Faculty Spotlights

Neutron Diffraction at Los Alamos National Laboratory
Verifies Existence of Some of the World’s Largest Gold Crystals

 JOHN RAKOVAN
Gold is frequently found in nature in its native state, i.e. as gold metal. Less commonly it is found in morphologically well-developed crystals, and those larger than one centimeter in size are quite rare. Thus, few studies have been made on the crystallinity of large gold crystals.  In 2006 I was given access to a collection of several dozen gold crystals, ranging in size from one to two centimeters, to study by X-ray diffraction. The samples were from placer deposits in the Icabaru area in Venezuela (Figure 1). One crystal in the batch was puzzling, showing a single crystal diffraction pattern in one orientation but a polycrystalline nature in all other orientations. This lead to the hypothesis that weathering and erosion of the gold crystals from their original source had altered their surfaces, although the overall single crystal morphology was preserved. To test this idea we needed to look at the interiors of the crystals, but without cutting them in half. To do this I turned to neutron diffraction because of the deeply penetrating nature of neutrons. The experiments were conducted Los Alamos National Laboratory’s (LANL) Lujan Neutron Scattering Center and the results were published in Rocks & Minerals (Rakovan et al. 2009). 

In January of this year I returned to Los Alamos to continue work on the crystallinity of gold specimens from Venezuela. Colleagues Heinz Nakotte, New Mexico State University professor and instrument scientist for the single-crystal diffraction (SCD) instrument, and Sven Vogel, instrument scientist for the high-pressure/preferred orientation (HIPPO) instrument, and I analyzed three of the world’s finest gold crystals, including an extraordinary octahedron that weighs 217.78 grams and measures 4.44 x 4.27 x 2.06 cm (Figure 2).  The structure or atomic arrangement of gold crystals of this size has never been studied before and we have a unique opportunity to do so.

The samples were all collected, over a thirty-year period, in the jungles and on the Gran Sabana, west of the Roraima Shield, Bolivar, Venezuela; home of the tepuis (table-topped) mountains and Angel Falls. This remote area was the inspiration and backdrop for Sir Arthur Conan Doyle's (1912) The Lost World. Interestingly, gold is not the only precious material found in the placer deposits of the Roraima Shield. There are also diamonds, and the two are sometimes found together, although their geologic origin is quite different

Using neutrons, that are created by smashing apart atoms (a spallation neutron source), we were able to “see” into the center of these large and very dense crystals to determine the nature of their crystallinity, and how they may have formed (Figures 3 & 4). Previous studies using X-rays were insufficient because the X-rays are so readily absorbed by heavy elements such as gold. X-rays will only penetrate the surface of these crystals to depths no greater than a few hundredths of a millimeter. Neutrons, however, are much more penetrating and can travel 1-2 cm into gold, which is just right for reaching the centers of these large samples.

The purpose of the study was twofold. The first was to do a bit of forensic science. Because of the spectacular size and perfection of these samples, their authenticity as real, naturally formed single crystals had been called into question. There was no doubt that they are gold; the question was whether Mother Nature was responsible for their beautifully faceted shapes. If they are natural---and we now know from our study that most of them are---there is great scientific interest in the nature of their crystal structures and how they may have formed, which was the second goal of the study. On April 7th a news release about the experiments (Ambrosiano 2014) was posted on the LANL online Newsroom, and more recently, the story went viral and has been reproduced on a large number of websites worldwide.

rakovanfig1

Hydraulic mining of a gold placer
Icabaru, Venezuela

rakovanfig2

The world's largest known
single crystal of gold

rakovanfig3

Moving the sample stage into the
expirement chamber of the HIPPO neutron
diffractometer, LANSCE, Los Alamos

rakovanfig4

Staff scientists Sven Vogel and Heinz Nakotte
with John Rakovan at the HIPPO neutron
diffractometer, LANSCE, Los Alamos

REFERENCES

Ambrosiano, N. (2014) World’s largest single crystal of gold verified at Los Alamos. Los Alamos National Laboratory News Releases: http://www.lanl.gov/newsroom/news-releases/index.php

Doyle, A. C. 1912. The lost world. London, U.K.:  Hodder and Stoughton.

Rakovan, J., N. Gasbarro, H. Nakotte,  K. Kothapalli, and S. C. Vogel. Characterization of gold crystallinity by diffraction methods. Rocks & Minerals 84:54-61.

Tremsin, A.S., Vogel, S.C., Mocko, M., Bourke, M.A.M., Yuan, V., Nelson, R.O., Brown, D.W., and Feller W.B. (2013) Energy resolved neutron radiography at LANSCE pulsed neutron facility.  Neutron News, 24:28-32.


Detecting Earthquakes Induced by Energy Development

Mike Brudzinski

    Enhanced recovery of oil and gas through hydraulic fracturing of shale reservoirs is having a remarkable impact on the geosciences, not to mention the US economy.  Nevertheless, there are causes for concern in regards to its potential impact on the environment.  It is a complicated debate, and one that my research had little to do with on until a few years ago.  I have spent most of my career working on subduction zones, and I was focusing on repeating earthquakes and their potential relationships to great earthquakes that can devastate coastal populations.  However, an unusual sequence of earthquakes began occurring in the previously aseismic area of Youngstown, Ohio in March 2011 and the earthquakes grew more frequent and larger by the end of that year.  The Ohio Department of Natural Resources (ODNR) became concerned that a nearby deep injection well designed to dispose of the wastewater left over from hydraulic fracturing might be the cause of the earthquakes, despite the fact that there were over 200 injection wells in the state without any similar earthquakes.  A local seismic network was deployed in November 2011, which showed the events were occurring in the shallow basement rock within a mile of the well.  Injection at the well was halted a little over a month later, but a magnitude 4 earthquake happened before the subsurface pressures had a chance to subside.

     My former doctoral student Steve Holtkamp and I were intrigued by this sequence so we decided to try adapting software we were using to identify repeating earthquakes in subduction zones.  This software uses the seismograms from an earthquake like a fingerprint and then scans for similar features in other recorded seismic data.  One of the real advantages is that it doesn't require a local network, so regional seismometers over 100 miles away could be used.  We were delighted to find the fingerprint scans for the Youngstown sequence discovered 100 additional smaller earthquakes that were previously undetected.  We enlisted Brian Currie to join our team to help make sense of the well history, and comparisons to the new earthquake catalog showed that 1.) events started almost immediately after injection began, 2.) earthquakes migrated away from the injection well over time, and 3.) the total number of earthquakes was proportional to the total amount of wastewater injected.  We even found that daily fluctuations in the injected wastewater translated to pulses in the number of earthquakes -- we could almost see the wastewater pushing the fault to move.  We weren't the only ones excited as Holtkamp's presentation of this work at the 2013 AAPG meeting was the first ever to be awarded both best student presentation and best meeting presentation (Matson Award). 

     Based on this success, we asked masters student Rob Skoumal to join the team and help optimize the fingerprint scanning technique to apply it to other potentially induced earthquakes.  He was able to utilize the Miami University Supercomputer to achieve over a million seismogram comparisons per second, such that we could complete a new earthquake fingerprint scan in about a half hour. This development enabled us to dig back through years of earthquakes in Ohio to look for other evidence of induced earthquakes, and we are starting to find similar cases in places like West Virginia and Oklahoma.  Yet the real value is in evaluating new earthquakes.  In essence, if the fingerprint scan reveals very few matches, then it is most likely the new earthquake is natural, but if it reveals dozens of similar earthquakes, it suggests the earthquakes may be human induced.  For example, a magnitude 3 earthquake occurred earlier this year about 10 miles southeast of the Youngstown sequence.  The fingerprint scans immediately revealed 50 similar earthquakes that began less than a week earlier, and no similar earthquakes in the preceding 4 years.  Surprisingly, there were no operating injection wells near these events, but instead we found active hydraulic fracturing in the area.  Further analysis revealed the earthquakes only occurred when well stimulations were within half a mile of a small previously unknown fault in the basement rocks.  These results are guiding the industry and regulators in how to proceed with future completion efforts.

     Earthquakes induced by wastewater injection or hydraulic fracturing still appear to be rare, with only ~1-2% of injection wells in Ohio tied to earthquakes and even lower rates from hydraulic fracturing itself (Figure 1).  Yet the fingerprint scanning and subsequent analysis by the Miami team provides an opportunity to identify the exceptions early such that changes in practice can be made to help manage long-term risks.  We are now actively partnering with ODNR and folks in the industry to help monitor for any future cases.

brudzinskifig1

Seismometer deployed in Eastern Ohio by
MU faculty & students to monitor for earthquakes
potentially caused by wastewater injection wells
or hydraulic fracturing

brudzinskifig2

Map of Ohio showing earthquakes
we have investigated with earthquake
fingerprinting scanning, as well as
two earlier reports of injection induced
seismicity in northeastern Ohio