NeuroImage 54 (2011) S30–S36

Contents lists available at ScienceDirect

NeuroImage
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y n i m g

Blast-induced electromagnetic fields in the brain from bone piezoelectricity
Ka Yan Karen Lee a,⁎, Michelle K. Nyein b, David F. Moore c, J.D. Joannopoulos d, Simona Socrate e, Timothy Imholt f, Raul Radovitzky b, Steven G. Johnson g
a

Department of Electrical Engineering, Massachusetts Institute of Technology, Cambridge MA 02139, USA Department of Aeronautics and Astronautics, Massachusetts Institute of Technology, Cambridge MA 02139, USA Defense and Veterans Brain Injury Center, Walter Reed Army Medical Center, Building 1, Room B207, 6900 Georgia Ave. NW, Washington DC 20309, USA d Department of Physics, Massachusetts Institute of Technology, Cambridge MA 02139, USA e Department of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge MA 02139, USA f Raytheon Co., 870 Winter St., Waltham MA 02451, USA g Department of Mathematics, Massachusetts Institute of Technology, Cambridge MA 02139, USA
b c

a r t i c l e

i n f o

a b s t r a c t
In this paper, we show that bone piezoelectricity—a phenomenon in which bone polarizes electrically in response to an applied mechanical stress and produces a short-range electric field—may be a source of intense blast-induced electric fields in the brain, with magnitudes and timescales comparable to fields with known neurological effects. We compute the induced charge density in the skull from stress data on the skull from a finite-element full-head model simulation of a typical IED-scale blast wave incident on an unhelmeted human head as well as a human head protected by a kevlar helmet, and estimate the resulting electric fields in the brain in both cases to be on the order of 10 V/m in millisecond pulses. These fields are more than 10 times stronger than the IEEE safety guidelines for controlled environments (IEEE Standards Coordinating Committee 28, 2002) and comparable in strength and timescale to fields from repetitive Transcranial Magnetic Stimulation (rTMS) that are designed to induce neurological effects (Wagner et al., 2006a). They can be easily measured by RF antennas, and may provide the means to design a diagnostic tool that records a quantitative measure of the head's exposure to blast insult. © 2010 Elsevier Inc. All rights reserved.

Article history: Received 25 January 2010 Revised 5 May 2010 Accepted 16 May 2010 Available online 12 June 2010

Introduction In this paper, we show that a blast pressure wave traversing the skull is a direct source of potentially intense electric fields in the brain, which may have significant neurological effects. This unexpected source does not appear to have been considered in studies of primary blast traumatic brain injury (TBI) (blast-induced neurotrauma (Cernak and NobleHaeusslein, 2009)). The mechanism is based on the fact that bone is a piezoelectric material: it polarizes electrically in response to an applied mechanical stress and produces a short-range electric field. A shockwave from an explosion generates large stresses in the skull and, consequently, large electric fields. Using computed stresses from fullhead-model simulations of typical shockwave exposures from improvised explosive device (IED) blasts, we calculate the induced charge density in the skull and estimate the resulting electric fields. We find fields in the brain on the order of 10 V/m in millisecond-scale pulses, more than 10 times stronger than the corresponding IEEE safety guidelines for controlled environments (IEEE Standards Coordinating Committee 28, 2002), and comparable in strength and timescale to

⁎ Corresponding author. E-mail address: kylkaren@mit.edu (K.Y.K. Lee). 1053-8119/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.neuroimage.2010.05.042

fields from repetitive transcranial magnetic stimulation (rTMS) that are known to have neurological effects (Wagner et al., 2006a). Independent of whether these fields play a role in TBI, they can be easily measured by RF antennas and may therefore provide a direct measure of the stresses on the skull: a ‘blast dosimeter’ that could be useful for diagnosis and study of blast-induced TBI. In order to assess the potential neurological impact of any blastinduced electromagnetic fields in the brain, we compare them to published safety standards and also to medical procedures where neurological effects are intentionally induced in the brain via electric fields. At millisecond timescales (ms pulses, ≈1 kHz frequencies) typical of IED-scale blasts, the IEEE safety standard for in-brain electric fields is 0.3 V/m for the general public and 0.9 V/m in controlled environments (IEEE Standards Coordinating Committee 28, 2002). Another point of reference is the medical procedure rTMS, which uses magnetic-field pulses to create electric fields in the brain that, in turn, induce currents which can disrupt brain activity in the short term or have long-term effects by stimulating the release of neurochemicals (Wagner et al., 2006b; Muller et al., 2000). A recent full-head finiteelement simulation of a typical commercial rTMS device found maximum in-brain currents densities of around 4.4 A/m2 (in msscale pulses) (Wagner et al., 2006a); the brain has a conductivity of about 0.28 A/Vm at kHz frequencies (Wagner et al., 2006a), and hence

we describe the methodology of our work. It is well known that the piezoelectric effect occurs in certain Fig. we conclude with future directions and some implications of our results.. 1990. which is measurable far away (Fine.. 1962. and describe the values used in this paper. however. Black and Korostoff.. / NeuroImage 54 (2011) S30–S36 S31 this generates in-brain electric fields of up to 4. Reinish and Nowick.2 m)2 ≈ 100 V/m.. This generates stress in the skull (middle: skull pressure). (Jackson. the key fact is that bone is known to be a strong piezoelectric material (Fukada and Yasuda. First. . Linde et al. 1982. We first review the phenomenon of bone piezoelectricity in Bone piezoelectric response section. given the conductivity of the brain and the fact that the ECT current travels over a distance of around 20 cm. Frankel and Toton. but a simple estimate suggests that the resulting fields are orders of magnitude below the IEEE safety limits for IED-scale explosions in air (see appendix). 1. 1949). 1.. In Methods section.K. 1965. 1966. Then we outline how we calculate the blastinduced charge densities (Blast-induced charge density) and electric field estimates (Electric field estimation). Lee et al. Bassett and Becker. Finally. For example. corresponding to kilohertz-scale (kHz) frequencies f (from the Fourier transform of a ms pulse). an incoming blast wave in the air hits the soldier's head from the side. This causes the bone to polarize and leads to an electric charge density (right) proportional to the stress (and its gradient). and this unfortunately also means that the millisecond pulses induced by IED-scale explosions are on a timescale that is neurologically relevant.02 kbar (Moore et al. a typical IED explosion has a millisecondscale high-pressure duration. summarize the available measurement data. 1996): even though a polarized piezoelectric material is neutral (no net charge) and the resulting fields are short range. In response to a shockwave from an explosion. the adjacency of skull bone to the cerebral cortex means that even short-range fields may be relevant to TBI if they are strong enough. 1988. Bur. 1996)). it is important to realize that such fields are reduced in the brain by a factor of roughly the relative permittivity of the brain matter.4/0. 1974. 1965. Here. Shkuratov et al. much larger EM fields can be generated when the high-pressure blast wave impacts a piezoelectric material (Reed et al. Boronin et al. the supersonic shock front is associated with a temperature increase that can ionize particles in the gas or other materials it passes through. 1994. Then. the field is a linear function of the applied stresses). In particular. Singh and Katz. a simple dimensional analysis gives an estimated in-brain electric field of (1 A) / (0.28 A/Vm) / (0.Y. are produced by electroconvulsive therapy (ECT). Harris and Presles. 1993. the initial explosion generates a flash of EM radiation (including visible light). Pfeiffer. 1975. Stresses are induced in the tissues in the head. This means that neither the wave nature of the light nor the induced magnetic fields are relevant on the scale of a human head: the electromagnetic response is accurately described by an electrostatic model in which purely electric fields are produced instantaneously in response to the charge density (Jackson. 1965.. Finally. 1996) Although EM pulses from this initial flash continue to be studied in many contexts. Methods Bone piezoelectric response Piezoelectricity is a phenomenon in which an electric polarization is generated in response to an applied mechanical stress: P = dσ ð1Þ where P. The net charge is zero. 2009)). Schematic generation of charges from a blast by bone piezoelectricity: (left) an incoming blast wave in the air impacts a soldier's head from the left. here we instead consider the generation of EM fields from the supersonic blast wave itself (outside of the detonation region where the initial flash is generated). (Harris. 1967. an electrical charge density is created via the bone piezoelectric effect due to the tensile and shear stresses. 1998). Anderson and Hagood. 1976. Kelly. Shkuratov et al.28 ≈ 16 V/m. 1989. First. In evaluating fields generated outside the brain. Hayes.... Hauver.. Cochran et al. and was originally thought to play a role in stimulating bone growth (Marino and Becker. at which the wavelength c / f of light is hundreds of kilometers in air (and at least hundreds of meters even in brain matter where the refractive index is 103 (Gabriel et al. 1957). and the stress tensor. Halperin et al. 2002). 2004).. such as retrograde amnesia. 2001. Aschero et al. and σ are the polarization density. 1977. 1996. 1979) but this effect should be negligible for IED-scale blasts where the pressures are much lower (0. an effect which has been observed to yield electric fields up to 200 V/m for very high pressures (100 kbar) corresponding to nuclearscale explosions. 2004. Adushkin and Soloviev. electric fields are produced in the brain via the process depicted schematically in Fig. 2008. 2004). which results in a ms-scale electric field pulse.K. (Gabriel et al. The piezoelectric effect of bone has been observed since Fukada (Fukada and Yasuda. d. 2004)). Results section presents the results we obtain: charge densities and estimated electric fields as a function of time and space for an IED-scale blast.. 1957. 1970). The time scale of these fields is directly related to the timescale of the shock (as described in more detail below. a highpressure blast front can polarize water molecules (and other polarizable particles). as well as an investigation of the sensitivity of the results to the precise value of the bone's piezoelectric response. Near. the piezoelectric tensor (a property of the material). Even stronger long-term neurological effects. Williams and Breger. An explosion can produce electromagnetic (EM) fields in several ways. respectively (Brainerd et al. which uses ms current pulses repeated several times a second with amplitudes up to 1 A (American Psychiatric Association Committee on Electroconvulsive Therapy. All of these procedures utilize millisecond pulses because that is the timescale of the neuron action potential (Bear et al.. but regions of large positive (red) and negative (blue) charge density are created which lead to short-range electric fields. 1998) which is about 105 at the neurologically important kHz frequencies. Allison. The remaining paper is organized as follows. Also. which polarizes even in response to low pressures (and are therefore used as pressure sensors and actuators for a variety of applications (Crawley and de Luis. 2006). 1987.

Thus. and blood-vessel walls (Fukada and Hara. and at this point we are mainly interested in order-of-magnitude estimates. Again. 1976). skin (Shamos and Lavine. Black and Korostoff. (1) and Eq. where the six columns correspond to the six degrees of freedom in the symmetric stress tensor σ. we use the standard encoding of the rank-3 tensor d as a matrix (Brainerd et al.. 2009. Lee et al. as well as on a human head protected by a kevlar helmet (Advanced Combat Helmet). collagen (Fukada et al.. 1975). an approximation that should be valid close to the skull where the field is most intense.. The simulation consists of fluid–solid interaction with a blast shockwave under open boundary conditions.. Note that the combination of Eq. The electric polarization produces a local charge density via the following formula (Jackson. we have chosen to take into account only bone piezoelectricity because the stresses in the bones are much higher than those of other biological materials in and around the brain (Moore et al.. we use Fig.. 1989. (2004) and Aschero et al. Nevertheless. 1949). material dispersion) (Pfeiffer. Blast-induced charge density We calculate the charge density using stress data on the skull from a finite-element full-head-model simulation (Moore et al. which leads in Results section to large charge densities around small features such as eye sockets where there are large stress variations.569 kg of TNT at a 0. we will see that the precise arrangement of the coefficients in d seems not to matter very much for the peak order of magnitude of the resulting in-brain electric field. live bone (Cochran et al.) This gives us the polarization vector P (in the fixed coordinate system of σ) at each vertex of the finite-element model. 2.) The computational head model differentiates 11 distinct biological materials characterized by different mechanical properties such as nonlinear viscoelasticity. (Since the coordinate system of σ is fixed in space. while the coordinate system of d is skullrelative. To obtain the polarization by the piezoelectric effect from the stresses. over the course of a typical IED-scale blast impact for the unhelmeted human head. Reinish and Nowick established that moisture level plays a role in bone piezoelectricity (Reinish and Nowick ).. and in other biomaterials such as tendons (Williams and Breger. 1974). the head model was generated with 1 mm3 resolution from magnetic-resonance images (Collins et al.2 pC/N (Fukada and Yasuda. In this work. The d14 coefficient in femur corresponds to bending the bone along its length. / NeuroImage 54 (2011) S30–S36 classes of crystals and ceramics. . we consolidate the available experimental data to obtain a realistic piezoelectric tensor: 0 d=4 0 3:8 2 0 0 3:8 3 0 10:5 −1:3 0 −14 0 −1:3 −10:5 0 5 × 10 C=N 0:5 0 0 0 typical IED-scale blast wave incident on an unhelmeted human head from the side. 1962). Therefore. For the coefficients. These values were all measured for animal or human tibias and femurs. we used values typical of those reported in a variety of measurements (Bur. we assume that the piezoelectric response in the skull is rotationally invariant around axis 3. 1969). as well as the corresponding peak electric field estimate. and Pfeiffer studied the effect of the frequency of the applied stress on bone piezoelectricity (i.K. The piezoelectric coefficient d is a rank-3 tensor (a “3d matrix”) encoding the polarization in response to different types and orientations of the stress. As we show in Results section. and so we chose the corresponding coordinate system for the skull so that d14 corresponds to bending the skull. where d14 = 0.S32 K. the largest coefficients are all on the order of 0. 1998) combined with a bone-windowed CT scan. As long as the orientation does not vary too rapidly in space (since rapid oscillations in d would yield canceling polarizations). As described in Moore et al. 1996). (1). anisotropy and strong strain rate dependence. 1976. Fukada and Yasuda demonstrated a piezoelectric response in dry bone at 2 kHz in 1957 and measured a typical piezoelectric coefficient d14 (described below) to be 0. fixed Frobenius norm ∥d∥F ). We compute ∇·P by linear interpolation of the polarization between the vertices of each tetrahedron (corresponding to first-order elements for P) to obtain the charge density on a dual mesh located at the center of each tetrahedron. (3) mean that not only large stresses but also large stress gradients give rise to large charge density ρ. 2009).e. Bur (1976) studied bone piezoelectric properties as a function of temperature and humidity in the range 10− 2 to 102 Hz. 1967). However. Aschero et al.Y. these simulations employed a fullhead tetrahedral mesh in the context of a blast with an overpressure of 30 atm (0. and we assumed rotational invariance with respect to bending around any axis in the tangent plane of the skull. 1957) (about one tenth of the value for quartz). This allows us to estimate the field near the skull in these regions. Subsequent measurements have shown that piezoelectricity is also present in wet bone (Bassett and Becker. Bur. we multiply d by σ according to Eq. 2009) of a In many regions. because the field of a sheet of charge is known analytically. (As described in Ref. 2009). 1998): ρ = −∇⋅P ð3Þ where ρ is the charge density per unit volume. equivalent to 0.03 kbar): the 99% lethal dose (LD99) for lung-injury survival in an unarmored person (albeit survivable with current personalprotection equipment)... these axes rotate relative to the location on the skull. (Here.. and more recent measurements were described in Halperin et al.6 m standoff distance. 1977). Electric field estimation ð2Þ where axes 1 and 2 are tangential to the plane of the skull and axis 3 is normal to the plane of the skull. (1996). we find (below) that the charge density in the skull locally resembles a sheet of charge. An extensive summary of these measurements can be found in Singh and Katz (1988).) This piezoelectric tensor possesses rotational symmetry in the tangent plane of the skull. the orientation is therefore not crucial in estimating the order of magnitude of the resulting electric field. (Moore et al. For simplicity. and the measured piezoelectric coefficients have some variability (within the same order of magnitude). Numerous efforts to measure such properties in biomaterials have been undertaken.105 pC/N is the largest coefficient as reported by Ref. we first rotate σ into the skull-relative coordinate system for each location on the skull. and then rotate back the resulting P... Average and maximum piezoelectric charge density magnitude |ρ| as a function of time.1 pC/N at kHz frequencies. the precise orientation and anisotropy of d appear to make little difference in the resulting charge density in the skull from a blast wave (no more than a factor of 4) as long as the magnitude of the components of d is the same (technically. In particular.

Therefore. we obtain the following estimate for the electric field magnitude near a charged region of the skull: jEj≈ ρh 2εeff ! ð5Þ ð4Þ 1 + 105 ≈ρh = 2ε0 2 ≈ρ⋅2300 Vm = C 2 ð6Þ Fig. For a surface charge density at the interface between two materials with permittivities ε1 and ε2.6 m. 1998).K. where ρs and ε are the surface charge density and the permittivity of the material. it can be shown from the method of images (Jackson. 3.K. and therefore can be approximated by the known field of an infinite charged plane. the corresponding surface charge density is ρs =ρh. that field has a magnitude of ρs /2ε. / NeuroImage 54 (2011) S30–S36 S33 this as a first approximation to estimate the maximum field magnitudes in the brain. along with the various labeled cross sections.Y. In particular. corresponding to 0. Lee et al. Computed charge density distributions in the skull region from front view (top left) and side view (middle left). The associated blast has an overpressure of 30 atm as it impacts the head. For the particular case of a volume charge density ρ in a thin sheet of thickness h.347 ms for the helmeted case.548 ms after impacting the head for the unhelmeted case. and 0. electric field very close to a sheet of charge is insensitive to the finite size of the charged region. 1998) that the corresponding field magnitude in both materials is ρs /2εeff where εeff = (ε1 + ε2)/2 is the average permittivity. In a uniform dielectric.569 kg of TNT at a distance of 0. respectively (Jackson. . The time is 0.

Fig. and find that the Fig.5. Future work should also consider the possibility that the effect of the electric fields on the brain is exacerbated by the mechanical blast injury. and so should be compared to safety standards and medical procedures for millisecond-scale pulses (equivalent to the standards for kHz frequencies (IEEE Standards Coordinating Committee 28. and compare with theoretical predictions. Conclusions From our investigation. Regardless of whether piezo-induced electric fields play a role in blast-induced TBI. In particular. Since the piezoelectric coefficient d has not been measured for human or animal skulls.) The results show that the precise details of the piezoelectric orientation in cranial bone are unlikely to change the order of magnitude of the peak charge density. especially since increased intracranial pressure is already known to increase risks in rTMS (Wassermann. with the average |ρ| being more than an order of magnitude smaller than the maximum |ρ|. First. our estimated field magnitude on the order of 10 V/m are about an order of magnitude larger than the IEEE safety standards and are comparable to the electric fields used in rTMS. so simply multiplying d by a constant will scale ρ and E by the same constant.) Second. These electric fields are estimated to be on the order of 10 V/m over millisecond-duration pulses. the same mechanism that produces fields . Lee et al. Here. and then d was rescaled to have a fixed Frobenius norm ∥d∥F = 0:159 × 10−12 C/N. Histogram of maximum charge densities calculated for 100 randomly oriented piezoelectric tensors. For example.85 ⋅ 10 − 12 F/m is the vacuum permittivity. 1996)). (Future experimental measurements of the piezoelectric effect in cranial bone should also determine any spatial variation of the piezoelectric coefficient d across the thickness of the bone or in different regions of the skull. Therefore. using uniform distribution to in [− 0. and there is some variation in measured data even for tibias and femurs. the biggest EM fields in the brain may come from an unexpected source – piezoelectricity of bone – that had not yet been studied in the context of blast TBI. h = 0. Results Using the methods described in the preceding section. This raises the possibility that piezoelectric fields produced by the blast wave impacting the skull may contribute to blast-induced TBI. Eventually. and concluding by investigating the sensitivity of these results to the precise details of the piezoelectric coefficient d. ε0 = 8. the charge density and electric field are clearly linear functions of d from Eqs. any rapid variation in the magnitude or orientation of the piezoelectric tensor d across sutures between cranial bones would lead to greatly increased ρ at these joints. we obtained the charge density in the skull and the resulting electric field magnitude estimates in the brain regions near the skull. In particular. determine the relevant piezoelectric coefficient. maximum |ρ| varies by around a factor of 2–3 from the |ρ| produced by our original value of d. we present these results. then the polarizations would be pointing in different directions and could mostly cancel. First. which will differ from the estimated fields above in regions where the charge density is varying rapidly in the skull. as long as the strength of the piezoelectric effect is the same. This maximum field is attained in a millisecond-scale pulse (corresponding to the duration of the high-pressure blast front). we assume that the orientation of the piezoelectric tensor is coherent (that the tensor orientation changes on the scale of at least centimeters. we plan to perform direct experimental measurements of shocked cranial bone in order to observe the resulting electric fields. We generate 100 random d tensors with this rotational symmetry. an assumption that has yet to be checked experimentally for the cranium but is consistent with measurements for femurs and tibias (which have a specific d orientation relative to the long axis of the bone). We proceed with our sensitivity analysis.5]. imposing rotational symmetry in the tangent plane of the skull. 3 that high charge densities are often concentrated in small (cm-scale) regions. as reviewed in the Introduction. we investigate how sensitive the charge densities and resulting electric fields are to the precise details of d. 1998). If d were rotating rapidly across the skull. a possibility that we believe merits further study. and the relative permittivity ε / ε0 of air and brain matter are 1 and roughly 105 (at kHz frequencies (Gabriel et al. 4. There are a number of questions that should be addressed by future research. (2). (1) and (3). we can see higher charge densities in the skull on the side facing the shockwave impact.. the lengthscale of coherence in our stresses and charge distributions). as well as the corresponding electric field estimates with respect to time over the course of the blast wave impact. We show in Fig. An immediate observation is that ρave ≪ ρmax.S34 K. / NeuroImage 54 (2011) S30–S36 where ρ is the charge density. The peak charge density and electric field are changed by less than 20% in the case of a helmeted head: a typical charge density for the helmeted case is shown in Fig. In this section. and using these to calculate the charge densities as in Blast-induced charge density section. The corresponding estimated maximum electric field from Eq. EM effects may need to be included in animal models and other research on blast-related TBI. they may be useful in research and diagnosis of blastrelated injuries. 4 the histogram of the maximum charge density |ρ|. We quantify this by fixing the Frobenius norm ∥d∥F (the root-mean-square value of the components of d). The only remaining question is the sensitivity with respect the “orientation” of d: the distribution of the components of the d tensor with a fixed “average” magnitude.5] assign values to the non-zero degrees of freedom in the tensor and then rescaling to keep the norm ∥d∥F fixed to that of the tensor in Eq. (The tensor components were chosen as uniform random numbers in [− 0. and randomly varying the distribution of the components.K. 0. for example.Y. both in the center of the side and also around the eye sockets where there are small features in the skull leading to large stress gradients (as mentioned in Blast-induced charge density section). 2002)). This is consistent with the observation from Fig. we are currently incorporating the computed charge density into a full three-dimensional finite-element solution of Poisson's equation − ∇ ⋅ (ε ∇)ϕ = ρ to find the full three-dimensional fields E = −∇ϕ.002 m is the thickness of the skull. respectively. electromagnetic solvers may need to be directly coupled with shockwave simulations. Ultimately. and displays similar peak densities but over a smaller area of the cranium. 2 displays the average and maximum charge density magnitude |ρ| in the skull bone. + 0. 3 (f).5. (4) is 6 V/m. exceeding IEEE safety standards by an order of magnitude and comparable rTMS procedures known to have neurological effects. starting with the time and space dependence of the charge density and the strength of the resulting fields. it seems that the precise details of the d orientation do not affect the order of magnitude of the charge density or the resulting in-brain electric fields.

. Biomech. J. Piezoelectric and related properties of hydrated collagen. 1158–1162.) References Adushkin. D. Becker. H. 3rd Edition. LA-UR-93-0983. Explosion and Shock Waves 40. 29. Salai. The detonation electric effect.. C95. from thermodynamic considerations. It turns out that the induced electric field in an air blast wave can be characterized in terms of the blast wave pressure. Standards on piezoelectric crystals. 777–780... Journal of Sound and Vibration 174. M. (For nuclear-scale blast waves with even higher temperatures in the thousands of K.L.S. 1993. Surv. V. (4)) and the potential will be V ∼pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi0.-N. 1965). Biomech. Rinaldi. Appl. Appl. and this separation requires energy qV where q is the value of an individual charge (e. Computational biology — modeling of primary blast effects on the central nervous system. P. 1814–1817... 739–741. 2111–2113. 1990. and substituted into the above equation for E to obtain: sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Pe−I = kT E∼ : ε0 significant. Tech. P. Phys. Soloviev.... Cady. Ionization of air by IED blast waves In order to eliminate the possibility that ionization of air by the blast wave can itself lead to neurologically significant electric fields. J. Atomic spectroscopy.6: IEEE Standard for Safety Levels with Respect to Human Exposure to Electromagnetic Fields. 1982.. Hauver.E. 649–657. Reed (Reed. 2006. C.. 4. Presles. 1998. E. Molotskii.. 1965.R. H. 597–602...C. Hara. Phys. J. D. A865780. 1969. Aschero. the fields from ionization quickly become more . and underground explosions. Fukada. Here. should be comparable to the thermal energy kT where k is Boltzmann's constant and T is temperature (Reif. Harris. F. 36.G. Med. 1979. Generation of electric potentials in bone in response to mechanical stress. 25.. F. P. they can be analyzed after the blast to provide a direct measure of the blast-induced stresses on the skull as a quantitative diagnostic metric [cite patent application]. Agronin... J. Combust. 37. 36.. Molecular. 2002. Murri. Appl. Dell..Y. Marino..G. 17. pp.. Phys.V. Yasuda. we present a conservative order-of-magnitude estimate on thermodynamic grounds (suggested by E. Johnson. Jackson. R.. 1253–1256. 41. Rep. L. V. R. On the piezoelectric effect of bone. 1996. 95.. Neuroscience: exploring the Brain. 1–12. 1949. Lit.. 1994.P. G.W. Kresse. Recall.... Phys. G.. 1967.W. Washington. Long-term repetitive transcranial magnetic stimulation increases the expression of brain-derived ð7Þ We now make a conservative estimate for E.B.. Jerusalem. E. 36. Mineev. Mango. D. Tech.G. M. Jaffee.. Jpn 12.A. and we use these parameters to estimate the fields from an IED-scale blast below. Army Research Laboratory.. 2002. The Effects of Nuclear Weapons. C. 1966. J. Gizdulick. S. Proceedings of the I. we obtain the charge density from the Boltzmann distribution: if the ionization energy of the gas is I and the number density is n.N. D. Science 137... G. Frankel. E. 2008)). 37. Hayes. Harris. J. Nyein. 2113–2118. W. Sci. 1976. et al. J. Krenev.C.. 2251–2269. M. 2009. Collins. 1974. I. Piezoelectric effect in blood vessel walls. and air-ionization energy alone. Allison. 911–918. S. 1977). Boronin. Conflict of interest statement The authors declare that there are no conflicts of interest. J. Woodbury.. V. March.N. Department of Defense. New York. 473. Ch. Gabriel. Journal of Cereb. Post. Even if the temperatures in the blast wave were to reach T = 1000 K (0. R... Muller. G. Phys.. First.. we consider the ionization by the blast wave in air only. 745–750. Keck. Mutchnik. with the ionization energy of air being I ≈ 10 eV (slightly smaller than the typical value for gases (Martin and Wiese. Measurements in the frequency range 10 hz to 20 ghz. Mechanism for the shock polarization of dielectrics.F.. Design and construction of a realistic digital brain phantom. Nano Lett. 495–507. if it is supplied by the blast wave. rep. Gabriel. 617–639..B.. independent of how the blast wave was generated. AIAA J.. S. Converse piezoelectric effect detected in fresh cow femur bone. Wiese.G. Ann.. IEEE Standards Coordinating Committee 28. Bur. Strain-related potentials in living bone. Bassett.086 eV) with pressures reaching P = 100 atm (107 Pa). Crawley. Imag. Second. Sykes. N. V. 10. Doran. This can now be combined with the ideal gas law P = nkT relating pressure P to n and T. Classical Electrodynamics. combustion... 1957. I. C. Rep... M. Cernak. Anderson. J.).. 1987. Soc. L. Glasstone. Washington. Explosion Shock Waves 26..J.E. Jensen. & Optical Physics Handbook. then the peak electric field will be E ∼ ρℓ / ε0 (similar to Eq... Toschi. thanks to the exponential dependence in the Boltzmann factor e − I / kT. M... R. Lee et al. Martin.W. M. M.. M. E. Appendix. 1965). A... Jpn 26.. Van Dyke. temperature. IEEE. Nature 228.T. Fukada. 238. A. Kapinos... J. W. NY..g.. P. A. S. Shock-induced polarization in plastics II. 1373–1385. September. 1063. Phys. Cochran.F. An electric field from ionization must be created by the separation of charges. Black. Piezoelectric effects and growth control in bone. 1996)). Moore.. Atomic.K. R. 2527–2532. Acad.D. We are grateful to Evan Reed for many helpful suggestions. The shock induced electrical polarization of water.P. Soc. W. 1378–1395. J. Training. Neuroimage 49 (Supplement 2). Wiley... Appl. P.. / NeuroImage 54 (2011) S30–S36 S35 inside the brain also generates short-range fields of similar magnitudes immediately outside the head. Chem.W. and such kHz-frequency fields are easily measured by radio-frequency (RF) antennas.. Korostoff. Phys. A. assuming the charges equilibriate on a shorter timescale than the blast wave duration (milliseconds).J. American Psychiatric Association. M. M... Radovitzky. J. N.. PRL 43. J. 16.. Biomech. Use of piezoelectric actuators as elements of intelligent structures... Fukada.. Kelly. E. E. that any field produced in air is reduced in the brain by roughly a factor of the relative permittivity (Jackson. A.A. then one should have ρ = qne − I / kT at equilibrium (Reif.E. this equation predicts electric fields of only ∼ 10 − 16 V/m.. A. N.E. 135–153.. Biol. Physical mechanism of electromagnetic field generation during the explosion of condensed explosive charges. W. M. S. 1996. Otter. 2nd Edition. Piezoelectric effect in human bones studied in nanometer scale. Paradiso. 2000. Simultaneous piezoelectric sensing/actuation: analysis and application to controlled structures. Cumming. 1965.J. Measurements of the dynamic piezoelectric properties of bone as a function of temperature and humidity. Los Alamos National Laboratory. Tech. Linde. 1962. 1977. D.. 29. The dielectric properties of biological tissues: II. 1998) of the brain tissue (∼ 105 (Gabriel et al. 77. M. also. we obtain Eℓ ∼ ρℓ2 / ε pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ℓ∼ kTε0 = qρ and peak fields E∼ρℓ = ε0 ∼ ρkT = qε0 . 463–469. Magnetic field generation by detonation waves. The Practice of Electroconvulsive Therapy: Recommendations for Treatment. 38.. Zijdenbos. R. Rosenman. Toton..W.. Lau. an electron) and V is the voltage. Kadaba. 22..H. Bear.. Romano.A... (Ed. If the blast wave has spatial width ℓ and the charge density is ρ.C. V.. Halperin. Experimental study of plexiglas and polystyrene. Kollokian.. and have been studied in more detail in that context (Glasstone and Dolan.. not the ionization within the detonation region (the initial flash). B. Biophys. New York. G. A. surface. 1970. A. J. E. Noble-Haeusslein. 1998. Med. and Privileging. 1996. R. In: Drake. pp.. Ueda. Acknowledgments This work was supported in part by the Joint Improvised Explosive Device Defeat Organization (JIEDDO) and the Institute for Soldier Nanotechnologies (ISN) through the Army Research Office. AIP. K. J. Traumatic brain injury: an overview of pathobiology with emphasis on military populations.R. Dolan. Palmieri.Y. IEEE Trans. de Luis. Becker. An improved design of electrodes for measurement of streaming potentials on wet bone in vitro and in vivo.. L. J. Noels. Combining this with qV∼ kT and solving for ℓ. 2004. 2001. Estimates of the Electromagnetic Radiation from Detonation of Conventional Explosives. 1965.G. B. This energy. Theory. Blood Flow Metab.. 3rd Edition. K. we estimate the electric field in terms of the charge density and temperature. 9. Generation of electric and magnetic fields by air. Phys. 0–3 kHz. 2004. 1989. Hagood.V. 2009. Urenski. B.. 1169–1173. The shock induced electrical polarization of water. 1976. Lippincott Williams & Wilkins. 507–511. S. Small RF antennas incorporated into protective headgear could therefore provide a form of “blast dosimeter:” if the field strengths are recorded. J. American Psychiatric Association Committee on Electroconvulsive Therapy.K.K. Phys. Connors. A.F. 5157–5164. D. EMP from Chemical Explosions. Appl. Fine. Brainerd. Shock-induced polarization in plastics I. 1996)). T10–T20.E.J.

2004. Piezoelectric actuator technology. Neurophysiol.F. Instrum. Sci. neurotrophic factor and cholecystokinin mrna. 2766–2769. 1975. Electromechanical properties of bone—a review.. Nowick. McGraw-Hill.. Rihs. J. 1998. L. F. Mosimann. J. 407–413. T.. 2717. 857–870. S. 1996. U. J.. New York. 014302.R. June 5–7. Talantsev. Repetitive transcranial magnetic stimulation of the dorsolateral prefrontal cortex affects divided attention immediately after cessation of stimulation.. L..J.. 315–321. 1–16. / NeuroImage 54 (2011) S30–S36 depolarization of lead zirconate titanate piezoelectric ceramics. 8. S. A... Armstrong. Wagner.H. Near. 1967. Pfeiffer. Eden. Neuroimage 30. T. L. 1965. Katz. Fregni. J.. C. Compact high-voltage generator of primary power based on shock wave . 246. M. Reinish. J. Ramos-Estebanez.. E. Temkin. Wassermann. U. PascualLeone. Proc. Piezoelectric properties of bone as functions of moisture content... Glownia.J. B. 1996. 2008. 53–57. but not neuropeptide tyrosine mrna in specific areas of rat brain. 75. A. Fundamentals of Statistical and Thermal Physics. Reed. Piezoelectricity in tendon and bone. 2008. Electric Fields in Air Shocks. 267–269. E.I.. 108. M. Biomech. 1977. PRL 101. H.A.. F.. Electroencephalogr.K. C.. J. Local piezoelectric polarization of human cortical bone as a function of stress frequency. SPIE: Smart Actuators I. 2006a.P. Singh. Menon. Transcranial magnetic stimulation and stroke: a computer-based human model study. G. Zahn. M. Breger.. Lee et al. Rev. W. Private Communications.. Nature 253 Shamos.. Res. Altgilbers. Baird.M.Y. Neuropsychopharmacology 23. 1988. H.. J. Atomic-scale time and space resolution of terahertz frequency acoustic waves. Clin... Psychiatr. Reif. Wagner. A. K.. 2006b.. Biomech.. Nature 213. Schlaepfer. E. Piezoelectricity as a fundamental property of biological tissues. 10. Fisch. 219–238.E.. Lavine. 205–215.L. p.D. 40. T.U. L. Reed... Bioelectricity 7. Shkuratov. Kim. S... E. Risk and safety of repetitive transcranial magnetic stimulation: report and suggested guidelines from the international workshop on the safety of repetitive transcranial magnetic stimulation.S36 K. Grodzinsky.. Williams.Y. M.. B.

Sign up to vote on this title
UsefulNot useful