Geochemistry, Mineralogy, Petrology & Volcanology

Is garnet going to be your best date?

Is garnet going to be your best date?

For geochronologists it is all about precision and accuracy. For petrochronologists it is more about precise positioning of those ages to trace back the evolution of a rock. For many, it is not just about getting a crystallisation or metamorphic age, it is about knowing when and at what conditions a given rock experienced certain petrogenetic processes. As we progressed from geochronology to petrochronology, many obstacles needed to be lifted, many of which related with analytical limitations and intrinsic factors. And why is that? Because in an ideal world we would have large, perfect crystals with no initial daughter isotope, but this is seldom the case. Many of the minerals that have significant amounts of radiogenic isotopes that can be used for dating are usually small, preserve chemical zoning and/or multiple crystallisation cycles, and more often than not incorporate some initial daughter isotope. Additionally, we also have isotope interferences during the measurements that further complicate things.

For many petrologists out there, garnet is perhaps the most incredible mineral that there is. It is not just the beauty of the faceted crystals or the mesmerising rainbow garnet andradite variety, but the powerful information that this mineral carries with it. How many metamorphic petrologists have relied on garnet’s chemistry to determine (multiple) P-T conditions of a rock? How many have used mineral inclusions shielded in garnet to determine a pre-existing, prograde mineral assemblage that aided inferring the P-T loop? How many have been using detrital garnet in provenance studies? It is undisputed that garnet is one of the most relevant minerals for petrologists, and this is why it was always very clear that obtaining ages from this mineral would be absolutely essential. But how can we date it?

In the 80’s, very relevant advances in solution isotopic dating applied to garnet were made. In the early 80’s, garnet was dated by Van Breemen and Hawkesworth, but also by Griffin and Brueckner using Sm and Nd isochrons. Indeed, garnet incorporates Sm and very little Nd, making it an ideal Sm-Nd clock. It was at the end of the 80’s when the first U-Pb dating was reported. This was done by Mezger and co-authors in 1989. These authors argued that Pb would be very slowly diffusing, such as in zircon, and therefore garnet would preserve peak metamorphic ages. Indeed, they compared their garnet ages with those of zircon from the same or similar rocks, demonstrating that they were right and that garnet U-Pb dating was possible. But, as they pointed out, the concentration of U in garnet is commonly very low, in the range of a few hundreds of ppb, making these analyses very challenging. Additionally, and in a similar way to garnet Rb-Sr dating, mineral inclusions can drastically change isotopic measurements and influence U-Pb final dates. It was only in the second half of the 90’s that garnet Lu-Hf dating was first reported by Duchêne and co-authors, owing it to garnet’s preference for Lu over Hf and the ability then to properly measure Hf isotopes.

Garnet is commonly dated by isotopic dilution of Sm-Nd using TIMS or of Lu-Hf using MC-ICP-MS, after crushing one or several garnet grains. It has also been dated using U-Pb by TIMS. Mineral inclusions and elemental/growth zoning are enemies of precise and accurate isotopic dating, especially when done by isotopic dilution techniques. Another complication arises from the multiple reactions that can produce garnet during a rock’s evolution, and thus potential population mixing during garnet fragment selection for analysis. These are commonly overcome by only integrating material from one single grain (e.g. using microdrilling), when they are very large, or by carefully examining and selecting multiple garnet grains or fragments (“bulk” garnet) under the binocular and scanning electron microscopes. An excellent review on garnet dating is given by Ethan Baxter and Erik Scherer in Elements (2013).

Garnet Isochrons

Garnet Isochron dating. a) systematics of Lu-Hf and Sm-Nd isochron dating; b) and c) are examples of how mineral inclusions can induce false isochrons and wrong age determinations. Modified from Cheng, 2019 (Journal of Geological Society of London Special Publications).

Until recently, analyses with sample introduction via laser ablation coupled to a mass spectrometer or by scanning ion mass spectrometry lacked resolution (for Nd, Hf and U). This constituted a setback in the capability of properly exploring and extracting useful information from a garnet capsule. Recent advances on in-situ U-Pb garnet dating have arisen from the interest of obtaining single ages without the need of mineral-mineral/whole rock isochrons. Due to the very low U concentrations found in garnet, commonly under the ppm level, a new focus over U-rich grandite (an andradite variety garnet) has resurfaced. In 2017, Seman, Stockli and McLean presented a methodology to date andradite-grossular garnet by U-Pb using LA-ICP-MS (ThermoFisher Element II, single collector, magnetic sector). This has set the room for many other similar studies that quickly followed, such was the interest and necessity from the skarn community to find a reliable petrochronometer. Yet, despite this awesome contribute, U-Pb dating of other garnet varieties, particularly pyrope or almandine, is not yet easily achievable (though it can be done, as recently reported by Millonig et al) due to the very low U concentrations and a lack of reference materials. So, how can we date those garnets? Well, the answer comes from a rather spectacular analytical solution: the combination of laser ablation with a reaction cell and two mass spectrometers (MS/MS).

Up until now, in-situ measurements of Hf have been hindered by Lu and Yb mass interferences on 176Hf. While precise and accurate Hf isotope measurements in minerals like zircon can be obtained by laser ablation coupled to a multi-collector ICP-MS, the low concentration of Hf in garnet prevents such an approach. In 2021, Alexander Simpson and co-authors have demonstrated the applicability of triple quadrupole mass spectrometry and that of a reaction cell to strip off the interferences that affect Lu-Hf garnet dating. These researchers have used NH3 gas to react Lu, Yb and Hf to shift the masses of those isotopes so that 176Hf could be measured with zero or very residual interferences*. According to these authors, their methodology can yield a precision of about 3 to 5% on the isochron age, which is 2 to 7 times larger than the corresponding precision obtained by ID-TIMS. These results are very promising and reveal that there is still room for improvement. We have now a relatively fast and hassle-free methodology to date individual garnet grains (or garnet zones) and thus make sense of more complex geological histories. One of the many applications of these developments has been published this year by Tamblyn et al. These researchers conducted large-scale, campaign-style garnet dating in the Western Gneiss Region in Norway. This inspiring work is now setting up the stage for what is coming up in the next years, with garnet becoming the best thing that you can date, either in petrological context, as xenocrysts or as detrital grains.

*If you want to know more about Triple Quadrupole ICP-MS technology from Agilent follow the link directing you to YouTube.

Avatar photo
I am a Research Fellow, working on improving petrogeochemical tools using accessory minerals such as titanite and rutile. I am interested to develop tools that allow us to assess metamorphic conditions from detrital minerals and improve our understanding on the evolution of crustal geothermal gradients through time. I love books. I love nature. I love Earth Sciences.

1 Comment

  1. Congratulations Inês for the excelent text.


Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>