Fundamental Research and Applications of ICP-OES and ICP-MS Toward Improved Understanding of Elemental Chemical Analysis: Part I
Inductively coupled plasma–optical emission spectroscopy (ICP-OES) and ICP–mass spectrometry (ICP-MS) are often considered mature techniques, but researchers know that some aspects of the techniques are still not fully understood. Probing the mechanisms involved can lead to more accurate results that, in some cases, may cast doubt on accepted explanations. John W. Olesik, who directs the Trace Element Research Laboratory (TERL) in the School of Earth Sciences at The Ohio State University, has spent many years studying the processes that control signals in ICP techniques, and yet he is still encountering surprises. He is also in a unique position to get involved in exciting applied research, because the TERL also provides elemental analysis and access to ICP-OES and ICP-MS instruments to groups throughout the university as well as to clients and collaborators outside Ohio State—leading to all sorts of collaborations. We recently interviewed him about his recent research and career. This is the first part of a two-part interview series.
You have reported on single-particle inductively coupled plasma–mass spectrometry (spICP-MS) for routine analysis of nanomaterials (1). This technique is being used for the detection and characterization of metal-containing nanomaterials in aqueous samples, and can measure the size, size distribution, particle number concentration, and major elemental composition. What led to the development of this technique and the concepts used to construct this method and instrumentation?
The concept of using time-resolved atomic spectrometry signals to measure particles by vaporizing or atomizing each particle to produce a cloud (or “flash”) of atoms or elemental ions was used as early as 1968 (Crider’s “spectrothermal emission aerosol particle analyzer [SEAPA]). DeGueldgre and coworkers first described colloid analysis using time-resolved single particle ICP-MS in 2004. Since then the huge growth in the use of engineered nanoparticles (NP), such as Ag NPs used as an antimicrobial in clothing, bandages, coatings, and so forth, and their fate in the environment (including in humans) has driven the interest and development of single-particle ICP-MS.
My research group had been investigating droplet desolvation, subsequent desolvated particle vaporization, atomization, and ionization in ICPs by injecting monodisperse droplets of solution into the ICP and monitoring the processes using microscecond time-resolved optical emission, laser light scattering and mass spectrometry starting in the early 1990s. So, while we were in effect measuring nanoparticles over more than 10 years, they were nanoparticles produced in the ICP following droplet desolvation rather than nanoparticles that were suspended in water. We started investigating single-particle ICP-MS of suspended nanoparticles shortly after reading DeGueldgre’s first publication.
What are the major breakthroughs and discoveries relative to ICP-MS and spICP-MS in terms of speed of data acquisition, signal analysis and processing methods, and the application of alternative mass analyzers that has provided the capabilities for characterizing nanoparticles?
In 2004, commercial ICP-MS instruments were capable of single-particle ICP-MS measurements although the number of measurements that could be made per second (~1000 Hz) was smaller than ideal and a delay in data transfer from each measurement (“dwell time”) necessitated even longer “dwell” times per measurement (10–20 ms) to minimize the probability of measuring only a portion of the signal from an individual nanoparticle. ICP-MS instrument manufacturers improved detection electronics and data storage so currently available instruments have data acquisitions rates up to 10,000 to 100,000 Hz without a delay in data transfer.
The vast majority of ICP-MS instruments are based on quadrupole mass spectrometers that measure one mass/charge at a time; bulk multielement analysis is obtained by rapidly peak hopping among the mass/charge of the elements of interest. Many engineered nanoparticles (including Ag, Au and SiO2) contain only one element measurable by ICP-MS so an ICP-quadrupole MS instrument is an effective element-selective nanoparticle analyzer. However, for more complex nanoparticles (including mineral nanoparticles or multielement engineered nanoparticles such as layered particles) a mass spectrometer capable of measuring the entire elemental mass spectrum from each nanoparticle is needed. This has driven the development of ICP-time-of-flight-MS (ICP-TOF-MS) instruments first by Detlef Gunther’s group at ETH-Zurich, then commercially available from TOFWERK and now also from Nu Instruments. These instruments can acquire and store complete ICP mass spectra, not just a single element, at 500 Hz to 30,000 Hz—fast enough to obtain a spectrum from each individual nanoparticle as long as the number of particles per milliliter is not too high.
Mass-dependent matrix effects in ICP-MS have been thought to be due to space charge repulsion in the positive ion beam just downstream of the skimmer cone based on experimental measurements published by Gary Horlick’s group in 1987 and theoretical calculations by Scott Tanner published in 1992 and 1997. Why did you decide to investigate mass-dependent matrix effects again more than 20 years later (2,3)? How do you explain the lack of dependence on analyte ion mass that you reported using an ICP-quadrupole-MS in 2017 and using an ICP-sector field MS instrument in 2020?
Good question because this was a case where: 1) the basic concept of space charge effects (two positive ions near each other would repel each other, exert the same electric field on each and therefore cause the lighter ion to move away farther), 2) the extensive experimental data from Horlick’s group, and 3) the sophisticated theoretical modeling by Tanner made perfect sense and were entirely consistent with each other. Our own group also published data showing, using time resolved ICP-MS measurements, repulsion of light ions (Li+) by heavier ions (Pb+) in individual ion clouds produced in the ICP from monodisperse droplets of sample.
The practical impact was that internal standards for multielement analysis by ICP-MS needed to be chosen to have similar masses as the analyte ions, so multiple internal standards were needed for multielement analysis. We did that while providing elemental analysis for clients who came to our lab. Then we noticed when looking at the data in detail from one set of samples that we could use the light internal standard, the mid-mass internal standard, or the high-mass internal standard for all the analytes and obtain accuracy that was nearly as good as when using an internal standard element with a mass similar to each analyte using our ICP-sector field MS instrument. That was surprising and didn’t make sense to us.
That finding led us to do a thorough literature search. I had often said (partially in jest) that the 1987 Tan and Horlick paper may have set the world record for most plots in one paper. We found no publications that were comprehensive enough, and certainly not as comprehensive as Horlick’s, using ICP-MS instruments beyond the first generation to address the question of mass-dependent matrix effects. There were some hints of the lack of analyte mass dependence for the same ICP-quadrupole-MS system we had in our lab in a 2000 publication from Eric Salin’s group. So, we decided to do measurements almost as comprehensive as Horlick’s on two of the ICP-MS instruments (one with a quadrupole MS, one with a sector field MS) in our lab. To our surprise, the matrix effects using both showed a lack of analyte ion dependence.
How do we explain our observations on the two instruments? As of now, we can’t. The matrix effects are more severe as the matrix mass increases, perfectly consistent with space charge. The matrix effects are not more severe as the analyte mass decreases, entirely inconsistent with space charge effects. So, the only two (completely unsatisfying) guesses I have now are either: 1) space charge effects are not responsible for mass-dependent matrix effects in ICP-MS (some other process is responsible, I don’t know what it could be), or 2) in addition to space charge effects there is some other (I don’t know what) offsetting process that results in the lack of analyte ion mass dependence.
Your research group has been studying the fundamental processes that affect ICP-OES and ICP-MS signals since the 1980s. More recently, you have been a co-author on papers in many, widely diverse disciplines (4–7). How and why did that happen?
I would say three things contributed to this. 1) Our own research, which focused on understanding the fundamental processes, provided us detailed insight into how the sample matrix and other variables could affect analysis accuracy (due to changes in plasma temperature, excitation, ionization or ion transport to the mass spectrometer or spectral overlaps) and potential means to overcome the potential errors. 2) Part of the TERL’s mission is to provide elemental chemical analysis for researchers in diverse disciplines, departments, and colleges at Ohio State as well as to teach others the fundamental concepts and practical considerations for elemental analysis. Many of those researchers are at the forefront of their disciplines. Together 1) and 2) provide a synergy that few other labs have. 3) We love challenges and have had tremendous opportunities to collaborate with researchers who rely on accurate and often challenging elemental analysis measurements.
Most analytical scientists are unfamiliar with the concept of studying the microchemistry of the otolith (a small structure in the inner ear) of fish to determine the natal origin of adult fish species in a freshwater lake. Would you briefly describe the intent of this study (4)? What special methods were used for sample preparation and analysis of the concentration of strontium [Sr] in these otolith samples?
Biogeochemically, “you are what you eat (or drink).” Trace element “fingerprints” can provide a history of where a fish was born and migrated during its life (from trace elements taken up from the water and incorporated into its bones), where oranges were grown (by relating the trace elements in the orange or orange juice to the trace element geochemistry of the soil the plant grew in), or what mine diamonds came from (depending on the geochemistry of rocks in the mine).
Fish otoliths grow over time (and even have bands analogous to tree rings). The trace elements in the otolith are dependent on the trace elements in the water when that portion of the bone was being formed. We have had a great collaboration with Profs. Elizabeth Marschall and Stu Ludsin of the Ohio State University Aquatic Ecology lab. They are interested in understanding fish populations in the Great Lakes, other lakes, and rivers including different stocks, their survival, and their propagation.
There is a community of fish ecologists throughout the world who use otolith microchemistry and laser ablation-ICP-MS measurements to obtain a history of the movement of fish. Important “simple” questions include: Did the fish return to the location where they were born in order to spawn? When stocking fish, how long should they be raised and grow before stocking and how does that affect their survival and propagation? What factors affect the fraction of fish born in a particular location that will survive?
Sample preparation involves sacrificing the fish, removing the two otoliths, cross sectioning them, polishing them, and measuring growth rates to relate distance from the core of the otolith where the bone first began to grow to time in years, months, or weeks and then mounting the polished sample on a microscope slide. Fortunately, our collaborators’ research groups do that part. For us, the challenge is providing precise enough trace element/Ca concentration ratios to be able to distinguish different locations where the trace element concentration ratios in the water are only slightly different, and in some cases, sampling with sufficient spatial resolution. Typically, 25 µm spatial resolution is sufficient but in one study our collaborators were interested in otoliths from larval fish; the otoliths were 25 to 50 µm across and we needed to avoid sampling the region at the core, where the trace element concentration ratios are very different. Actually, the most challenging part of that experiment was for our collaborators to find, extract, and mount the very small otoliths.
In another research paper you described a method for the accurate analysis of trace elements, such as manganese (Mn), iron (Fe), copper (Cu), and zinc (Zn) in the Lyme disease spirochete (5). What special analytical requirements were needed for using ICP-sector field-MS for this analysis? What was learned from the analytical results?
This is another example of a great collaboration. Prof. Xin Li, then at Ohio State University, now in the Department of Medicine at Tufts Medical Center in Boston, called me one day about measuring Fe in the Lyme disease spirochete (spiral-shaped bacteria) Borrelia burgdorferi. Biological experiments her group had done indicated that an iron metalloprotein was important in the Lyme disease spirochete but the elemental analysis results she received from another laboratory did not seem to make sense.
Furthermore, her observation was met with resistance because a paper had been published about 10 years earlier stating that an iron metalloprotein was not involved. We looked back at that paper and found the description of the instrument used to be unusual (an “ICP optical mass spectrometer”). Although there was at one time a commercial instrument designed to simultaneously measure optical emission and mass spectrometry signals from the same ICP, that concept had a fundamental problem. The optimum location to measure the optical emission signal was downstream of the optimum location to measure the mass spectrometry signal (which we knew based on fundamental investigations). However, downstream of the sampling cone of the mass spectrometer was not optically accessible and even if it was, the rapid expansion in the interface would greatly “dilute” the sample atoms and ions. Very few of that instrument model were sold.
We investigated further and found that the lab that did the measurements had an ICP-OES system (not the ICP-OES-MS nor an ICP-MS instrument). After talking with Prof. Li about the expected Fe concentrations we realized the detection limits for Fe by ICP-OES were barely below the Fe concentrations expected in the sample so it would be very difficult or impossible to obtain accurate enough results by ICP-OES.
After working with Prof. Li’s group to be sure that the reagents they used in their biological experiments were pure enough so that Fe contamination in the reagents would be much smaller than the Fe concentration from the metalloprotein, we received a set of samples from their experiments. A couple days after we sent the results to Prof. Li she called back very excited and said, “The results make perfect sense…. but they mean my lab messed up the biology experiment”. Her group conducted another experiment avoiding the mistake they had made and sent us another set of samples. We used ICP-sector field-MS with enough spectral resolution to avoid signal from the intense ArO+ signal also nominally at mass 56 (like the most abundant Fe isotope but at a slightly higher mass), and with enough sensitivity and low enough background for the concentration of Fe we needed to measure to be several orders of magnitude above the detection limit. The next call I got from Prof. Li was to say, “we have proven that Fe is important (contrary to the previously published conclusion) and the biological experiments are entirely consistent with that result.”
After more experiments to confirm the result, a manuscript was submitted and quickly rejected (reviewers said a previous publication already proved Fe was not important). Prof. Li’s group did more experiments and we made more ICP-SFMS measurements. They all once again proved that Fe was important. We submitted the paper to a different journal along with a detailed explanation of why the previous measurements, done 10 years earlier, were unable to accurately measure the Fe. The paper was accepted and published. Other researchers confirmed the results after seeing our “controversial” manuscript.
References
(1) M.D. Montaño, J.W. Olesik, A.G. Barber, K. Challis, and J.F. Ranville, Single particle ICP-MS: advances toward routine analysis of nanomaterials, Anal. Bioanal. Chem. 408(19), 5053–5074 (2016).
(2) J.W. Olesik, and S. Jiao. Matrix effects using an ICP-MS with a single positive ion lens and grounded stop: analyte mass dependent? J. Anal. At. Spectrom. 32(5), 951–966 (2017).
(3) S. Jiao, and J.W. Olesik, Characterization of matrix effects using an inductively coupled plasma-sector field mass spectrometer, J. Anal. At. Spectrom. (2020). DOI: 10.1039/d0ja00207k.
(4) K.Y. Chen, S.A. Ludsin, B.J. Marcek, J.W. Olesik, and E.A. Marschall, Otolith microchemistry shows natal philopatry of walleye in western Lake Erie, J. Great Lakes Res. (2020). https://doi.org/10.1016/j.jglr.2020.06.006.
(5) P. Wang, P., Lutton, A., Olesik, J., Vali, H., & Li, X. (2012). A novel iron- and copper-binding protein in the Lyme disease spirochete. Molecular Microbiolgy 86(6), 1441-1451.
John Olesik is a Research Scientist and Adjunct Associate Professor in the School of Earth Sciences at The Ohio State University. He earned a B.S. in Chemistry in 1977 from the University of Rochester and a Ph.D. in Chemistry from the University of Wisconsin-Madison in 1982 with John P. Waters. After a postdoc at Indiana University with Gary Hieftje, he was an Assistant Professor of Chemistry at the University of North Carolina-Chapel Hill before moving to Ohio State, arriving on the day his daughter was born. John and his group, the Trace Element Research Laboratory (TERL), are focused on: 1) investigating the fundamental chemistry and physics to convert samples to atoms, ions, and signals, 2) enhancing capabilities provided by plasma-based instruments for elemental analysis, and 3) applying plasma spectrometry for analysis of biogeochemical, geological, biological, medical, pharmaceutical, and environmental samples including micro- and nano-particles. John serves on the Editorial Advisory Boards of Spectrochimica Acta Part B and Spectroscopy and the Advisory Board of the Journal of Analytical Atomic Spectrometry. He received the Society of Applied Spectroscopy Lester Strock Award in 2001 and the Spectrochemical Analysis Award from the American Chemical Society in 2009.
