APPLIED PHYSICS LETTERS 99, 204101 (2011)
a)
Ra u´ l A. Baragiola, Catherine A. Dukes, and Dawn Hedges
Engineering Physics, University of Virginia, Charlottesville, Virginia 22904, USA
(Received 10 August 2011; accepted 26 October 2011; published online 14 November 2011)
We report the production of up to 10 ppm ozone during crushing and grinding of typical terrestrial
crust rocks in air, O and CO at atmospheric pressure, but not in helium or nitrogen. Ozone is
2
2
formed by exoelectrons emitted by high electric fields, resulting from charge separation during
fracture. The results suggest that ground level ozone produced by rock fracture, besides its
potential health hazard, can be used for early warning in earthquakes and other catastrophes, such
Earthquakes can inflict enormous costs in lost lives and
property damage. Although they are unpredictable due to
their chaotic nature, earthquakes may be detected early
through multiple phenomena that precede them by up to sev-
eral days. Researchers have been actively seeking reliable
identification of seismic precursors with the goal of using
them to enable early warnings to help minimize tragedy. So
far, phenomena associated with rock deformation or fracture,
including the emission of electromagnetic (EM) waves,
rough rocks with the vice, in addition to the inconstant
amount of pre-existing cracks in the samples. Therefore, we
devised a more controlled method of fracture, systematic
grinding, a method known to result in the emission of EM
7
waves.
Fracturing was done with a 7.1 mm silicon carbide
grinding stone attached to a vertical high-speed (30 k rpm)
Dremel. Similar results were obtained with an alumina bit.
The rocks, fixed in a polytetrafluoroethylene jig, were
abraded for 260 s, after which the chamber atmosphere was
pumped into the ozone detector through a 5-6 lm PTFE par-
ticle filter. Figure 2 shows the results of the grinding experi-
ments on rocks of differing petrology. Initially, the ozone
signal increases due to the transit time of the gas to the detec-
tor, then peaks and decays as gas is pumped from the cham-
ber. The strongest measured signal after 260 s grinding was
10 600 ppb, from rhyolite (for context, an extended exposure
1
atmospheric ions, and radon, have been unsuccessful as
2
early warnings to minimize tragedy. On the other hand,
there is ample (mostly anecdotal) evidence that animals can
3
–5
anticipate earthquakes, but the cause of this sensitivity is
elusive, with no conclusive correlations with physical pre-
cursors that could be used for earthquake early warning.
Could the link between pre-seismic animal behavior and
electrical phenomena be ozone? This gas, easily detectable
6
8
at ꢀ7 ppb by humans, is a by-product of electrical dis-
to 100 ppb ozone is considered unhealthy for most humans
9
and lowers the yield of agricultural crops ). The total amount
charges in air, which could accompany rock fracture. In fact,
ozone has been reported to appear months prior to an earth-
of ozone produced in the 260 s grinding can be estimated
from the integral of the peaks and the pumping speed and are
(in micro moles): 0.12 (basalt), 0.1 (gneiss), 0.4 (granite),
0.7 (rhyolite), and 0.3 (schist).
3
quake in the vicinity of the epicenter.
Here, we present evidence for ozone generation during
rock fracture in air, advance a model that accounts for its
production, and propose that rock fracture may be a frequent
source of ground ozone pollution in different environments.
We fractured single igneous and metamorphic rocks in
air at atmospheric pressure within a small (ꢀ5 L) acrylic
The fact that similar levels of ozone were produced both
in compression fracture with the vice and, by grinding with
two different bit materials, show that fracture, and not bit-
rock rubbing, is the cause of ozone production. It also rules
out heating during grinding as a significant factor in ozone
production, besides the fact that heating is not a plausible
mechanism of ozone production from rocks.
2
chamber with a 0.2 cm vent. Chamber gases were pumped
at 1 L/m into a dual-beam photoelectric ozone monitor (2B
Technologies 205) before, during, and after fracture. The de-
tector has an accuracy of 2% or 1 ppb, and a sensitivity of
To determine if the production of ozone is dependent
primarily on atmospheric oxygen, silicate material, or gas
released during rock fracture, we measured the effects of
varying the gas environment. Gasses with and without oxy-
gen were chosen for the study. For these experiments, we
ground rhyolite for 260 s at 1 atm, detecting ozone during
rock fracture only in air and in oxygen, and to a lesser extent,
in carbon dioxide atmospheres. No ozone was detected when
grinding in nitrogen or helium, as shown in Fig. 3. This
implies that oxygen in the environment of the fracture region
is necessary for the production of ozone.
0.1 ppb; it is crosschecked with an ozone calibrator (Eco
Sensors OG-3). The time constant for ozone decay is ꢀ300 s,
determined by the chamber volume, pumping speed and
adsorption, desorption and reactions from walls, and the
sample and rock fragments. Background ozone is measured
to fluctuate around 1.5 ppb.
Initial experiments were done by compression in a vice
until fracture, and recording the ozone emission in real time.
Fig. 1 shows results for a piece of granite, ꢀ30 mm ꢁ 30 mm
ꢁ
15 mm in size. We found a strong variability in ozone pro-
duction (a factor of 4) likely due to uneven contact of the
We propose the mechanism for ozone generation is frac-
toemission of electrons, a processes initiated by large electric
fields induced by charge separation due to fracture, which
a)
Electronic mail: raul@virginia.edu.
0003-6951/2011/99(20)/204101/3/$30.00
99, 204101-1
VC 2011 American Institute of Physics
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
31.156.157.31 On: Mon, 24 Nov 2014 16:39:31
1