For Group C systems, the new USP This column has highlighted the main (10) “USP <1058> Analytical Instrument <1058> divides software into three changes in the new version of USP Qualification in Process Revision,” Phar- types: <1058> on analytical instrument quali- macopeial Forum 42(3) (2016). • Nonconfigurable software fication that became effective on August (11) General Chapter <1058>, “Analytical In- • Configurable software 1, 2017. We then discussed three of the strument Qualification,” in United States • Configurable software with custom main changes: roles and responsibilities, Pharmacopeia 40, 1st Suppliment additions changes to the 4Qs model, and the much- (United Stated Pharmacopeial Conven- As can be seen from Figure 3, these improved approach to software validation tion Rockville, Maryland, 2017). three subtypes can be mapped to GAMP for Group C systems. (12) ISPE, Good Automated Manufactur- software categories 3, 4, and 4 plus cat- In general, the USP is moving toward ing Practice (GAMP) Guide Version 5 egory 5 modules. These changes now full life cycle processes. When the new (International Society of Pharmaceutical align USP <1058> closer to, but not version of <1058> became effective in Engineering, Tampa, Florida, 2008). identically with GAMP 5. The main August 2017, it is likely that a new revision (13) ISPE, Good Automated Manufacturing difference is how firmware in Group B cycle will be initiated. If this occurs, a full Practice (GAMP) Good Practice Guide: instruments is validated—directly with life cycle will be the centerpiece of this A Risk-Based Approach to GXP Compli- GAMP 5 or indirectly when qualifying revision. ant Laboratory Computerized Systems, the instrument with USP <1058>. Map- Second Edition (International Society ping of GAMP 5 software categories to Acknowledgments of Pharmaceutical Engineering, Tampa, the new USP <1058> groups has been I would like to thank Chris Burgess, Mark Florida, 2012). published (16) for those readers who Newton, Kevin Roberson, Paul Smith, and (14) P. Smith and R.D. McDowall, LCGC Eu- want more information harmonizing Lorrie Schuessler for their helpful review rope 28(2), 110 –117 (2015). <1058> and GAMP approaches. This comments in preparing this column. (15) C. Burgess and R.D. McDowall, Spec- chapter is much improved and closer in troscopy 28(11), 21–26 (2013). approaches, but not quite there yet! References (16) L. Vuolo-Schuessler et al., Pharmaceuti- However, the bottom line is that (1) General Chapter <1058>, “Analytical In- cal Engineering 34(1), 46–56 (2014). software validation of Group C systems strument Qualification,” in United States under the new USP <1058> should be Pharmacopeia 35–National Formulary the same as any GxP system following 30 (United States Pharmacopeial Con- R.D. McDowall is GAMP 5. One item that is not men- vention, Rockville, Maryland, 2008). the Principal of McDowall tioned in the new <1058> is a traceability (2) AAPS White Paper on Analytical Instru- Consulting and the direc- matrix. For Group B instruments, it will ment Qualification 2004, (American As- tor of R.D. McDowall Lim- be self-evident that the operating range sociation of Pharmaceutical Scientists, ited, as well as the editor of a single parameter will be tested in the Arlington, Virginia). of the “Questions of Quality” column for LCGC Validation of Chroma- OQ. However, this process changes with (3) R.D. McDowall, Europe, Spectroscopy’s Group C systems, especially because tography Data Systems: Ensuring Data sisterit magazine. i DDirect correspondence to: software and networking are involved; a Integrity, Meeting Business and Regu- [email protected] www.spectroscopyonline.com September 2017 Spectroscopy 32(9) 31

IR Spectral Interpretation Workshop The Carbonyl Group, Part I: Introduction

The carbonyl or C=O group is the perfect functional group for detection by infrared (IR) spectroscopy because its stretching vibration peak is intense and is located in a unique wavenumber range. In this introduction to the IR spectroscopy of the carbonyl group we explore why the peak is intense, and see how to apply that knowledge to the analysis of the spectra of ketones. Brian C. Smith

cross all the installments I have written so far the IR spectrum. This area is sometimes referred to as there has been an overarching structure: Each the carbonyl stretching region as a result. The carbonyl A large section has been devoted to the infrared stretching peak is perhaps the perfect example of a group (IR) spectroscopy of a specific chemical bond. We wavenumber, a peak that shows up intensely in a unique started with the C-H bond, moved on to the C-O bond, and relatively narrow wavenumber range. This attribute and now it is time to discuss the spectroscopy of the makes C=O stretches one of the easiest IR peaks to spot C=O or carbonyl group. These bonds are very com- and assign. Because of this trait, an IR spectrometer is, mon and are found in ketones, aldehydes, esters, and in some respects, the perfect carbonyl detector. carboxylic acids, among others. The types of materials Carbonyl bonds can be divided into two classes de- where you will find carbonyl groups include polymers, pending on what type of carbons are attached to the proteins, fats, solvents, and pharmaceuticals. carbonyl carbon. A saturated carbonyl group has two The carbon in a C=O bond is referred to as the “car- saturated carbons attached to the carbonyl carbon. An bonyl carbon” as shown in Figure 1. aromatic carbonyl group has one or more aromatic car- Carbonyl bonds are highly polar because of the large bons attached to the carbonyl carbon. An example of an electronegativity difference between carbon and oxy- aromatic carbonyl group is shown in Figure 3. gen. As a result, the carbonyl carbon has a large par- Saturated and aromatic carbonyl groups can be dis- tial positive charge and the oxygen has a large partial tinguished from each other using IR spectroscopy. In negative charge as denoted in Figure 1. Recall that in a general, aromatic C=O stretching peaks fall ~30 cm-1 chemical bond when there are two charges separated by lower than saturated C=O stretching peaks. This is be- a distance it forms what is called a dipole moment (1). cause aromatic rings, such as benzene, contain p-type Additionally, remember that one of the characteristics orbitals with electron density that sticks up out of the that determines the intensity of infrared peaks is the plane of the molecule (2) as illustrated to the left in Fig- change in dipole moment with respect to bond length, ure 3. The carbonyl group also contains a p-type orbital dμ/dx, during a molecular vibration (1). that points through space toward the orbitals on the Since the carbonyl group has a large dipole moment, aromatic ring, Figure 3. The orbitals are close enough when this group stretches and contracts the value of that they somewhat overlap, allowing some of the dμ/dx is large, thereby giving an intense peak. This electron density from the carbonyl bond to be pulled vibration is illustrated in Figure 2. off into the benzene ring in a phenomenon known as Carbonyl stretching peaks generally fall between conjugation, which is illustrated at right in Figure 3. 1900 and 1600 cm-1 (assume all peak positions hereafter Conjugation weakens the C=O bond, lowers its force are in wavenumber units), a relatively unique part of constant, and hence its peak position is lowered, on 32 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com

Figure 4. Note that the carbons attached to the carbonyl carbon are denoted Carbonyl O “alpha carbons.” carbon A common example of a ketone- CO containing molecule is acetone (di- C methyl ketone), whose IR spectrum is shown in Figure 5. The carbonyl stretch of acetone Figure 1: The chemical structure and charge Figure 2: The stretching vibration of a carbonyl falls at 1716, and is labeled A in Fig- distribution in a carbonyl bond. bond. ure 5. In general, for saturated ketones this peak appears at 1715 ± 10. Note how large this peak is, easily the biggest in the spectrum. This is O O common for the carbonyl stretches of many functional groups. C C It is tempting to assume that the carbonyl stretching peak for ketones uniquely identifies them, but this is not the case. Saturated esters, aldehydes, and carboxylic acids, Figure 3: The p-type orbitals and conjugation within an aromatic carbonyl group. amongst others, have their C=O stretching peaks around 1715. Thus, we will see that the C=O stretching peak is diagnostic for carbonyl O groups but not for anything else. We will be dependent upon other Alpha peaks in IR spectra besides the C=O carbon C Alpha carbon stretch to distinguish the different CC types of carbonyl containing functional groups from each other. Figure 4: The structural framework of the ketone functional group. For ketones, the peak that can help distinguish them from other functional groups is the C-C-C stretch vibration, which is illus- 1.0 1716 O trated in Figure 6.

0.8 C Note that this vibration involves CH CH the two alpha carbons stretching 3 3 1222 1362 0.6 asymmetrically about the carbonyl C carbon. This vibration gives rise to A B 0.4 an intense peak between 1230 and Absorbance

530 1100 for saturated ketones. This 1422 0.2 peak appears in the spectrum of ac- 3005 1093 903 etone at 1222 and is labeled B. Note 0.0 it is the second largest peak in the spectrum. 3500 3000 2500 2000 1500 1000 500 Wavenumber (cm–1) Normally a C-C stretching vibration peak is small because the Figure 5: The infrared spectrum of acetone, C3H6O. electronegativity difference between two carbon atoms is often negli- average about 30 cm-1 (1). Thus, The IR Spectroscopy of Ketones gible, giving nonpolar bonds and for almost every carbonyl contain- Perhaps one of the most common small values of dμ/dx. However, the ing functional group we discuss I chemical structures to contain a car- carbonyl carbon in the C=O bond will quote two carbonyl stretching bonyl group are ketone molecules. has a large positive partial charge regions, one for the saturated and Ketones consist of a C=O with two on it, as shown in Figure 1. This one for the aromatic versions of the carbons attached, one to the left charge polarizes the bonds to the functional group. and one to the right as illustrated in two alpha carbons, and when these www.spectroscopyonline.com September 2017 Spectroscopy 32(9) 33

Table I: The infrared group wavenumbers for ketones O Wavenumber Vibration Range C Saturated C=O stretch 1715 ± 10 C C Aromatic C=O stretch 1700–1640 Saturated C-C-C stretch 1230–1100 Figure 6: The C-C-C stretching vibration of ketones. Aromatic C-C-C stretch 1300–1230 discover

1.0 1686

O 1266 A B 0.8 C

CH3 0.6 C 761 1360 691 589 0.4 Absorbance 1599 1449 0.2 solve 3064 0.0

3500 3000 2500 2000 1500 1000 500 Wavenumber (cm–1)

Figure 7: The IR spectrum of acetophenone (phenyl methyl ketone), C6H8O. C-C bonds stretch dμ/dx is large, acetone because of conjugation. A giving the typically large ketone summary of the group wavenum- assure C-C-C stretching peak labeled B in bers for ketones is shown in Table I. Figure 5. Note that both the C=O stretch The chemical structure of an and the C-C-C stretch are sensitive aromatic ketone, acetophenone, as to whether a ketone is saturated is shown in Figure 7. Note that al- or aromatic, so either one of these though there is an aromatic and a peaks can be used to distinguish saturated carbon attached to the saturated and aromatic ketones What did you carbonyl carbon in acetophenone, from each other. The C=O and it is considered an aromatic ketone. C-C-C stretching peaks are the best This is because it only takes the peaks to use to determine if there is do today? presence of a single aromatic sub- a ketone in a sample. -;09ₔ509ₔ9(4(5 stituent to engage in conjugation and lower the carbonyl stretching Methyl Ketones and peak position. The IR spectrum of Their Umbrella Mode acetophenone (phenyl methyl ke- A methyl ketone is a structure where tone) is also shown in Figure 7. one or more of the substituents on The carbonyl stretch, labeled A, is the carbonyl carbon is a methyl at 1686, while the C-C-C stretch, la- group. Both acetone and acetophe- Find out more at beled B, is at 1266. For aromatic ke- none are examples of methyl ketones, generally the C-C-C stretch tones. Previously we learned (2) that [OLYTVÄZOLYJVT falls between 1300 and 1230. Note methyl groups in alkyl chains have ZVS]LPZ that these are the two most intense an umbrella mode around 1375, and peaks in the spectrum. Structurally we also observed that this peak was -VY9LZLHYJO

Quiz Section: Answer to the July Quiz and a New Interpretation Challenge

The Last Quiz those cases. However, narrowing down the millions of pos- I apologize for the X marks through two of the peaks, but sible structures to just two is still useful. A library search those two peaks are misleading. or looking up the relevant spectra in an IR atlas would allow final identification of the sample. As it turns out the alkyl chain has three methylenes. The molecule is named 632

2971 1-hexyne because it has six carbons, and the triple bond is at 1.5 2947 carbon 1, making it monosubstituted. The chemical structure of 1-hexyne is seen in Figure ii. 1.0 1778 1249 3328 2886 1140 1460 1376 1706 0.5 1166 HC C CH CH CH CH

Absorbance 2 2 2 3 3582 2103

0.0 Figure ii: The chemical structure of 1-hexyne, the answer to last installment’s 4000 3500 3000 2500 2000 1500 1000 500 infrared interpretation challenge. Wavenumber (cm–1) A summary of the assignment of the peaks in the spectrum of 1-hexyne is shown in Table i. Figure i: The infrared spectrum of a 1-hexyne, C6H10. Ignore the peaks with X’s through them. Table i: Proper assignment for peaks in the spectrum of 1-hexyne. As I have taught, you should read an IR spectrum from Peak Assignment left to right like a sentence in a book (1). The first peak we 3328 H-C≡C stretch come to is at 3328. Normally we would assign this peak 2971 CH asymmetric stretch position as an OH stretch (3). However, as we have learned, 3 2947 CH asymmetric stretch OH stretches are normally broad because of hydrogen bond- 2 2886 CH symmetric stretch ing, and this peak is narrow. Recall from last time (4) that 3 monosubstituted alkynes have an unusually intense, narrow, 2106 Monosubstituted C≡C stretch

high-wavenumber C-H stretch that falls between 3350 and 1376 CH3 umbrella mode 3250. The peak at 3328 meets this description and is properly 632 H-C≡C- wag assigned as the C-H stretch of a monosubstituted alkyne. As we also discussed last time (4), monosubstituted alkynes have New Quiz a C≡C stretching peak of variable intensity from 2140 to Using everything you have learned in this and previous col- 2100. The feature at 2103 in Figure i corresponds to this peak. umns, determine the functional groups present and try to Lastly, the H-C≡C bond of monosubstituted alkynes has an discover the chemical structure of this compound. intense C-H wagging peak in the 700 to 600 range, which is shown in Figure i at 632. 100 Now that we have established the presence of the H-C≡C- moiety, the final question is what is the structure of the 80 3413.9 995.4

single substituent. There is a group of three C-H stretches 517.5 760.0 60 1086.2 just below 3000, indicative of the presence of CH3 and CH2 589.8 1205.5 groups (3). Specifically, 2971 and 2886 are the methyl asym- 945.2 40 metric and symmetric stretches, and 2947 is a methylene 2882.2 Transmittance (%) Transmittance

asymmetric stretch. The umbrella mode peak at 1376 con- 1416.7 20 2939.6

firms the presence of the methyl group (5). This all means 1459.9 1717.8 2978.5 1172.3 the substituent is an alkyl chain. The methyl and methylene 0 1366.0

asymmetric stretch peaks are about the same size, indicat- 4000 3500 3000 2500 2000 1500 1000 500 –1 ing the CH2/CH3 ratio is two or greater (6). Recall (5) that if Wavenumber (cm ) an alkyl chain has four or more methylenes in a row, it will Figure iii: The IR spectrum of a liquid. Sampling technique: capillary thin exhibit a CH2 rocking peak at 720 ± 10. There is no such peak in Figure i, which means the alkyl chain has one, two, film. or three methylenes in it and a methyl group. Now, since we Table ii: Peaks that need to be identified know the CH /CH ratio is 2 at minimum, the possibilities 2 3 2979 1459 are narrowed down to a chain with two or three CH2 groups. As has been mentioned before (5), when interpreting IR 2939 1366 spectra it is not always possible to obtain the exact alkyl 2882 1172 chain length from a spectrum by itself. This quiz is one of 1717 — publish faster

discover

advance research +?9_P9HTHU0THNPUN 4PJYVZJVWL

reveal chemistry evolve materials What did you do today?

Whether you’re discovering new materials, solving analytical problems or assuring WYVK\J[X\HSP[``V\YZWLJ[YVTL[LYULLKZ[VKLSP]LY[OLKLÄUP[P]LHUZ^LYZ`V\»YL SVVRPUNMVY·MHZ[;OLYTV-PZOLY:JPLU[PÄJNVLZIL`VUK`V\YL_WLJ[H[PVUZ^P[O a full line of FTIR, NIR and Raman spectroscopy systems to help you move from sample to answer, faster than ever before.

>P[O[OL;OLYTV:JPLU[PÄJ™ DXR™_P9HTHU0THNPUN4PJYVZJVWL`V\NL[HU easy-to-use system that rapidly delivers accurate data – so you spend less time JVUJLU[YH[PUNVUTLHZ\YLTLU[HUKTVYL[PTLVUZJPLU[PÄJKPZJV]LY`

Discover. Solve. Assure. [OLYTVÄZOLYJVTKPZJV]LYK_Y

-VY9LZLHYJO

This intense CH3 umbrella mode is tached to the carbonyl carbon. They SSoftware, a company indicative of a methyl ketone. Thus, if exhibit diagnostically useful C=O and tthat sells chemometric, your analysis of a spectrum indicates C-C-C stretches, both of whose peak mmultivariate analysis, that a ketone and methyl group are positions are sensitive as to whether aand process control present, and the CH3 umbrella mode the ketone is saturated or aromatic. software. Before joining CAMO, Dr. Smith is unusually large, it is a good bet Methyl ketones display a unique high- ran his own FT-IR training and consulting you have a methyl ketone. This peak intensity CH3 umbrella mode. business for more than 20 years. Dr. is diagnostically very useful. After a Smith has written three books on infrared ketone is identified the next step is References spectroscopy: Fundamentals of FTIR and to determine the two substituents at- (1) B.C. Smith, Spectroscopy 30(1), Infrared Spectral Interpretation, both tached to the carbonyl carbon. The 16–23 (2015). published by CRC Press, and Quantitative large methyl ketone umbrella mode (2) B.C. Smith, Spectroscopy 31(3), Spectroscopy: Theory and Practice tells us that one of the substituents is a 34–36 (2016) published by Academic Press. He can be

CH3 group. (3) B.C. Smith, Spectroscopy 32(1), reached at: [email protected] 14–21 (2017). Conclusions (4) B.C. Smith, Spectroscopy 32(7), The carbonyl group, C=O, has a large 18–24 (2017). For more information on this topic, dipole moment and thus an intense (5) B.C. Smith, Spectroscopy 30(7), please visit our homepage at: stretching vibration whose peak 26–31, 48 (2015). www.spectroscopyonline.com generally appears from 1900 to 1600. (6) B.C. Smith, Spectroscopy 30(9), When an aromatic ring is attached to 40–45 (2015). www.spectroscopyonline.com September 2017 Spectroscopy 32(9) 37

The 2017 Emerging Leader in Molecular Spectroscopy Award

Russ Algar, an assistant professor in chemistry at the University of British Columbia (UBC), in Vancouver, Canada, has won the 2017 Emerging Leader in Molecular Spectroscopy Award, which is presented by Spectroscopy magazine. This annual award recognizes the achievements and aspirations of a talented young molecular spectroscopist, selected by an independent scientific committee. The award will be presented to Algar at the SciX 2017 conference in October, where he will give a plenary lecture and be honored in an award symposium, both on Monday, October 9.

Megan L’Heureux

magine your doctor taking out a cell phone one day and Algar “a rising star” on the Canadian using the camera to diagnose an illness or to run a series academic scene. “The development of I of tests that lead to a more accurate result than current novel spectroscopy tools and materials technology allows. If those dreams of “personalized medi- for chemical analysis is an important cine” become a reality one day, it will be in part, thanks to the research field with potentially major research that Russ Algar, an assistant professor in chemistry impacts in the analytical chemistry and at the University of British Columbia (UBC), in Vancouver, biotechnology sectors,” said Boudreau. Canada, has been doing. In many respects, Algar is leading “Professor Algar is already one of the the charge to combine analytical chemistry and personalized most prominent and promising young medicine by developing smartphone-based point-of-care di- actors in this field.” Russ Algar agnostics in which photoluminescence from semiconductor Algar has certainly achieved a great quantum dot (QD) nanocrystals is measured by a smartphone deal in his short career. He received his PhD in 2010 (from camera. He is also working on fluorescence assays that target the University of Toronto, under Ulrich J. Krull), and has been protein biomarkers and cell types, and QD-based concentric an assistant professor at the University of British Columbia Förster resonance energy transfer (cFRET) probes with mul- since 2012. In that short time, he has published more than 75 tifunctional detection capabilities. manuscripts and several book chapters with more than 3000 This article highlights the work Algar has accomplished so citations according to the ISI Web of Science. He has also de- far and his future research plans, and shares some insights into livered more than 50 presentations at national and interna- Algar and his work from friends and colleagues. tional conferences. In addition, Algar has been the recipient of several awards such as the Fred Beamish award (see photo on A Stellar Candidate next page) and an Alfred P. Sloan Research Fellowship in 2017. The Emerging Leader in Molecular Spectroscopy award rec- Algar is also active in the broader scientific community ognizes the achievements and aspirations of a talented young through his work with various conferences, editorial advisory molecular spectroscopist. Algar was highly recommended for boards, and scientific societies. For example, this year Algar the 2017 award by Chris Orvig, a professor of chemistry and was the co-organizer for two Canadian Society for Chemistry pharmaceutical sciences and director of the Medicinal Inor- (CSC) symposia. Algar is also a member-at-large on the CSC ganic Chemistry Group at UBC. Orvig explained that Algar Analytical Chemistry Division and a member on the board of was hired at UBC to strengthen their analytical division with directors for the Chemical Education Fund—among others. spectroscopy expertise, and he truly excelled in that endeavor. Given his body of work, it’s easy to see why Algar was “He is a stellar candidate for this new young researcher award nominated for our award. But how did he first become inter- and I am delighted to recommend him in the strongest pos- ested in analytical chemistry and molecular spectroscopy? sible terms,” said Orvig. Algar said his broad interest in analytical chemistry evolved Denis Boudreau, a professor at the Université Laval, calls naturally as part of his undergraduate studies and the way 38 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com

available in rural and remote communities without specialized labs,” Algar said. “This benefits both the patient and the health care budget.” But a smartphone assay is not a magic bullet, Algar is quick to point out. “The full solutions to health care challenges will have many moving parts,” he notes. “But the concept embodied by the smartphone assay is likely to be an important component.” How exactly will smartphone-based assays work? Algar and his group’s initial research in this area focused on finding and characterizing the best materials for fluorescence-based assays with smart- Algar receiving the Fred Beamish Award (sponsored by Thermo Fisher Scientific) from Mark phones. They began focusing on QDs, James of Thermo Fisher Scientific. At left: Canadian Society for Chemistry (CSC) President Rui which are much brighter than conven- Resendes. (Photo courtesy of CSC, specifically Gale Thirlwall.) tional fluorescent dyes, making their fluorescence easier to detect and translating his mind works. He also cited the di- key to earlier diagnosis of diseases and into a more sensitive assay. It’s even pos- versity that can be found in analytical more effective interventions, resulting sible, Algar explained, to use blue light chemistry and molecular spectroscopy in better health,” he said. “An important filtered from the camera flash to excite as something that attracted him to those concept within this context is personal- measurable fluorescence from a QD. And fields. “Analytical chemistry and molec- ized medicine, where medical interven- the emission of QDs can be precisely ular spectroscopy embrace technology, tions are tailored to the specific mani- tuned to match the built-in color filters enable new fundamental discoveries festation of a disease in an individual of a smartphone camera (red-green-blue about how nature works, strive to create rather than the population average, and or RGB), to help improve sensitivity and practical inventions, and easily interface adjusted over time while monitoring the facilitate the use of the different RGB with other disciplines in chemistry, bi- specific response of that individual to the color channels to detect different assay ology, physics, material science, and en- intervention.” targets. “In short, the properties of a QD gineering,” he said. “I really like having These concepts are great in theory, but can make fluorescence measurements on such a wide range of space in which to there is a measurement problem in mak- a smartphone not only possible, but also work and explore.” ing them a viable reality: Many measure- practical,” he said “From this point, it is ments need to be made at multiple time possible—at least in principle—to take A Passion for Exploration points. “The model of having a doctor any existing fluorescence-based assay Exploration seems to be a passion for order tests and collect a sample, or hav- with QDs and adapt it to a smartphone.” Algar, as he delves into new research ing the patient go to a specialized labora- Algar is also researching QD-based areas with the excitement and thorough- tory, is not going to be sufficient,” he said. concentric Förster resonance energy ness of an early adventurer. For Algar, Instead, what’s needed is a model were transfer (cFRET) probes with multifunc- an idea can open a whole new world of nurses can do some tests at the hospital tional detection capabilities. A conven- possibility. For example, his work creat- bedside, doctors can do some at their pri- tional FRET probe is monofunctional ing fluorescence assays to target protein vate practice, and patients can do some (that is, it detects one target) whereas a biomarkers and cell types was inspired tests at home. “Smartphone-based assays cFRET probe is multifunctional (that is, by the concept of molecular medicine, for protein and cell biomarkers can po- it detects multiple targets). The conven- which focuses on understanding and as- tentially enable this new model,” he said. tion with FRET probes is to use n col- sessing pathologies at the level of molec- Smartphone-based assays also have ors of QD to detector or label n different ular mechanisms and cellular processes the potential to address the discrepancy targets, but his group wondered if they rather than the systemic observations of in health care provided to people in could use one color of QD to detect n dif- classical medicine. urban and rural communities. People ferent things. The answer they came up For example, rather than looking at in rural or remote communities often with was cFRET with n = 3. “We were blood pressure, temperature, coughing, must travel far for medical care, and very happy to hit n = 3 because most and other symptoms of disease, Algar about half of the health care budget in cellular processes occur as a cascade of explained, in molecular medicine, one one of Canada’s northern territories is multiple steps and often share molecular would look at the levels of certain pro- dedicated to patient travel. “If molecu- machinery with other processes,” he said. teins, genes, and cell types. “Many people lar assays are available in the form of “With n = 3, a good number of critical believe that molecular medicine is the a smartphone assay, then they can be steps in those cascades can be detected www.spectroscopyonline.com September 2017 Spectroscopy 32(9) 39 with a single QD and using a cFRET design of high-contrast time-gated FRET probe is likely to be more effective than probes, including photonic logic gates,” using three separate FRET probes, al- he says. though we still need to prove this point.” Algar’s goal with the LLCs is to create Algar’s cFRET research is currently some probes that can do “dirty work” moving from the test tube to a live bio- in the sense of being robust enough to logical system, in the form of cultured be added directly to a clinically relevant cells. “Our first big goal is to quantita- sample, such as serum or, even better, tively track a multistep enzymatic cas- whole blood directly from a patient. In cade in single cells,” he said. “It’s exciting the case of photonic logic, Algar’s long- because we’re going to see if cFRET can term aim is a probe or system of probes enable the new insights and discoveries that can make an autonomous decision that we think it can.” about the levels of multiple biomarkers Algar hiking in Zermatt, Switzerland. Algar and his group are also studying other novel materials and configurations for biosensing. Examples include lumi- nescent lanthanide complexes (LLCs) and semiconducting polymer dots (Pdots), which may be useful materials for point-of-care diagnostics, cellular imaging, and biophotonic Boolean logic operators for rapid “yes or no” screening assays for multiple pathogens or disease states. Because Pdots are exceptionally bright, even more so per particle than QDs, Algar’s group is looking at how these materials can be used as an al- ternative to QDs in their point-of-care diagnostic research. But Pdots are less developed than QDs, so Algar’s group is developing new chemistries to help facilitate the applications they envision for Pdots in point-of-care diagnostics. “Our Pdot research is in the early stages, but it looks promising, and I think we will have some good results by the spring,” he said. LLCs, are special, Algar says, because of their very long excited-state lifetimes, 3D Raman image of a pharmaceutical ointment. which permit time-gated measurements. Time-gating is the idea of having a flash of excitation light, then a delay of mi- croseconds to milliseconds before measuring the luminescence; this approach 3D Raman helps suppress scattering and fluores- gg cence from anything but the LLC. “In my discoveries opinion, the concept is well-established Turn ideas into but under utilized, perhaps because LLCs have not typically been very bright,” he Let your discoveries lead the scientiﬁ c future. Like no other system, WITec’s confocal 3D Raman microscopes allow for said. But some newer LLCs are brighter, cutting-edge chemical imaging and correlative microscopy and Algar’s group has been able to do with AFM, SNOM, SEM or Proﬁ lometry. Discuss your ideas some interesting spectroscopy with with us at [email protected]. them, such as elucidating a “sweet spot” for time-gated FRET, which balances two competing factors—higher FRET efficiency and the concomitant reduction Raman • AFM • SNOM • RISE www.witec.de in LLC donor lifetime. “We’ve leveraged MADE IN GERMANY this idea to show some new twists on the 40 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com and generate a binary “healthy/sick” out- to FRET-relays and concentric-FRET maximizes productivity.” put, where a “sick” output could be used present outstanding advances for the Algar’s peers have also noted his drive as a decision point for further biomedical molecular spectroscopy community,” for success. Algar’s PhD advisor, Ulrich testing. In both cases, the goal is to design he said. Krull, of the University of Toronto, said probes that may be complex, but are sim- Boudreau, of Université Laval, said it is rare to see anyone so driven to suc- ple to use and provide simple readouts. that Algar excels at tailoring QD emis- ceed on so many fronts. “Algar starts “Our research toward this concept is sion to combine various colors for his day at 6 a.m. with an intense physi- in its infancy,” Algar said. “We’re still multiplex and FRET-based assays that cal workout, then he brings enormous learning how to design systems that can respond selectively to the presence of creativity to both his research and teach- generate a single, clean, binary output certain biomolecules such as DNA ing, and contributes to his community while detecting multiple biomarker tar- strands, enzymes, and antibodies. “The in leadership roles,” he said. gets, in the context of different types of multiple functionalities that Algar and Algar’s published work has also re- biomarkers.” his group are able to embed in these ceived praise. “I also really appreciate nano-constructs lower the requirements the thorough set of investigations that he Impact in the Field: on the instruments needed to optically has performed and published on, which Praise from Experts read the assays, as demonstrated by his essentially provide a series of how-to Algar’s peers in the scientific commu- ingenious work on smartphone-based manuals for the interested user of these nity have taken notice of his work. His optical readers,” he said. “In this field, techniques,” said Medintz. friends and colleagues have described he is one of the best—if not the best—on Lucy mentioned two papers in par- his personal character and work ethic the planet.” ticular, which Algar published as a in glowing terms such as honest, profes- Others, such as Jean-Francois Mas- graduate student, one of which involved sional, altruistic, humble, hard-working, son, a professor at the Université de the assembly of a modular fluorimeter enthusiastic, relentlessly innovative, mo- Montréal, said it is difficult to pinpoint (1). “In addition, last year he published tivating, scholarly, and rigorous. What only one example of Algar’s work as his Journal of Chemical Education papers do they think is his greatest contribution most impressive achievement, as there on a guided inquiry experiment where to molecular spectroscopy so far? are many to choose from. Ultimately, he students built their own photometer, Igor L. Medintz, who is the U.S. Na- narrowed it down to Aglar’s use of a cell and another that demonstrates the color vy’s Senior Scientist for Biosensors and phone for measuring kinetic reactions wheel using LEDs and indicator dyes,” Biomaterials at the U.S. Naval Research using fluorescence spectroscopy. “In said Lucy (2,3). Laboratory and was Algar’s post-doc- particular, the use of the RGB channels Hildebrandt recalled a story that il- toral advisor, believes that Algar’s work of the cell phone camera for monitoring lustrates what Algar brings to the field. developing different biosensing modali- different reactions is clever,” he said. “It Hildebrandt has been a collaborator ties based on QDs and energy transfer is also an outstanding example of mul- with Medintz’s lab for a long time and he can be considered seminal. In particular, tidisciplinary research, where nanoma- always wanted to perform double-step he lauded Algar’s work with time-gated terials, biochemistry, instrumentation, FRET experiments in his own lab, but multiFRET sensing, concentric mul- and software are combined in one work- it never panned out. When Algar joined tiFRET around QDs with up to three ing sensor. This is an example of high- Medintz’s lab, within only a few months different dye acceptors, and incorpo- quality analytical chemistry.” he had set up all the experiments and rating both FRET and electron transfer Charles Lucy, a professor emeritus of successfully performed very complicated at the same time into a QD-based sen- chemistry at the University of Alberta, double-step temporal and spectral reso- sor. “As Algar has amply demonstrated, focused on a different aspect of Algar’s lution FRET for multiplexed biosensing. these sensors can detect multiple targets contribution: his passion for teaching, “In other words, Russ is someone who (multiplexing) or even directly moni- and his innovative approach to it. can do highly sophisticated research in tor coupled biological processes such Medintz agreed. “Algar firmly be- a few months, for which others would as enzyme activity while only requir- lieves in the societal role of professors take years, and he obtained excellent re- ing a single QD-based construct,” said and what they should continually strive sults that resulted in a joint publication Medintz. This capability, he notes, can for in terms of teaching, mentoring new in the Journal of the American Chemical considerably ease the required instru- scientists, and adding to the body of Society,” he said (4). mentation and materials needed for knowledge,” he said. such experimental formats. “Moreover, Algar himself said that his students’ Future Work the data that they provide is essentially theses and research projects are the So, what does the future hold for this orthogonal, with almost no cross talk.” best part of his job and the basis of his bright, young scientist? Niko Hildebrandt, a professor at the research program. “My students, post- In Masson’s view, the answer to this Université Paris-Sud, agreed that Algar’s doctoral fellows, and I work together question depends on a combination of work has significantly advanced the un- on the research,” he said. “I think—and Algar’s goals and the opportunities that derstanding and application of QDs for hope—that such a team approach pro- arise. “Our careers are not predefined FRET. “In particular, his contributions vides a good training environment and and an event can easily trigger changes www.spectroscopyonline.com September 2017 Spectroscopy 32(9) 41 in our path, whether it is toward a new ticularly for students, is to try not to research field, a different institution, or be overwhelmed by the accomplish- being involved in administrative duties,” ments, knowledge, or experience of he said. “However, there is one variable people at later stages of their educa- that is sure: Algar has a tremendous po- tion or career. “You can’t instantly tential to contribute at all levels in our become those people, but each day community.” you can try to learn something new, See Others felt more confident in pre- improve in some way, or move some- dicting some future outcomes for thing forward (even a little bit), and Algar. “Without a doubt, he is already eventually end up at the same point a major player in the discipline of an- as them,” he said. things alytical chemistry and bioanalytical It’s obvious that Algar lives by that chemistry,” said Boudreau. “I see him advice. becoming an influential leader and clearer More About Algar a role model in Canadian academia, bolt not only in research and bioanalysis An in-depth interview with Russ Algar with Co but also for education and public out- focused on his research, challenges, Microwave Digestion | Clean Chemistry | Mercury Analysis reach.” and accomplishments will be pub- Krull agrees that Algar will con- lished in the October 3, 2017, edition tinue to be recognized as a creative of our newsletter, The Wavelength. leader, specifically in nanoparticle technologies. “I anticipateHELPIN that he willG 2018 Emerging Leader in help move these technologiesCHE M ISTS from re- Molecular Spectroscopy: search to the bedside for medical and Now Accepting Nominations clinical applications, he said. For information about how to nominate someone for the 2018 award, please see High Performance Lasers Conclusion the call for nominations on our website: From his work with QDs and cFRET www.spectroscopyonline.com/emerg- AtQ Singl Me friequenclestoney , We Help Chemists. to the newerMicro researchwa vwithe DigestionPdots and ing-leader-molecular. Q UV-VIS-MIR LLCs, Algar seems determined to 30 Years.Q CW & Q- 5sw0itched Patents. 20,000 Global Users. achieve innovative goals that the sci- References Q Unprecedented reliability entific and medicalClean communities Chemistry have (1) W.R. Algar, M. Massey, and U.J. Krull, J. with HTCure™ been striving to accomplish for a long Chem. Educ. 86, 6871 (2009). time. He is Mercuryquick to point Analysis out, how- (2) J.J. Wang, J.R.R. Nunez, E.J. Maxwell, and ever, that he does not do it alone, say- W.R. Algar, J. Chem. Educ. 93, 166–171 ing that he shares his success with his (2016). spouse, Melissa, who is also a scien- (3) W.R. Algar, C.A.G. De Jong, E.J. Maxwell, tist, as well as with his students, post- and C.G. Atkins, J. Chem. Educ. 93, 162–165 doctoral fellows, and collaborators. (2016). Visit Our Updated Website We asked Algar what advice he (4) W.R. Algar, D. Wegner, A.L. Huston, J.B. would give to someone that wanted to Blanco-Canosa, M.H. Stewart, A. Armstrong, follow in hismilestonesci.com path. He broke it down to P.E. Dawson, N. Hildebrandt, and I.L. Med- three key points: First, he pointed to intz, J. Am. Chem. Soc. 134, 1876−1891 866.995.5100 • milestonesci.com work ethic, citing the Thomas Edison (2012). quote, “Genius is 1% inspiration and 99% perspiration.” Megan L’Heureux is the managing “I think I got ahead by outworking editor of Spectroscopy and LCGC North people rather than by outsmarting America magazines in Iselin, New Jersey. them,” said Algar. Direct correspondence to: Second, he said it is important to [email protected] ◾ learn to be confident in the face of criticism. “It’s not easy, but I think the path to success is learning from, embracing, and even asking for critical— For more information on this topic, and hopefully constructive—feedback please visit our homepage at: rather than avoiding it,” he said. www.spectroscopyonline.com cobolt.se His third recommendation, par- 40 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com

and generate a binary “healthy/sick” out- to FRET-relays and concentric-FRET maximizes productivity.” put, where a “sick” output could be used present outstanding advances for the Algar’s peers have also noted his drive as a decision point for further biomedical molecular spectroscopy community,” for success. Algar’s PhD advisor, Ulrich testing. In both cases, the goal is to design he said. Krull, of the University of Toronto, said probes that may be complex, but are sim- Boudreau, of Université Laval, said it is rare to see anyone so driven to suc- ple to use and provide simple readouts. that Algar excels at tailoring QD emis- ceed on so many fronts. “Algar starts “Our research toward this concept is sion to combine various colors for his day at 6 a.m. with an intense physi- in its infancy,” Algar said. “We’re still multiplex and FRET-based assays that cal workout, then he brings enormous learning how to design systems that can respond selectively to the presence of creativity to both his research and teach- generate a single, clean, binary output certain biomolecules such as DNA ing, and contributes to his community while detecting multiple biomarker tar- strands, enzymes, and antibodies. “The in leadership roles,” he said. gets, in the context of different types of multiple functionalities that Algar and Algar’s published work has also re- biomarkers.” his group are able to embed in these ceived praise. “I also really appreciate nano-constructs lower the requirements the thorough set of investigations that he INDUSTRIES Impact in the Field: on the instruments needed to optically hasMI performedLESTONE and published on, which Praise from Experts read the assays, as demonstrated by his essentially provide a series of how-to Algar’s peers in the scientific commu- ingenious work on smartphone-based manuals for the interested user of these nity have taken notice of his work. His optical readers,” he said. “In this field, techniques,” said Medintz. friends and colleagues have described he is one of the best—if not the best—on Lucy mentioned two papers in par- his personal character and work ethic the planet.” ticular, which Algar published as a in glowing terms such as honest, profes- Others, such as Jean-Francois Mas- graduateHELPING student, one of which involved ACADEMIA CANNABIS CLINICAL sional,COSMETI altruistic,CS humble,ENE hard-working,RGY ENVI RONMENTson, a professorAL at the Université de theC assemblyHEMIST of a modularS fluorimeter TESTING enthusiastic, relentlessly innovative, mo- Montréal, said it is difficult to pinpoint (1). “In addition, last year he published tivating, scholarly, and rigorous. What only one example of Algar’s work as his Journal of Chemical Education papers do they think is his greatest contribution most impressive achievement, as there on a guided inquiry experiment where to molecular spectroscopy so far? are many to choose from. Ultimately, he students built their own photometer, Igor L. Medintz, who is the U.S. Na- narrowed it down to Aglar’s useD of iscovera cell and another how that demonstratesMilest onethe color vy’s Senior Scientist for Biosensors and phone for measuring kinetic reactions wheel using LEDs and indicator dyes,” FOOD & FEED METALS NUTRACEUTICAL PHBiomaterialsARMA/USP at the U.S.POLYME NavalR SResearchSPE usingCIALTY fluorescence spectroscopy. “In said Lucy (2,3). CLaboratoryOMPLIANC andE was Algar’s post-docC- HEMIparticular,CALS the use of the RGBcan channels help Hildebrandt your recalledlabora a storyt ory.that il- toral advisor, believes that Algar’s work of the cell phone camera for monitoring lustrates what Algar brings to the field. developing different biosensing modali- different reactions is clever,” he said. “It Hildebrandt has been a collaborator PRODUCTS ties based on QDs and energy transfer is also an outstanding example of mul- with Medintz’s lab for a long time and he can be considered seminal. In particular, tidisciplinary research, where nanoma- always wanted to perform double-step he lauded Algar’s work with time-gated terials, biochemistry, instrumentation, FRET experiments in his own lab, but multiFRET sensing, concentric mul- and software are combined in one work- it never panned out. When Algar joined tiFRET around QDs with up to three ing sensor. This is an example of high- Medintz’s lab, within only a few months different dye acceptors, and incorpo- quality analytical chemistry.” he had set up all the experiments and rating both FRET and electron transfer Charles Lucy, a professor emeritus of milestonesci.comsuccessfully performed very complicated DIGESTION EXTRACTION MERCURY at theCLE sameAN time intoSYNTHESIS a QD-based sen- ASHINGchemistry at the University of Alberta, double-step temporal and spectral reso- ANALYSIS CHEMISTRY sor. “As Algar has amply demonstrated, focused on a different aspect of Algar’s lution FRET for multiplexed biosensing. these sensors can detect multiple targets contribution: his passion for teaching, “In other words, Russ is someone who (multiplexing) or even directly moni- and his innovative approach to it. can do highly sophisticated research in tor coupled biological processes such Medintz agreed. “Algar firmly be- a few months, for which others would as enzyme activity while only requir- lieves in the societal role of professors take years, and he obtained excellent re- ing a single QD-based construct,” said and what they should continually strive sults that resulted in a joint publication Medintz. This capability, he notes, can for in terms of teaching, mentoring new in the Journal of the American Chemical considerably ease the required instru- scientists, and adding to the body of Society,” he said (4). mentation and materials needed for knowledge,” he said. such experimental formats. “Moreover, Algar himself said that his students’ Future Work the data that they provide is essentially theses and research projects are the So, what does the future hold for this orthogonal, with almost no cross talk.” best part of his job and the basis of his bright, young scientist? Niko Hildebrandt, a professor at the research program. “My students, post- In Masson’s view, the answer to this Université Paris-Sud, agreed that Algar’s doctoral fellows, and I work together question depends on a combination of work has significantly advanced the un- on the research,” he said. “I think—and Algar’s goals and the opportunities that derstanding and application of QDs for hope—that such a team approach pro- arise. “Our careers are not predefined FRET. “In particular, his contributions vides a good training environment and and an event can easily trigger changes www.spectroscopyonline.com September 2017 Spectroscopy 32(9) 41 in our path, whether it is toward a new ticularly for students, is to try not to research field, a different institution, or be overwhelmed by the accomplish- being involved in administrative duties,” ments, knowledge, or experience of he said. “However, there is one variable people at later stages of their educa- that is sure: Algar has a tremendous po- tion or career. “You can’t instantly tential to contribute at all levels in our become those people, but each day community.” you can try to learn something new, See Others felt more confident in pre- improve in some way, or move some- dicting some future outcomes for thing forward (even a little bit), and Algar. “Without a doubt, he is already eventually end up at the same point a major player in the discipline of an- as them,” he said. things alytical chemistry and bioanalytical It’s obvious that Algar lives by that chemistry,” said Boudreau. “I see him advice. becoming an influential leader and clearer a role model in Canadian academia, More About Algar not only in research and bioanalysis An in-depth interview with Russ Algar with Cobolt but also for education and public out- focused on his research, challenges, reach.” and accomplishments will be pub- Krull agrees that Algar will con- lished in the October 3, 2017, edition tinue to be recognized as a creative of our newsletter, The Wavelength. leader, specifically in nanoparticle technologies. “I anticipate that he will 2018 Emerging Leader in help move these technologies from re- Molecular Spectroscopy: search to the bedside for medical and Now Accepting Nominations clinical applications, he said. For information about how to nominate someone for the 2018 award, please see High Performance Lasers Conclusion the call for nominations on our website: From his work with QDs and cFRET www.spectroscopyonline.com/emerg- Q Single frequency to the newer research with Pdots and ing-leader-molecular. Q UV-VIS-MIR LLCs, Algar seems determined to Q CW & Q-switched achieve innovative goals that the sci- References Q Unprecedented reliability entific and medical communities have (1) W.R. Algar, M. Massey, and U.J. Krull, J. with HTCure™ been striving to accomplish for a long Chem. Educ. 86, 6871 (2009). time. He is quick to point out, how- (2) J.J. Wang, J.R.R. Nunez, E.J. Maxwell, and ever, that he does not do it alone, say- W.R. Algar, J. Chem. Educ. 93, 166–171 ing that he shares his success with his (2016). spouse, Melissa, who is also a scien- (3) W.R. Algar, C.A.G. De Jong, E.J. Maxwell, tist, as well as with his students, post- and C.G. Atkins, J. Chem. Educ. 93, 162–165 doctoral fellows, and collaborators. (2016). We asked Algar what advice he (4) W.R. Algar, D. Wegner, A.L. Huston, J.B. would give to someone that wanted to Blanco-Canosa, M.H. Stewart, A. Armstrong, follow in his path. He broke it down to P.E. Dawson, N. Hildebrandt, and I.L. Med- three key points: First, he pointed to intz, J. Am. Chem. Soc. 134, 1876−1891 work ethic, citing the Thomas Edison (2012). quote, “Genius is 1% inspiration and 99% perspiration.” Megan L’Heureux is the managing “I think I got ahead by outworking editor of Spectroscopy and LCGC North people rather than by outsmarting America magazines in Iselin, New Jersey. them,” said Algar. Direct correspondence to: Second, he said it is important to [email protected] ◾ learn to be confident in the face of criticism. “It’s not easy, but I think the path to success is learning from, embracing, and even asking for critical— For more information on this topic, and hopefully constructive—feedback please visit our homepage at: rather than avoiding it,” he said. www.spectroscopyonline.com cobolt.se His third recommendation, par- 42 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com

In Situ Raman Spectroscopy Monitoring of the Reaction of Sulfur Trioxide with Polyethylene Fibers in Chlorinated Solvents

Sulfur trioxide (SO3) is a highly reactive oxidant, and its reactions with polyethylene fibers in chlorinated solvents can be used to functionalize and stabilize the polyethylene fibers for subsequent carbonization to produce carbon fiber. In this study, the apparent reaction kinetics between SO3 and polyethylene were investigated in various halogenated solvents using in situ Raman spectroscopy with an immersion Raman probe. This work demonstrates the power of in situ Raman spectroscopy to monitor hazardous reactions, and the results show that both solvent and polyethylene properties have a large influence on the reaction kinetics.

Xiaoyun Chen, Jasson Patton, Bryan Barton, Jui-Ching Lin, Michael Behr, and Zenon Lysenko

arbon fiber is being used in an increasingly wider sis, because SO3 readily reacts with water to form sulfuric range of applications, such as in light structural acid (H2SO4), and traditional titration methods cannot materials and polymer composites (1–5). Polyeth- differentiate between SO and H SO . Raman spectros- C 3 2 4 ylene (PE) has been suggested to be a lower-cost pre- copy was found to be a powerful tool that allowed in situ cursor material to carbon fibers than the traditional monitoring of key reaction species in the solution phase precursor polyacrylonitrile (PAN). Polyethylene fibers such as SO3 and sulfur dioxide (SO2). can be converted to carbon fibers by a stabilization Raman spectroscopy has previously been applied step (by sulfonation) followed by a carbonization step to study the molecular structure and oligomerization

(1,6–10). Sulfur trioxide (SO3) in halogenated solution of SO3 (12,13) and SO2 (14–16) Furthermore, a Raman was proposed to be an effective way for the stabilization method has been developed to determine the concen- and sulfonation of polyethylene (11). tration of SO2 and SO3 in the gas phase. This article The main hypothesized sulfonation reaction is shown describes a calibration method that was first developed in Figure 1a. Various side reactions can also occur as to quantify SO3 and SO2 in halogenated solvents based shown in Figures 1b and 1c, and will sequester SO3 that on a classical least squares (CLS) fit of Raman spectra. is then unavailable for the main sulfonation reaction (Fig- This calibration was then applied to monitor the SO3 ure 1a). Knowledge of the SO3 solution concentration is of reaction with polyethylene fibers with a goal to under- vital importance because it affects the reaction rate and stand the influence of various experimental variables process economics. However, the high reactivity of SO3 such as the polyethylene density, branching, and crystal- makes this reaction difficult to monitor by off-line analy- linity (three properties interrelated to each other), and www.spectroscopyonline.com September 2017 Spectroscopy 32(9) 43 the reaction-solvent type. Scanning electron microscopy and X-ray energy-dispersive spectroscopy (SEM- (a) -CH -CH - + 2 SO -CH=CH- + SO + H SO XEDS) was used as a complementary 2 2 3 2 2 4 tool to directly obtain the radial sulfur distribution from thin sections of polyethylene fibers as a function (b) of sulfonation time. xy xy O O SO Experimental O SOSO Reagents and Materials OH n H Polyethylene fibers were prepared OH O from three types of Dow proprietary polyethylene resins with dif- O O O O O ferent crystallinity and density. (c) HO S OH HO SSO OH HO SSO OH These three fibers will be referred n to by their densities, with PE-950, O O O O O PE-926, and PE-917 having a density of 0.950, 0.926, and 0.917 g/mL, respectively. Polyethylene fibers were Figure 1: Sulfonation reactions: (a) the hypothesized main reaction, and (b, c) potential side reactions. melt spun using a Hills Inc. com- mercial melt spinning line. The re- geometry. The signal was dispersed Results and Discussion sulting fibers were ~20 μm in diam- by a grating and projected onto a The polyethylene fiber surface eter with a tenacity of ~2 g/denier. charge-coupled device (CCD) cam- turned brown and then black dur-

SO3 was received from Aldrich in era for spectral collection. The body ing its reaction with SO3, and the 40-g lots in solid form in glass bot- of the immersion optic was made of solution remained clear. The sur- tles. To be able to transfer the SO3 Hastelloy C alloy and the tip of it face color change is believed to be in a safe fashion, the bottles were was sealed with a sapphire window. mostly due to the formation of the stored in a low-temperature oven in No corrosion was observed even conjugated double bond system a vented hood at 37–39 °C with the after multiday experiments. Typical shown in Figure 1a. Attempts to col- appropriate high-temperature cut- laser power ranged from 100 mW lect Raman spectra from the fiber off safety features (monomeric SO3 to 400 mW, and typical spectral surface were made but strong fluo- boils at 45 °C). When SO3 is in its acquisition conditions were 1-s ex- rescence overwhelmed the Raman liquid form, it is easily manipulated posure time with 10 accumulations signal. As a result, all the Raman in a well-vented hood (significant and with the cosmic ray filter on spectra were collected from the so- fuming is always evident) for dis- (20 s per spectrum). In some reac- lution-phase composition change to solution into inert solvents. SO3 is tions, strong fluorescence was ob- monitor the reaction progress. The exceedingly hazardous: inhalation served and the exposure time was majority of the reactions were car- or contact with skin results imme- shortened correspondingly. All ex- ried out in DCE. Several other halo- diately in severe burns. periments were carried out at room genated solvents were tested. Some Several halogenated solvents were temperature (unless otherwise speci- of them showed a fast reaction with investigated in this study: 1,2-di- fied) with stirring inside a custom- SO3, and others showed negligible chloroethane (ClCH2CH2Cl, DCE), ized glass reactor. Care was taken to reaction. Several other inert sol- methylene chloride (CH2Cl2), chlo- ensure that the laser beam was not vents were also tested as reaction roform (CHCl3), carbon tetrachlo- focused onto a sulfonated fiber; oth- solvents. The calibration results ob- ride (CCl4), and 1,1,2,2-tetrachloro- erwise overwhelmingly strong fluo- tained in the DCE solution are dis- ethane (CHCl2CHCl2). All solvents rescence would result. cussed below. As a complementary were acquired from Aldrich and tool, SEM-XEDS was used to directly were used as received. Data Analysis measure the sulfur distribution on Spectral visualization was performed the cross sections of partially sulfo- Raman Spectroscopy using OMNIC 8.2 software (Thermo nated polyethylene fibers, and those A Kaiser RXN1 Raman spectrom- Fisher Scientific Inc.). Because of results are also discussed below. eter with 785-nm laser excitation spectral overlapping, quantitative was used with a fiber-optic probe. analysis was carried out using a CLS Calibration Experiments A Kaiser 0.25-in. diameter short- algorithm as implemented in MAT- A quantitative method was needed focus immersion optic was used to LAB. Details of the CLS method can to monitor the SO3 consumption and collect signals in the backscattering be found elsewhere (17–19). SO2 generation in the solution phase 44 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com

υ the 1 mode of SO3, and the strong -1 υ 1145 cm peak from the 1 mode of SO (16,20–22). DCE was the initial 2.5 2 solvent selected for this study be-

2 cause of its compatibility with SO3. Its Raman spectrum is consistent 1.5 with that previously reported (23) -1 1 and had a peak at 1054 cm that

SO3/DCE weight ratio -1 y x interfered with the SO 1066 cm 0.5 = 0.0552 + 0.0016 3 R2 = 1 peak, and also another peak at -1 0 1145 cm that interfered with the 0 5 10 15 20 25 SO3/DCE CLS response ratio -1 SO2 1145 cm peak. 1066 Because of the severe spectral interference from DCE, a CLS model was developed to deconvolute the 1145 spectra for quantitation of SO and SO3 3 SO2 concentrations. The CLS algo- Raman intensity (arbitrary) rithm uses a linear combination of

SO2 all the pure component spectra to reproduce a reaction spectrum by minimizing the sum of squares of DCE the differences (19,24). The CLS re- 200 400 600 800 1000 1200 1400 1600 sponse, also referred to as the weight, Raman shift (cm–1) is proportional to the concentration of the corresponding species. The spectral range between 1000 and Figure 2: Standard Raman spectra of DCE (purple), 3 wt% SO2 DCE solution (green), and 3.7 wt% SO3 1250 cm-1 was used for the CLS anal- DCE solution (red). The inset plot shows the calibration for SO3 in DCE; a linear correlation is found for ysis with the spectra of three pure the low concentration range between the SO3/DCE weight ratio and the CLS response ratio. components: DCE, and the SO3 and SO2 solution spectra after subtrac- tion of solvent contributions. For the

quantitation of SO3 and SO2, their CLS responses were normalized by t = 0 the DCE solvent CLS response. The normalized CLS response ratios t = 20 min were found to be linearly proportional to the weight ratio of SO3 and SO to DCE solvent within a certain t = 4 h 2 t = 4.5 days concentration range (<10 wt%). The reason for such a limited concentration range is that in addition to free

1040 1100 1160 monomer, SO3 may exist in forms such as dimer, trimer, tetramer, or polymers, and each form has different Raman features (13). An example Raman intensity (arbitrary) of such a correlation is shown in the

inset plot in Figure 2. When the SO3 concentration was sufficiently high, 400 600 800 1000 1200 1400 Raman shift (cm–1) a nonlinear correlation behavior was observed because of dimer and oligomer formation. Figure 3: Representative spectra from in situ Raman monitoring of the solution-phase reaction. Raman Monitoring of the during the polyethylene and SO3 in Figure 2. At such concentrations, SO3 Reaction with Polyethylene reaction. Standard spectra of pure SO3 and SO2 exist mostly in their The solution-phase SO3 and SO2 con- DCE solvent, 3 wt% SO2 in DCE, free monomer form as indicated centrations can be determined using -1 and 3.7 wt% SO3 in DCE are shown by the strong 1066 cm peak from the method described above. Raman www.spectroscopyonline.com September 2017 Spectroscopy 32(9) 45 spectra collected as a function of reaction time could thus be used to monitor the sulfonation reaction of 120 polyethylene fibers. Representative PE-917 reaction spectra from an example reaction are shown in Figure 3. In 100 this experiment, 106 mg of PE-926 fiber (0.025 g/m) was reacted with 80 PE-926

1.75 g of SO3 dissolved in 14.54 g of DCE. Upon contact, the fiber surface 60 quickly darkened in color (25), and the solution remained mostly color- 40 PE-950 less. The consumption of SO3 and Conversion (%) 20 generation of SO2 can be seen clearly in the inset spectra in Figure 3. The CLS model and calibration were ap- 0 plied to the reaction spectra so that 0 0.5 1 1.5 2 –20 Time (h) the concentrations of SO3 and SO2 could be calculated. Based on the main reaction shown in Figure 1a, the amount of polyethylene should Figure 4: SO2 formation kinetics observed for three types of polyethylene fibers. lead to the consumption of only approximately 35% of the starting SO3. curred. The observed concentration the overconsumption of SO3 readily The experimentally observed SO3 increase of SO2 was also greater than could be explained by the side reac- consumption was much higher, in- the 1.5 % concentration expected tions (Figures 1b and 1c), the reaction dicating that a significant consump- if only the main reaction shown in mechanism responsible for overgen- tion resulting from side reactions oc- Figure 1a was operable. Although eration of SO2 remains to be studied. 46 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com

similar diameter) in terms of SO2 production. It is obvious that the PE-950 fiber, with the highest crys-

100 tallinity, exhibited the slowest reac- 90 tion, and a reaction with an overall lower extent was noted when the 80 DCE+CH Cl 2 2 profiles plateaued. 70 Cl CHCHCl Similarly, the influence of sol- 60 2 2 vent type on the reaction kinetics 50

intensity DCE with CH Cl soaking was investigated. Only solvents that 3 40 2 2

SO showed negligible reactivity toward 30 DCE SO3 (see next section) were used. 20 Five reactions using the same type 10 CH Cl 2 2 of polyethylene fiber and a similar 0 0 0.5 1 1.5 2 2.5 starting amount of SO3 and poly- Time (h) ethylene were carried out in various inert solvents (see the next section). Figure 5 compares the percentage of

SO3 remaining (after normalizing Figure 5: Reaction kinetics observed for the same polyethylene fiber reacting with SO3 in various solvents. against the starting concentration). The type of solvent appeared to have a large influence on the reaction kinetics. The same type of meth- odology was used as outlined above

16 for the SO3–DCE mixture system PE-926 25 min PE-926 25 min to develop CLS models for the SO3 PE-950 25 min 14 PE-950 60 min quantitation for each solvent system (or using peak area if a solvent 12 spectrum did not interfere with the -1 SO3 1145 cm peak), but the details 10 are not shown here. In all five reac-

tions, the SO3 solution was added 8 into the reactor shortly after time 0. S/Cl ratio 6 The consumption rate appeared to be the slowest with DCE as the sol- 4 vent. When methylene chloride was used as the solvent, a significantly 2 faster reaction rate was observed. It was hypothesized that such a rate 0 increase might be a result of par- –15 –10 –5 0 5 10 15 tial swelling by methylene chloride Position (μm) (26,27). To test this hypothesis, another experiment was performed by pre-immersing polyethylene fibers

Figure 6: S/Cl profiles of PE-926 and PE-950 fibers after 25 and 60 min in a 1.58 M solution of SO3 in DCE. in methylene chloride and then immediately carrying out the reaction

With such a method, it became occurred. The conversion can then with SO3 in a DCE solution. Almost possible to gain a deeper under- be calculated based on the amount identical reaction kinetics were ob- standing of the influence on the re- of polyethylene, which was always served, disproving the swelling hy- action kinetics of various key pro- the limiting reagent (that is, the pothesis. In another experiment, a cess variables such as crystallinity, number of moles of SO3 is always mixture solvent with 1:1 (w/w) DCE fiber diameter, and solvent type. To more than twice that of -CH2- re- and methylene chloride was used; further aid the comparison of reac- peating units in polyethylene). Fig- the initial kinetics profile resides tion kinetics, the SO3 and SO2 con- ure 4 compares the reaction con- between that of the pure DCE and centrations were converted to extent version for three experiments with of pure methylene chloride. Among of conversion, assuming only the similar conditions but with three all the solvents tested, 1,1,2,2-tetra- main reaction shown in Figure 1a different polyethylene fibers (with chloroethane appeared to facilitate www.spectroscopyonline.com September 2017 Spectroscopy 32(9) 47 the reaction most effectively. It was found that solvent type and (16) Y. Song, Z. Liu, H.-K. Mao, R.J. Hemley, polyethylene crystallinity, density, and D.R. Herschbach, J. Chem. Phys. SEM-XEDS Characterization and branching have a substantial in- 122(17), 174511 (2005). of Sulfonated Fiber Cross-Sections fluence on the reaction kinetics. The (17) D.M. Haaland and R.G. Easterling, Appl. SEM-XEDS analysis was used to exact mechanism behind such in- Spectrosc. 36(6), 665–673 (1982). directly measure the radial sulfur fluences remains to be investigated (18) D.M. Haaland, R.G. Easterling, and D.A. distribution within the polyethyl- and is likely related to diffusion and Vopicka, Appl. Spectrosc. 39(1), 73–84 ene fiber as a way to monitor the mass transfer limitation. Comple- (1985). extent of sulfonation with time. mentary SEM-XEDS measurements (19) D.M. Haaland and R.G. Easterling, Appl. This information is less quantita- of sulfur distribution within the Spectrosc. 34(5), 539–548 (1980). tive but complementary to the solu- polyethylene fiber as a function of (20) R.A. Nyquist and R.O. Kagel, Handbook of tion concentration provided by the time were also obtained to further Infrared and Raman Spectra of Inorganic Raman results. Fibers withdrawn confirm these results. This study Compounds and Organic Salts: Infrared after various lengths of sulfonation clearly demonstrates the power of in Spectra of Inorganic Compounds (Aca- time were immediately placed into situ Raman spectroscopy to monitor demic Press, Cambridge, Massachusetts, a water bath to quench the reaction these hazardous reactions. 2012). and then were prepped for analysis. (21) E.T.H. Chrysostom, N. Vulpanovici, T. Ma- Figure 6 shows a comparison of References siello, J. Barber, J.W. Nibler, A. Weber, A. radial distributions of S/Cl ratio for (1) E. Frank, F. Hermanutz, and M.R. Buch- Maki, and T.A. Blake, J. Mol. Spectrosc. PE-926 and PE-950 fibers sulfonated meiser, Macromol. Mater. Eng. 297(6), 210(2): 233–239 (2001). for 25 and 60 min after immersion in 493–501 (2012). (22) A.J. Dorney, A.R. Hoy, and I.M. Mills, J. a 1.58 M solution of SO3 in DCE. The (2) E. Thostenson, W. Li, D. Wang, Z. Ren, and Mol. Spectrosc. 45(2), 253–260 (1973). slower rate of sulfur concentration T. Chou, J. Appl. Phys. 91(9), 6034–6037 (23) S. Mizushima, T. Shimanouchi, I. Harada, increase in the PE-950 fiber relative (2002). Y. Abe, and H. Takeuchi, Can. J. Phys. to the PE-926 fiber is clearly shown, (3) W. Jonda, “Lightweight structural part 53(19), 2085–2094 (1975). suggesting that the sulfonation reac- formed of carbon fiber-reinforced plas- (24) K.R. Beebe, R.J. Pell, and M.B. Seasholtz, tions for PE-950 are slower. In fact, tic” (Google Patents, 1978). Chemometrics: A Practical Guide (Wiley- the overall magnitude of sulfur in (4) M. Tavakkolizadeh and H. Saadatmanesh, Interscience, Secaucus, New Jersey, the PE-926 fiber after only 25 min J. Struct. Eng. 129(1), 30–40 (2003). 1998). was greater than that observed in (5) D. Chung, Carbon Fiber Composites (But- (25) D. Fischer and H.H. Eysel, J. Appl. Polym. the PE-950 fiber after 60 min of terworth-Heinemann, Oxford, UK, 2012). Sci. 52(4), 545–548 (1994). sulfonation. These observations are (6) J.M. Younker, T. Saito, M.A. Hunt, A.K. (26) A. Peterlin, J.L. Williams, and V. Stan- consistent with the differences in Naskar, and A. Beste, J. Am. Chem. Soc. nett, J. Polym. Sci. A Polym. Phys. 5(5), reaction rates observed via Raman 135(16), 6130–6141 (2013). 957–972 (1967). spectroscopy as shown in Figure 4. (7) J. Ihata, J. Polym. Sci., Part A: Polym. Chem. (27) J.L. Williams and A. Peterlin, J. Polym. Sci. These results clearly demonstrated 26(1), 167–176 (1988). A Polym. Phys. 9(8), 1483–1494 (1971). that the reaction kinetics between (8) A.R. Postema, H. De Groot, and A.J. Pen-

SO3 and polyethylene are sensitive to nings, J. Mater. Sci. 25(10), 4216–4222 the polyethylene material’s density, (1990). branching, and crystallinity, which (9) X. Huang, Materials 2(4), 2369 (2009). Xiaoyun Chen, Jui-Ching are three properties closely related (10) S. Horikiri, J. Iseki, and M. Minobe, “Pro- Lin, and Michael Behr are to each other. cess for production of carbon fiber,” Pat- with Analytical Sciences at The Dow ent US4070446 A, 1978. Chemical Company in Midland, Conclusion (11) J.T. Patton, B.E. Barton, M.T. Bernius, X. Michigan. Jasson Patton, Bryan In situ Raman spectroscopy was Chen, E.J. Hukkanen, C.A. Rhoton, and Z. Barton, and Zenon Lysenko used to monitor the SO3 consump- Lysenko, “Processes for preparing carbon work in the organic polymer and tion and SO2 generation in DCE fibers using sulfur trioxide in a haloge- organometallic group at The Dow and other chlorinated solvents dur- nated solvent,” Patent WO2014011462 Chemical Company. ing polyethylene and SO3 reactions. A1, 2013. Direct correspondence to: CLS analysis was used to quantify (12) S.-Y. Tang and C.W. Brown, J. Raman [email protected] ◾ the spectral contribution of SO3 and Spectrosc. 3(4), 387–390 (1975). SO2 in the presence of strong solvent (13) R.J. Gillespie and E.A. Robinson, Can. J. interference. The reaction kinetics Chem. 40(4), 658–674 (1962). data could thus be obtained from (14) R.G. Dickinson and S.S. West, Phys. Rev. For more information on this topic, the solution-phase measurements 35(9), 1126–1127 (1930). please visit our homepage at: when different types of polyethylene (15) H. Gerding and J.W. Ypenburg, Recl. Trav. www.spectroscopyonline.com reacted with SO3 in various solvents. Chim. Pays-Bas 86(4), 458–462 (1967). 48 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com PRODUCTS & RESOURCES Sensor coatings I CP argon line filter Acton Optics’ Metachrome and The washable Guard- Unichrome coatings are designed ian In-Line Non-Return to extend the detection capabilities gas filter from Glass of charge-coupled devices, charge- Expansion is designed injection devices, and complementary to safeguard ICP nebu- metal oxide semiconductor sensors lizers by preventing into the ultraviolet portion of the clogging from stray spectrum. According to the company, particles in the argon lines. According to the company, the filter has a each coating can be deposited on built-in one-way valve that prevents liquid from siphoning back into the most bare chips or packaged devices, instrument gas box. provided technicians can access the Glass Expansion, sensor. Applications include OES, ICP Pocasset, MA; spectroscopy, and LIBS. www.geicp.com/Guardian Princeton Instruments, Acton, MA; www.actonoptics.com

Benchtop fluorometer I CP and IC calibration standards HORIBA’s Aqualog benchtop Stock ICP and IC calibration stan- fluorometer is designed dards from Inorganic Ventures are with the company’s A-TEEM available in 30-mL volumes. Accord- technology for fingerprinting ing to the company, its small product molecules with high specific- bottles reduce chemical waste, save ity and ultrahigh-sensitivity to on shipping costs as non-hazardous identify, quantify, and under- volumes, and are manufactured stand dynamics of molecules under its ISO 17025 and ISO Guide in mixtures, and identify 34 accreditation. fluorescing and absorbing Inorganic Ventures, molecular states. According to Christiansburg, VA; the company, the spectrometer includes an ultrafast CCD and uses absor- www.inorganicventures.com bance, transmittance, and excitation–emission matrix data to fingerprint molecules in seconds. HORIBA Scientific, Edison, NJ; www.horiba.com

Raman probe head I CP-OES system The RamanRxn probe head PerkinElmer’s Avio 500 ICP- from Kaiser Optical Systems OES system is designed with is designed with a fixed-focus simultaneous background cor- optical design intended to rection and is intended for low provide long-term measure- and high concentration testing. ment stability and high According to the company, the signal-to-background ratios system’s applications include and is compatible with a environmental, petrochemical range of optics. According (lubricants and used oils), geo- to the company, noncontact chemical, food, pharmaceutical, optics with working distances and industrial (including batter- of 0.3–17 in. can be used to measure through windows or bottles. In situ ies) markets. immersion optics for reaction vessels, laboratory reactors, and process PerkinElmer, Waltham, MA; streams reportedly are also available. www.perkinelmer.com Kaiser Optical Systems, Ann Arbor, MI; www.kosi.com

Autosampler SEM Raman system The YZ autosampler from PIKE Renishaw’s scanning electron Technologies is designed for microscope (SEM) system sample transmission measure- is designed for the acquisi- ments where precise mapping tion of both SEM and Raman is required. According to the data from the same area on company, the autosampler’s a sample. According to the Y and Z plane movement (no company, users do not have rotation) makes it suitable for to transfer the sample to a dif- spectroscopic measurements ferent measurement location of samples having a spectral or instrument, ensuring rapid signature influenced by orientation, such as a polarized film. correlative analysis. PIKE Technologies, Renishaw, Madison, WI; Hoffman Estates, IL; www.piketech.com www.renishaw.com www.spectroscopyonline.com September 2017 Spectroscopy 32(9) 49

FT-IR microscope Portable Raman spectrometer Thermo Fisher Scientific’s Nicolet B&W Tek’s iRaman Pro iN5 FT-IR microscope is designed ST portable Raman for particulate identification in spectrometer is designed the laboratory. According to the to measure through company, the microscope has an a variety of materials. optical setup that allows users According to the company, to simultaneously examine a the spectrometer can sample and collect chemical measure through clear or information, and it is suitable semitransparent containers, for applications in food safety, white and red plastics, pill manufacturing, and academia. coatings, yellow and manila colored paper, white packaging envelopes, Thermo Fisher Scientific, and glass. Madison, WI; B&W Tek, www.thermofisher.com/in5 Newark, DE; http://bwtek.com/products/i-raman-pro-st/

785-nm lasers Multispectral imaging camera Compact, narrow linewidth The PIXELTEQ brand SpectroCam VIS- 785-nm frequency-stabilized SWIR 640 multispectral imaging camera diode lasers from Cobolt from Ocean Optics is designed for short- are designed with an output wavelength infrared imaging in tissue power of up to 500 mW, and and biologic material analysis for medical include an integrated optical and forensics applications. According to isolator. According to the com- the company, the device has an InGaAs pany, integrated electronics sensor for increased sensitivity, full allow for easy incorporation frame 640 x 512 pixel resolution and into high-end Raman-based 15-μm pixel pitch, and can produce live systems or other analytical processed images of six spectral bands instrumentation. at a rate of up to 25 frames per second. Cobolt AB, Ocean Optics, Stolna, Sweden; Dunedin, FL; www.cobolt.se www.oceanoptics.com

ULF Raman filters Benchtop XRD system BragGrate Raman filters from Rigaku’s MiniFlex benchtop X-ray dif- OptiGrate are designed to fraction system is designed for phase enable access to Stokes and identification and quantification, and the anti-Stokes Raman bands in the determination of percent (%) crystal- ultralow terahertz frequency linity, crystallite size and strain, lattice range down to 5 cm-1. parameter refinement, Rietveld refine- According to the company, laser ment, and molecular structure. Accord- line cleaning and light rejection ing to the company, the system can be notch filters are provided, used in research in material science and and the filter production line chemistry, and in industry for research is extended to cover many and quality control. standard and custom laser wavelengths from 405 nm to 1550 nm. Rigaku Corporation, OptiGrate, Tokyo, Japan; Oviedo, FL; www.rigku.com www.optigrate.com

Liquid flow cells Deep-UV certified reference material Spectral Systems’ Super-Sealed A certified reference material liquid flow cells are designed and (CRM) from Starna Scientific manufactured for continuous flow is designed to qualify a UV infrared transmission sampling spectrophotometer in the applications. According to the deep UV. According to the company, the large port diameter company, the CRM can provide permits higher flow rates to reliable qualification data speed feedback of component below 200 nm, and a new concentration. reference material (TS8) was Spectral Systems, developed with suitable spectral Hopewell Junction, NY; characteristics in the region of 190–230 nm. www.spectral-systems.com Starna Cells, Inc., Atascadero, CA; www.starna.com 50 Spectroscopy 32(9) September 2017 www.spectroscopyonline.com ®

Low Frequency Raman Filters Stokes and anti-Stokes modes at 5 cm-1 and up

Available wavelengths 400-2500 nm 405, 442, 458, 473, 488, 491, 514, 532, 552, 561, 568, 588, 594, 632, 660, 785, 830, 980, 1064, 1550

Ad Index ADVERTISER PG# ADVERTISER PG# Acton Optics & Coatings 28 New Era Enterprises, Inc. 50

Amptek 3 Ocean Optics, Inc. CV2, 58

B&W Tek, Inc. 27 Ondax 17

CASSS 18 OptiGrate Corporation 50

CEM Corporation CV TIP, 53 PerkinElmer Corporation 11, CV4

Cobolt AB 41, 57 PIKE Technologies 13, 59

Eastern Analytical Symposium 22 Princeton Instruments 5

Electro-Optics Technology 6 Renishaw, Inc. 15

Glass Expansion 54, 55 Rigaku GMG 10, 56

HORIBA Scientific 7 Spectral Systems 6, 60

Inorganic Ventures IV Labs 9 Starna Cells, Inc. 4, 22

Kaiser Optical Systems, Inc. 45 StellarNet, Inc. 21

Materials Research Society 19 TechnoSpex Pte Ltd. 36

Mightex Systems 4 Thermo Fisher Scientific 33, 35, 61

Milestone, Inc. Insert Wasatch Photonics, Inc. 29

MKS Instruments, Inc. 23 WITec GmbH 39, 62 Supplement to

September 2017 www.spectroscopyonline.com

THE APPLICATION NOTEBOOK 52 Table of Contents THE APPLICATION NOTEBOOK – SEPTEMBER 2017 THE APPLICATION NOTEBOOK Atomic Spectroscopy 53 Digestion of Difficult APIs in Accordance with USP <233> Bob Lockerman, CEM Corporation

54 Considerations When Analyzing Real-World Samples by ICP Spectrometry Ryan Brennan and Jerry Dulude, Glass Expansion

56 Total Mercury in Coal Using Direct Mercury Analysis Nippon Instruments Corporation

Molecular Spectroscopy 57 Blink and You Miss It—Catching Fleeting Catalytic Intermediates by High-Speed 785 nm NIR Raman Spectroscopy Duenpen Unjaroen*, Wesley R. Browne*, and Elizabeth Illy†, *University Groningen and †Cobolt AB

58 SERS Methods Detect Trace Levels of Cocaine, Heroin, Methamphetamine, and THC Anne-Marie Dowgiallo, PhD, Ocean Optics

59 Surface-Sensitive Spectroelectrochemistry with ATR-SEIRAS Tyler A. Morhart and Ian J. Burgess, Department of Chemistry, University of Saskatchewan and PIKE Technologies

60 Diamond Anti-Reflection Coating Spectral Systems LLC

61 Qualifying ATR Accessories for Pharmaceutical Applications Steve Lowry, Thermo Fisher Scientific

62 RISE™ Raman Imaging and Scanning Electron Microscopy Investigation of Twisted Bilayer Graphene WITec GmbH

Also 63 Call for Application Notes THE APPLICATION NOTEBOOK – SEPTEMBER 2017 Atomic Spectroscopy 53

Digestion of Difﬁ cult APIs in Accordance with USP <233> Bob Lockerman, CEM Corporation

The new USP Chapters <232> and <233> will become ofﬁ cial on January 1, 2018 and bring signiﬁ cant changes in sample preparation and analysis of pharmaceutical samples both in excipients and APIs. Certain APIs pose challenges to acid digestion because they are very stable compounds and not easily broken down.

SP methods <232> and <233> call for total digestion of Upharmaceutical samples and quantifi cation of individual elements. Many pharmaceutical materials can be easily digested but APIs with multiple aromatic ring structures can be very diffi cult to completely break down and obtain a clear digestion, as prescribed in the new chapters. Microwave digestion can provide a solution to many of these compounds. Th is application note will focus on the use of the CEM MARS Figure 1: Clear solutions after digestion and dilution. Note the 6 microwave digestion system and new high performance iPrep yellow color of the third sample is due to the palladium catalyst vessels to completely digest very diffi cult APIs. Th e ring structures in the TrixiPhos Sample. of the APIs are shown to illustrate the high stability of these compounds. Th e sample sizes listed represent the maximum allowable iPrep is a simple to use three-piece vessel which uses a simple in order to achieve a complete digest. hand torque device to seal. Th e patented dual-seal design provides for higher temperatures and fi ne control of the vent and reseal Experimental Conditions process necessary for these sample types. Th ree APIs were prepared using a CEM MARS 6 microwave Each API sample was weighed and added to an iPrep liner with digestion system equipped with iWave technology and iPrep vessels. 9 mL HNO3 and 1 mL HCl. Th e vessels were capped, assembled, iWave is a novel Light Emitting Technology™ that measures the and placed in the MARS 6 for digestion. Digestion parameters temperature of the actual sample solution inside the vessel and does are recorded in Table I. not require an internal probe. Results Th e MARS 6 with iPrep vessels was able to completely digest each Table I: API with structure and approximate sample weight of the API materials and provide a clear solution. Of note is that Ramp Hold the TrixiPhos digestate is a bright yellow color. Th is is a result of Max Sample Tempera- Sample Name Structure Time Time Weight (mg) ture (°C) the Pd catalyst used when synthesizing the compound. Each API (min) (min) sample was run in duplicate in order to confi rm the success of the sample preparation. All samples were completely digested as shown in Figure 1. TrixiPhos Pdt-Bu P t-Bu 100 25 30 250 O

SPdNHO 2 O Conclusions OH Th e MARS 6 system with iPrep vessels is an ideal option for Sudan OrangeN 500 25 30 250 N working with these diﬃ cult pharmaceutical materials. OH CEM Corporation 4-Fluorophenyldi- O F S + P.O. Box 200, Matthews, NC 28106 phenylsulfonium F3C S O 250 25 30 250 triﬂ ate O tel. (704) 821-7015, (800) 726-3331 Website: www.cem.com 54 Atomic Spectroscopy THE APPLICATION NOTEBOOK – SEPTEMBER 2017

Considerations When Analyzing Real-World Samples by ICP Spectrometry Ryan Brennan and Jerry Dulude, Glass Expansion

Dealing with a sample matrix high in total dissolved solids (TDS) is a common challenge for ICP spectrometry. 120% Most “real samples” contain considerable concentrations 100% of TDS including soils, sludge, waste water, brines, high 80% acid digests, and fusions. The challenges encountered 60% 40% include interrupted runs, signal drift, clogged nebulizers, 20% shortened torch life, and greater interferences. To Nebulizer gas ﬂow (%) 0% overcome these challenges, it is essential you select 01234567891011 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Time (minutes) a proper sample introduction system so that you can Nebulizer without elegra Nebulizer with elegra achieve the best performance your ICP-OES or ICP-MS can provide. In this article, a complete sample introduction Figure 1: Elegra nebulizer high TDS stress test (25% NaCl). package is presented to improve ICP performance in a high TDS sample matrix.

6 Hours of running 10% NaCl n argon humidifi er is essential and one of the most important A accessories for ICP-OES and ICP-MS analyses involving samples with high TDS. Th e added moisture in the nebulizer gas helps to reduce salt deposits in the nebulizer and torch injector, allowing uninterrupted operation. In addition to the argon humidifi er, a large bore injector is recommended, typically, greater than 1.5 mm ID for Quartz outer tube radial view ICP-OES instruments and greater than 2.0 mm ID for axial view ICP-OES and ICP-MS instruments. Due to the rapid devitrifi cation of quartz in the presence of high TDS, a demountable ICP torch is preferred to reduce replacement costs. A nebulizer capable of handling high TDS should be chosen and combined with a baffl ed Ceramic outer tube cyclonic spray chamber to eliminate large droplets from reaching the torch. Below we will describe in detail the characteristics of each of Figure 2: A comparison of resistance to devitrifi cation when exposed to high salt matrix. these essential high TDS components.

Argon Humidifi er As expected, without the Elegra, the nebulizer was completely clogged Th e ElegraTM argon humidifi er, released in February 2016 (1), utilizes after only 5 min. In contrast, the same nebulizer with the Elegra, held highly effi cient membrane technology to add moisture to the argon as a relatively constant gas fl ow throughout the entire test (over 30 min). it fl ows through the inert metal-free construction. Th e device runs at A similar result was also observed with the injector (1). atmospheric pressure and does not require heating or electric power, as required by other manufacturers. Th e Elegra is also confi gured with a Torch Selection bypass switch so that the operator can turn humidifi cation on and off Th e combination of high temperature and salt deposits causes a quartz without connecting or disconnecting any tubing. Th e elegant design torch outer tube to devitrify, resulting in frequent replacement of the results in a very compact accessory that is simple to install. entire torch. Th e D-TorchTM off ers an aff ordable, robust demountable A nebulizer stress test was completed aspirating a 25% NaCl ICP torch design which greatly reduces consumable costs. With the solution while monitoring the nebulizer gas fl ow with and without D-Torch the outer tube can be replaced when worn and the optimum the Elegra (Figure 1). A decrease in gas fl ow is a good indicator that a injector can be selected for each application. Th e D-Torch also nebulizer is clogging. Note a nebulizer with a TDS tolerance rating of incorporates a ceramic intermediate tube for greater robustness. only 5% was used to illustrate the advantage of the Elegra. As a worst- Unique to the D-Torch is an optional ceramic outer tube, case test, the salt solution was aspirated continuously, with no rinsing. which does not suff er from devitrifi cation. In Figure 2, a quartz THE APPLICATION NOTEBOOK – SEPTEMBER 2017 Atomic Spectroscopy 55

and aspirated at 1 mL/min for 9 h with no rinsing. Th e combination of the SeaSpray DC nebulizer, Twister spray chamber, ce- Elegra - seaspray DC - twister - D-Torch 1.10 ramic D-Torch, and Elegra argon humidifi er provided exception- 1.08 1.06 al stability (Figure 3). A measurement was taken approximately 1.04 every 3 min over a period of 9 h, while a precision of less than 1% 1.02 1.0 was maintained throughout the experiment. 0.98 Wavelength Precision (% RSD) 0.96 Zn 213 0.8 0.94 Mn 257 0.6 Conclusions 0.92 Mg 280 0.6 Normalized intensity 0.9 An easy way to eliminate drift, interrupted runs, and frequent 19 78 153 228 303 378 453 528 maintenance to your ICP sample introduction system is to add Time (minutes) Mg 280.270 Mn 257.610 Zn 213.857 the Elegra argon humidifi er. Th e D-Torch off ers an aff ordable, robust ICP torch design while greatly reducing torch consumable Figure 3: Long term stability test, 9 h analysis of 1 ppm standard costs. Th e optional ceramic outer tube for the D-Torch provides a in 3.5% NaCl. signifi cant improvement in torch life and stability in the presence of high TDS. Combining the SeaSpray nebulizer with the Twister outer tube is compared to a ceramic outer tube, both of which spray chamber will provide optimum sensitivity and exceptional have been exposed to 6 h of running 10% NaCl. You can clearly long-term precision even in the presence of high TDS. see that there is no change to the integrity of the ceramic outer tube, whereas the quartz is severely devitrifi ed. References A detailed performance evaluation of the fully ceramic D-Torch (1) Glass Expansion February 2016 Newsletter, “The Elegra Argon Humidifier: was presented in a 2010 application note (2). In this report, Uninterrupted and Maintenance-Free ICP Operation.” the D-Torch exhibited exceptional stability in the presence of (2) Thermo Fisher Scientific Technical Note #43053, “Radial Demountable 3% NaCl and provided equivalent analytical performance to a Ceramic Torch for the Thermo Scientific iCAP 6000 Series ICP standard quartz torch. Spectrometer” (2010). (3) Glass Expansion October 2014 Newsletter, “ICP Spray Chamber Update.” Spray Chamber Selection Glass Expansion spray chambers are specifi cally designed to provide high performance and fast washout (3). Th e TwisterTM cyclonic spray chamber is the ideal choice for high TDS applications due to its central transfer tube (baffl e). Th e baffl e acts as a secondary droplet fi lter to reduce droplet size and plasma loading without compromising detection limits yet improving precision (%RSD). Th e reduced sample load helps to increase torch life, slow salt build-up at the injector tip, and decrease the frequency of cleaning ICP-MS interface cones.

Nebulizer Selection Th e SeaSprayTM nebulizer comprises a unique self-washing tip designed to prevent crystal growth, providing a tolerance of up to 20% TDS. Th e SeaSpray’s popularity is due to its ability to oﬀ er outstanding nebulization eﬃ ciency for trace level analyses Glass Expansion Inc. combined with excellent precision and a high tolerance to TDS. 4 Barlows Landing Rd, Unit 2A, Pocasset, MA 02559 To evaluate the performance of the SeaSpray DC nebulizer a 1 tel. (508) 563-1800, fax (508) 563-1802 ppm multielement standard was prepared in a 3.5% NaCl matrix Website: www.geicp.com 56 Atomic Spectroscopy THE APPLICATION NOTEBOOK – SEPTEMBER 2017

Total Mercury in Coal Using Direct Mercury Analysis Nippon Instruments Corporation

This application note demonstrates the ability of the Experimental Conditions NIC MA-3000 direct mercury analyzer to analyze coal Coal samples are ground with a mortar to avoid segregation eff ect. samples with accuracy and precision, in compliance Calibration is done using certified aqueous ionic-mercury standard with ASTM D-6722-01 (Standard Test Method for solution diluted to the required concentration. Th e least-squares Total Mercury in Coal and Coal Combustion Residues regression method is used to create and complete the calibration curve. by Direct Combustion Analysis). Analysis was performed using the NIC MA-3000 dedicated direct mercury analyzer that selectively measures total mercury by thermal ince mercury occurs naturally in coal and other fossil fuels, when decomposition, gold amalgamation, and cold vapor atomic absorption Sthese fuels are burned for energy, the mercury becomes volatilized spectroscopy. Th e MA-3000 analyzer measures mercury in virtually and airborne into the atmosphere. In the United States, power any sample matrix—solid, liquid, and gas—and off ers quick results plants that burn coal to create electricity account for about half of all without time-consuming or elaborate sample preparation processes. man-made mercury emissions. In nature, elemental mercury can go through a series of chemical transformations that convert it to highly Results toxic forms that are concentrated in fish and birds. Method Th e most toxic form of mercury is methylmercury, an organic form created by a complex bacterial conversion of inorganic mercury. Sample Conditions Mercury from the food chain is known to bioaccumulate in humans, Atomize1: 150 °C, 1 min as bioaccumulation in fish and birds carries over into human STD solution Atomize2 – , – Atomize3: 800 °C, 2 min populations, where it can result in mercury poisoning. Mercury is Atomize1: – , – dangerous to both natural ecosystems and humans because it is highly Coal Atomize2: 180 °C, 2 min toxic, especially because of its ability to damage the central nervous Atomize3: 850 °C, 2 min system. Mercury poses a particular threat to human development in Result utero and in early childhood. Th us, to prevent mercury poisoning, STD addition Sample Sample size (mg) N Conc. (μg/kg) CV (%) it is necessary to accurately quantify total mercury in coal so that Recovery (%) mercury emissions into the atmosphere may be carefully controlled. Sample 42–44 3 31.9 7.7 103

Conclusion Th is report demonstrates that NIC MA-3000 is able to reproduce good STD addition recovery of coal. NIC MA-3000 analyzes coal samples with accuracy and precision. It is a perfect solution to today’s increasing laboratory demand for easy, fast, and accurate mercury measurements.

References (1) WHO HP: http://www.who.int/ipcs/assessment/public_health/mercury/en/. (2) US EPA: https://www.epa.gov/air-emissions-inventories/2011-national- emissions-inventory-nei-technical-support-document. (3) USGS: https://pubs.usgs.gov/fs/fs095-01/fs095-01.html.

Nippon Instruments Corporation Figure 1: NIC MA-3000 Direct thermal decomposition mercury 14-8, Akaoji-cho, Takatsuki-shi, Osaka. 569-1149 analyzer tel. +81-72-694-5195, fax 81-72-694-0063 Website: www.hg-nic.com THE APPLICATION NOTEBOOK – SEPTEMBER 2017 Molecular Spectroscopy 57

Blink and You Miss It—Catching Fleeting Catalytic Intermediates by High-Speed 785 nm NIR Raman Spectroscopy Duenpen Unjaroen*, Wesley R. Browne*, and Elizabeth Illy†, *Univ. Groningen and †Cobolt AB

Raman Scattering

It is only in the last two decades that Raman spectroscopy has begun to realize its potential as an almost uni- versally applicable analytical technique from materials and life sciences applications to point of care analysis. This is primarily thanks to the availability of compact laser sources, high sensitivity cameras and high resolution compact spectrometers. In this application note the use of a stable high power NIR 785 nm laser is shown to be critical in achieving high time resolution and high signal to noise ratios while allowing for both fast and slow Figure 2: Typical spectrum of the Cobolt 08-NLD laser (FWHM processes to be captured simultaneously. <40 pm), output power 500 mW.

Catching Transient Species 100 ms per spectrum can easily be achieved.

The weakness of detectors in the NIR region (800–1050 nm) with As an example, a oxidation reaction in which H2O2 is added to quantum effi ciency dropping off over this range from ca. 40% to a solution containing a precatalyst is chosen. In this reaction only 0% is a key drawback of 785 nm Raman spectroscopy. Typically 0.25 mM precatalyst is present (far below that which can be detected long exposure times and spectral averaging are required to achieve a by Raman spectroscopy but its absorption at 785 nm means that good signal to noise level, however this approach seriously reduces its spectrum is resonantly enhanced) and H2O2 is added to give a the time resolution achieved (to minutes) and means that weak fi nal concentration of 250 mM which decreases slowly over time. signals for compounds that are present for only a few 100s of mil- Within 1 s of addition of H2O2, however, the precatalyst becomes liseconds cannot be observed. Using more effi cient NIR detectors activated with an intermediate that is present only with a maximum (such as back thinned CCDs) does not usually solve this problem concentration of 0.05 mM for 0.5 s and has an absorption band near due to etaloning that can be diffi cult to suppress. Th is drawback can 785 nm which results in enhancement of its Raman spectrum by ca. be circumvented by a combination of effi cient collection optical ar- 10,000 times through resonance. However, it is observed only be- rangements and above all high laser light fl uxes. With a stable high cause high signal to noise spectra can be recorded at 200 ms intervals. powered laser (500 mW at sample) and a large confocal volume for collection to limit sample damage, time resolutions as low as Lasers for Raman Spectroscopy By far the most popular wavelength used for Raman spectroscopy is 785 nm, as it off ers the best balance between avoiding fl uorescence, H O 0.20 2 2 absorption of the laser light (and Raman scattering) by the sample 0.30 0.15 intermediate 0.25 0.10 and therefore heating eff ects, and the limits to detector sensitivity. 0.05 intermediate 0.20 Intensity (AU) 0.00 Th e lasers available at 785 nm are diode lasers, however, a narrow 0.15 12 345 -1 Times (s) line width (<3 cm for condensed phases such as solids and so- 0.10 Intensity (AU) H O signal lutions and even narrower for gases) is essential for Raman spec- 2 2 Intermediate complex precatalyst 0.05 troscopy in order to resolve individual Raman bands. Th e Cobolt 0.00 0204060 80 100 120 800 1000 1200 1400 1600 08-NLD 785 nm laser addresses all of the important performance Times (s) Raman shift (cm–1) features in a compact footprint and ensures reliability thanks to the Figure 1: (Left) Graph showing intensity of bans of H O (black) 2 2 proprietary manufacturing technique called HTCure™. and precatalyst intermediate (red) shows that the precatalyst intermediate is only present for about 1 s during the reaction but Cobolt AB that the H2O2 concentration decreases steadily over 120 s. (Right) Vretenvägen 13, 17154 Solna, Sweden Selected spectra immediately before, during, and after addition of tel. 46 8 545 912 30 H O showing precatalyst, catalyst intermediate, and H O . 2 2 2 2 Website: www.coboltlasers.com 58 Molecular Spectroscopy THE APPLICATION NOTEBOOK – SEPTEMBER 2017

SERS Methods Detect Trace Levels of Cocaine, Heroin, Methamphetamine, and THC Anne-Marie Dowgiallo, PhD, Ocean Optics

urface enhanced Raman spectroscopy (SERS) is an extension of SRaman spectroscopy in which gold or silver nanoparticles amplify the Raman signals. Th e technique works via an electromagnetic eﬀ ect Silver substrate + 100 ppm cocaine 700 Raman excitation: 785 nm, 3 s, 15 mW where molecules come into proximity with gold or silver particles. Spot 1 When incident laser light strikes the nanoparticulate surface, 600 Spot 2 localized surface plasmons can be excited, greatly enhancing Raman Spot 3 500 signals. Th e enhancement can be signiﬁ cant, making SERS well- Average suited to trace level detection of illicit drugs such as cocaine, heroin, 400 methamphetamine, and tetrahydrocannabinol (THC). To test how well Ocean Optics SERS substrates can detect 300 Intensity (counts) trace drugs, measurements were performed using gold and silver 200 nanoparticles with an Ocean Optics modular Raman setup. As we discovered, detection of several illicit drugs using gold 100 nanoparticles is a rapid, reliable technique that requires only a 0 few milliwatts of laser power. 400 600 800 1000 1200 1400 1600 1800 Raman shift (cm–1)

Sample Preparation Figure 2: Raman signal enhanced with silver substrates. Heroin hydrochloride, cocaine hydrochloride, Δ9-tetrahydrocannabinol, and (±)-methamphetamine were prepared in 100 ppm Experimental Setup solutions in methanol. For measurements with the gold and silver SERS substrates, we used a For the Ocean Optics SERS substrates, which comprise analyte- modular Raman system comprising the QE Pro high-sensitivity spec- sensitive nanoparticle chemistries applied to a treated glass surface, we trometer, a 785 nm laser for Raman excitation, and sampling optics. used 15 μL of 100 ppm solutions for the gold and silver nanoparticle Th e 785 nm excitation produces excellent Raman spectra for most substrates. We tested at 100 ppm to determine feasibility, although chemicals, with limited interference from ﬂ uorescence. Th ese systems results suggest lower concentrations are possible. also oﬀ er very good spectral resolution, making them a preferred wavelength choice for Raman spectroscopy of chemicals and organic materials. Th ree measurements for each sample were taken from multiple spots on the same substrate, which then were averaged. Gold substrate + 100 ppm cocaine 4000 Raman excitation: 785 nm, 3 s, 15 mW Results 3500 As demonstrated in the spectra, SERS methods utilizing Ocean Optics substrates can detect ppm levels of illicit drugs, which Raman 3000 Spot 1 Spot 2 techniques alone would not be able to accomplish. 2500 Spot 3 Average Also, the gold nanoparticles appear to work best for detecting these

2000 drugs. Certain peaks have higher Raman cross-sections, and each peak is enhanced by the gold nanoparticles diﬀ erently depending on how 1500 Intensity (counts) the molecule is oriented with respect to the gold surface.

1000

500

0 Ocean Optics 400 600 800 1000 1200 1400 1600 1800 Raman shift (cm–1) 8060 Bryan Dairy Rd., Largo, FL 33777 Figure 1: The Raman spectrum of pure cocaine is dominated by tel. (727) 733-2447 peaks at 872, 999, 1026, 1273, and 1597 cm-1. Website: www.oceanoptics.com THE APPLICATION NOTEBOOK – SEPTEMBER 2017 Molecular Spectroscopy 59

(a) (b) 1 0.005 2 0.005

1 Surface-Sensitive Spectroelec- 3 trochemistry with ATR-SEIRAS absorbance

Tyler A. Morhart and Ian J. Burgess, Department of 4 Chemistry, University of Saskatchewan 1800 1700 1600 1500 1400 1300 1800 1700 1600 1500 1400 1300 Wavenumber (cm–1) We monitor potential-induced changes at a Au fi lm Figure 2: ATR-SEIRAS spectra of 4-methoxypyridine (MOP). (a) MOP electrode using the ATR-SEIRAS sampling technique. adsorbed at open circuit conditions. (b) MOP potential difference Benefi ts are discussed. spectrum. Sample: +200 mV (MOP adsorbed). Reference: -700 mV (MOP desorbed). ttenuated total refl ectance surface-enhanced infrared absorption to ensure potential control. In the cell body, a hole was bored to A spectroscopy (ATR-SEIRAS) provides in situ surface-sensitive fi t a reference electrode arm. IR spectra of molecules at the electrode-solution interface. A Au We sputtered a ca. 30 nm Au fi lm electrode on a 60° angled Si thin fi lm electrode is deposited on the refl ecting plane of the ATR ATR crystal. We then cycled the potential of the fi lm in 50 mM optic. Excitation of surface plasmon modes in the fi lm provides up KClO4 supporting electrolyte between -100 and +1100 mV versus to 10-fold electric fi eld enhancement at the interface, enhancing IR Ag/AgCl. Th is cleans the fi lm and textures the surface to improve cross-sections of adsorbed species up to 100-fold. Controlling the SEIRAS intensity. Th e VeeMAX III was set to 60° to match the applied potential on the fi lm allows vibrational characterization of crystal face angle. FT-IR spectra were collected at 4 cm-1 resolution, potential-induced changes. Such studies reveal fundamental details 128 scans, and 40 kHz scanning velocity using an MCT detector. of electrochemical processes and inform the development of next generation energy and sensing solutions. Results We used the potential-dependent adsorption of 4-methoxypyri- Experimental Conditions dine (MOP), a nanoparticle shape-directing ligand, as a test sys- Th e VeeMAX III with ATR and electrochemical cell option (PIKE tem. We collected a reference spectrum of supporting electrolyte, Technologies, Madison WI) was used as the FT-IR sampling then a sample spectrum with 0.1 mM MOP, both at the open accessory. Th e electrochemical cell is designed to be easily modifi ed circuit potential. Th e result is shown in Figure 2a. We observe two to meet varied requirements in the fi eld. We modifi ed the cap of strong ring vibrations (1) and the methoxy C-O-C asymmetric the VeeMAX III electrochemical cell to accept a Au coil counter stretch (2). A third ring vibration (3) has its transition dipole mo- electrode, a glass bubbler for Ar purging, an exhaust bubbler, and ment (TDM) mostly parallel to the electrode-solution interface a Au bead electrode. Th e Au bead was pressed against the Au fi lm when adsorbed, so (3) is very weak. Only vibrations with their TDM normal to the interface are IR active. Th is confi rms the surface sensitivity of the measurement. Figure 2b reveals changes in interfacial composition upon potential changes. Th e negative absorbance band (4) represents loss of interfacial water. Th is band is absent in (a) because, at open circuit conditions, water adsorbs in a fl at state on Au, rendering

the H2O bending mode inactive in SEIRAS. We see also the characteristic MOP bands, which shows that we have potential control over the adsorption of MOP.

Conclusion With PIKE’s variable angle VeeMAX III ATR accessory and electrochemical cell, we have demonstrated a surface-sensitive technique which can resolve minute changes in interfacial composition.

PIKE Technologies 6125 Cottonwood Drive, Madison, WI 53719 tel. (608) 274-2721 Figure 1: PIKE VeeMAX III accessory with electrochemical cell option. Website: www.piketech.com 60 Molecular Spectroscopy THE APPLICATION NOTEBOOK – SEPTEMBER 2017

Diamond Anti-Reﬂ ection Coating Spectral Systems LLC

iamond is an important material choice for sampling optical Delements in Fourier transform-infrared (FT-IR) spectroscopy. Th e multi-phonon absorption in the mid-infrared, although temperature dependent, is weak for thin optical elements and can 80 be referenced to an empty cell background. However, the cost of diamond optics requires the sampling area to be very small. As a result, the signal-to-noise ratio (SNR) advantages of the FT- 60 IR systems are frequently stretched to the limit and more energy 40 throughput is usually quite helpful. Transmittance Frequently, the sampling technique of choice is attenuated total refl ection (ATR) for its benefi ts of ease of analysis. ATR elements 20 can be manufactured from most optical materials of relatively high refractive index. Ge, Si, ZnSe, ZnS, KRS-5, and diamond 4000 3000 2000 1000 are often used for this application. For most of these materials, a Wavenumbers broad band anti-refl ection (BBAR) coating can be applied to both Figure 1: Transmission spectrum of uncoated diamond window surfaces of a window or the entrance and exit facets of an ATR shown in blue. Transmission spectrum of Diamond XP-BBAR coating element and can signifi cantly increase the energy throughput of on both sides of a diamond window shown in green. the sampling accessory or system. Since diamond and ZnSe have similar refractive indexes a similar BBAR can be used on diamond once the binding layers of the coating are appropriately modifi ed. Spectral Systems has developed its Diamond XP-BBAR™ coating with spectral performance as shown in Figure 1. It is seen in Figure 1 that the energy and thus the SNR is about 30% better through most of the analysis region of a typical FT-IR instrument. Since there are no suitable coating materials for the very long wavelength portion of the spectrum the coated element is not as good, but still usable, below 600 cm-1. If a liquid nitrogen cooled MCT detector is used, this portion of the spectrum is usually not recorded by the spectrometer.

Spectral Systems 35 Corporate Park Drive, Hopewell Junction, NY 12533 tel. (845) 896-2200 Website: www.spectral-systems.com THE APPLICATION NOTEBOOK – SEPTEMBER 2017 Molecular Spectroscopy 61

Qualifying ATR Accessories for Pharmaceutical Applications Steve Lowry, Thermo Fisher Scientiﬁ c

Recent advances in single-bounce ATR accessories provide an easy way to acquire a high-quality infra- 3.0 35 micron polystyrene ﬁlm red spectrum from very small amounts of sample in 2.8 Polystyrene Diamond ATR less than 1 min. This has resulted in increased use of 2.6 this technique in the pharmaceutical industry, mak- 2.4 2.2 ing accessory qualiﬁ cation a concern. We suggest 2.0 an adaptation of the Pharmeuropa guidelines as an 1.8 excellent starting point. 1.6 1.4 Absorbance 1.2

1.0 T-IR spectrometry is a key technique used throughout the 0.8 0.6 F pharmaceutical industry and most of the pharmacopoeias 0.4 in the world contain a section on how to qualify an infrared 0.2 0.0 3000 2500 2000 1500 1000 spectrometer for use. However, there is no clear guidance from Wavenumbers (cm–1) regulatory agencies on how to qualify an FT-IR spectrometer Figure 1: Report from workfl ow based on new ATR proposed peak system with an ATR accessory. In fact, there is some question locations with the transmission and ATR spectra from a NIST trace- whether this falls under “Analytical Instrument Qualifi cation” able polystyrene reference. or “System Suitability.” While the US Pharmacopeia defi nes appropriate measurements for wavelength accuracy for spectra from a polystyrene fi lm in transmission (blue) and the transmission spectroscopy, it clearly states that ATR spectra are ATR spectrum (red) acquired with the diamond iD5 Accessory diff erent. Th e European Pharmacopoeia recommends the same on a Th ermo Scientifi c™ Nicolet™ iS™5 FT-IR spectrometer. polystyrene peak positions but adds a resolution test. A recent Th e results (targets and actual) of the proposed pharmacopoeia Pharmeuropa report (29.3) describes some proposed changes to measurements are tabulated (inset) for the diamond ATR the European Pharmacopoeia and actually proposes ATR peak spectrum. While we obtained good results using the peaks positions for polystyrene as well as pointing out that a diff erent described in the Pharmeuropa proposal, the peak near 3060 cm-1 resolution test is also necessary. One challenge with ATR spectra is is quite weak, a situation which would be even worse for Ge-ATR the relative diff erences in peak intensity. While some of the peaks (shallower depth of penetration). in a transmittance spectrum of 35 μm thick polystyrene fi lm are We conclude that the guidelines recommended by Pharmeuropa highly absorbing and not appropriate for verifying wavelength provide an excellent starting point for this long overdue accuracy, they are much weaker in an equivalent ATR spectrum, development. However, we strongly recommend replacement particularly in the C-H stretching region. In this application note, of the 3059.7 cm-1 peak with either of the stronger signals near we will describe an evaluation method based on the Pharmeuropa 3025 cm-1 or 2920 cm-1, which will provide a more robust proposal that we believe will verify the performance on an ATR measurement, regardless of the ATR crystal type. accessory in FT-IR instruments.

Experimental Conditions Questions about this application note may be directed to: Spectra were acquired on various FT-IR spectrometers with steve.lowry@thermofi sher.com. diff erent single-bounce ATR accessories at 2 cm-1 instrument resolution using a NIST traceable polystyrene standard that had been qualifi ed in a calibration laboratory. A workfl ow was created using the four peak locations proposed for ATR qualifi cation. We chose to use the full width at half maximum on polystyrene peaks Thermo Fisher Scientifi c at 1154 cm-1 and 906 cm-1 to determine spectral resolution. Th e 5225 Verona Rd., Madison, WI 53711 instrument resolution was confi rmed using water vapor peaks tel. (800) 532-4752 with the accessory mounted in the system. Figure 1 shows the Website: www.thermofi sher.com 62 Molecular Spectroscopy THE APPLICATION NOTEBOOK – SEPTEMBER 2017

240 240

180 180 G G

CCD cts 120 2D CCD cts 120 2D

60 D 60 ™ D 1400 1600 1800 2000 2200 2400 2600 2800 RISE Raman Imaging and 1400 1600 1800 2000 2200 2400 2600 2800 rel. wavenumbers (cm–1) rel. wavenumbers (cm–1)

G Scanning Electron Microscopy R@1514 240 240 50

180 180 30 20

Investigation of Twisted CCD counts G 10 CCD cts CCD cts 120 D’@1625 2D 120 R@1514 1400 1500 1600 1700 1800 1900 2000 relative wavenumbers (cm–1) Bilayer Graphene 60 60 D

WITec GmbH 1400 1600 1800 2000 2200 2400 2600 2800 1400 1600 1800 2000 2200 2400 2600 2800 rel. wavenumbers (cm–1) rel. wavenumbers (cm–1)

2D onfocal Raman imaging is an ideal method for studying 2D 240 Cmaterials because it reveals their molecular characteristics 180 G through a nondestructive and fast procedure. It is used to describe CCD cts 120 the orientation of layers and to investigate defects, strain, and 60 D functionalization of these materials as their Raman properties 1400 1600 1800 2000 2200 2400 2600 2800 are determined by molecular bonds, relative orientation, and rel. wavenumbers (cm–1) number of layers. Th eir morphological details can be visualized Figure 2: Raman spectra of twisted bilayer graphene. Variations with scanning electron microscopy (SEM). By correlating the in peak intensity ratios of G/2D Raman bands, changes in the FWHM of the 2D bands as well as the appearance of a 2D’ band information of both approaches with RISE™ (Raman Imaging at 1625 and an R (resonance) band at 1514 relative wavenumbers and Scanning Electron) microscopy, the sample can be more allow for the determination of the stacking order and the twist thoroughly analyzed at high resolution. angles of the graphene layers. For comparison, the spectrum of Here the stacking, folding, and twisting of bilayer graphene monolayer graphene is depicted as a dashed line. was investigated. Th ese factors strongly inﬂ uence its phononic and electronic properties. Correlative Raman-SEM (RISE™) imaging was performed with a combined WITec/ZEISS microscope for the analysis of bilayer graphene that was obtained by chemical vapor deposition and placed on a Si/SiO2 substrate. Th e SEM image shows that the sample consists of monolayer and multi-layered graphene areas (Figure 1). Whether and how the multi-layers are twisted was established by analyzing the position of the G and 2D peaks of graphene, the peaks’ intensities, and their full widths at half maximum (FWHM) (Figure 2). Additional 2D′ and R bands allowed for the determination of the twist angles. Figure 3 shows the color-coded RISE® image derived from the spectra. 10 μm Figure 3: RISE image of twisted bilayer graphene, the correlated overlay of complete Raman and SEM analyses. The entire sample is covered with graphene. Monolayer graphene (ML) is shown in yellow. Identiﬁ cation of bilayer graphene by typical Raman bands: • Blue: non-twisted, AB stacking, more intense G band than ML • Purple: twisted by 0–3°, smaller 2D band than ML • Green: twisted by 3–8°, D’ peak at 1625 relative wavenumbers • Cyan: twisted by >12°, resonance peak at 1514 relative wavenumbers, extremely intense G band • Red: twisted by 20°, very intense 2D band. Raman image parameters: 22,500 spectra, 50 ms integration time per spectrum.

WITec GmbH Lise-Meitner-Straße 6, 89081 Ulm, Germany 10 μm tel. +49 (0) 731 149 700, fax +49 (0) 731 149 70 200 Figure 1: SEM image of twisted bilayer graphene. Website: www.witec.de THE APPLICATION NOTEBOOK – SEPTEMBER 2017 ADVERTISING SUPPLEMENT 63 THE APPLICATION NOTEBOOK Call for Application Notes

Spectroscopy is planning to publish the next is- Format enough to remain legible after reduction. Pro- sue of Th e Application Notebook in December • Title: short, specifi c, and clear vide labels and identifi cation. Provide fi gure 2017. As always, the publication will include • Abstract: brief, one- or two-sentence abstract captions as part of the text, each identifi ed paid position vendor application notes that • Introduction by its proper number and title. If you wish to describe techniques and applications of all • Experimental Conditions submit a fi gure or spectrum, please follow the forms of spectroscopy that are of immediate • Results format of the sample provided below. interest to users in industry, academia, and • Conclusions government. If your company is interested • References Tables in participating in this special supplement, • Two graphic elements: one is the company Each table should be typed as part of the main contact: logo; the other may be a sample spectrum, text document. Refer to tables in the text by fi gure, or table Roman numerals in consecutive order (Table Stephanie Shaff er, Publisher, • Th e company’s full mailing address, I, etc.). Every table and each column within (774) 249-1890 telephone number, fax number, and the table must have an appropriate heading. Stephanie.Shaff [email protected] Internet address Table number and title must be placed in a All text will be published in accordance continuous heading above the data presented. Edward Fantuzzi, with Spectroscopy’s style to maintain uni- If you wish to submit a table, please follow Associate Publisher, formity throughout the book. It also will the format of the sample provided below. (732) 346-3015 be checked for grammatical accuracy, al- [email protected] though the content will not be edited. Text References should be sent in electronic format, prefer- Literature citations must be indicated by ara- Application Note Preparation ably using Microsoft Word. bic numerals in parentheses. List cited refer- It is important that each company’s material ences at the end in the order of their appear- fi ts within the allotted space. Th e editors Figures ance. Use the following format for references: cannot be responsible for substantial Refer to photographs, line drawings, and (1) T.L. Einmann and C. Champaign, Science editing or handling of application notes graphs in the text using arabic numerals in 387, 922–930 (1981). that deviate from the following guidelines: consecutive order (Figure 1, etc.). Company Each application note page should be no more logos, line drawings, graphs, and charts must For more information on this topic, than 500 words in length and should follow be professionally rendered and submitted as please visit the following format. .TIF or .EPS fi les with a minimum resolution www.spectroscopyonline.com of 300 dpi. Lines of spectra must be heavy

Table I: Factor levels used in the designs

160 Factor Nominal Value Lower Level (−1) Upper Level (+1) 140 Gradient profile 1 0 2 120

100 Buffer concentration 40 36 44

Golf score 80 Mobile-phase buffer pH 5 4.8 5.2 60

40 Detection wavelength (nm) 446 441 451

$150,000 $200,000 $250,000 $300,000 $350,000 $400,000 Triethylamine (%) 0.23 0.21 0.25 Cost Dimethylformamide 10 9.5 10.5 Figure 1: The ion suppression trace is For more information, contact: shown on the bottom. Column: 75 mm Stephanie Shaffer at (774) 249-1890 or [email protected] × 4.6 mm ODS-3; mobile-phase A: 0.05% Ed Fantuzzi at (732) 346-3015 or [email protected] heptafluorobutyric acid in water; mobile- phase B: 0.05% heptafluorobutyric acid in Materials Due for December Issue: October 31, 2017 acetonitrile; gradient: 5–30% B in 4 min. Direct submissions to: Peaks: 1 = metabolite, 2 = internal standard, Cindy Delonas, Associate Editor 3 = parent drug. [email protected] Copyright © 2017 PerkinElmer, Inc. 400373_02 All rights reserved. PerkinElmer® is a registered trademark of PerkinElmer, Inc. All other trademarks are the property of their respective owners. WH F HI T G - Spectrum TwoN™ FT-NIRSpectrometer N H- ERE I R PER

T H

F YOU OR AT M

GOE ANCE

GO S L Two N:Itgoeswhereveryoursamplesleadyou. modules providesadditionalpeaceofmind.TheSpectrum simple, powerfulsoftwarewith21CFRPart11compliance sampling –makeitthemostversatilesmallFT-NIRever.Plus, Module withatriggeredsolidsprobeforthrough-container Module withmultiplevialsizes,andaRemoteSampling Near InfraredReflectanceModule,aHeatableTransmission flexible, portablesystem.Threenewfeatures–aplug-and-play QA/QC oradvancedchemometricanalyses–allonone performance, sensitivity,andresolution,whetherforsimple the SpectrumTwoN™systemdelivershighestlevelsof From pharmaceuticalstofood,polymerspetroleum, performance This earn

transporta more

www.perkinelmer.com/SpectrumTwoN

for b

le all

y t y stem pes

of is b Q ig A/

Q on C

samples