Downsizing and layoffs have affected nearly every facet of the American workforce during the last year, and the spectroscopy job market is no exception. Despite this, the spectroscopy field has seen some increases in salary, as well as workloads and stress levels.   

This year, Spectroscopy's salary survey had 553 respondents, from a broad range of job sectors and locations. Here are a few statistics about this year's results: 					

28% of respondents are from the Midwest region; 27% are from the Northeast region; 21% are from the Southeast region; 21% are from the Southwest region; and 3% are from the Northwest.

20% of respondents are 50–54 in age; 14% are 40–44 in age; 14% are 45–49 in age; 12% are 55–59 in age; 11% are 35–39 in age; 10% are 60–64 in age; 7% are 30–34 in age; 6% are 25–29 in age; 5% are 65+ in age; and 1% are 20–24 in age. 

73% of respondents are male and 27% are female.

43% of respondents hold doctoral degrees; 34% hold bachelor's degrees; 20% hold master's degrees; and 4% hold associate degrees.

62% of respondents received their highest degree in chemistry; 11% received it in biology; 8% received it biochemistry/biophysics; 8% received it in physics; 3% received it in environmental engineering/science; 2% received it in chemical engineering; 2% received it in geology/earth science; 2% received it in polymer science; and 2% received it in pharmaceutical science.

64% of respondents work in the private sector; 18% work in academia; 14% work in government/national lab; and 4 % work in other areas.

93% of respondents are employed full-time; 3% are postdoctoral researchers/graduate students; 2% are employed part-time; 1% are contract employees/ consultants; and 1% are unemployed. 

20% of respondents are senior scientists/researchers/research fellows; 15% are chemists/spectroscopists; 14% are laboratory directors/managers; 10% are staff scientists/researchers/research fellows; 5% are full professors; 5% are laboratory technicians/technologists; 5% are analysts; 4% are research assistants/associates; 2% are assistant professors; 2% are associate professors; 2% are process engineers; 2% are technical directors; 1% are process chemists; 1% are CEO/presidents; and 12% are other.

According to this year's survey, the mean salary for all respondents is $80,778, which is an increase from last year's mean of $78,807 — an increase of $1971. This is also the largest salary we've seen over the last decade (see Table I).  

Table I: 2001â2010 average salaries			

As seen in Table I (with the exception of 2008), salaries have risen continuously for spectroscopists by a few thousand dollars each year. The largest increase reported was in 2006, when the average salary rose by nearly $5000. When evaluating the data this way, it is not surprising that the salaries continued to rise despite the troubled economy. 

Another factor playing into salaries again this year was respondents' level of experience in their respective fields. An estimated 40.9% of respondents had between 21 and 35 years of experience; 17.6% had between 10 and 15 years of experience; 11.8% had between 16 and 20 years of experience; 11.6% had less than 5 years of experience; 9.6% had between 5 and 9 years of experience; 5.3% had between 36 and 40 years of experience; and 3.3% had 41 or more years of experience. As with most industries, the more professional experience a person has, the higher their salary. 

One Small Step for Womankind, One Giant Leap for Gender Equality

As in previous years, there is still a significant earnings gap between males and females. However, this year women have made small strides to close in on their male counterparts' salaries. According to the survey, composed of 73% males and 27% females, males are earning an average salary of $83,054 and females are earning an average salary of $69,356. While this is still a gap of $13,698, it is a significant improvement over 2009's gap of $15,578, 2008's gap of $16,000, and 2007's gap of $14,218 (see Figure 1). 

Figure 1: Salary gap between males and females.			

 covers a broad range of industries, but the three big players are academia, government, and the private sector. This year's survey was composed of 64% of respondents from the private sector; 18% from academia; 14% from government/national lab; and 4% from other areas. Private sector employees reported an average salary of $84,799, which is slightly less than last year's average. Respondents working in academia reported an average salary of $66,448,  about a $4000 increase over last year's average. Finally, scientists working in government/national labs reported an average salary of $75,373—slightly smaller than last year's average but still higher than 2008's (see Figure 2). 

Figure 2: Salary summary by job sector, 2008â2010.			

While everyone loves a salary increase, it almost never comes for free and this year is no exception, as our respondents reported an increase in workload and stress level.  

According to our survey, 62% of respondents had an increase in workload; 30% had their workload remain the same; and 7% had a decrease in workload (1% responded with not applicable). As possible causes for increased workload, most respondents said staffing cuts or an increase in business. Other reasons cited for increased workloads were new equipment additions to the lab, new handling techniques and procedures, and new regulations. Respondents who reported a decrease in workload cited automation through new equipment, increased staffing, and new technology that increased productivity and quality as the main causes, while continuing education, specialization of laboratory personnel, and improved workplace attitudes were also possible reasons. 

Respondents also reported a 54% increase in stress level; 41% reported their stress level remained the same; and 5% reported a decrease in stress level (see Figure 3). The most common reasons chosen for increased stress level were business uncertainty, staff management uncertainty, and negative workplace attitudes. Those who reported a decrease in stress level cited improved workplace attitudes, increased staffing, and continuing education as the reasons. 

Figure 3: Stress levels, 2009â2010.			

Despite the increased workloads and stress levels, 38% of respondents reported feeling satisfied with their current position (down from last year's 42%); 33% felt very satisfied (down from last year's 47%); 14% were not satisfied (up from last year's 12%); 12% felt extremely satisfied; and only 3% were not at all satisfied (see Figure 4). Although this is good news for employers, compared to last year, the satisfaction levels have decreased and dissatisfaction has increased, which is not too surprising when looking at the entire picture. 

Figure 4: Job satisfaction levels, 2010.			

In accordance with people feeling satisfied in their current positions, 39% of respondents reported feeling satisfied with their current employer; 28% reported feeling very satisfied; 17% reported feeling not satisfied; 12% reported feeling extremely satisfied; and 4% reported feeling not at all satisfied. 

Regardless of individual levels of satisfaction, not many people are too keen on getting out there in this type of job market. According to our survey, 59% of respondents are not considering making a job change in the next year; 22% are not sure; and 19% are considering a job change. Of those respondents considering a job change, the main reasons are professional advancement, income, job environment, and intellectual challenge. 

Professional advancement is an item on the top of most people's to-do lists, and knowing your next step is a key aspect of moving up. However, it's not always as clear-cut as it may seem, especially if people are looking to crossover into a new aspect of their job or a new field entirely. Our respondents offered us a mixture of answers on their professional goals with 41% reporting that they don't know or are not sure of their next career step; 20% said lab director/manager; 20% said retirement; 6% said company executive; 5% said consultant; 3% said full professor; 3% said assistant professor; and 2% said associate professor.  

Another aspect of professional advancement is having a timeline for career goals. For example, it's unrealistic to go from laboratory technician to company executive within two years, but if you spread that out over a larger time frame and take career advancements in that direction, it becomes a much more realistic goal. According to our survey, 25% of respondents think it will take more than five years to take their next career step; 23% think it will take 1–2 years; 19% think it will take 2–3 years; 12% think it will take 4–5 years; 11% think it will take less than a year; and 10% think it will take 3–4 years. 

Employers also play a major role in career advancement for their employees, with the opportunities they present. 45% of respondents reported that their companies are supportive of career advancement; 20% reported their companies are not supportive; 20% reported their companies as very supportive; 8% reported their companies as extremely supportive; and 7% reported their companies as not at all supportive. Compared to last year's results, it seems that companies are becoming more supportive of their employees career-advancement goals, with an increase from 44% of companies reported as supportive and a decrease from 25% of companies reported as not supportive or not at all supportive. 

The fringe benefits a company offers to its employees also can have a positive impact on career goals and overall job satisfaction. According to the survey, 63% of respondents reported that their companies offered in-house training and 38% reported that their companies did not. Of those that did offer in-house training, tuition reimbursement, training courses, and seminars were the primary focus. Another benefit that companies often offer to employees is covering the cost of conferences and web seminars, and 77% of respondents reported their companies covering such projects, while 23% reported their companies do not cover them. 

Overall, the relationship between employer and employee is a very complex dynamic, but the spectroscopy community seems to have a firm grasp on how to keep their companies running and employees satisfied. 

Spectroscopists from across the United States participated in this survey, and in the category of "Region by Salary," the Southwest was back on top with the highest average income this year, after a brief fall to second place in 2009 (see Figure 5). This year, the Southwest region reported an average salary of $94,496, a significant jump from last year's average of $79,111. The respondents' level of education might help explain the large salary jump since 44% have doctoral degrees, 19% have master's degrees, 33% have bachelor's degrees, and 4% have associate's degrees. 

Figure 5: Salary summary by geographical region, 2010. (Southwest: CA, HI, AZ, NM, NV, CO, UT, OK, TX; Northeast: NY, MA, CT, PA, VT, NH, RI, ME, NJ; Midwest: OH, IN, IL, IA, MO, MI, MN, WI, KS, ND, SD, NE; Southeast: LA, MS, AL, GA, FL, TN, KY, NC, SC, AR, DE, VA, WV, MD, DC; Northwest: OR, WA, ID, MT, WY, AK.)			

The Northeast has fallen from the top spot to a close second, with the region reporting an average salary of $94,473, a difference of a mere $23. This is nearly a $9000 increase from last year's average for this region, but again, education levels could help explain why. Respondents from this region reported 45% with doctoral degrees, 23% with master's degrees, 30% with bachelor's degrees, and 2% with associate's degrees. 

In third place this year is the Midwest, with an average salary of $77,080, an increase of about $2600 from last year's average. Respondents in the Midwest also had high education levels with 40% holding doctoral degrees, 17% holding master's degrees, 39% holding bachelor's degrees, and 4% holding associate's degrees. 

Figure 6: Level of education by geographical region, 2010. 			

Falling to fourth place this year is the Southeast, with an average salary of $77,021, a mere $59 short of third place. This is a slight decrease from last year's average ($77,094). Although this does not explain the decrease in salary, respondents reported education levels of 45% doctoral degrees, 21% master's degrees, 31% bachelor's degrees, and 3% associate's degrees. 

Last, but not least, is the Northwest, with an average salary of $68,406. This region had the least amount of participants and saw the largest decrease in salary from previous surveys. In 2009, the average salary was $74,403, a difference of about $6000. This region also had a low percentage of participants with doctoral degrees (41%) and master's degrees (6%). The majority of respondents had bachelor's degrees (53%), which could explain the decrease in salary. 

Salary Survey with the assistance of an outside vendor. A link for the online survey was sent to the entire circulation

The invitation was successfully sent to 24,465 subscribers. A total of 553 individuals responded. The field time of the survey was December 2 through December 15, 2009 (a reminder email was sent). To encourage participation, all subscribers who responded to the survey were entered into a drawing to win one of five $100 gift cards.  

Articles

Article

This year brought some increases in salary, workload, and stress level for the spectroscopic community.

2010 Salary Survey

In the field of materials analysis, it is no secret that the topic of food safety is currently garnering more and more attention, with instrumentation from nearly every technique area being employed in the fight against contaminations and poisonings. FT-IR, ICP-MS, and IRMS (to name just a few techniques) have been featured in many recent columns and articles geared toward food safety in the pages of this journal and have been featured at conferences and symposia around the world as well. 

However, at the recent IFPAC Conference, held last month in Baltimore, Maryland, an interesting shift in the food safety discussion could be sensed, as Dr. Glenn Black, Director of Science Operations of the Grocery Manufacturers Association, delivered an insightful talk as part of the opening plenary session. Dr. Black signaled that a new phase of the ongoing worldwide food safety battle may have been reached. "Food safety" was replaced by "food defense" as Dr. Black described the need to defend the food chain from start to finish, definitely a more aggressive posture with regard to preventing future contaminations and poisonings. After the dog food, baby food, and peanut butter scares of recent years, it is clear that the public's tolerance for food safety crises is becoming lower and lower, so this more aggressive stance is a wise one to say the least. 

It will be interesting to see how this issue continues to unfold, and as always, be sure to look for more on this topic in future issues of 

, where food safety continues to generate a great deal of research and commentary. Last month's feature on the validation of extra virgin olive oil and international food products in general generated a great deal of buzz, as the team of researchers from Ohio University focused on an issue of increasing importance, and additional food safety features are planned throughout 2010. 

In the meantime, we hope you enjoy the issue. 

From the Editor

In the field of materials analysis, it is no secret that the topic of food safety is currently garnering more and more attention, with instrumentation from nearly every technique area being employed in the fight against contaminations and poisonings.

Defending the Food Chain

Somewhat unique among molecular spectroscopy techniques, the application of nuclear magnetic resonance (NMR) for industrial process analysis is driven primarily by the petroleum industry. Although the market for process NMR was heavily affected by the recent recession, it is a very valuable niche technique that will see demand rebound. 

Process NMR demand by industry in 2009.			

Although it falls into the category of molecular spectroscopy, NMR spectroscopy differs from other techniques in the category because it is not an optical technique. Instead, it uses a combination of magnetic and radio frequency (RF) radiation for analysis. This eliminates issues associated with optical techniques, while maintaining the advantages of being nondestructive, and providing continuous analysis, both of which are key to online instrumentation. 

Process NMR is used most heavily by far in the petroleum refining industry, which is an area in which it was first applied in the mid-1990s. Process NMR, which can be applied at many locations within a refining process, can provide real-time data used for both feed-forward and feedback control. As sulfur content restrictions have tightened, refiners have become increasingly conservative, but by using process NMR, they are able to produce more product while controlling sulfur concentration. This can translate into additional profits of many thousands of dollars per day in an average-sized refinery. Other industries in which process NMR is used include in polymers for the monitoring of fluorinated materials and online process control, the monitoring of phosphorous in the mining and chemicals industry, and in the power generation industry. 

All of the industries in which process NMR is used contracted considerably during the recession, resulting in global demand of around $5 million in 2009. However, demand should rebound fairly quickly, being led by the construction, expansion and improvement of refineries in rapidly growing markets such as China. 

Contact: Stuart Press, Strategic Directions International, Inc., Los Angeles, CA (310) 641-4982, 

Market Profile

Somewhat unique among molecular spectroscopy techniques, the application of nuclear magnetic resonance (NMR) for industrial process analysis is driven primarily by the petroleum industry.

Market Profile: Process NMR

In libraries, patrons now often find that shelves of printed references have been replaced with computer terminals that access the array of electronic resources on the library intranet, as well as access to resources, information collections, and unsupervised sites and musings of the external web. Should casual readers look a bit more carefully through the reference lists for "Mass Spectrometry Forum" over the past year, they would find at first a smattering, and then recently more references cited by their URL (universal resource locator) or by their DOI (digital object identifier). Some resources might be found in both printed and electronic form. The contrast between availability and portability of the forms is dramatic. But more and more, resources that support our continuing education in mass spectrometry are found solely in electronic form, and, through web access, are available virtually anywhere.  

The web, however, is more than a simple portal to educational resources for mass spectrometry (MS) that can be delivered and displayed electronically, but closely resemble the classic printed forms. Search engines provide the means to discover new resources, in both printed and visual form. Links between sites create the network of interrelated resources. Alerts can be set in a user profile so that newly posted resources (applications, technical notes, announcements, and new research publications) are brought to the attention of the users. Further, the web brings not only text, but images, sound, video, and interactivity to the user. None of this is any surprise in 2010; the expansion and maturation of the web has infiltrated modern life quite expertly. It is perhaps only when we hold a mass spectrometry book in our hand, and remember how eager we were to first purchase it and read it, that we might appreciate just how fundamental a change has occurred in our access to information resources.  

A previous column (1), "Electronic Resources for Mass Spectrometry," appeared in 1996, which seems like (and is) ancient electronic history. The American Society for Mass Spectrometry (ASMS) was still publishing abstracts from the annual meeting in printed form. Recently, ASMS renamed the elected position of Member-at-Large for Measurements and Standards to Member-at-large Digital Communications to reflect changing responsibilities. The 

 is available electronically, conference papers are submitted electronically, and the Society creates webcasts for some sessions. An ASMS member can vote electronically for the elected positions, pay dues electronically, and update member information via the web. One of the challenges facing the member elected to this new position will be how all such electronic information should be archived, and how long it should be available, and to whom. It might seem that there is no substantial investment in maintaining electronic information, but there is in fact a substantial commitment to secure storage, ready access, and the availability of information in a digital form that will remain widely readable into the indefinite future.  

It is time to revisit the web to survey the resources available to those interested in learning about MS, at all different levels. We begin with the obvious disclaimer that this overview cannot be comprehensive, but must be selective. Resources on the web are extensive, but there is no guarantee that they are not duplicative, and no assurance that the information presented is accurate. Of course the same is true for printed media, but while a poorly composed book can reach an audience of a few thousand readers, a website is available at the stroke of a key to millions. The price for accessibility is the speed with which inaccurate information can be dispersed, and the risk that such information cannot be traced or recalled. For commercial and other purposes, there are many sites that do not offer original content, but present collections of material from other sites, sometimes presented without proper attribution. The accuracy of the information on such aggregator sites must be evaluated carefully. Finally, and in contrast to the permanence of the printed page, the web is transient. Not only can the content of a web page change, but the availability of links also changes. The information might still exist on the web, but it is not clickable and can be accessed only with the full URL. As an example with which readers are surely familiar, search engines may index a site that is changed by the time a reader attempts to view it. Examination of cached sites and looking through the Wayback machine (2) provides some rudimentary tools to uncover information from the web that is not current.  

In this not-so-random column, first, I have chosen specifically to include mostly noncommercial sites. This choice is not only for reasons of objectivity (as I would invariably fail to include all or to treat all equally), but also for reasons of practicality, as these are the sites (often with professional resources) that tend to update more often. Additionally, some require subscription or the entry of information that is properly of interest for sales and marketing. Second, I have not included tutorial sites in MS, such as might be associated with particular university courses. A future column will look at these sites, but will require a broader perspective than can be gained here. In particular, there is some reason to be concerned about overt errors that might appear on such sites. Textbook errors have long been a particularly vexing issues, as they cannot be corrected until the authors are made aware of them (through reviews or individual letters), and not until a new printed edition appears. One can assume that a website might be more easily corrected, but the proper and courteous process to do so is not yet agreed upon. In this column, we'll first visit websites for several professional societies and MS discussion groups, and then tour Murray's MS site (this does have a commercial connection) to show you the breadth of modern web tools. Finally, we will finish with a peek into a few sites on which you might encounter unexpected information. After all, it's often the accelerated threaded search from site-to-site that's the web parallel to informed and relaxing browsing through books in a library  

Websites for professional societies provide a wealth of information not only to members, but to others interested in MS. As an example, consider the website for the International Mass Spectrometry Foundation (3). This Foundation oversees the International Mass Spectrometry conferences. Dates for conferences through 2016 are described. Conference reports are available on the site for meetings for the 18th Conference (2009) back to the 14th Conference in 1997. This snapshot of the conference provides metrics on attendance, and oral and poster presentations, and also recaps award winners for that particular conference. For the most recent conference, the abstracts for oral presentations (day by day), poster presentations, and keynote lectures can be accessed from the site. A separate link provides addresses and contact information for authors. A link to the most recent Bremen conference provides the ability to search through the text of the program (4). Commercial sponsor documents are available for download. For some sessions, and often through subscription, the full slide presentations for plenary and keynote lectures are available. In summary, the modern website provides the definitive record of the event, starting from well before the event to after it has been completed.  

The websites of professional societies in MS around the world are generous in linking to each other. A list of societies is provided in Table I. Any given compilation found on the web may or may not link directly to the current hosting site, so user caution and patience are appropriate. Aside from information relevant to each individual society, these sites host many common links and often provide much the same basic information. However, each site can provide something unique. For instance, the Indian Society for Mass Spectrometry has published a directory of mass spectrometers, last updated in 2003. This listing by instrument throughout the country seems fairly unique and would be invaluable when attempting to identify and correct an instrumental problem via communication with users rather than directly with the manufacturer. The Dutch Society for Mass Spectrometry has a special tab for "History." Click on that link and you will be brought to a talk called "History of Mass Spectrometry in the Netherlands" by M.C. ten Noever de Brauw. All of the slides for this talk are presented, and describe a fascinating history of MS, with rare archives and photographs. Pay no attention to the definition of a mass spectrometrist found on slide 65. The website for the American Society for Mass Spectrometry provides free to all readers the "What is Mass Spectrometry?" resources, and these are duplicated on many sites devoted to specific university courses on MS or analytical spectroscopy. 

Table I: A compilation of professional mass spectrometry societies as listed on the website for the International Mass Spectrometry Foundation. At any given time, a small percentage of the accompanying websites may be inactive or out-of-date.			

Professor Kermit Murray of Louisiana State University oversees a "Mass Spectrometry blog" (5) described as "a web log of MS sites and other links and items of interest." This site is an excellent example of just how powerful combined web sources can be in providing descriptions of MS at all levels, from basic tutorials to the latest research and applications. Accordingly, we'll step through a few features (randomly because I have a personal interest in some of them and admit to not fully understanding others). The following assumes that the appearance of the site does not change radically in the time between composing this column and its appearance in print format.  

On the right of the blog site is that invaluable "search" tool. What appears on the screen from the blog is limited. One can scroll down through individual windows, but presumably (hopefully) the search for results from the blog will automatically include past archived material that exists on the server without a direct appearance in the current window. The window below promises access to archives that extend from January 2002 through May 2004, although none of these links seem to be active. The blog does link directly to various Wiki feeds relevant to MS, and to "feeds" from journals of MS and analytical chemistry in general. Links to podcasts would qualify as "other items of interest" as these might not be related directly to MS. Murray's blog also provides you with presorted collections of tags to groups (MS and otherwise), companies, resources, software, news (via a Google search on "mass spectrometry"), and finally, to other blogs on MS (unfortunately, not active). A selection of articles from several current issues for MS journals is included, as well as a fascinating subsection of the site that details the latest changes to the Wikipedia entries relevant to MS. Murray also includes a section on job openings currently available for mass spectrometrists. 

Murray's selections of news on MS and allied topics are laid out in the righthand window of the blog site. This author's understanding is sometimes stretched (expanded), such as the information on the "mass spectrometry word cloud" of October 18, 2009. This scrolling window contains direct links to video and animations, as well as slide shows. On April 3, 2009, I learned about Slideshare, and clicked through a slide presentation (complete with transcript) on the basics of tandem MS (MS-MS). Looking a little deeper on the slideshare site for this particular resource, I found that 7925 users had viewed all or part of this presentation, and that it had been downloaded 251 times. There was one comment for this particular contribution. Each of Murray's news entries provides for comments. Most had none, and a few had only a few. One would assume that more abundant comments and ample discussions appear on the LinkedIn site devoted to MS (see Murray's link on April 2, 2009). It is not characteristic of mass spectrometrists to be so quiet.  

From spending time scrolling through the site, it should be clear that it takes an extraordinary amount of time and effort to create and maintain a site as useful as Murray's blog. It is a service to the community that is probably not given full credit by colleagues and users. Several hours a day could be devoted to its maintenance and its updates, and that level of effort is beyond the level of an individual. We are led to two conclusions: There will be a time delay and a level of incompleteness in any aggregator site, and we should exhibit both patience and offer assistance; and responsibility for such sites, including the still larger and more complex sites of the future, will either be drawn from commercial groups, or groups of professionals dedicated to the field of MS. The latter concept is an interesting parallel with the small groups of professionals that formed the first users for MS in the 1950s and 1960s. 

Finally, an explorer of the web might find sites that unexpectedly link to MS, or that contain novel information. No links are provided here; this task is designed to be completed by readers. Knowing a bit of history, and with the ability to surf the web, you might find the site that describes "Isotopes and Aston's Mass Spectrograph" in the Museum of the Cavendish Laboratory. You might find the Wasserburg mass spectrometer, as acquired by the National Museum of American History (part of the Smithsonian museums). You might also find a "Vintage Mass Spectrometer" at the Schenectady Museum. One can visit the site of a bookseller such as Amazon to create a list of all books that include the term "mass spectrometry," or to determine which textbook is listed as bestselling on that site, or which is available as a download as a Kindle book. There are too many sites to mention that include lists of "trivia" questions related to MS, mentions of MS on television shows, and 2840 sites that hit the search for both "mass spectrometer" and "Star Trek." Photography has gone digital, and images (and video) are now as easily uploaded to the web as is text. One can search the web for images related to search terms, and by so doing uncover a website that celebrates the visit of Professor R.G. Cooks to the Dalian Institute of Chemical Physics in July 2009 (6), or images from his attendance at the International Mass Spectrometry Conference in Bremen in September 2009 (7). Finally, while electronic translation abilities are still primitive, especially for constrained technical language, a web resource can be translated from one language to another. 

Mass spectrometers are connected to each other in our laboratories over an intranet of some sort. Various research laboratories within an individual company are interconnected, and data streams from site to site, as needed. Traditional print publications describing research and applications of MS have evolved into complementary print and electronic forms, and the continued shift into the web and virtual worlds seems assured. The last great conservative bastion is the use of web resources and technology in teaching and education. There has been much research, much debate, more than a few grand schemes, and, not surprisingly, a few experiments that concluded not so well. MS sites on the web always will be a display of diversity, and increasingly will incorporate images, video, animations, comments, and discussions as well as text. Their links will become more extensive and multilayered. There will be more direct access to original data as well as its interpretation. The web can host both a more visible reasoned debate as well as more acrimonious disputes. Professional MS societies might wish to consider means to certify accuracy and completeness for information as a service both to their members but also to the general public.   

 maintained a collection of reprints for his scientific publications. The half-life of a publication is short, and it became tiresome to maintain paper reprints in the file. Moving boldly into the future a mere 10 years late, he recently scanned all past publications into pdf format, and recycled the paper. The oeuvre is now strictly electronic, and the author can now be wiped out in a single hard-drive failure. Contact the author (who still maintains a flesh-and-blood existence) at 

. The author is solely responsible for the contents of this column.  

http://www.imsc-bremen-2009.de/imsc2/submissions.php?navid=14

http://english.dicp.cas.cn/ns/icn/200909/t20090907_36392.html

http://www.imsc-bremen-2009.de/imsc2/bildergalerie/galerien/Friday__Sep__4__2009/index.php?navid=

Mass Spectrometry Forum

The Web contains a diverse collection of resources (tutorials, texts, images, videos, and animations) of interest to mass spectrometrists.

A Random Walk Through the Web of Mass Spectrometry

Raman microscopy has been attractive because of its ability to characterize materials on a spatial scale commensurate with optical microscopy. Typically, the lateral spatial resolution is quoted as determined by the Airy disk (1), which is 1.22λ/NA. λ is the wavelength of the illuminating light, and NA is the numerical aperture, which is equal to 

 is the index of refraction of the medium (1.0 in the case of air) and θ is the angle subtended by the optics. However, it is interesting to this aging Raman spectroscopist that in recent years, the quoted spatial resolution of a Raman microscope has been getting smaller and smaller as more instrument manufacturers enter into the business of Raman microscopy. In fact, the spatial resolution has not been improved; one cannot redefine the laws of physics. We believe that the equations selected to "define" spatial resolution have been selected to give a smaller number. In the end, the only true way is to test the spatial resolution is to define well-controlled measurement conditions. This will be the focus of this column. 

For a start, we can say that the Airy disk description cannot be correct for a Raman microscope. The Airy disk assumes uniform illumination of the focusing optic, and the laser profile is anything but. In addition, in some instruments, the Gaussian laser profile is not well matched to the aperture of the focusing objective. At any rate, this article is going to concentrate on the depth resolution of the Raman microscope. Optical calculations for depth resolution of an optical microscope state that the it is proportional to λ/(NA)

. The essential point to recognize is that the spatial resolution of any Raman microscope depends upon the detection optics as well as the focusing optics. How effectively does the optical system collect the Raman signal excited in the laser focal spot, and reject the signal from the surrounding volume that is illuminated by the laser but not in focus? 

The essential question is how to evaluate the depth resolution experimentally. Historically, people have used a piece of a polished Si wafer for these tests. But this material was chosen more for the repeatability of any measurement of its signal rather than its appropriateness for answering the questions of depth resolution. The advantages of using silicon are that it is readily available, the optical properties of a polished surface are perfect, and the signal generated from a particular orientation using a particular laser wavelength and instrument parameters is reproducible. This last point is not to be taken lightly. Raman scattering is a "single-beam" technique, and as such there are many factors affecting the absolute magnitude of the signal. Using silicon as the test sample enables one to track the performance of an instrument and to compare signals taken under different conditions (for example, with different gratings or with the sample in a different orientation) because the signals are reproducible when a set of measurement conditions are controlled. This is a point that I have made in one of my earlier columns.  

After we had determined that silicon provided a good test sample, it was realized that it could be used to test the depth resolution. However, this is where its use becomes a bit questionable. Figure 1 indicates what the problems are. Figure 1a shows the beam path for the laser focused precisely at the surface of the silicon. After the light hits the surface in focus, it will continue to penetrate the silicon as it diverges. How far it penetrates depends upon the wavelength of the laser. It turns out that in the visible part of the electromagnetic spectrum the depth of penetration is not negligible. At 633 nm, the wavelength of the HeNe laser, the light will penetrate 3 μm, but this number changes rapidly with wavelength. At 785 nm it is about 12 μm, at 532 nm it is 0.7 μm, and in the UV it can drop to 5–10 nm. What Figure 1 shows is that when the laser is focused on the sample surface, this point is reimaged on the confocal hole and entrance to the spectrograph. However, the light that diverges from the point of focus also can generate Raman light, and the path that light will take through the collection optics is shown in orange. For the point selected in the figure, the Raman light will be focused before the slit and off-center. As such, it will have little chance of passing the entrance hole. But the question is, how much is little? If we excited with 785-nm radiation, Raman light can be generated as far as 12 μm below the surface. How well is this light rejected? What about light that is 6 μm below the surface? How can this be quantitated? 

Figure 1: (a) Laser beam path focused at the silicon surface, (b) microscope stage scanned in a vertical direction. 			

The solution was to perform a "Z scan." Figure 1b illustrates what is happening in such a measurement. In this case, the microscope stage is scanned in the vertical direction so that the surface passes through the plane of focus of the laser beam. Then you can record the intensity of the spectrum as a function of 

 position. The problem with this measurement (aside from the fact that the depth of penetration is far from infinitely narrow) is that as the sample is moved away from the focal plane, a larger area is being excited by the laser beam. So the signal that is measured is a convolution of losses due to the surface not being in focus, and gains due to a larger volume being illuminated.  

Is there a better way to determine the confocal properties of a Raman system? Having thought about it for a while, we decided that microscopic polymer beads might provide a solution. In this case, when the microscope stage is raised, the lateral extent from which the signal can be generated is controlled by the particle size. We will show measurements taken from Si as well as polymer beads (2 μm and 0.5 μm diameter) using the 532 nm laser beam (minimizes the depth of penetration of the laser into silicon so that the effect of the lateral illumination of the surface can be separated from the volume effect). 

Figure 2: Depth profile of (001) Si. 532 nm excitation, 25 Î¼m confocal hole.  FWHM = 2.6 Î¼m.			

Figure 2 shows a depth profile of the silicon signal when the confocal hole was set to 25 μm. The profile was fitted to a function whose full width at half maximum (FWHM) was about 2.6 μm, which is considerably more that the depth of penetration of the light. To gain insight into what is happening, a series of confocal measurements were made on a different day. In this case, the focal plane was identified visually in the monitor, and the confocal hole was opened from 0 to 300 μm in 10-μm steps, at 

 = –1, 0, 1, and 2 μm relative to the visual focus. These results are shown in Figure 3. Figure 3a shows the confocal scans displayed showing the real observed intensities; Figure 3b displays the same data, but in relative units. 

Figure 3: Confocal profiles of silicon signal recorded at Z = -1 (red), 0 (green), 1 (blue), and 2 Î¼m (violet), with the 0-Î¼m plane defined visually. (a) profiles of the absolute intensities; (b) profiles scaled so that the maxima for all traces were identical.			

 = –1 and 0 μm, the signal was identical until the hole was opened to more than 55 μm. It actually is not easy to determine what the exact focal plane is by visual inspection. But, if we establish a criterion of best focus as the plane in which the confocal trace turns on most rapidly (relative to its maximum intensity, see Figure 3b) then, in this case, 

 = –1 μm appears to be the better choice for focus. What is also interesting is that at 

 = –1 and 0 μm, the signals are higher than at 

 = 1 and 2 μm for hole sizes less than 125 μm. But for larger hole sizes, the detected signals are greater for 

 = 1 and 2 μm (Figure 3a). Because we know that there cannot be a volume effect in this sample at this wavelength, this has to be due to the increased illumination area. In addition, Figure 3b shows that as the hole size is increased, it takes longer to saturate the signal, which is consistent with a larger image at the hole because the sample plane is not the laser focal plane.  

We have recorded confocal scans (varying the height of the sample and the hole size) of polystyrene spheres with nominal diameters of 2 and 0.5 μm. Figure 4 illustrates the depth profiles of the scans of the 2-μm polystyrene spheres as the hole size is varied between 25 and 105 μm. Each of the profiles was fit to a symmetric function. The FWHM varied only between 4.7 and 5.1 μm. 

Figure 4: Confocal profiles of a 2-Î¼m sphere of polystyrene (using the 620-cm-1 band) as the sample plane (Z) and the size of the confocal hole is varied.  The left side of the figure is a full representation of the data, while the right side shows horizontal cuts through the data. 			

Similar measurements were made on 0.5-μm polystyrene spheres. Those depth profiles are shown in Figure 5. For a particle of such a small volume, the polystyrene signal was weaker than in the larger beads, which accounts for the noise in the depth profiles. In this case, we do observe a systematic sharpening of the profile as the confocal hole is closed. This can be visualized more readily by displaying the profiles with the same maxima, as seen in Figure 5b. 

Figure 5: Depth profiles of 0.2-Î¼m polystyrene sphere recorded with the confocal hole set to 200, 175, 150, 125, 100, 75, 50, and 25 Î¼m (top to bottom).  (a) Profiles shown with absolute intensities, displaced vertically for viewing; (b) profiles overlaid with same min/max intensity to visualize the change in profile width.			

The right-hand part of the figure shows that the intensity starts to appear at the same 

 position (at the top of the sphere), but when the hole is enlarged, the intensity drop-off at the bottom of the sphere is delayed, resulting in a larger profile, and an apparent displacement of the peak position.  

These profiles were also fitted, resulting in Table I. 

Table I: Depth profiles of 0.5-Î¼m polymer bead			

Results: Silicon Under 0.-5 -μm Polystyrene Sphere

These measurements held a surprise for us. The 0.-5 -μm polystyrene spheres had been deposited on Si (whereas the 2-μm spheres were on Al). So we were able to plot the depth profiles of silicon under a sphere and compare them to profiles of silicon. Figure 6 shows the depth profiles under the bead. 

Figure 6: Depth profiles of silicon signal when measured under a 0.5-Î¼m polystyrene sphere. Confocal hole (from top to bottom) was 200 to 25 Î¼m in 25-Î¼m increments. Left: profiles in absolute intensity; right: profiles displayed with the same min to max. 			

Figure 7 shows the depth profiles, with a 25-μm hole, of bare silicon versus silicon under a bead. Note that these profiles were measured sequentially on the same day under identical conditions. We observe several things in this measurement: The signal under a bead was higher than on the bare silicon, and the FWHM was tighter under the bead. 

Figure 7: Sequential measurements of depth profile on silicon, measuring through/under a 0.5-Î¼m polystyrene bead (red) versus measuring on bare silicon (purple). The confocal hole was 25 Î¼m in both cases. 			

The conclusion to be drawn is that the bead is acting as a solid immersion lens that increases the effective NA of the focusing, which, in turn, decreases the spot size, and increases the collection efficiency! 

Because of the confocal nature of Raman measurements there are a number of applications that can be addressed uniquely with this technology. They include diffusion of one chemical species into or through another, and the sharpness of an embedded interface. Once anyone tries these measurements, it becomes clear that below-surface confocality is compromised with the objectives that usually are used. This was recognized quite early in a technical note (2) but followed up on in more detail in publications from a number of laboratories (3). In particular, these publications deal with the effect of spherical aberrations, introduced when using dry (metallurgical) objectives to study materials below the surface. These effects have not been discussed in this column, but it is clear that if one is monitoring a subsurface interface, one must include consideration of both effects. The best objective for the sample must be selected, and a confocal hole consistent with the desired spatial resolution must be used.  

The experimental determination of the confocal depth from which Raman signal is collected is determined by instrumental parameters, sample parameters, and, sometimes, interactions between the sample and the optics. While a sample like silicon is ideal for reproducibility, it is not ideal for defining depth resolution because as the sample is defocused, a larger illuminated area provides a larger signal. By making measurements at 532 nm, where the depth of penetration of the light is less than a micrometer, we were able to document this effect. When the confocal hole was 200 μm, the FWHM of the depth profile was greater than 6 μm, but at 25 μm the FWHM was not much greater than 2 μm, which was close to the number recorded for the 0.5-μm sphere of polystyrene. Because the 0.5-μm spheres of polystyrene were mounted on silicon, it also was possible to compare silicon depth profiles measured on bare silicon and silicon under a sphere; under the sphere, the silicon signal was stronger and the depth profile was sharper, leading to the conclusion that the polymer sphere is functioning as a solid immersion lens. The bottom line is that confocal measurements on a Raman microscope must be made with an understanding of the optical behavior of both the instrument and the sample, and their possible interactions. As a follow-up to this discussion, my next column will consider the role of sample characteristics in the choice of sampling optics and the consequential spatial resolution. In particular, we will consider the interesting situation in which one needs to minimize spatial resolution to achieve a macroscopic signal representing the true sample average.

 is the Worldwide Raman Applications Manager for Horiba Jobin Yvon (Edison, NJ). She can be reached by email at 

Principle of Optics: Electromagnetic Theory of Propagation, Interference and Diffraction of Light

, 7th Edition (Cambridge University Press, Cambridge, UK, 1999). 

(2) "Raman chemical depth profiling with automated Z focus" (ISA Technical Note R-27, June 1998). 

Molecular Spectroscopy Workbench

The experimental determination of the depth resolution of a Raman microscope is described.

Depth Resolution of the Raman Microscope: Optical Limitations and Sample Characteristics

The origin of the characteristic spectral lines of the elements in the optical and X-ray regions of the electromagnetic (EM) spectrum is all about electron transitions. Excitation of outer-shell electrons to an excited state and their subsequent return to ground state results in electromagnetic radiation in the optical (ultraviolet–visible–infrared) region with wavelengths from about 100 to 1000 nm. Excitation of inner-shell electrons produces EM radiation in the X-ray region with wavelengths from about 0.01 to 10 nm. 

Figure 1 shows the two X-ray spectral lines and some prominent optical emission lines of magnesium (Z = 12). Note the logarithmic scale for wavelength. These spectral lines are listed by both wavelength and energy in Table I. Some relative spectral line intensities are also shown (2). The designation "U" for the optical emission lines refers to spectral lines of the neutral atom and "V" for the singly-ionized atom. The number that follows indicates the relative intensity with "1" generally being the strongest spectral line, depending upon the excitation condition. 

Table I: Prominent spectral lines of magnesium 

Some additional comments on Figure 1 and its associated table:   

The "U" and "V" nomenclatures are somewhat antiquated, included here solely to preserve the connection to the source, the MIT Wavelength Tables (2). Currently, spectral lines in the optical region are designated as follows: A capital "i" (I) for the neutral atom, II for the first ionized state, and so forth. For example, today we write Mg I 285.21 nm and Mg II 279.55 nm in referring to these spectral lines.

Please note that the X-ray and optical relative intensities of Table I are on different scales (3). 

Figure 1: Spectral lines of magnesium in the EM spectrum.  

The electron energy levels of atoms are grouped together based upon the principal quantum number 

. These groups differ from each other by about a factor of 10 in binding energy. Figure 2 shows a generic electron energy level diagram for the first three levels (n = 1, 2, 3) and the associated sublevels. The standard electron energy level designations are shown along the left side along with those used in X-ray fluorescence. On the right side are shown the associated quantum numbers: the principal quantum number "n," the orbital angular momentum quantum number, "l," and the magnetic quantum number "m." This diagram sets the stage for those to follow. 

The electron energy transitions resulting in X-rays are easier to portray. The innermost electron shell is called the K-shell; the next farthest out from the nucleus is called the L-shell, followed by the M-shell, and so forth. 

To produce the characteristic spectral lines in the X-ray region, the excitation energy (E) must be such that it is equal to or greater than the energy required to remove an electron from its shell, known as the "binding energy" (EB). That is, we must have: 

An outer shell electron will "drop down" to fill the void created in the inner shell and there is a certain probability that an X-ray characteristic of that atom will be emitted. This process is called X-ray fluorescence (XRF) and is shown in Figure 3 (4). The characteristic X-rays are called "K lines" if they result from an electron filling the innermost, or K-shell, and "L lines" if they result from filling the next electron shell out, the L-shell. Figure 4 shows an X-ray energy level diagram for magnesium. 

In the optical emission spectroscopy (OES) process the electron transition is somewhat different. In this case, the excitation energy must provide the necessary energy for the following processes:  

Move an electron of the neutral atom from its ground state to an excited state or,

Remove an electron from the atom (ionization) or,

Excite an electron of the ion into an excited state of that ion.

Upon the return of the excited electron to lower energy levels, energy in the form of light is given off. 

Figure 4: Energy level diagram for magnesium (X-ray). 

Of course, equation 1 still holds true. Or does it? In the case of ionization, it certainly does. For the first and third processes noted earlier, a better formulation might be E ≥ E

 is the appropriate excitation energy (atomic or ionic).  

An electron energy level diagram for magnesium is presented in Figure 5. It shows the ground state of the neutral Mg atom, the first and second ionization potentials and several of the prominent spectral lines in the optical wavelength region. 

Figure 5: Energy level diagram for magnesium (optical). 

Probably the first thing the reader will note in comparing Figures 4 and 5 is the vast difference in energy scales (

axis). Note that the binding energy of the Mg K-shell electrons is about three orders of magnitude greater than the OES ionization–excitation energies. 

In the XRF process the binding energy of the innermost electrons is the critical quantity as regards our subject here. Taking the example of the Mg atom, if the energy of the incoming excitation is insufficient to eject a K-shell electron, there are no characteristic X-rays produced.  

Consider, for example, an 1100-eV photon impinging upon the Mg atom. Where does incoming energy go? The authors have never seen this question raised in traditional texts on X-ray fluorescence (5–7). Perhaps an L-shell electron is ejected, or an M-shell electron is ejected, producing optical transitions. Compton scattering and heat would seem to be other possibilities. 

In the OES process, there are two quantities of interest: The first is the "excitation energy" to transit an outer shell electron to a short-lived (about 1 ns) "excited state." The "decay" of the electron in the "excited state" can take several paths, resulting in the emission of different spectral lines. The second is the ionization energy (or ionization potential). Only the ionization energy has a direct correlation with the binding energy of XRF as regards removing electrons from the atom. 

There is no clear way to combine the two diagrams (Figures 4 and 5). However, let's just superimpose Figure 5 on top of Figure 4. The result is the very odd-looking Figure 6. 

Figure 6: Energy level diagram for magnesium (combined OES and XRF). 

Figure 6 is certainly problematic for several reasons, all of which relate to the 

The mixing of log and linear energy scales on the 

-axis is clearly unconventional from a mathematical point of view.

To add insult to mathematical injury, the graph begins at +1300 eV, goes to zero and ends at about 25 eV. This is most irregular.

In the OES process, the 3s2 energy level (M shell) is considered the zero level for the various excited levels and ionization of the atom (8,9). This is not a problem when restricted to the OES transitions. However, this is clearly not acceptable for the XRF energy diagram since it would mean that these electrons are not bound to the nucleus! 

However, this is a tutorial intended to show the similarities between X-ray and optical emission electron transitions. As such, with proper explanation, this figure is perhaps of pedagogical interest despite the mathematical "irregularities." 

The transition of electrons between their energy levels provides the basis for elemental spectrochemical analysis. Electrons are excited to higher energy levels or leave the atom (both possible in optical transitions, but only the latter in XRF). The same electron may return to a lower energy level (optical) or others may take up the vacancy  (X-ray).  

, a retired Senior Researcher at Usiminas Steelworks R&D Analytical Division, is presently an ICP Application Chemist for SPECTRO SSA, Brazil. As a hobby, he enjoys reading subjects related to anthropology and playing the piano. E-mail: 

, a physicist by training, has some 30 years of experience in elemental spectrochemical analysis (OES and XRF). He is currently a consultant in this area from his home in Atibaia, São Paulo, Brazil. His other interests include mineralogy and history of science. Occasionally, he still plays the blues harmonica. He can be reached at 

 (MIT Press, Cambridge, Massachusetts, 1969). 

 (Lawrence Berkeley National Laboratory, 2001). Online at  

Atomic Perspectives

This article is a follow-up to an earlier "Atomic Perspectives" column on the spectral lines of hydrogen (1).

Electron Transitions in Optical Emission and X-Ray Fluorescence Spectrometry

Spectroscopy-based instrumentation is an important sector of the analytical and life science instrument industry. Over the years, the overall industry, and spectroscopy in particular, have proven to be very resilient businesses, servicing the scientific requirements of a vast array of applications in every conceivable industry. Growth has been fairly steady for the last decade, averaging approximately 6% each year until recently. Compared to the double-digit growth seen during the 1980s and 1990s, most would agree that our industry is both mature and stable.  

The demands of the life science marketplace, environmental concerns, the search for new materials, and security and defense concerns are important driving forces. The regional focus has now shifted to developing countries, especially in Asia where growth far outpaces that of North America, Europe, and Japan. But even as application requirements evolve and markets ebb and flow, 2009 proved to be a major challenge for every industry, including the laboratory instrument industry, and indeed, the spectroscopy segment as well. The global economic recession of the last 18 months was the worst the world has seen since the Great Depression of the 1930s. Fortunately, the bad economics news appears to be ebbing in terms of global economic development, although unemployment statistics are quite glum and will likely stay so for most of 2010. Unprecedented government intervention, in the form of expansionary fiscal and monetary policies and economic stimulus programs, has provided the basis for a renewal of global economic growth. 

Figure 1: 2009 spectroscopy market by sector ($8.5 billion).			

With these developments in mind, and after reviewing various industry data for 2009 and the prospects for 2010, 

 (IBO), a publication of Strategic Directions International, Inc. (SDi), a consulting and market intelligence firm specializing in this market, has estimated the total worldwide market for analytical and life science instrumentation revenues at slightly under $37 billion for 2009. This represents an unusual decrease in instrument industry revenues of 3.0% from 2008, a result not seen since the early 1980s. SDi now forecasts that the analytical and life science instrument industry will rebound in 2010 and will expand at the rate of 5.2%. While not particularly exciting, it is a level of expansion that would be very welcome in many sectors of the global economy. Furthermore, 2011 is expected to return to the more familiar growth rate of about 6.5%. So the future looks immeasurably better. 

Although one could assume that an economic recession might not be the best time to become acquisitive, 2009 certainly saw an increase in merger and acquisition activity in the analytical and life science instrument industry. Agilent Technologies and Danaher Corporation announced the two biggest deals, which also involved companies deeply involved in spectroscopy markets. Agilent Technologies announced its intention to acquire Varian, a company with significant spectroscopy capabilities, including its key nuclear magnetic resonance (NMR) business, of special interest to Agilent. Regulatory approvals have been received, but include the need to divest certain businesses including Varian's inductively coupled plasma–mass spectrometry (ICP-MS) and gas chromatography (GC)–MS-MS product lines. At $1.5 billion, this was the biggest transaction for our industry in some time. Danaher also announced a twofold transaction, which would bring into a single entity the mass spectrometry businesses of the MDS Analytical Technology Division and Life Technologies' mass spectrometry business that came with its acquisition of Applied Biosystems in 2008. Combined, this transaction is valued at $1.1 billion. 

Table I: Spectroscopy product growth prospects for 2010			

Interest in aftermarket products and services, including such items as consumables (especially chemicals), software updates, and accessories is very keen. As the installed base of instruments grows and the market matures, the aftermarket becomes especially important. In fact, in selected technology areas, the aftermarket is a very significant product segment that is growing faster than instrument system revenues and is often more profitable. It is also true that for some technologies, instrument systems (hardware) actually might be declining in unit sales while the aftermarket provides at least minimal growth. So it should be noted that all the revenue projections in this article include the total mix of revenue sources: instrument system sales, aftermarket, and service. 

The spectroscopy market can be segmented into three distinct sectors: molecular spectroscopy, atomic spectroscopy, and mass spectrometry. The applications of the techniques included in each sector and their prospects differ significantly, especially within each sector. Worldwide revenues for all spectroscopy segments is estimated at about $8.5 billion for 2009, or about 23% of the entire laboratory analytical and life science instrument market. 

The overall spectroscopy market is estimated to have declined 3.4% in 2009, a disappointing level of performance compared to the almost 7% increase in 2008 over 2007. SDi projects spectroscopy market growth in 2010 to exhibit renewed strength and increase at the rate of 7.2%. Such a rebound more than makes up for the decline experienced last year and puts the market at almost 4% higher than that in 2008. In addition to the positive affects of a general return to global economic growth, government stimulus programs in many countries, but especially in the U.S., China, and the EU, will give a real boost to the instrumentation industry. Likewise, various industrial markets such as metals, semiconductors, automotive production, and chemicals will be reinvigorated. In combination with continued growth from academia and government expenditures, the prospects look very good for both 2010 and 2011. 

Figure 2: 2009 molecular spectroscopy market by technique.			

Molecular spectroscopy techniques account for almost 40% of the spectroscopy market, although many diverse techniques are represented that address a wide range of applications in almost every industry category. Atomic spectroscopy is the second largest sector at over 33%, while MS represents 27% of spectroscopy revenues. Please note that liquid chromatography (LC) and GC front-ends to mass spectrometers, and their associated aftermarket and service revenues, are not included in this analysis. 

In 2009, the global molecular spectroscopy market contracted by 3.3% to about $3.4 billion, but it is expected to see a solid rebound in growth in 2010 of more than 8% growth, with similar growth to follow in 2011. The relatively high proportion of demand that comes from government and academic laboratories, as well as its association with life science applications, has helped to insulate the molecular spectroscopy market significantly from the impact of the deep recession. 

Figure 3: 2009 atomic spectroscopy market by technique.			

Most molecular spectroscopy techniques experienced approximately flat growth in demand in 2009, with the major exceptions being color measurement, which saw a whopping 25% drop, and Raman, which showed better than 6% growth. The color analysis market is not significantly tied to the rest of the molecular spectroscopy market, either in terms of end-users or vendors, and currently is experiencing decreased demand of more than 30% from its peak in 2007 of around $375 million. Even with a return to growth in the high single-digits, it will be some time before the color analysis market approaches its previous high-water mark. 

Raman spectroscopy was the only molecular spectroscopy technique to see significant positive growth in demand in 2009, due almost exclusively to the explosion in demand for handheld and portable instruments. Portable infrared instruments saw even stronger growth than handheld Raman, but were not nearly enough to offset a small percentage drop in the much larger overall infrared spectroscopy market. To a lesser extent, the same was true of handheld near infrared (NIR) demand in the near-infrared market. In 2010, growth in demand for Raman easily should be back in the double-digits as the technique grows into a more mature market. The more broad-based infrared and molecular spectroscopy markets will see annual growth in the 6–8% range. 

Figure 4: 2009 MS market by technique.			

The NMR market, which accounts for nearly 30% of the molecular spectroscopy market, saw a slight pullback in demand in 2009 as many end-users cancelled or put on hold such big-ticket orders. However, the large installed base helped drive demand for service and aftermarket products, such as probes and cryogenic gases, which still grew slightly. Some of the held-back demand for systems in 2009 will be converted to orders in 2010, thus leading a return to double-digit annual growth, which NMR often achieves. Because of the fact that such a high proportion of NMR demand comes from academic and governmental laboratories, the stimulus spending initiatives put in place over the past year will undoubtedly have a disproportionately large impact on the NMR market. 

The ellipsometry market, which is tied heavily to the semiconductor and electronics industry, has seen sliding demand since 2007. However, the technology industry, which looks to be leading other industries out of the recession, should drive a significant rebound in demand somewhat more quickly than other molecular spectroscopy techniques. 

Figure 5: Spectroscopy market by product segment.			

Both the UV–vis and polarimeter and refractometer markets saw very steady demand from 2008 through 2009, and should return to mid-single-digit growth in 2010. Both of these markets consist of instruments that are generally lower in cost, most of which are under $10,000. As relatively small-ticket items, they were less of a consideration to end-users who had to cut capital budgets in 2009. Demand for microvolume UV–vis, which is related primarily to the biotechnology industry, will continue to see very strong double-digit growth, more than offsetting generally flat to declining segments of the UV–vis market such as single-beam and visible-only instruments. 

Fluorescence instrumentation is also tied to the life science market, but standalone instruments continue to exhibit sluggish demand. This is primarily because of microplate readers that often include a fluorescence detection mode and offer productivity improvement potential for many applications. 

One very new area within molecular spectroscopy is terahertz spectroscopy, which currently represents less than $10 million in annual global demand but is poised to develop into another significant segment of the molecular spectroscopy market. Several vendors have recently introduced, or are about to introduce, new terahertz systems, which initially are being targeted at the pharmaceutical and hospital–clinical markets, although significant potential lies in other areas such as security. 

Figure 6: Estimated growth by spectroscopy sector for 2010.			

A very large number of suppliers are involved in the design, manufacture and sale of molecular spectroscopy instrumentation. Overall market share position, however, tends to be a function of dominance in particular product segments. Bruker is the overall leader based upon its very significant NMR–electron paramagnetic resonance (EPR) business, as well as its involvement in infrared technology. Varian is also a major NMR vendor with a leading role in UV–vis. Thermo Scientific and PerkinElmer, in the third and fourth positions, offer a wider array of molecular spectroscopy instruments, but with very strong positions in infrared and UV–vis. Foss has recently moved into fifth position as the result of increased interest in food and agricultural for which its NIR and IR instruments are in demand.  

Specialty suppliers like Hach Danaher in UV–vis for water analysis and Horiba in Raman and fluorescence are key participants in their respective areas of focus. Color measurement companies such as X-Rite, formerly the number five supplier, and Datacolor have both suffered from diminished demand in the auto, packaging, textiles, and clothing sectors.  

Following on a disappointing 2009, the atomic spectroscopy market is forecast to grow at 5.3% for 2010. Last year, SDi speculated that environmental applications would help to keep demand for several of the relevant atomic spectroscopy techniques stable in 2009 due to rising international interest in green technologies and environmental safety. Unfortunately, this turned out to be only partially the case. Although particular elemental analyzers and sum parameter instruments like total organic carbon (TOC) did continue to achieve growth, the workhorse environmental techniques of atomic absorbance (AA) and ICP spectroscopy experienced significant declines in 2009, as industry and environmental testing laboratories managed to achieve greater workflow with their current instrumentation and delayed replacements of older instruments. 

AA and ICP are among the larger markets within atomic spectroscopy, and their decline contributed to the overall decline of 4.3% in this technology segment in 2009.  Although AA and ICP will achieve positive growth for the coming year, these markets are forecast to fall short of the 2008 demand. ICP-MS, the most sensitive atomic spectroscopy technique, remained essentially flat in 2009 and should experience much better growth for 2010, aided not only by additional environmental demand, but also other important applications in pharmaceuticals, semiconductors, and electronics. The aforementioned elemental analyzers, particularly TOC and other sum parameter instruments, also should achieve better than average growth for the sector. 

Inorganic elemental analyzers will not improve quite as much as TOC or the organic analyzers. Part of the reason for the difference is the metals industry, which is a healthy consumer of many specialized metals analysis instruments in the inorganic elemental analyzers category. Although global metal production has not recovered fully, producers are beginning to ease cutbacks and this should provide a moderate beneficial effect to that market, as well as the market for arc-spark optical emission spectroscopy, which is almost entirely dependent upon metals analysis. This also applies to the X-ray fluorescence (XRF) market, which has utility in metals analysis. But the XRF market has become a very diverse market, with applications ranging from metals and semiconductors, to environmental analysis of soils or paint, to the analysis of consumer products for hazardous substances.  

Restrictions on Hazardous Substances (RoHS) and similar directives have helped grow the XRF market into the largest individual technique over the past few years. New regulations on lead, cadmium, and other hazardous elements in toys from China are continuing to help expand the market in the short term. Although XRF as a whole has been subject to many conflicting trends and drivers, the short-term outlook is for significant growth in 2010. X-ray diffraction (XRD), should achieve the greatest growth for 2010 at 7.4%. Quite unlike the rest of the atomic spectroscopy segment, the major driver for XRD is in life science: research on the crystallography of proteins and small molecules. 

The top five vendors make up a little more than half the market. Thermo Scientific holds the leading position with 15% of the market, a position it has attained due to its participation in all of the individual technology areas, and its great strength in several individual markets. PerkinElmer, with a 10% share, competes primarily in AA, ICP, and ICP-MS, but also has elemental analyzers. The next three companies, Spectris, Rigaku, and Bruker, are involved principally in XRF and XRD. The sixth largest vendor, Varian Instruments, is being acquired by Agilent, which itself has a leading market share in the ICP-MS market. Considered together, Agilent and Varian would hold the third position in the overall market, with a total share of about 9%. More focused players dominate in some markets, such as Spectro Analytical (AMETEK), which leads the arc-spark market, and LECO, which rules both inorganic and organic elemental analysis.  

The spectroscopy market and its major segments, atomic and molecular spectroscopy and mass spectrometry, is analyzed and forecasted for 2010.

Spectroscopy-03-01-2010

2010 Salary Survey

Defending the Food Chain

Market Profile: Process NMR

A Random Walk Through the Web of Mass Spectrometry

Depth Resolution of the Raman Microscope: Optical Limitations and Sample Characteristics

Electron Transitions in Optical Emission and X-Ray Fluorescence Spectrometry

Spectroscopy Bounces Back

Products

Vol 25 No 3 Spectroscopy March 2010 Regular Issue PDF