3.8 Ga Impact Zircons?: Technique Development

Timothy O'Brien
University of Rochester
Dept. of Earth and Environmental Sciences

OPT407: Electron Microscopy
Spring 2014

1. Introduction
2. Methods
3. Results and Discussion
           3.1 Zircon Crystal Surface: Secondary Electrons & Backscatter Electrons
           3.2 Zircon Interior: BSE and X-ray Analysis
4. Conclusion
5. Future Work
6. References
7. Acknowledgements

 

1. Introduction

The Late Heavy Bombardment (LHB) was a period of high frequency meteorite impacts on Earth during the early Archean Eon between 3.8 and 3.9 Ga that was first reported from Lunar impact crater dating (Cohen et al., 2000; Tera et al., 1974). Thus far, no terrestrial evidence has of such a period has been reported, likely a consequence crustal recycling via plate tectonics that has effectively erased the record of any such impact craters. Ancient detrital zircon crystals have potentially preserved evidence of LHB through specific shock fracture sets that have been associated with other well studied meteorite impacts (Erickson et al., 2013). Zircons themselves are extremely stable minerals that are not only capable of preserving such structures, but of being preserved themselves with minimal changes through time. An extensive survey of zircons from the Jack Hills region of Australia, dating from the LHB period was conducted by Montalvo et al. (2014) in an effort to locate terrestrial evidence of 3.8-3.9 Ga meteorite impacts. Despite evaluating over 1400 crystals, the survey failed to produce any impact shocked zircons.

Figure 1: Location of Wanderer Conglomerate, Zimbabwe (from Dodson et al., 1988)


The Wanderer Conglomerate (WC) offers a rare opportunity to study Archean zircons that have not been subjected to metamorphic conditions beyond low to mid-Greenschist facies. Part of the 3.5 Ga Sebakwian Group, zircons from the Wanderer Conglomerate have been reported by Dodgson et al. (1988) to date from ~3.8 Ga, the approximate end of LHB. The aim of the work conducted herein is to serve as a pilot study to determine the most effective techniques for observing the presence or absence of impact associated fracturing of zircons, similar to those discussed by Erickson et al. (2013) in their study of the 2.5 Ga Vredefort Dome impact structure of South Africa. Optical microscopes do not effectively provide the surface contrast that Scanning Electron Microscope (SEM) images can, which is required to fully map and differentiate fracture patterns within individual crystals (Figure 2A). Multiple imaging methods (Secondary Electron Detector, In-Lens Detector, Backscatter Electron Detector, Energy Dispersive X-Ray Spectroscopy) have been evaluated to determine those that quickly and clearly display surface fractures across zircon crystals. An additional goal of great importance for this study was determining an effective means of mounting the zircons. Zircons of the Wanderer Conglomerate present a challenge for mounting and handling due to their size and delicateness that needed to be addressed to permit imaging of both the complete crystal and polished cross section of the crystal in the SEM.

It is important to note that this work is by no means as thorough as would be required for a formal study. A larger quantity of zircons is required to better characterize the population within the Wanderer Conglomerate. Instead, this represents an initial assessment of the quality of zircons preserved in the Wanderer Conglomerate, while fine-tuning some of the preparation and analysis techniques that will be required for a more thorough future survey.

(back to top)

 

2. Methods

Sample Collection

Mounting zircons presented the most challenging problem for this study. As previously discussed, the size of zircons being studied (<100 um, Figure 2A) prohibits extensive handling and the small quantity available for study necessitates maximization of analyses that can be conducted on each zircon. The zircons represent material weathered from an older crystalline source (~3.8 Ga) and deposited into younger sedimentary rocks. Obtaining the zircons requires the time intensive task of crushing the host sedimentary rock and carefully searching through the crushed material (Figure 2C) using delicate tools that permit manipulation of individual crystals/grains (Figure 2B). Only a few zircons were located within the material one of which shattered during attempted mounting. As a result only one complete and one partial zircon were successfully mounted for this study. Such is the nature of working with zircons and the reason why handling of the crystals must be minimized.

Figure 2: A) Zircon from Wanderer Conglomerate, B) Eye-lash Sorting Tool, C) Zircon-containing rock crush


Sample Mounting

Initially, the zircons were to be mounted to a stainless steel stub using conductive carbon tape. However, the surface topography of the tape possess hemispherical depressions larger than the crystals themselves that would have resulted in the potential loss of crystals into the carbon tape. Additionally, any attempts to orient mounted crystals would also result in the sample being engulfed in carbon. The required mount needed to provide a secure surface for the zircon while permitting further preparations for additional analyses. As such, the zircon was mounted to a standard glass slide using a clear two-stage epoxy (Figure 3). Optical clarify of the epoxy is key to readily allow sample location. A circle of conductive liquid paint was placed around the mounted zircon and the entire slide sputter coated with gold to provide a conductive coating to the zircon for preliminary imaging. The gold was applied using a Denton Desk-II sputter coater to a thickness of ~90 Å. The glass slide was mounted to a stainless steel stub using carbon tape and grounded with a strip of copper tape.

Figure 3: Zircon crystal embedded in epoxy and grounded with copper tape

Analyses

The zircons were imaged using Secondary Electrons (both SE2 and In-lens detectors) and Backscatter Electrons (BSE) to determine which provided the best image of the crystal surface. Imaging was conducted using a Zeiss Auriga CrossBeam SEM-FIB at the University of Rochester's Institute of Optics. After imaging the crystal surface, additional epoxy was added to fully encase the whole zircon and allowed to set for 24 hours. The epoxy encased zircon was then polished down using polishing sheets of decreasing grit size (15 um, 6 um, 3 um, 1 um) to expose a cross sectional view of the crystal for imaging and Energy Dispersive X-Ray Spectroscopy (EDS). The analysis was performed with the aim of determining if any fractures that did not express themselves on the crystal surface could be more clearly distinguished in the crystal interior via BSE imaging and to determine if any layering from crystal growth (similar to that of hail) was present. The layering would result in geochemical differences that could potentially be recorded using EDS elemental mapping and subsequently evaluated for offsets within the layers, indicating fine fractures through the crystal (similar to offset of bedding along a fault).

Image Post-Processing

All post-image processing was conducted using the open source software GIMP v2.8 and Inkscape, as well as Adobe Photoshop CS6.

(back to top)

 

3. Results and Discussion

3.1 Zircon Crystal Surface: Secondary Electrons & Backscatter Electrons

Figures 4-7 provide preliminary information regarding both the characteristics of the Wanderer Conglomerate zircon population, as well as which method of imaging provided the most information regarding crystal fracture sets (if any). Erickson et al. (2013) produced detailed images of "shocked" zircons from Vredfort Dome impact crater in South Africa that serve as a visual guide to characteristic fracture sets. As the partial zircon (Figure 6) displays no visible fractures, the who zircon provides the best means of a preliminary assessment for these features. A side-by-side comparison of the Wanderer zircon (Figure 7) with a sample of the Erickson et al. (2013) zircons immediately reveals the lack of fracture intensity in the Wanderer zircon that is clearly visible in the Vredfort Dome "shocked" zircon. The observed fracture set outlined in the Wanderer zircon (Figure 7) can likely be attributed to the crush process by which the zircons were separated from their host sedimentary material.

Despite no observable "shock" fracture sets being observed in the Wanderer Conglomerate zircons, the process of imaging the two zircons via secondary electrons (Figures 4 and 6) and backscatter electrons (Figures 5 and 7) shows BSE images as superior in this instance for providing clear contrast of surface fractures of the zircons.

Figure 4: SE image of zircon
Figure 5: BSE image of zircon

Figure 6: Composite SE image of partial zircon (In-lens & SE2 Detectors)

Figure 7: False-color BSE image of zircon with major fracture set highlighted with dashed lines



Figure 8: Zircon from Vredfort Dome, South Africa displaying "shock" fractures(From Erickson et al, 2013)



3.2 Zircon Interior: BSE and X-ray Analysis


Having evaluated the exterior surface of the crystals for evidence of impact associated fracture sets, the whole zircon crystal remained in-situ to its mounting and was encased in epoxy for polishing. The aim was to develop a means of allowing a seamless transition from imaging the external surface of whole crystals to polishing them to reveal a cross-section of the interior for imaging and x-ray analysis. Maintaining the crystals in their original positions prevents potential loss of crystals during handling and allows one to better maintain a record of data associated with each crystal when working with numerous crystals at once. With the intent of performing EDS on the zircon, the sample was carbon coated using a thermal high vacuum evaporator. This was done to avoid any overlap that could occur between characteristic peaks of zirconium (Zr) and a gold (Au) sputter coating.

Figure 9: BSE image of polished displaying concentric growth rings. 1-4 denote the individually identifiable growth rings (BSE image)
Figure 10: BSE image with high contrast displaying close-up of zircon concentric growth rings. A-C identify the boundaries between individual growth rings.

Polishing the zircon revealed a series of growth rings (Figures 9 and 10) that could be used to evaluate offset across annealed fractures. The growth rings are attributed to compositional changes of the magma during formation of the zircon. In this instance, further polishing would be required to produce a cleaner surface for evaluating the fractures. It is still clear that minimal fractures penetrate the interior of the zircon and the observed fractures can likely be attributed to the polishing process.
The rough final surface of the zircon, while clearly displaying the presence of growth rings, does not allow a clear view of the rings across the entire surface, making it difficult to study them for potential offsets across fractures. With future analyses of Wanderer zircons likely, an additional analysis was performed in an attempt to take advantage of the potentially distinguishable differences in chemistry between growth rings for evaluating offset across annealed fractures. EDS was used to develop a series of elemental maps of the zircon. It was hoped that strong differences in major elemental composition between the growth rings (if present) would provide a clearer view of the rings themselves. The elemental maps themselves (Figure 11) failed to produce such a result, which can be attributed to multiple possibilities: 1) no significant difference in major element composition between growth rings, 2) Selecting the entire crystal for mapping did not allow for a high enough resolution evaluation of growth ring geochemistry, and 3) Major geochemical differences present in trace elemental or isotopic information.

Figure 11: Zircon Elemental Maps
Figure 12: Zircon Energy Dispersive X-Ray Spectra

Despite the failure of EDS elemental maps to differentiate between the zircon's growth rings, an EDS line spectra was collected sub-parallel to the crystal's long axis (Figures 13 and 14). Subtle changes in the spectra could possibly be attributed to chemical differences between growth rings, but further investigation would be required to confirm this.

Figure 13: Zircon with approximate location of EDS line spectra.
Figure 14: Zircon spectra line

(back to top)

 

4. Conclusion

While the zircons evaluated in this study do not possess fracture patterns associated with impact events, two crystals are certainly not representative of the entire population of zircons within the Wanderer Conglomerate. The most important result of the study was determining the advantage of Backscatter Electron images for crystal surface structure in this instance over those produced using Secondary Electrons. Additionally, a preliminary technique was established to permit imaging of the crystal and subsequent polishing to reveal a cross sectional view, without having to relocate the crystal from its mount. This limits the amount of handling required for the easily displaced and damaged zircons.

(back to top)

 

5. Future Work

As a pilot study, there is obvious room for improvement and future work. Aside from the obvious continuation of searching the zircon population of the Wanderer Conglomerate for impact zircons, the methodology of mounting zircons requires further work. While the methods employed for the study were a first step and did ultimately work, it would be greatly beneficial to improve the technique so as to minimize the adhesive epoxy employed for each zircon. This will ensure complete exposure of the crystal for imaging. A circular glass slide with rounded edges would also be beneficial and make sample polishing simpler, without the risk of an edge catching on the polishing wheel.

(back to top)

 

6. References

Cohen, B.A., Swindle, T.D., and D.A. Kring (2000) Support for the Lunar Cataclysm Hypothesis from Lunar Meteorite Impact Melt Ages. Science, v.290, p 1754.

Dogdson, M.H., Compston, W., Williams, I.S., and J.F. Wilson (1988) A search for ancient detrital zircons in Zimbabwean sediments. Journal of the Geological Society, v.145, p 977-983.

Erickson, T.M., Cavosie, A.J., Moser, D.E., Barker, I.R., and Radovan, H.A. (2013) Correlating planar microstructures in shocked zircon from the Vredefort Dome at multiple scales: Crystallographic modeling, external and internal imaging, and EBSD structural analysis. American Mineralogist, v.98, p 53-65.

Montalvo, P.E., Cavosie, A.J., and Valley, J.W. (2014) A Constraint on Shocked Mineral Abundance in the Jack Hills Zircon Suite. In Proceedings, 45th Lunar and Planetary Science Conference, March 17 – 21, The Woodlands, Texas.

Tera, F., Papanastassiou, D.A., and G.J. Wasserburg (1974) Isotopic Evidence for a Terminal lunar Cataclysm. Earth and Planetary Science Letters, v.22, p 1-21.

(back to top)

 

7. Acknowledgements

I would like to thank Brian McIntyre for all of his assistance and suggestions in preparation of samples and use of the SEM.

(back to top)

Please enter any comments, criticisms, questions, etc. below.

Your name:

Email address: