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
5
1
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 effec- tive use of spectroscopic technology as a practical research and measurement tool. With recognized expertise in a wide range of technique and application areas, board 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 publica- tion 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 spec- trum 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 fol- lowing 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 Inc West 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 correla- tion 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 lat- tice 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 configura- tion. 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 fol- low 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 spec- trum 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 sig- nificantly 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 varia- tions 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 theo- retical 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 inci- dence 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 re- gion 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 envi- sion a distribution of angles of inci- dence that is dependent on the numer- ical 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 objec- tive with higher numerical aperture is used. This phenomenon can be attrib- uted 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 crys- tallographic axis. Therefore, one needs to be cognizant of the numerical aper- ture 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 qualifica- tion (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
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 qualifica- tion (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 qualifi- cation. 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 acquisi- tion, 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 inter- face 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) spectros- copy 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 ke- tones 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 stretch- ing 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 func- tional 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 vi- bration 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 ke- tones, 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