On generating a Geological Model for Hydrogen Gas in the Southern Taoudeni Megabasin (Bourakebougou area, Mali) Denis Briere, Chapman Petroleum Engineering Ltd., Suite 700 1122 4 St SW Calgary, T2R 1M1, Canada Tomasz Jerzykiewicz, Geoclastica Consulting Ltd., 68 Hawkwood Hill, NW Calgary, T3G 3C6, Canada Wojciech Śliwiński, Wroclaw University, Institute of Geological Sciences, M.Borna Pl. 9, 50-204 Wroclaw, Poland An occurrence of 98% pure hydrogen gas has been discovered north of Bamako in a water well drilling surprise. A proven volume of hydrogen makes the discovery very important locally and perhaps it might even be of global significance. From the villagers in the mud huts of Bourakebougou to the green energy geovisionaries around the world, the question becomes: “What does this H2 gas discovery mean?” The objective of this paper is to explore the hydrogen discovery using conventional geological methods. Extensive surveys and technical reports were made to produce a working geological model. It is proposed that this new model provides predictive potential for discovering more naturally occurring hydrogen gas in the basin and elsewhere in the world. The discovery well is located in Tambaoura Basin (the southwestern part of Taoudenni Megabasin) in the Block #25 (Fig. 1). Figure1. Geological setting of the Block #25 The block covers a large area in the Tambaoura Basin north of its faulted margin build up with the early Proterozoic granite which belongs to the West African Craton. Our geological model is based on detailed sedimentological studies of two fully cored 2500m stratigraphic wells, with subsurface to surface field stratigraphy correlations, and shallow seismic survey interpretations (Jerzykiewicz, 2012). Bedrock within the block consists of the Neoproterozoic sedimentary cover and Mesozoic intrusive rocks, largely dolerites. The sedimentary cover of the basin belongs to the Neoproterozoic-to-early-Paleozoic succession of clastics and intrusive rocks assigned to Mesozoic (Jurassic?). The Neoproterozoic sediments are subdivided into three groups of formations: the Sotuba (St), the Souroukoto (So) and the Bakoye (Ba), Figs. 2 and 3). Figure 2. Stratigraphic classification of the south-western part of the Taoudeni Megabasin according to Deynoux et al., (2006). The crystalline basement of the Tambaoura Basin belongs to the Leo Shield of the West African Craton (Fig. 1). The lower part called Liberian is of the Lower Precambrian (Archean) age and it is highly metamorphosed, migmatized and granitized about 2700 Ma ago. The Archean is represented by granites, migmatites, anorthosites, charnockites and schists with hyperaluminous and ferruginous horizons (magnetitic quartzite). The upper part of the basement is called Birrimian and it is of Mid Precambrian age (metamorphosed and granitized at about 1800 Ma ago). We have identified two prospects named Bakoye and Sotuba-Souroukoto in the Proterozoic rocks of the Tambaoura basin (the Southern Taoudeni Megabasin). Their geological setting is like that of the Atar petroleum system of Mauritania (Fig. 3). Both prospects contain petroleum source rock (black shale with algal structure) and reservoir quality sandstones of 15% porosity. Deposition of the organic-rich strata in the basin was controlled by climate changes. Warm greenhouse conditions elevated the global sea level and led to the deposition of vast algal mats (most notably stromatolites) in the intertidal environment. The Bakoye strata represent the foreland stages, and the SotubaSouroukoto strata represent the rifting in a perisutural basin that developed in the West African Craton as a result of plate collision. The shallow seismic surveys show flower structures in dolerite that resulted from strike-slip tectonic deformation. Hydrogen gas concentrations occur in these flower structures. Figure3. Regional cross-section of the Block #25 (Tambaoura Basin). 1 - Lower Souroukoto strata (Koulouba Sandstone) with the Sotuba Sandstone at base , 2 – Middle Souroukoto strata (Nkomi Shale), 3 – Upper Souroukoto strata (Kati Sandstone), 4 – Bakoye strata, 5 – Nioro strata (Cambrian), 6 – Dolerite. The mere occurrence of such significant volumes of pure hydrogen in the Bourakebougou trap raises questions concerning the source of hydrogen gas and the conditions that facilitated such a high preservation potential of H2. The actual source of hydrogen gas in Bourakebougou is still to be determined though there are competing theories currently in play as follows: 1) 2) 3) 4) 5) 6) 7) Mantle degassing from deep faults. Rock crushing along fault lines. Serpentinization in mafic rocks exposed to ground water. Anaerobic corrosion Anaerobic fermentation. Spontaneous Potential natural electrolysis Petroleum cracking 1. Mantle degassing from deep faults Hydrogen gas has been observed in volcanoes, geothermal springs, and deep-rooted faults (Sano Y., et al. 1993). Russian and French scientists (Larin 1975; Larin et al., 2015) are currently progressing this model of deep mantle degassing which manifests itself as a continuous flux from circular patterned surface vents. In Bourakebougou, there is a nearby deep fault occurrence into the Lower Proterozoic basement which could be the delivery system for mantle degassing in the Mali hydrogen discovery. To date, a continuous gas venting has not yet been recorded in the village. Continuous hydrogen gas flowed from the discovery well when it was first drilled, and also now, when the master valve is opened. To test this theory, a year long program for wide area surface mapping of hydrogen gas eminations must be made to verify any continuous flux venting effects that are observed like in other parts of the world from circular sinkhole patterns. 2. Rock crushing along fault lines Rock crushing along fault lines could also be responsible for the generation of hydrogen gas in the nearby deep fault occurrence of Bourakebougou. Freund et al., (2002) has found that hydrogen molecules diffuse out of freshly fractured mineral surfaces. He proves that hydrogen molecules can be derived from small amounts of water dissolved in minerals in the form of hydroxyl OH- ions whenever minerals crystallize in an H2Oladen environment as follows: - 2OH O2 2- + H2 It appears that whenever a deep fault undergoes tectonic slippage, hydrogen gas is produced at the crushed rock face. In Bourakebougou, there is evidence of nearby deep fault intermittent activity from the results of passive seismic surveys performed by Petroma. The presence of intermittent activity from a small cave called “La Maison du Diable” on a western escarpment in the Bourakebougou village supports the theory that gas is being generated randomly maybe from rock crushing along fault lines through the dolerite sill, the Kati Sandstone, and even down to the basement. It has been found by Freund (op. cit) in his laboratory experiments that H2 molecules cannot easily diffuse out of mineral grains where the intergranular space is saturated with water. In this case the out-diffusion of H2 will proceed only to the point where the H2 concentrations inside the minerals reach their equilibrium with the H2 concentration in the water film at grain boundaries. Hydrostatic pressure increases the H2 solubility in water only slightly. In the presence of an intergranular water film, a large fraction of the H2 is retained inside the mineral grains. The H2 solubility in pure water at 1 bar and 0°C is 0.022 cm3 of H2/cm3. This decreases to less than 0.014 cm3 of H2/cm3 in water with a salinity of 22 to 11 ppm (rainwater). Well water in the village is averaging a salinity of 44 ppm in 34 wells, which further inhibits H2 solubility, and presumably acts to retain the hydrogen gas in place. To test this theory, passive seismic recorders should be installed in La Maison du Diable along with trace gas detectors over a one year program to verify that seismic activity triggers gas generating occurrences. 3. Serpentinization in mafic rocks exposed to ground water The process of serpentinization of ultramafic rocks is one of the possibilities which should be taken into account as a potential source of hydrogen. The process of hydrogen production could take place according to the following reaction (Sleep et al., 2014): 3Fe2SiO4 + 2H2O → 3SiO2 + 2Fe3O4 + 2H2 fayalite (olivine) + water → quartz + magnetite + hydrogen In such a process the final product is pure hydrogen. The olivine could be either from the ultramafic rocks of the deep basement (Lower Precambrian) or directly from the dolerite dyke. Supercritical water must be available for this reaction. The igneous sill at Bourakebougou would have satisfied this temperature requirement long ago, but not today. To test this theory, future drill cuttings into the Archean must demonstrate the presence of olivine, serpentinite, and magnetite from the same borehole. 4. Anaerobic corrosion Another possible source of hydrogen could be from the anaerobic corrosion of hyperaluminous and ferruginous horizons in the basement rocks. This thermic interaction without air, could lead to the degradation of iron which is slowly oxidized by the protons of water into making ferrous hydroxide (green rust) which can be described by the following reaction: Fe + 2 H2O → Fe(OH)2 + H2 Iron water Ferrous oxide Further under anaerobic conditions, the ferrous hydroxide (Fe(OH)2) can also be oxidized by the protons of water to form magnetite and molecular hydrogen. This process is described by the Schikorr reaction: 3 Fe(OH)2 → Fe3O4 + 2 H2O + H2 ferrous hydroxide → magnetite + water + hydrogen This reaction could be followed possibly by oxidation of magnetite to hematite with the release of more hydrogen: 2Fe3O4 + H2O → 3Fe2O3 +H2 magnetite hematite These processes could also occur during the anaerobic corrosion of iron in oxygen free groundwater and in reducing soils below the water table. The final reaction with the release of hydrogen takes place at low temperatures (<200oC) hence it may continue into the present time (Mayhew et al. 2013). This reaction of the oxidation of magnetite to hematite is relevant to the case in question because a lab tested dolerite sample measured ferromagnetism at 1% SI bulk magnetic susceptibility. This reaction could explain the presence of hydrogen gas in the Bourakebougou trap in such large volumes to this day despite a perceived low preservation potential of H2 in the 100m overburden. Though hydrogen can be produced this way when water reacts with fresh mineral surfaces and oxidizes ferrous iron, this reaction usually depends upon the exposure of fresh rock surfaces from episodic opening of cracks and fissures. Perhaps the rainy season provides this effect by changing the water table depths and squeezing rainwater into the fractures of the rock flower structures that have been seen on shallow seismic. To test this theory, many future drill cuttings to the base of the Souroukoto must demonstrate a geologic mapping of anaerobic corrosion indicators that show a migration path for the hydrogen gas to the existing well at Bourakebougou. 5. Anaerobic fermentation In the mud brick village of Bourakebougou, hydrogen gas might possibly be produced naturally by bacteria that can convert organic biomass into biogas through dark fermentation of organic materials (Hallenbeck and Benemann, 2002) as follows: C12H22O11 + 5H20 → 4CH3COOH + 8H2+ 4CO2 Glucose Acetic Acid Dark fermentation (Chen, W.H., 2006) is a catabolic process in which bacteria converts sugar and protein to carboxylic acids, hydrogen gas, carbon dioxide, and organic solvents including ethanol (C2H6O), acetone (C3H6O), propanol (C3H8O), and butanol (C4H9OH). There are times during the year when the villagers report that their water from certain wells tastes bad. A pH value no lower than 5 is required for dark fermentation. In all of the many wells of the village, drinking water pH ranges from 6 to 8. This is close to the Canadian water quality guidelines of drinking water pH of 6.5 to 8.5 and it is correct to support dark fermentation. To test this theory, the village biowaste sludge should be tested in the laboratory for hydrogen bioproduction, and fermentation micro-organisms must definitely be detected either in the ponds or be found to exist in the laterite topsoil surrounding the village. 6. Spontaneous Potential natural electrolysis The electrolysis of water1 is the decomposition of water (H2O) into hydrogen gas (H2) and oxygen (O2) due to an electric current being passed through the water. The overall reaction is as follows: 2 H2O(l) → 2 H2(g) + O2(g) Liquid gas gas Pure water is a fairly good insulator and therefore it conducts an electric current poorly unless a very large potential is applied to cause an increase in the auto-ionization for the electrolysis to proceed in low conductivity. If a water-soluble electrolyte is added, then the conductivity of the water rises considerably and electrolysis is allowed to continue at lower voltages. The electrolytes with Li+, Rb+, K+, Cs+, Ba2+, Sr2+, Ca2+, Na+, and Mg2+ are frequently used, as they form inexpensive soluble salts. The soil geochemistry performed in Bourakebougou identified all nine of these electrolytes within twelve inches of the laterite topsoil. For natural water electrolysis to occur, a standard potential of 1.23 V at 25 °C with pH 7 is needed. Taking losses into account however, voltages over 1.48 V are required for the reaction to proceed at practical current densities. 1 Electrolysis of water, from Wikipedia.org, the free encyclopedia During the surface resistivity surveys in Bourakebougou, a spontaneous potential measurement was performed which revealed an east-west naturally occurring isocontour of 0.3 Volt potential difference within one metre into the topsoil laterite. This result indicates the presence of a cathode wall under the village where hydrogen would normally appear if the voltage were higher, and also the presence of an anode wall where oxygen would normally appear along on the western escarpment. To test this theory, a shallow and deep SP survey should be recorded continuously for a year to observe whether increased voltages occur when hydrogen gas sensors in the village show activity at the same time when oxygen gas sensors in La Maison du Diable show activity. 7. Petroleum cracking Cracking is the process where long chain molecules are broken down into simpler molecules by the breaking of carbon-carbon bonds. Cracking and its end products are strongly dependent on temperature and on the presence of catalysts. Petroleum cracking has been an industrial process since 1891. The cracking process typically runs at temperatures around 400 oC and requires a catalyst such as alumina which promotes asymmetric breakage of chemical bonds. Alumina occurs naturally in the mineral Corundum which is found in ultramafic intrusions like those in Bourakebougou. Consider the ways of how a butane molecule (CH3-CH2-CH2-CH3) might be cracked: 48% probability: break at the CH3-CH2 bond: CH3* / *CH2-CH2-CH3 → CH4 + CH2=CH-CH3 Methane Propylene 38% probability: break at a CH2-CH2 bond: CH3-CH2* / *CH2-CH3 → CH3-CH3 + Ethane CH2=CH2 Ethylene 14% probability: break at a terminal C-H bond: H/CH2-CH2-CH2-CH3 → CH2=CH-CH2-CH3 1-Butene + H2 Hydrogen gas: In and around the village of Bourakebougou, after the rainy season, a blue film can be seen on the pond water. This water with its blue film has been sampled and tested in the laboratory. Results show that the blue film contains complex hydrocarbon chains with 0.6 mg/L of (C10 - C16), and 4.6 mg/L of (C16 - C34), and also 0.8 mg/L of (C34 - C50) carbon chain molecules. Like serpentinization, petroleum cracking needs high temperatures. This would have been satisfied long ago during the igneous intrusion event, but not today unless the cracking is from much deeper in the Proterozoic. This theory also requires that hydrocarbons are necessarily present at those deep depths. As of this date, no oil has been discovered in Mali, but there is one encouragement from the nearby village of Didieni because long ago they named their village:”water that burns”. To test this theory, deep drilling around Bourakebougou must prove the existence of oil in the Proterozoic basement. References Chen, W.H., 2006: Biological hydrogen production by anaerobic fermentation. Retrospective Theses and Dissertations. Iowa State University, Paper 1863. Deynoux, M., Affaton, P., Trompette, R. and M. Villeneuve, 2006: Pan-African tectonic evolution and glacial events registered in Neoproterozoic to Cambrian cratonic and foreland basins of West Africa Journal of African Earth Sciences 46, 397–426 Freund F., Dickinson J.T., Cash M., 2002: Hydrogen in rocks: an energy source for deep microbial communities. Astrrobiology Spring; vol. 2(1):83-92. Hallenbeck, P.C.; Benemann, J.R. 2002: Biological hydrogen production; fundamentals and limiting processes. International Journal of Hydrogen Energy. vol. 27 (11): 11851193(9). Jerzykiewicz, T., 2012: Towards a comprehensive geological model of the Tambaoura basin and its hydrocarbon system (Block #25 – Bamako Area, Mali). Report written for Petroma Ltd. Available from the Geoclastica Consulting Ltd., 68 Hawkwood Hill, NW Calgary, T3G 3C6, Canada Larin N., Zgonnik V., Rodina S., Deville, Prinzhofer A., Larin V.N., 2015: Natural Molecular Hydrogen Seepage Associated with Surficial, Rounded Depressions on the European Craton in Russia. Natural Resources Research, vol. 24 (3): 369-383 Larin, V. N., 1975, The Hypothesis of Primordial Hydride Earth. Nedra, Moscow. Mayhew L.E., Ellison E.T., McCollom T.M., Trainor T.P., & A. S. Templeton A.S., 2013: Hydrogen generation from low-temperature water–rock reactions. Nature Geoscience, vol 6, 478-484. Sano Y., Urabe A., Wakita H., 1993: Origin of hydrogen-nitrogen gas seeps, Oman. Applied Geochemistry. vol 8(1) : 1-8 Sleep N.H, Meibom A., Fridriksson Th., Coleman, R.G. & Bird D.K., 2014: H2-rich fluids from serpentinization: Geochemical and biotic implications. PNAS vol. 101, no. 35, 12818–12823.