Earthquake and Electrochemistry: Unraveling the Unpredictable

Earthquakes are measured using well defined seismic parameters such as seismic moment (Mo), moment magnitude (Mw), and released elastic energy(E). How this tremendous amount of energy is accumulated silently deep inside the earth's crust? The most obvious question in seismic research remains unanswered. We found an inherent and intriguing connection between the released energy in an earthquake and electrochemical potential induced in an ultra-thin metal oxide electrode immersed in an aqueous pH solution, which leads us to understand the origin of the energy accumulation process in an earthquake. A huge electrochemical potential is accumulated from numerous electrochemical cells formed in a unique layer structure of hydrated clay minerals (predominantly smectite), which resulted in a lightning-like discharge in the lithosphere (hypocenter). The subsequent thunder-like massive shockwave is produced, which initiates tectonic plate movement along a fault line, probably through acoustic fluidization (AF), and resulting seismic energy is transmitted as primary wave (P-wave), secondary wave (S-wave), and surface waves. The presence of electrical voltage in the hypocenter directly supports the seismic electric signal (SES), further strengthening the VAN method of earthquake prediction. Our finding is supported by a plethora of research and observation devoted to seismic science. This study will indeed find its significance if immediate action is implemented to monitor the evolution of electrochemical potential, seismic electrical signal (SES), and ionic activity in the fault zone at lithosphere as well as in the ionosphere for predicting an impending earthquake for saving human lives as early as possible.


INTRODUCTION
Earthquake is the most complex natural calamity claiming human lives and mass destruction of manmade structures. A natural earthquake occurs due to a sudden slip or collision of tectonic plates in a fault zone deep in the earth's crust, and the stored energy is radiated as seismic waves causing ground motion and acceleration. With tremendous progress of modern science and instrumentation, seismometer and earthquake source theories are matured enough to provide quantitative source parameters than the only magnitude. A well-established moment magnitude (Mw) scale (1,2) has been used to quantify shallow and deep earthquakes based on radiated wave energy. This magnitude scale derives from the concept of the seismic moment (M0= μAD), which is directly related to the measurable parameters in a seismic event such as the area of the rupture(A) along the geological fault, the average of displacement (D) or slip during the rupture and shear modulus (μ) or elastic constant. The famous Richter-Gutenberg energy-moment magnitude relation(3) of a seismic event is Log E=4.8+1.5M w where E is associated with elastic energy in Joule which drives the earthquake to progress from the hypocenter (foci), Mw is the moment magnitude of an earthquake. How this tremendous amount of energy is accumulated deep in the earth's crust? We have no clues on the origin of such an energy accumulation process occurring silently inside deep in the earth's lithospheric region. Most of the seismic study only quantify parameters after an actual seismic event occurred. The ultimate goal of earthquake research is to understand the seismic source's nature and a timely forecast of an impending earthquake by finding suitable pre-seismic precursors.
Significant geochemical precursors, i.e., the anomalous concentration of dissolved ions (4) and gases in groundwater, have been measured before intermediate and large earthquakes. Radon ( 222 Rn) gas anomaly in the environment (soil) and groundwater before the earthquake was detected in many cases (5). A wide range of pre-earthquake phenomena has been reported involving ground and satellite-based observations. Atmosphere-ionosphere response a few days before the M9 great Tohoku Japan earthquake revealed a rapid increase of emitted outgoing longwave radiation (OLR) of infrared (10-13 µm) and increased in a variation of total electron content (TEC) in the ionospheric region above the epicenter (6). Thermal infrared (TIR) and ionosphere anomaly was detected in the Iran earthquake (7), 2001 7.6 Bhuj earthquake (8), and 2017 Mw6.5 Jiuzhigou earthquake (9). Such pre-seismic anomalies are possibly linked to the energy accumulation process before the actual earthquake strike. The most debated and criticized VAN method (10) has been used in Greece and other countries for earthquake prediction. The method is based on detecting geo-electric potential or seismic electrical signals (SES) that appear prior to earthquakes.
While working on an electrochemical sensor, we found a gradual increase of electrochemical potential in a solid-state electrode, which reminds us of the gradual increase of energy with an earthquake's magnitude. What a connection and consequences! We found a close match between estimated electrochemical potential induced in a solid-state electrode and energy released in an earthquake while looking more deeply into earthquake energy release patterns corresponding to the magnitude scale. With this clue, we investigated the origin of the energy accumulation process through an electrochemical reaction involving clay minerals and water.

Derivation of Electrical Potential from Released Energy in an Earthquake
Richter-Gutenberg energy-magnitude provides an accurate estimation of seismic energy corresponding to the moment magnitude scale.
E is the released energy in Joule, and Mw is the moment magnitude of an earthquake event. The magnitude Mw= 0 in the moment-magnitude scale has the equivalent energy of 63.09573×10 3 Joule. The electrical potential could be derived from released energy from the simple calculation. The electrical potential can be calculated as follows, We assign SEP as seismic electrical potential, which only increases exponentially with the increase of magnitude scale. Another factor, i.e., 3.6×10 6 will remain constant in Mw (0~1.9). Similarly, we can calculate the electrical potential for other earthquake moment magnitudes. Table 1 shows details of earthquake magnitude Mw (0~1.9) and corresponding electrical potentials. The magnitude scale Mw (0~1.9) is considered as an initial basic set for calculation. Other sets like Mw (2~3.9), Mw (4~5.9), Mw (6~7.9), Mw (8~9.9) will follow (see Supplementary Materials, Table  S1~S4) the same trend of Mw (0~1.9) with increasing multiplication factor. One may simply correlate such huge electrical potential because of the fault zone's energy accumulation process before an actual earthquake strike. The following section will reveal an intriguing connection between earthquake electrical potential and potential originated from a spontaneous electrochemical reaction.

Electrochemical Potential at the Electrode-Aqueous Interfaces
The metal-oxide electrode responds to the aqueous pH buffer solution through a reversible electrochemical reaction. The redox electrochemical reaction occurring at the electrode-aqueous interface can be written as, 2 where A may represent a simple metallic ion and B the corresponding metal. The electrode potential for the above reaction is given below: The nH+ is the number of moles of hydrogen ion chemisorbed (chemical reaction) at the electrode surface, and its values are 1, 2, 3, …, n. The ne-is the number moles of electron transfer, which is the same as the change of oxidation number of the material (Z) involved in the redox (A→B) reactions. Therefore, 'x' has a precise stepwise discrete value, which controls the ultrathin oxide electrode's potential in a similar discrete manner. The ion exchange factor 'x= nH+/ne-' will determine the exact electrochemical cell potential in a spontaneous electrochemical reaction with an aqueous solution. We assign NP as the Nernst potential (0.0591 V pH -1 ), which is a special case of equation (9) owing to symmetric ion-exchange factor (x=1). Electrode potential due to asymmetric ion exchange (x≠1) can be assigned as Pourbaix Potential or PP as the idea of asymmetric ion exchange was originally formulated by M. Pourbaix through pH-potential formulation (12). We could relate PP and NP by a simple equation, (10) x PP NP x = The electrode potential could be expressed as 0 (11) x E E PP pH = −  E is electrochemical potential due to half-cell reaction occurring at the working electrode-aqueous interface. Another half-cell reaction (E o ) is occuring at the reference (Ag ‫|‬ AgCl) electrode, which is well understood non-interfering rection. The half-cell reaction of each working electrode could be added as a regular electrochemical cell (See Supplementary Materials, Figure S1). For N number of electrochemical cells (NCell), the total electrochemical potential will be, The abundance of SiO2 and Al2O3 in the earth crust leads us to focus on materials with Z=+3, +4) and associated electrochemical potential in aqueous solutions with different ionic exchange factors. In laboratory condition, the various electrochemical potential for Al2O3 and SiO2 was reported. The minimum value for Al2O3 was 17 mV for low temperature grown oxide (13) whereas the ideal value of 40-55 mV is expected (13 Very high electrode potential 300 mV pH -1 was achieved in 3D nano-porous P-Si structure (16) which could be due to ion exchange factor of x=21/4 (PP21/4=310 mV). It is indeed a beautiful example of interfacial engineering for achieving high electrode potential in Si-SiO2 system. Trend of increasing electrode potential from 30 mV to 300 mV for a unit electrochemical cell in Si-SiO2 system clearly indicates that the value could be much more in case of extreme engineered 3D surfaces. 5

Correlation between Electrochemical Energy and Elastic Seismic Energy
Based on the experimental evidence of electrode potential in Si-SiO2 system and Pourbaix formulation of higher-order ion-exchange in several materials system (Ga-Ga2O3, Ir-Ir2O3, Cr-Cr2O3, V-V2O3, and many more), it seems logical to generalize the Pourbaix Potential for any material system. We could extrapolate the trend of higher-order ion exchange by considering chemisorption of an increasing amount of hydrogen ion (H + ) in the electrical double layer (EDL) formed at the electrode/aqueous interface. Considering the lowest order ion-exchange factor x=1/4 for Si-SiO2 and x=1/3 for Al-Al2O3, we calculated the PP for both materials.
At this point, we found a striking similarity between seismic electrical potential (SEP) and Pourbaix Potential (PP). In fact, the average of PPx=1/4 and PPx=1/3 is 0.017255, which is very close to the SEPMw=0.0=0.01752. Other factors from equation (12) i.e., pH×NCell could be seen as equivalent of 3.6×10 6 of equation (4). Pourbaix Potential for different ion-exchange factor for Al-Al2O3 (Z= +3) and Si-SiO2 (Z=+4) are presented as separate MS Excel datasheet (see Supplementary Materials-External Database). After careful observation, we choose several values closely match with seismic electrical potential (SEP) (see Table 1). Selected Pourbaix potential is shown in Table 2. To understand the correlation between SEP and PP, both SEP (V) vs. Mw (0.0~1.9) and Log (SEP) vs. Mw (0.0~1.9) curves ( Figures 1A and 1B) are plotted accommodating the closely matched PP values from Table 2. All curves are surprisingly overlapped, which indicates an inherent link between SEP and PP. The theoretical temperature effect on the electrode potential is 0.02 mV per degree Kelvin (17). Considering temperature of 200 o C at 75 km depth (18) in the earth crust, the change of electrode potential will be T×0.02 The Elovich equation has been widely used in adsorption kinetics, which describes the chemisorption (chemical reaction) mechanism in nature through a multilayer adsorption site (19)(20)(21). According to this equation, the adsorption site increases exponentially with adsorption. It is noteworthy to mention that the maximum Pourbaix potential for each cell is 12.40 V, and the value is linked with two fundamental constants, i.e., Planck constant and speed of light in a vacuum (hc=12.39 eV×100 nm). We believe that this the fundamental limit of hydrogen ion adsorption in the electric double layer (EDL) of electrode/aqueous interface. A compact table (Table 3) is prepared to show the moment magnitude scale and corresponding SEP or PP and pH×NCell values.
As the pH of water usually varies from 1~12 in pH scale, the number of electrochemical cells (NCell) will vary accordingly, keeping the total electrical potential constant as per the moment magnitude-energy relation. At this point, we strongly believe that the electrochemical potential accumulated from numerous electrochemical cells is the origin of earthquake energy. This electrical energy transforms into mechanical energy (seismic energy) to drive the entire earthquake process. 6

DISCUSSION
Originally earthquake energy is calculated from the concept of stored elastic energy in the rockforming materials. The extracted seismic electrical potential (SEP) closely match with Pourbaix Potential (PP). Is this an accidental coincidence, or are they inherently linked? How to validate an utterly different origin 'a huge electrochemical potential' as an earthquake triggering source? We strongly believe that the accumulation of electrical potential and electric field in the earth's crust prior to any earthquake event could provide a unified model to explain most of the existing observation and findings devoted to seismic sciences unequivocally. "Extraordinary Claims Require Extraordinary Evidence" (ECREE), a famous phrase was made popular by astronomer Carl Sagan. This aphorism is at the heart of scientific skepticism, a model for critical and rational thinking. In the following sections, we will try to validate our claim by systematic analysis and logical clarifications of related findings directly or indirectly linked to our study.

Origin of Electrochemical Potential in Fault Zone and Role of Clay Minerals
The electrochemical potential obtained from redox reaction of Al2O3 (Z=+3) and SiO2 (Z= +4) with aqueous pH buffer in different ion exchange factor closely match with the earthquake electrical potential (SEP). Naturally, we must focus primarily on the source of the SiO2 and Al2O3 in the earth's crust and earthquake fault zone. In nature, during the weathering process, rocks react with water and produce clay minerals. Clay minerals are phyllosilicates, a stacked layer of two dimensional (2D) sheets of hydrated aluminosilicates (SiO2, Al2O3, H2O) found in geologic deposits, terrestrials weathering environment, and marine deposits (22). Water is present in a variable amount as part of the structure of minerals. Based on the stacking of silicon tetrahedra (T) and alumina octahedra (O) sheets in the crystal unit or layers, clays are categorized into three different groups, i.e., 1T:1O, 2T:1O, 2T:2O type clay minerals. A schematic representation of different clay minerals is shown in Figure 2. Clay minerals have a wide range of cation-exchange capacity (CEC) due to the crystal lattice's overall charge imbalance. This property arises when either Si (IV) or Al (III) is substituted with a lower valent metal cation such as Fe (III) or (Al (III) or Mg (II)), respectively, resulting in a net negative charge region in silicate layer.
Kaolinite (2SiO2Al2O32H2O) is the most prominent 1T:1O clay mineral (Figure 2A). It is theoretically composed of 47% SiO2, 39%Al2O3, and 14% H2O. The layers are held together with strong hydrogen bonding. Interstitial cations and water do not enter between the structural layers mineral particle when the clay is wetted. The effective surface of kaolinite group minerals is restricted to its external surface for an interfacial reaction. Smectite minerals are 2T:1O type, which is characterized by the octahedral sheet (1O) sandwiched between two tetrahedral (2T) sheets ( Figure 2B). It is theoretically composed of 67% SiO2, 28%Al2O3, and 5% H2O (without interlayer water). Owing to the weak van der Waals force between interlayer, waters, and exchangeable cations (K+, Ca+, and Mg 2+ ) are easily entered into the interlayer spacing of smectite, causing expansion and swelling when the clay is wetted. Vermiculites are also 2T:1O type minerals with very high negative charge associated with these minerals owing to tetrahedral substitution by a lower valent metal ion ( Figure 2C). Water molecules with other cations are strongly adsorbed in the interlayer spacing. The cation exchange capacity (CEC) of vermiculite is higher than all other silicate clays, including smectite, because of the very high negative charge in the tetrahedral sheet.
The internal surface exceeds the external surface of clay crystal for both smectite and vermiculite.
Illite is a non-expanding (2T:1O) clay minerals, where potassium ion (K+) acts as a binding site preventing the expansion of crystals ( Figure 2D). It is theoretically composed of 12% K2O, 45% 7 SiO2, 38%Al2O3, and 5% H2O. The hydration, cation adsorption is less intense in fine-grained illite than in smectite but more than kaolinite due to the presence of interstratified layers of smectite or vermiculite. Chlorites are 2T:2O type non-expanding clay, which is iron magnesium silicate with some aluminum present. The crystal unit contains two silica tetrahedra sheet (2T) and two magnesium dominated octahedral (2O) sheet ( Figure 2E). The effective surface of illite and chlorites are also restricted to its external surface for interfacial reactions. A comparison of essential properties is presented in Table 4(23).
In reality, specific clay mineral does not occur independently; instead, it is common to find several clay types in an intimate mixture. Such minerals are known as a mixed layer or interstratified such as "chlorite-vermiculite, "mica -smectite," etc. The clay mineral structure is fascinating with attractive and large surface areas for interfacial reactions in aqueous environments. In general, it is estimated that a cubic centimeter (1cm 3 ) of clay has a reactive surface of around 2800 m 2 , which is equivalent to the area of a football field! For an analogy, a centimeter-thick pad of paper includes about 100 sheets, whereas a centimeter-thick layer of clay minerals includes about 100,000 sheets. Such a huge number of clay sheets gives us a sense of the number of NCell required to satisfy the values in Table 3. Smectite and vermiculite will be the perfect candidate for harnessing such huge electrochemical potential owing to cation exchange capacity in the interlayer spacing and outstanding adsorption properties. Dissociative chemisorption of water(24) on silica surface and formation of hydronium ion could explain the increasing number of hydrogen ion adsorption in clay surface. It seems logical to think that each interlayer tetrahedral surfaces may serve as one active electrochemical cell separated by an octahedral sheet. The concept of generation of electrochemical potential in unique layer (surface) structure of clay minerals is based on several previously reported experimental results of Al-Al2O3/aqueous and Si-SiO2/aqueous system together with Pourbaix formulations of pH-Potential diagram of Si-SiO2 and Al-Al2O3 system.

Electrochemistry in Clay mineral-aqueous interface and Role of Water in the Earthquake Triggering
We realized that the electrochemical reaction between clay minerals and water is the source of electrical potential, which slowly buildup and finally trigger an earthquake. Our claim is supported by several observations and scientific studies where water plays a vital role in triggering an earthquake. Reservoir induced seismicity (RIS) is an active area of seismic research to figure out the role of water and water pressure in earthquake triggering. Gupta  that the reservoir water slowly diffuses to the underneath clay minerals of the nearby fault zone and create a perfect condition for a gradual buildup of electrochemical potentials for triggering an earthquake.
Water injection to stimulate hot and dry rock of the proposed geothermal project in Basel, Switzerland, triggered excessive seismic activity putting the project on hold (29). The possibility of the tidal triggering of large earthquakes has been extensively studied to find the correlation between the earthquake and tidal stress (30). A tidal variation on earth's water occurs in each new moon and full moon days due to the combined gravitation pull of moon and sun. The enhanced gravitational pull on the water body near the fault zone will increase the possibility of electrochemical reactions which could be directly linked to earthquakes. Spike in earthquake following the typhoon Morakot(31), Taiwan could also be linked to our findings. Himalayan earthquake is less frequent in the monsoon season, but the number increases significantly in the winter. Michas et al. (32) recently reported the correlation between seismicity and water level fluctuation in Polyphyto Dam, North Greece. We believe that water level variation could create suitable conditions for electrochemical reactions.

Evidence of Electrical Potential
Accumulation of electrical potential is directly linked to the pre-seismic precursor observed and reported in many literatures.

Ionospheric Perturbations
Planet earth works as a fabulously complex system, and the lithosphere is inherently connected to the atmosphere. We may not know how the earthquake affects the atmosphere. Variation in ionospheric Total Electron Content (TEC) was observed before many earthquakes (>Mw5.0) through the analysis of global positioning satellite data. Lithosphere-Atmosphere-Ionosphere coupling (LAIC) is a well-known model for understanding the perturbation in the ionosphere. The appearance of geo-electric voltage in the seismically active region is believed to be one of the possible sources of ionospheric perturbations (33). Pulinets et al. (34)proposed that due to anisotropy of atmospheric conductivity at a height greater than 60 km, the large-scale highintensity vertical electric field appearing at a seismically active region few days before mainshock can penetrate the ionosphere and create ionospheric anomalies. TEC anomaly is usually observed above large thunderstorms, whereas no detectable localized TEC variation is observed for a thunderstorm-quiet night(35).

Terrestrial Outgoing IR Emission
Emission of infra-red radiation of wavelength 8-13 µm was detected (36,37) from ground-based observation satellite before several large earthquakes. The electrical potential and associated electrical field before any earthquake will excite the SiO4 tetrahedral stretching electronic bond of clay minerals or rock-forming silica, which could emit huge infrared radiation typically in the range 800-1300 cm -1 (7.7-12.5µm) (38, 39).

Ground Water Chemistry, Radon Gas Emission and Earthquake Light
Evidence of variation of dissolved species in groundwater before an earthquake was observed and reported in many literature (4). Radon gas anomalies in groundwater and soil gas was studied aiming to predict earthquake (5). Earthquake light (EQL) was observed just before or during strong earthquake events (40). Electrical field-induced ionization of different crustal elements could easily dissolve in groundwaters, leading to a change of groundwater chemistry. A trace amount of Uranium or Uranium oxide is always present in the rock-forming materials (41). Electronic configuration of Uranium could be written as [Rn] 5f3 6d1 7s2 which has six valence electrons. A strong electrical field could accelerate the spontaneous decay process of Uranium and facilitate the formation of more stable Radon, which we usually detect before an impending earthquake. Similarly, the electric field-induced ionization of environmental gases like nitrogen, oxygen could explain the strange luminous phenomena 'EQL' observed and recorded during strong earthquakes.

High Energy Electrical Discharge and Generation of Acoustic Shock Wave
The electrical potential will continue to rise exponentially in an earthquake preparation zone until it reaches its maximum electrical breakdown voltage of the surrounding materials/medium. Naturally, a high energy electrical discharge will occur in the medium, which produces an acoustic shockwave. For an analogy, natural lightning and thunder will better visualize this initial phase of the earthquake. Thunder like roaring (rumble) sound was heard immediately prior to the onset of felt vibrations caused by an earthquake (42). Sometimes the sound is compared with an explosion or sound of a freight train (see Supplementary Materials, additional information). Sometimes no shock was reported after an intense roar was noticed (43). Sample collected from PSZ [Taiwan Chelungpu-fault Drilling Project (TCDP)] shows features relating to melting or amorphous materials under SEM and TEM observations (25), which could be a signature of lightning-like discharge. Similar amorphous rock fulgurites(44, 45) are formed due to lightning strikes on the quartz-rich soil/mountain. Okazaki et al. (46) studied lawsonite dehydration as an earthquake triggering source in subducting oceanic crust. This dehydration could be resulted due to the immense heat produced during electrical discharge. It is generally accepted that weak fault materials cannot accumulate a large amount of elastic strain energy. The presence of smectite or interstratified weak clay minerals in the primary seismic slip zone excludes the possibility of storing huge elastic strain energy for earthquake triggering. Acoustic fluidization (AF) is the most plausible physical process for dynamical fault weakening (47)(48)(49). Fault starts to slip as long as the high-frequency acoustic shock waves remain sufficient to lead AF before dissipating within lowacoustic impedance fault gouge (50,51). Thus, AF induced mechanical fault-slip radiates seismic waves (P-wave, S-wave, and surface wave). Lin et al. (52)studied naturally occurring thunder as a seismic source. The acoustic shockwave from strong thunder could be coupled to the ground and excite P and S waves near the surface, causing seismicity (53). Finally, a schematic illustration is presented in Figure 3 to describe the whole earthquake process based on the previous and present understanding.

CONCLUSIONS
We have derived electrical potentials from seismic energy in the moment magnitude Mw (0.0-1.9) scale. We found a close match between seismic electrical potential (SEP) and Pourbaix Potential (PP) which leads us to find an electrochemical origin of earthquake energy. Weak clay minerals found in several scientific deep drilling projects excludes the possibility of stored strain energy as an earthquake triggering source. We have identified smectite and vermiculite as strong candidates for the generation of substantial electrochemical potential due to its excellent adsorption and cation exchange capacity in the silicon tetrahedral-aqueous interface. The exponential increase of Pourbaix Potential is supported by the Elovich model of chemisorption kinetics in clay minerals' multilayer adsorption site. We strongly believe that huge electrochemical potential is the possible earthquake triggering source that drives the entire earthquake process through lightning-like discharge and thunder like shockwave propagation. Acoustic fluidization (AF) is the possible physical process for the mechanical fault slip releasing seismic radiation through P-wave, S-wave, and surface waves. Several pre-seismic phenomena (SES, TEC, IR, EQL, and Radon gas) could be explained considering accumulation of electrical potential and associated electrical field in the hypocenter. This finding is a way forward to understand the true origin of an earthquake and paves the way for predicting an impending earthquake.

Demonstration of Electrochemical Cell using Liquid metal Potentiometric Sensor
Recently A. Das et al. (11) demonstrated a potentiometric sensor using an ultrathin Ga2O3-liquid metal system. The potentiometric sensor is fundamentally inherited from the electrochemical cell, where one half cell reaction occurs in a working electrode-aqueous interface. Another half-cell reaction is a well understood non-interfering reaction occurring at the reference (Ag‫|‬AgCl) electrode. Here we demonstrated that the half-cell reaction in each working electrode (LMPD) could be added like a regular electrochemical cell ( Figure S1). The measured potential for one LMPD is -1.100 V in pH 10 aqueous buffer solution where the voltage is doubled (-2.200 V) for properly connected two LMPD electrodes in series. For N number cells (NCell), simply the resultant potential will be NCell×1.100 V. The measured potential from a single cell (-1.100 V) can be understood from the following electrochemical reaction and associated electrode potential. The possible electrochemical response at the LMPD/aqueous pH buffer (pH 10) solution as follows, 3