MSA at 100 and why optical mineralogy still matters: The optical properties of talc

MSA at 100: Introduction

MSA celebrated its 100th anniversary in 2019. Planning for the event occupied MSA Council meetings for several years prior to the event, and we were fortunate that we could celebrate this occasion in person as Covid-19 hit shortly afterward. There were several events in which we came together as a Society to celebrate the MSA Centennial, and my goal here is to memorialize them in one place.

Heaney and Gunter (2019) reviewed past MSA celebrations and detailed current issues that MSA should address. Rakovan and Gunter (2019) had a slightly different twist on their article dealing with the Centennial as they discussed Alex Speer, MSA’s executive director for 25 years, while pointing out he had been a member for 50 years; Alex stepped down as director in 2019. The photo in Figure 1b is of Alex and the new executive director Ann Benbow, taken at the 2019 MSA award luncheon at the annual GSA meeting in Phoenix.

Our feature event of the Centennial was the “MSA Centennial Symposium” with organizing co-chairs Peter Heaney and Steven Shirey. Proceedings of the two-day event can be found at: http://www.minsocam.org/msa/Centennial/MSA_Centennial_Symposium.html. The following is a summary of the event written by the organizers that appeared in Elements 15, p. 285.

On 20–21 June 2019, 160 mineral enthusiasts gathered in the newly renovated Carnegie Institution for Science (Washington, D.C., U.S.A.) building to celebrate the 100th anniversary of MSA through moderated presentations of exciting advances in the solid Earth sciences. The 14 themed colloquia were proposed by MSA members, and they beautifully illustrated the broad reach and profound impact of mineralogy today. The opening session on sustainability included sobering messages from Gordon Brown and Michael Hochella regarding the lasting legacy of open-pit mining in the western U.S.A. and the role of incidental nanomaterials in controlling contaminant dispersal in mine wastes. A related theme by David Singer and Michael Schindler emphasized the need to characterize soil horizons across multiple length scales, with evidence that processes at the nanoscale do not extrapolate simply from larger size regimes.

Ross Angel and Lucie Tajcmanov offered perspectives on the future of metamorphic petrology by stressing that state-of-the-art characterization techniques require a commensurate understanding of the complex physics and chemistry that produce metamorphic textures. In their session on mineral analysis, Michael Wiedenbeck provided a historical overview of SIMS, while Simon Jackson impressed the audience with current capabilities in trace element mapping by LA-ICP-MS. Othmar Müntener and Roberta Rudnick next walked the audience through models of the formation of the lower and upper crust based on field studies and experimental petrology. Fabrizio Nestola and Graham Pearson took the audience even deeper through a review of the exotic inclusions and isotopic compositions that have been observed in diamonds. Kim Tait and Aaron Celestian closed Thursday’s session with their perspectives on how mineral museums can thrive in the next century by aligning their missions through close collaborations with the research community.

Friday began with a COMPRES-sponsored overview of synchrotron-based studies in mineral physics. Przemysław Dera discussed surprising 5- and 6-coordination states for Si at high pressure, and Jin Zhang described anisotropy in omphacite as a means of detecting eclogite in the Earth’s mantle. Elizabeth Rampe and Harry McSween next offered revelations into the early history of Mars through rover-based in situ rock analyses and characterization of the >100 meteorites that originated on Mars. In a session sponsored by Rob Lavinsky, Shaunna Morrison, and Simone Runyon challenged the audience to imagine the power unleashed by connecting the dots in the enormously large mineralogic and petrologic data sets that geologists have amassed over the last century. Alexandra Navrotsky closed the morning with an announcement of

the next chapter in her multifaceted career as director of a new Materials of the Universe program at Arizona State University. Gilberto Artioli tugged us back in time to consider the earliest uses of minerals in ceramics from 18 000 BCE, and his talk was followed by Michael Tite’s history of the earliest Pb-Sn oxide glazes that were innovated in the Middle East in counterthrust to Chinese porcelain. The Gemological Institute of America funded the following session, with Wuyi Wang describing the emergence of synthetic gem diamonds over the last 15 years and Mandy Krebs illustrating trace element and isotope approaches to provenance colored gems. Supported by C2/m Mineralogy, John Hughes and Jill Pasteris expounded on the essentiality of apatite as a pillar for both our civilization and our bodies, and Ann Wylie and Matthew Sanchez concluded the meeting with presentations on the real, and supposed, health hazards of mineral dusts.

Figure 1

Photographs to celebrate MSA’s 100th. (a) The outgoing MSA executive director Alex Speer and incoming executive director Ann Benbow. (b) Greeting signs for the 2019 annual GSA meeting in Phoenix announcing MSA’s 100th. (c) A memento given to MSA by GSA. (d) Former MSA presidents and my mentors, and (e) some of my former Ph.D. students. The photos in d and e are meant as a metaphor to show how we arrived at our career and how we help others to obtain theirs.

The inspirational science was enhanced by a spectacular evening reception among the stunning gem and mineral exhibits in the Smithsonian National Museum of Natural History, highlighted by President Mickey Gunter’s toast with specially embossed champagne glasses. The symposium proved that, despite the diversity of our interests, there is more that unites than divides us thanks to our common foundation in the minerals and rocks that support our existence.

The final events of our centennial were held at the annual GSA meeting in Phoenix. GSA was very helpful and accommodating to our centennial; they even surprised us by posting signs stating it was our 100th (Fig. 1b). They allowed us to have back-to-back sessions on Monday with a celebratory lunch between them. A highlight of that lunch was the presentation to MSA by the current president and executive director of GSA of a memento of this occasion (Fig. 1c). MSA also recognized Alex Speer for these 25 years of service as our Executive Director. The scientific portion of the all-day session was titled “Mineralogical Society of America (MSA) at 100: Reflections, refractions, diffractions, intrusions, subductions, reactions, etc. from MSA past presidents.” Sixteen former presidents submitted abstracts to this session, and in a break from tradition, the current presidential address was also given. Abstracts for this session are available at: https://gsa.confex.com/gsa/2019AM/meetingapp.cgi/Session/47772 and https://gsa.confex.com/gsa/2019AM/meetingapp.cgi/Session/48800 sites.

President Letters re: MSA at 100

What follows are slightly modified versions of my MSA president letters that appeared in Elements over the 100th year of the Society. These letters were written with the intent to form a portion of this paper and all directed to “MSA at 100” with different themes:

A look backward. It was indeed an honor for me to be the 100th MSA president as MSA turned 100; a time in which we celebrated our successful past and plan for a similar future. Figure 1d shows (from left to right) MSA presidents, numbers 100, 62 (Jerry Gibbs), 58 (Don Bloss), and 68 (Paul Ribbe). I was blessed to have had them as my mentors. Looking back, each of them also had at least one past MSA President as their graduate advisor. In what would seem like a strange diversion at this point—a story seems fitting. At the 2002 IMA meeting MSA President 83 (Rod Ewing) started to introduce me to MSA charter Executive Director (Alex Speer). I mentioned to Rod, who I had met in 1981, that I had met Alex two years sooner. Rod replied “our world is small, perhaps too small.” Of course, this was intended as a joke, but could be taken negatively. However, this statement has stuck with me for years, and I have concluded it is the mentoring, learning, friendships, and all the other positive things in life that does result from this “small world,” as essentially, we are a worldwide network of mineralogists who have very close ties. In fact, it came back again in this study as one of the talc samples was collected by Al Chidester; turns out he was Don Bloss’ teaching assistant in optical mineralogy when Don was an undergrad at the University of Chicago pre-World War II. A corollary to this is the motto used by Google Scholar “stand on the shoulders of giants.” I truly believe this is a reason our society has lasted 100 years, and if we, the members, continue the efforts put forth by those who have come before us, our Society can look forward to a bicentennial celebration in 2119; of course sans any of us reading this letter. (Above modified from Gunter 2018b.)

Professional service. It is easy to overlook professional service as part of our “job”; however, I would view it as one of the most important things we do. And although all of us might not teach or do research, we all should be performing professional service. My former department chair once told me that for every paper I submit, I should review two to three—this made sense as I knew my papers would be reviewed by two or three other scientists. My largest service effort was to co-organize the 2005 Goldschmidt Meeting in Moscow, Idaho. Although it might be easy to determine how many papers we should review, I think it is harder to determine how many international meetings we should organize! Regardless, meetings do not organize themselves, nor do professional societies run themselves, nor do journals publish themselves; these all depend largely on the volunteer efforts of members or dedicated staff.

A major part of MSA’s organization consists of 31 committees with 148 individuals filling them. So, for a society of roughly 2100 members, one can see that at any one time, 7% of us are involved in the operation of the society. Most of these committees are filled on a three-year term, so much of the “memory” in our system resides in our Executive Director of going on 25 years (yes, 25% of MSA’s life!), Alex Speer. As many of you know, Alex will be retiring at the end of 2019, and I have the pleasure to announce Alex’s successor—Dr. Ann Benbow. Ann comes to us most recently as the Executive Director of the Archaeological Institute of America, and before that, she had spent over a decade as Director of Geoscience Education and Public Understanding as well as the Director of Education, Outreach, and Development at the American Geosciences Institute. Her official start date is March 1, 2019 so there will be a 10 month overlap with Alex.

I have always viewed service as the most important role of an academic, or as I stated at the outset, all of us, as we give our most important commodity—our time. MSA clearly is a society run by dedicated volunteers, and I would like to take this opportunity to thank all of them and encourage all of you to become more involved with MSA. Finally, along with giving time, giving money is also a good thing! As such, my major goal as president will be to establish a centennial fund so MSA can fully fund our outreach activities (i.e., the lecture program, Mineralogy-4-Kids, MSA-talk, etc.). Much of this will be done in collaboration with our incoming executive director as Ann has had considerable experience in this area—so stay tuned for more on that. And really finally just in case you are wondering how often you should serve as president of MSA, based on our current membership, it would be once in 2100 years! (Above modified from Gunter 2019a.)

Teaching mineralogy. This president’s letter may seem more fitting for a Triple Point column as an opinion piece, but based upon a career of teaching and learning basic and applied mineralogy, I think it is a subject we all need to consider regardless of our professional status. I stated that I thought it was important to “return the pure study of mineralogy back to geological sciences curricula” in my write-up when I ran for vice president and president. Many of the more senior readers of this article will recall when mineralogy was taught as a year-long course, often with mineralogy the first semester followed by optical mineralogy. Not to leave out the petrologists, as petrology used to be a full-year course as well. And now, many schools have condensed two years’ worth of courses into one half year, while our knowledge in these areas have only grown!

Perhaps we are a victim of our own success, as over the last 100 years our science has matured to the level where, although there are still unanswered questions, we do have a deep understanding of our field. For instance, look no further than the Reviews in Mineralogy and Geochemistry series, where there are entire volumes dedicated to mineral groups from A to Z (i.e., amphiboles to zeolites). We have also been able to design entire families of instruments that allow us to accurately and precisely determine the composition of minerals and their internal structure at several scales, as well as send a miniaturized version of one of those to Mars and to understand the theory of how these instruments work. So, it might seem to many there is less of a need for a year-long course in mineralogy because we have made all of these accomplishments in the last 100 years. But they could not be further from the truth, as a basic understanding of mineralogy underlies most of the geosciences. In fact, this might also explain why “pure” mineralogy is being taught less, as there are so many non-mineralogists who can teach mineralogy more as a means to an end rather than an end in itself. In fact, one of my petrology colleagues used to joke that my mineralogy course should be taught as an introduction to petrology course and named likewise as it was required for his petrology course.

Specifically, let us take optical mineralogy and use of the polarized light microscope as an example, although most certainly we could pick other areas such as crystallography (Gunter 2004). A few years back, I conducted a non-scientific survey on MSA-Talk. Two-thirds of those who responded stated their institution used to teach a semester-long course in optical mineralogy, while only one-third do now. Although there might be many reasons (e.g., inclusions of other courses, courses on more “modern” instrumentation, reduction in credits hours for the major, etc.), the end result is fewer students learn this much-needed basic skill. And to “prove” it is much-needed, look no further than the private sector where several companies teach week-long courses in polarized light microscopy; the cost of these courses is approximately $2000. Of course, there are many other reasons to teach optical mineralogy, and the use of the polarized light microscope is often pointed out in lively discussions on MSA-Talk.

Although there are new things to learn over the next 100 years, there most certainly needs to be a return to teaching more of the fundamental principles of mineralogy that have been removed from our curriculum, and I only need to point to the example of optical mineralogy. MSA most certainly has teaching resources to help accomplish this. But as we teach those new things, let us not forget the old things that are the foundations for building the new things! (Above modified from Gunter 2019b.)

Mineralogy research. The orientational dependence of the physical properties of minerals is one of the answers I give when someone asks about my research interests. The main physical properties that interest me are the refractive indices, and I was fortunate enough to study how they change as a function of composition and structure for andalusite. This resulted in my first paper in American Mineralogist back in 1982, and also showed how an orthorhombic mineral could be optically isotropic (Gunter and Bloss 1982). This paper contrasted with the first paper published in the American Mineralogist in 1916: “The occurrence of lamellar calcite in Rhode Island”; although they could measure optical properties, composition was done with wet-chemical methods and diffraction of X-rays by crystal had just been discovered in 1912.

“X-rays” were not mentioned in an American Mineralogist title until 1927 when Stockwell (1927) used them to determine the unit cell of garnets and in turn use it to help predict the garnet’s composition. Before that, refractive index and specific gravity had been used; thus, this paper showed for the first time how the physical properties of minerals relate to their structure. At this point, indirect methods to determine composition were very useful as it was easier to perform them than to determine the composition by wet-chemical methods. Following the next year was a paper titled “The oscillation method of X-ray analysis of crystals” Gruner (1928a); I suspect there are few dues-paying members of MSA who have ever taken an oscillation photo, although several senior members may have. Regardless, this was the first method used to obtain crystal structure information. The same year Gruner (1928b) used this method to determine the crystal structure of analcite as isometric with a = 13.64 Å and “16 molecules of NaAlSi₂O₆·H₂O” in its “unit cube.” Among other things, these data allow us to calculate the density instead of determining the specific gravity. And 91 years later his formula and cell edge remain the same; all that has changed is we now call this mineral analcime.

Turning to the composition of minerals, when our society was formed it was not uncommon to see ideal formulas for minerals written in terms of their oxides and instead of the now-used chemical formulas. For example, analcime would be written as 1/2Na₂O·1/2Al₂O₃·2SiO₂·1H₂O. This practice was followed for two reasons: (1) at this point we lacked the understanding of how elements would fit into a mineral as we did not know their structures, and (2) compositions were determined by wet-chemical methods. Many of us will recall the inorganic chemistry lab where we were given a solid unknown, dissolved it, then precipitated portions of it, and very carefully weighed each; at the point I most certainly did not realize I was determining the composition of the material as I was so concerned on performing the “technique.” Fortunately, just like X-rays allowed us to determine structures, the development of electron beam techniques, especially Castaing’s development of the electron microprobe—which occurred unbeknownst to me as I was in grade school—allowed for precise and accurate determination of the composition of micrometer-sized mineral grains in polished grain mounts or thin sections.

We can now use synchrotron radiation to understand the orientational dependence of absorption spectra (for example Dyar et al. 2002). To study these, we use a spindle stage, whose precursor was described in a 1924 American Mineralogist article by Kerr (1924) entitled “A simple rotation apparatus.” Of course, it would be helpful if we could calculate these spectra from first principles, but we have yet to accomplish that or to even be able to calculate refractive indices. We have also yet to fully understand what external shape minerals will take as they crystallize, which is basically the question asked in the paper published in 1914. However, as noted above, we have made many advancements in the first 100 years of our Society, and there are many yet to come in the second 100! (Above modified from Gunter 2019c.)

The other IMA and... I write this letter a few days after returning from MSA’s Centennial. The first session at that meeting discussed mineral resources, which we could relate to IMA. To a mineralogist, the acronym IMA means the International Mineralogical Association; however, there is another meaning closely related to what many of us do—the Industrial Minerals Association (https://www.ima-na.org). It is an umbrella organization helping industry to provide the very materials we use daily. Some of these minerals are familiar to all of us (e.g., feldspars, talc, etc.), whereas others are less well known (e.g., barite, wollastonite, etc.). Regardless, in developed countries, we consume about 130 pounds of these minerals per day per person. And, while the 842 pounds of lunar material returned to earth during the Apollo missions may provide more excitement than the ~70 million tons of industrial minerals we use in the U. S.A. yearly, the latter are much more important for daily lives.

Over my career, I have worked on projects dealing with several industrial minerals, from aluminosilicates used as refractories, to zeolites used in water purification systems and radioactive waste treatment. I viewed this research as proactive and positive with the potential to help society. Over the past 10–20 years, my efforts in this field have turned more to defending this industry against claims of purported asbestos content of their products. Asbestos contamination of commercial products is most certainly of concern, but what is critical in this area is something as straightforward as the correct identification of common rock-forming minerals. For example, shards of talc are often misidentified as anthophyllite asbestos because their Mg/Si ratios (3/4 vs. 7/8, respectively), easily overlap in semiquantitative EDS spectra. Also, they both yield a 5.2 Å repeat in electron diffraction, where that repeat is used to “prove” a particle is an amphibole in routine asbestos methods. The last talk of the Centennial session gave several examples of incorrect identification being used against the industry in the ongoing talc litigation. We also hear in these cases that miners cannot tell talc from country rock, so the talc gets “contaminated.” Yet I have spent time in mines where there is no question of the contact between ore and “country rock,” it is not that hard to tell talc from marble!

While many reading this article might find this hard to believe, you will find it harder to believe the following sampling of mineralogically incorrect statements I have read in legal and regulatory documents: Anthophyllite and chesterite are polymorphs; one cannot distinguish monoclinic from orthorhombic amphiboles by electron diffraction; or better (really worse) yet, that amphiboles cannot be distinguished from pyroxenes in soils, and finally “Mica is mica. If it has aluminum silicates and phosphorus, phosphates, that’s mica. It is just aluminum silicates is a type of clay.” In Chapter 19 of our book (Dyar and Gunter 2019) titled “Mineral Identification” we state, when handed a mineral: “Do you want me to guess what it is or tell you? If you want me to tell you, it will take a little work, but we’ll know for sure.” Our community can correctly identify minerals with the appropriate choice of analytical methods.

The final participant comment in the last session in D.C. was a plea that we “take back mineralogy.” This was a fitting way to end the meeting. But how do we do this? It should be clear from my above comments that mineral misidentification is a common practice in the “real world.” Perhaps it is time we mineralogists consider professional licensing, as it exists for geologists, engineers, and many other fields. No doubt both MSA and IMA can support this to aid in something as simple as proper mineral identification. (Above modified from Gunter 2019d.)

Looking forward. As I compose my final President’s letter, I ponder the future of our Society and discipline. As I look at the smiling faces of three of my former Ph.D. students above, it gives me great hope (Fig. 1e). There are many like them without gray hair who will carry on with the service, teaching, and research in mineralogy that the gray-haired ones among us did. What I fret about can also be represented in the photo by the word “asbestos.” It is not the fact that asbestos has presented health problems in modern society, but more how we tend not to be involved in some of these problems, as they can be very controversial.

What I have noticed over my career is a trend, especially among us academics, to look for problems more than to solve them. This hit home when I was working with an industrial hygienist for an industrial minerals company, and he said “no problems, no funding.” This was further reinforced when we went to visit a small industrial mineral mine, and the president of the company did not want to let us in as we were professors. It was not always this way, as “in the good old days” we used to work more closely with such industries. Of course, in the U.S.A., federal funding and university expectations play a role in what is “valued.” Regardless, I think we need to work to help industry solve its problems and not create more for them.

What I had thought less of over my career is how MSA has been the catalyst for many other professional societies and journals. The outcome of this has been to reduce our number of members. This might be good or bad overall for the geosciences, but it is important for us to be aware of. Something I do believe to be bad is, along the way, how we have given much less attention to what are sometimes referred to as “amateur mineralogists.” They really are the heart of MSA, as they are the ones who collect the minerals we study.

Yes, this letter reads more like what one might see on the opinion page of a newspaper. In fact, I have never written one of those and have only written one thing that might be close—A Triple Point article in Elements a decade ago on, you guessed it, asbestos (Gunter 2009). Regardless, for our society and discipline to exist over the next century, I think we will need to put more effort toward rebuilding our ties to industry and the collecting community. Also, we need to change the priorities of the funding agencies and universities to allow us to accomplish this.

Finally, I thank all the members, past and present, of MSA for a successful 100 years, for the honor it has been to be the 100th president, and for all you have done for mineralogy in the last century. (Above modified from Gunter 2019e.)

Why optical mineralogy still matters

As noted above, and well known to many, we have decreased the teaching of optical mineralogy at the college level in the United States. As a result, the teaching of this much-needed discipline is taken on by the private sector. At the same time, active research in this area is also on the decline. For those interested in the capabilities of optical mineralogy research see the final chapter in Bloss (1981), the forward in Bloss (1999, 2020), and Bloss’s MSA president’s paper (Bloss 1978). Obviously Bloss championed optical mineralogy in both teaching and research, and the most updated results of Bloss and coworkers research are given in Gunter (2020). Specifically, for the spindle stage methods, Steven and Gunter (2017, 2020) present an Excel spreadsheet capable of many optical mineralogy calculations (e.g., those performed by EXCALIBR). That program and other spindle stage related materials can be found on MSA’s website at: http://www.minsocam.org/msa/Monographs/#Spindle_Stage.

Unbeknownst to many of us the use of the PLM is widespread in industry and the regulatory world, specifically for screening materials for asbestos (McCrone 1974; ISO 2012). Herein I use as the definition of asbestos as the asbestiform variety of serpentine (i.e., chrysotile), crocidolite, amosite, tremolite, actinolite, and anthophyllite when the latter three occur with an asbestiform habit. Thus, this is a two-part definition: (1) first identify one of the six mineral species and (2) the mineral must occur in an asbestiform habit. The former is more-or-less a straightforward task for a mineralogist but can be challenging for others. The latter presents several challenges, but definitions and methods to ascertain morphology were developed last century to resolve this issue (e.g., Campbell et al.1977; Crane 1992; Perkins and Harvey 1993; Yamate et al. 1984). However, confusion does occur when certain regulations for asbestos used in an occupational setting are applied to the natural environment or areas where commercial asbestos was not used (Gunter 2010), specifically for talc (Gunter et al. 2018). The application of some of these regulations to the natural environment would make much of the Earth asbestos “contaminated” (Thompson et al. 2011; Gunter 2018a). The main issue in the natural setting deals with the mineral group amphiboles (Gunter et al. 2007); both their occurrence in commercial asbestos as well as their natural occurrence composing 5% of the Earth’s crust.

The purported asbestos content of talc has been a concern for decades but has gained recent attention based on litigation directed toward cosmetic talc. I believe the role of the mineralogist in this area is as straightforward as the correct identification of talc and minerals that might be associated with it. Examples of these issues are given in Buzon and Gunter (2017) and Gunter et al. (2018). However, the ongoing misidentification issues made by others are not published in the peer-reviewed literature but occur in documents used for litigation support. Lacking is a thorough study of the optical properties of talc, which will help refine methods to determine the purported asbestos content of talc. The issues are twofold: (1) there is not a clearly defined range of refractive indices for talc ores from different locations, which are used in several products, and (2) dispersion data are unavailable and needed to aid in identification with the dispersion staining method. Rutstein et al. (2020) propose a method using a combination of PLM and XRD for bulk talc samples, and these data will find immediate use in those regulatory efforts.

Optical properties of talc

The need. Over the past decade, we have characterized talc with one goal being to better understand its potential asbestos content. One of the many analytical methods we have used is PLM, as the optical properties of talc differ from potential asbestos minerals. For example, chrysotile has lower refractive indices, whereas amphiboles have higher. Even though the differentiation is easy for those who understand mineralogy and PLM, talc can be misidentified for chrysotile as the α value of talc (i.e., the smallest refractive index value for talc, which is perpendicular to the talc elongation) is similar to chrysotile; thus, a talc particle viewed on edge can be confused with chrysotile. Of course, the talc index changes on stage rotation to β or γ (i.e., the large refractive index, N), as this direction is parallel to the elongation of the talc. Another problem occurs with the acceptable range of refractive index deemed to match talc. A similar issue occurred in identifying the winchite and richterite amphiboles in Libby, Montana. The compositional range of these amphibole species was limited, as were their refractive indices (Bandli et al. 2003), but the full range (i.e., the high-Fe end-members) could cause labs to identify low-Fe pyroxenes as amphiboles. Sanchez and Gunter (2013) discuss this problem at length and provide a solution.

To better understand the optical properties of talc, a literature review is first presented, as well as noting the refractive index values given in reference books. Table 1 lists several references and the ranges given. Values of β and γ are very similar and often lumped together. The unanswered question is, where did these values come from? Apparently, many of them came from the Doelter and Dittler (1912) paper. Table 2 lists specific values; the α value, in general, does not agree with Table 1, whereas the γ ones do, and β is rarely given. Chidester (1962) does provide compositional data and values for γ from talc samples he collected in active Vermont talc mines, but there was no correlation of increasing γ to Fe concentration. Forbes (1969) likewise provided γ values and Fe concentration for synthetic samples, but again no correlations occurred. In both cases, this might be due to small data sets, low precision measurements, or small variation in composition. Dispersion data are only provided in two studies (McCrone and Delly 1973; McCrone et al. 1979), and these data are not self-consistent. Dispersion data are important as discussed below in methods used to identify talc by commerciallabs. Finally, Phillips and Griffen (1980) do provide a plot of γ increasing as a function of Fe content, but no details are given on the source of their data.

It is somewhat understandable why no thorough optical study of talc has been conducted. In fact, my attempts to determine the optical properties of talc on single crystals with the spindle stage failed as talc edges tend to scroll, making Becke line determinations difficult to impossible. However, talc particles will sometimes lie “on edge” in grain mount. When this happens, they appear as a fiber or an elongate particle. In this case, the long direction coincides with either β or γ and the perpendicular direction is α. Given these types of particles, the refractive indexes perpendicular and parallel to the elongate talc can be measured in grain mounts, and these are referred to as n and N, pronounced “little n” and “big N” (Bloss 1961) to distinguish them from α, β, and γ.

Sample selection. Twenty samples were selected, 18 of the 98 characterized by EPMA in Buzon (2016) and two from McNamee (2013), to span the observed compositional range in these two studies. The remaining sample of Johnson’s Baby Powder was purchased in Moscow, Idaho, on May 20, 2020. In these previous studies, talc varies very little in composition, with Fe and F being the elements most affecting the optical properties and the only elements to correlate to them. Table 3 lists the selected samples from high- to low-Fe concentration. The table also lists the sample number from Buzon (2016), a brief description, location, and who collected them. In addition to compositional variation, efforts were made to select samples that represent several of the talc-producing regions in the world. Worldwide deposits are discussed in general in McCarthy et al. (2006), and Buzon (2016) characterizes samples from many of these locations. Chidester et al. (1964) review talc deposits in the United States, and Van Gosen et al. (2004) discuss them with a focus on their potential amphibole content as related to formation conditions and protolith. More information on the geological formation and mineralogical characterizations for certain deposits of samples used herein are provided in Gouverneur Talc mining district (Engel 1962; Ross et al. 1968; McNamee and Gunter 2013, 2014; Gunter et al. 2018), southwestern Montana (Berg 1979; Buzon and Gunter 2017), Death Valley region, California (Wright 1968), Vermont (Gunter et al. 2018), and Val Germanasca, Italy (Sandrone and Zucchetti 1988).

Analytical methods. Methods to determine the refractive index of minerals were first discussed in the American Mineralogist by Emmons (1928) by use of Becke lines, which required determining a “match” by correct interpretation of Becke line colors in polychromatic light (for examples, see Fig. 2). These were further refined in Emmons and Gates (1948) by what is now known as the double variation method, whereby one changes the temperature of the immersion media, thus changing its refractive index, and finds a “match” in monochromatic light that makes the grain actually disappear. Bloss (1981, 1999) describes the double variation and refinements made with advancements in temperature controls and wavelength variation. Computer programs were developed by Su et al. (1987) to increase the precision and accuracy of this method, precision being increased by multiple measurements at different temperatures with associated matches at different wavelengths, and accuracy being increased by calibration of refractive index liquids and measurements of glass standards with refractive indices known to the sixth decimal place in the visible wavelengths.

►Figure 2

Back cover of Bloss (2020) showing both colored Becke lines and central-stop dispersion staining colors for liquid and solid matches at different visible wavelengths. These colors are often described in texts, but actually seeing them is more helpful.

Figure 3 shows the PLM used in this project with a heating/ cooling stage mounted on the rotating stage and a slide monochromator below. Figure 4 shows images of three different minerals from sample 14 in cross-polarized polychromatic light with the first-order waveplate inserted. Also shown are three inserts in plane polarized where the refractive index of the liquid was changed to the values shown by heating or cooling to match the three separate grains. The range of heating/cooling allows for a variation in the liquids of ~0.060. Thus, there is no need to make separate grain mounts with different liquids to aid in mineral identification if the mineral’s refractive indices do not differ by more than 0.060.

Figure 3

A polarizing light microscope equipped with a heating/ cooling stage and monochromator. This is the setup used herein to employ the double variation method for precise refractive index determination.

Figure 4

An example of the use of a heating/cooling stage to identify minerals with differing refractive indices in the same grain mount for sample 14. A near index-matching fluid (i.e., 1.58) was chosen for talc, then cooled to 1.61 to match tremolite and heated to 1.55 to match serpentine. (Modified from Gunter et al. 2018.)

As noted above, dispersion staining is also used to aid in mineral identification discussed in American Mineralogist by Winchell (1929), Dodge (1948), and Wilcox (1983), especially for asbestos identification (Bloss 1999; McCrone 1974); dispersion staining colors are shown in Figure 2. Figure 5 shows dispersion staining images of sample 3 in a 1.545 liquid (Fig. 5a) and a 1.595 liquid (Fig. 5b). For both figures, note the elongate, blueish particle near the center of the field of view; this dispersion staining color indicates a near match between the particle and the liquid at 589.3 nm. Because the lower polarizer is east-west (i.e., horizontal), the particle in Figure 5a is a near match to 1.545, the n of the grain, while that in Figure 5b is a near match to 1.595, the N of the grain (NB: Values for N and n are, as usual, give for 589.3 nm). In Figure 5a, many of the plate-like particles, which are talc sheets, appear yellow, which indicates the grain is greater than the liquid, while in Figure 5b, they appear bluish, indicating a near match. While matching Becke lines colors or dispersion staining colors in polychromatic light aids in mineral identification, measurements in monochromatic light are more precise and accurate and used to characterize the optical properties of minerals.

Figure 5

Central-stop dispersion staining images of sample 3 in a 1.545 liquid (a) and 1.595 (b) with an E-W polarizer. In a, the liquid matches n (i.e., the vibration direction perpendicular to the long axis), as indicated by the blue vertically orientated particle in the photo’s center, while in b the refractive index of the fluid is higher, thus matching N (i.e., the vibration direction parallel to the long axis). Also, the talc plates in a appear yellow indicating their refractive index exceeds the liquid, while in b they are blueish indicating a near match (refer to Fig. 2 for a guide to dispersion staining colors).

Figure 6 shows a higher magnification view of the particles at the center of the field of view in Figure 5b, with two elongate talc particles in different orientations, labeled 1 and 2, and a larger calcite grain. Again, the sample is immersed in a 1.595 liquid, but the temperature is raised so the liquid matches particle 1 based on the Becke line color (Fig. 2). The match for N is obtained when the talc’s long axis is parallel to the lower polarizer by slightly lowering the stage (Fig. 6b). Particle 2 also matches when it is rotated E-W (Fig. 6d). To precisely find a match at 589.3 nm, the Becke lines are observed in monochromatic light. Figure 7a shows a magnified image of particle 1 from Figure 6. In Figure 7b, the temperature of the liquid is changed until the grain disappears; this would be the precise match for 589.3 nm. In Figure 7c the liquid is greater than the grain, so a light-colored Becke line enters the liquid when the stage is lowered, whereas in Figure 7d, the light-colored Becke line enters the grain. For the measurements herein, the temperature of the liquid was adjusted until the grain had a near match around 550 nm, then heated in 5° increments to obtain talc/liquid matches to below 486.1 nm and then cooled to obtain them to at least 656.3 nm. With these data dispersion equations could be obtained for each sample, and refractive index values calculated over the range from 486.1 to 656.3 nm. For samples 2 to 21, a 1.595 liquid was used for N and a 1.545 liquid for n, whereas for sample 1, 1.605 and 1.555 liquids were used, respectively.

Figure 6

The same area is imaged as in Figure 5b in a 1.595 liquid, except at a higher magnification and in plane-polarized light showing two elongated talc particles in different orientations (labeled 1 and 2) and a larger calcite grain. The temperature of the liquid is raised so N of the talc particles matches the liquid at 1.5919. In a, particle 1 is parallel to the lower polarizer and shows very low relief (i.e., is a near-match to the liquid), while particle 2 shows high relief (i.e., does not match the liquid). In b, the stage is slightly lowered and particle 1 exhibits colored Becke lines for a grain liquid match at 589.3 nm, while a light colored Becke line goes into the liquid for particle 2. In c and d, the stage is rotated so as to bring particle 2 parallel to the lower polarizer, thus showing its N matches that of the liquid.

Figure 7

These four images are of single elongate talc particle from sample 7 immersed in a 1.595 liquid. The temperature of the liquid is raised until N of the grain matches that of the liquid, for this sample—1.5867. While this match can be based on the color of the Becke lines in polychromatic light as seen in a, a more quantitative method to determine this match is made in monochromatic light where the grain disappears in b at 589.3 nm. In c, the liquid is greater than the grain as a light colored Becke line goes into the liquid, while in d the liquid is less the grain. This is the visual example of the double variation method and precise determination of the refractive index of a solid.

Figure 8 shows an example of colored Becke lines and dispersion staining colors for sample 16 for wavelength matches at N (Figs. 8a and 8b), Nf (Figs. 8c and 8d), and Nc (Figs. 8e and 8f). Compare these to the colors shown in Figure 2; although both methods are used to obtain a match between the liquid and grain, the double variation method, as shown in Figure 7, allows for more precise refractive index determination.

Figure 8

These six images are of a single elongate talc particle from sample 16 immersed in a 1.595 liquid, with the temperature changed so the grain and liquid match at different wavelengths. The images in the left column are in plane-polarized light showing diferent Becke line colors, and those on the right show different dispersion staining colors. (a and b) Matches at 589.3 nm, (c and d) 486.1 nm (or Nf in Table 5), and (e and f) 656.3 nm (or Nc in Table 5).

Results and discussion. Table 4 lists the N and n values for the 21 samples in this study, the calculated birefringence (δ), and compositions (i.e., Fe and F) for 20 of the samples; note the Fe is expressed in FeO% as well as Fe/(Fe+Mg). As is

typical, the refractive index values given in Table 4 are for Na light (i.e., 589.3 nm). From the literature review, there are scant optical data on talc from known locations. However, samples 14 and 16 correspond and agree with the data in Table 2 for the samples from New York of Wright (1960) and Ross et al. (1968). Sample 3 is from the same mine as the samples measured by Chidester (1962), and have similar values for N or his γ. Finally, the data from McCrone et al. (1979) are from the same deposit as sample 2 with similar optical data.

Table 5 provides observed refractive index values at 486.1 nm (i.e., Nf and nf) and 656.3 (i.e., Nc and nc). Nf-Nc, the dispersion coefficient, is then calculated for both N and n values. The dispersion coefficients are required to assign a matching refractive index to a material based on the dispersion staining colors as described in detail in Su (1993, 2003). The only other published values for dispersion for talc are given in Table 2. For both references, the value for α (which correlates to n in this study), appears incorrect, too high (McCrone and Delly 1973) and too low (McCrone et al. 1979). Also, the value for γ (which correlates to N in this study) appears too low (McCrone and Delly 1973), whereas β and γ are in agreement with the data herein (McCrone et al. 1979). Regardless, their data are not self-consistent, so the data presented in Table 5 appear to be the only reliable dispersion data for talc.

Figure 9 shows plots of N and n as a function of FeO%. As would be expected, the refractive indices increase as Fe replaces Mg in talc [see Chapter 18 of Dyar and Gunter (2019) for a more thorough discussion of the relationships of optical properties and composition]. For each plot, 16 of the samples are shown as black dots, whereas samples 14 and 16 are plotted as red plus signs and samples 19 and 20 are plotted as green boxes. For N (Fig. 9a), the plus signs follow the trend, but for n (Fig. 9b) they plot above it; the green boxes plot below the trend for both n and N. In Table 4 samples 19 and 20 have both the lowest values for Fe and the highest for F. Although Fe increases the refractive index as it replaces Mg because it increases the electron density, F will lower the refractive index as it forms more ionic bonds. For this reason, these samples depart from the trends in Figure 9. The larger n values for samples 14 and 16 relate to how these talc crystals formed and the associated remnants of randomly orientated elongate anthophyllite between (i.e., parallel to) the talc layers. McNamee et al. (2015) discuss this in detail and show images of talc formation on the {210} planes of anthophyllite. Although these have an effect on n, they have little to no effect on N because the preferential orientation to light vibrating parallel to the layers is affected more than for that normal to the layers. Finally, the values obtained for N and n of these samples agree with those obtained by Wright (1960) and Ross et al. (1968).

Figure 9

Plots of N (a) and n (b) vs. FeO% for the 20 samples in Table 4, with associated best-fit linear regression line. The green block symbols are for samples 19 and 20, which deviate from the regression lines based on high-F content. The red plus signs are for samples 14 and 16 which deviate from the regression line for n, probably because these samples contain remanent anthophyllite as explained in the text and shown in McNamee et al. (2015). Finally, the slope of N is greater than n.

Based on the above samples 19 and 20 are excluded from the regression data in Figure 9a and 9b, whereas 14 and 16 are excluded from Figure 9b. A regression analysis then yields the following two equations:

There is a statistically better fit for N than n, but both are statistically significant. Also, the slope for n is less than for N, which makes structural sense as Fe would preferentially increase the electron density within the sheets has compared to perpendicular to them. Similar relationships are shown for andalusite (Gunter and Bloss 1982), where Fe increases the refractive index at a faster rate parallel to edge-sharing chains of octahedral, and for olivine as Fe substitutes for Mg parallel to octahedra in the β direction (Dyar and Gunter 2019).

In general, sheet silicates are optically negative because the electron density is greater within the sheets than perpendicular to them. This is similar to other minerals with ring structures or channels in that the majority of the bonding—and thus electron density—is parallel to the rings or channels (Palmer and Gunter 2000). Also, trioctahedral sheet silicates are often nearly uniaxial negative, while dioctahedral ones are not (Dyar and Gunter 2019). Curiously, a literature search found no other published study showing the relationship of the optical properties of sheet silicates to composition, except for chlorite (Albee 1962), which at first glance appears not to follow this trend. Figure 10 shows a plot of the trend lines for Albee’s α, β, and γ values vs. composition as modified by Bloss (1985). Chlorites plotting on the left side of the diagram (i.e., high Mg) are optically positive, whereas those on the right side (i.e., high Fe) are optically negative. Also, in this illustration, the three lines interest near the middle and, in turn, mark the change in optic sign, whereas the refractive index values parallel to the a-axis and b-axes increase at a faster rate than parallel to c-axis.

Figure 10

A graph of refractive indices vs. composition (Albee 1962) for chlorite group minerals with the three refractive index relationships relabeled for this biaxial mineral by Bloss (1985). Mg-rich chlorites (those plotting on the left of the diagram) are optically positive, so the largest refractive index is perpendicular to the layers. Chlorites change to optically negative near the middle of the graph as higher atomic number cations replace Mg. Because the structure of chlorite has O-H bonds near parallel to the c-axis in the octahedral layer, this results in the largest refractive index value for Mg-rich chlorites, but with increased electron density in the sheets Mg-poor chlorites become optically negative like most sheets silicates.

The reason for the different rates of change is explained by consideration of the crystal structure of chlorite (Fig. 10b). The chlorite structure is similar to that of talc but with a layer of octahedra between each “talc layer.” The octahedral layer has H atoms (small black spheres) bonded perpendicular to it. The OH bonds contribute to the refractive index direction along c-axis, and for the Mg end-member, cause this direction to have the largest refractive index. As Fe substitutes for Mg the refractive indices parallel to the sheets increase at a faster rate than perpendicular—the same effect was seen for talc, but in this case, the trends cross and Fe-rich chlorite becomes optically negative, like all other sheet silicates. Gunter and Ribbe (1993) show a similar effect for the natrolite group zeolites and how the orientation of channel H₂O has a differential impact on refractive index as a function of orientation. Palmer and Gunter (2000) also show how the mean refractive index of heulandite group zeolites can be used as a proxy to determine the H content in cation-exchanged samples. Thus H, OH, and H₂O all play a major role in the interpretation of the refractive index of minerals.

Final thoughts

It was most certainly one of the highlights of my professional career to be able to serve MSA as president during its centennial year. But I am only one of now a 100 presidents and hundreds of additional members to serve MSA in its first century. I also feel fortunate to have selected optical mineralogy as my specialty. I recall being told the world only needed three optical mineralogists when I chose this field for graduate studies. Based on my experiences, the need for optical mineralogists, especially outside of academics and geology, is greater than in any other area in geology. At least I know at this point the demand for our skills far exceeds the people with those skills, and, as mentioned above, the private sector is trying to meet this need with week-long “training” courses designed to convey only the most basic principles of the method, but of course one cannot hope to learn optical mineralogy in a week! I think the main role of optical mineralogy, or mineralogy in general, is the correct identification of minerals. This, of course, can only be done once the minerals are correctly characterized, which is most certainly also our realm. Perhaps it is time we spend more effort on outreach and education and less time on our research, as recent trends show us that it seems hard for society to just “follow the science”; maybe we should try and lead in that area—just a final thought.