TY - JOUR
AU1 - Zhang,, Xueang
AU2 - Yang,, Zhichao
AU3 - Li,, Xiaoyan
AB - SUMMARY Uranium is an important energy material and a key exploration object in the field of radioactive geophysical exploration. Through field study, it was found that fractured uranium is one of the most abundant sources of uranium, therefore it is an important target of energy geophysical exploration. However, the fractured structure increases the difficulty of uranium exploration and analysis. Therefore, in order to identify the distribution and analyse the content of fractured uranium deposits, it is necessary to study the influence of fracture parameters on the occurrence state of uranium ore and its manifestation status in neutron logging response. Considering the fracture structure from different aspects, this study simulated the uranium ore occurrence state in the pore-fracture medium environment and obtained corresponding logging response results by setting two fracture parameters (number density and angle). The study generated several findings, namely: (1) high angle and high porosity fracture type uranium strata have the greatest influence on thermal neutrons; (2) under the condition of high porosity, the neutron energy spectrum displays a dispersion peak and moves towards the high energy region. It is noteworthy that in a high number density environment, medium-low angle uranium ore has the strongest thermal neutron absorption in a low energy range. (3) energy decay rate is the highest in the high angle fracture environment and (4) the higher the density of the fracture number, the more susceptible the counting is to the influence of the angle factor. These findings can be used to quantitatively analyse the effects of fracture parameters on uranium deposits. These results have important guiding significance to the evaluation of a fractured uranium reservoir. Fracture and flow, Downhole methods, Numerical approximations and analysis, Numerical modelling, Statistical methods 1 INTRODUCTION Uranium is an important mineral for the production of clean energy, and it has remained one of the key research focuses in geophysical exploration activities. Uranium deposits are directly or indirectly surveyed and quantified by different geophysical methods. (Animesh et al. 2015; Hajnal et al. 2015; Chen et al. 2018). Mandal et al. analysed the characteristics of low-density uranium mineralization zones by gravity, magnetism and electricity, and defined the target zones of uranium mineralization zones in the corresponding areas. The Shea Creek area was studied by Hajnal et al., using high-resolution seismic exploration methods, to evaluate the potential of surface seismic reflection technology to detect uranium zone markers. Chen et al. identified the blind uranium deposits based on the geochemical characteristics of the sandstone rocks extracted from the cores and the multivariate statistical assessment of compositional data. Due to the radioactive decay of uranium deposits, many research methods for uranium exploration are also radioactive (Asfahani 2017; Bilal & Petrescu 2017). Essaid et al. used inductively coupled plasma emission spectroscopy (ICP-AES) to analyse the surface water of a uranium reactor in Romania and determined the formation and source of natural uranium and thorium in the mine area. In another study, Jamal et al. estimated the porosity and conductivity of water in the basalt aquifer using radioactive logging methods and electrical logging techniques. Using the radiological exploration method, neutron logging can directly quantitatively analyse the uranium reserves and state, pulse neutron logging, fast neutron inelastic scattering ray energy spectrum logging and neutron capture ray energy spectrum logging (Abdel-Fattah & Mohamed 2015; Ajayi et al. 2015a,b; Dance & Paterson 2016; Ortega et al. 2015, 2017; Bogdanovich et al. 2017; Tian et al. 2018), among others. Along with these studies, Abdel-Fattah et al. used a variety of radioactive logging methods (e.g. gamma logging and neutron logging) to analyse the influence of a sedimentary environment on reservoir physical properties. Ajayi et al. estimated the formation elements and mineral composition using the gamma spectral inversion method and the fast numerical simulation method, respectively. In their study, Dance et al. observed the carbon dioxide saturation distribution and capture relationship using core analysis and neutron logging, respectively. Ortega et al. used thermal neutron logging and gamma time-decay logging to undertake two-dimensional inversion interpretation of formation lithology and rock physical properties, and studied the intrusion shoulder bed and borehole inclination effect through neutron and gamma logging. Bogdanovich et al. improved the logging efficiency by simultaneously analysing the neutron field and gamma field formed in the well, and Tian et al. measured the salinity of formation water using the prompt gamma neutron activation analysis logging method (PGNAA). With the development of neutron logging technology in uranium geophysical exploration, various types of uranium deposits have been analysed and detected. Fractured uranium ore is one of the most common occurrence states of uranium ore; however, the complexity of the fracture structure makes it difficult to detect. In the field of radioactive geophysical logging, there is limited study on fractured uranium strata. To the best of our knowledge, no reports have been published on the identification of uranium strata in a fractured medium for neutron logging. Based on the numerical simulation of pulsed neutron logging (Yang et al. 2016; Zhang et al. 2018) we investigated the occurrence state of uranium and uranium deposits under a fracture medium (Hudson 1980, 1991; Hudson et al. 1996; Zhang & Wang 2015; Tün et al. 2016; Zhang et al. 2017) influence law of logging response, through numerical simulation which can eliminate the influence of natural environment factors on the logging results and interference. This quantitative work allowed for a more comprehensive analysis of the fractured uranium and provided better predictive guidance for geophysical uranium mining. At the same time, the numerical simulations improved the accuracy of the exploration results and the efficiency of geophysical exploration efforts, thereby resulting in economic benefits for uranium mining. 2 WELL-SITE ENVIRONMENT SIMULATION In this study, the data and model structures used to establish the numerical uranium environment are based on a real uranium environment, in particular, the occurrence characteristics of sandstone type uranium ore. This kind of uranium ore environment is suitable for detections by the neutron logging method, which can directly quantify uranium ore with high accuracy. Therefore, the selection of the formation environmental parameters for numerical simulation has strong practicability and great application value, and the obtained simulation results can be used to guide the neutron logging in actual sandstone type uranium deposits. Neutron uranium logging technology detects fission neutrons generated by neutron excitation of uranium in the formation, which can directly reflect the uranium content in the formation. In this study, the fast neutrons generated by the pulsed neutron source react with uranium in fracture formation, following which the fracture parameters and logging response are obtained by statistical reaction law. The 235U is a fissionable nucleus with different fission reaction cross sections (probabilities) under different neutron energy conditions. Neutrons emitted during fission are collectively called fission neutrons, and Ninety-nine per cent of fission neutrons (called prompt neutrons) are released at the instance of fission (about 10−14s). Part count of the prompt neutron is proportional to the uranium content in the formation. Therefore, the study of the response relationship of neutron logging to uranium is a direct and straightforward uranium measurement method, which makes it easier to correct the accuracy of uranium content than other indirect uranium measurement methods. In the actual formation environment, it is necessary to conduct environmental correction for formation factors that have an impact on neutron distribution. In this study, the influence law of fracture parameters on pulsed neutron logging response is analysed to provide reference and theoretical support for environmental correction of neutron distribution in the field uranium measurement. A 14 MeV pulsed neutron source was used in this study to send fast neutrons into the formation (Zhang et al. 2018): $$\begin{eqnarray} _1^3H + _1^2H \to _2^4He + _0^1n + 17.588\,{\rm MeV}. \end{eqnarray}$$(1) The |${}_0^1n$| moderated into thermal neutrons and induced 235U fission to detect the variation law of fission neutrons and the response results of fracture parameters. The general reaction formula of fission is as follows: $$\begin{eqnarray} {}_{92}^{235}U + {}_0^1n \to {}_{Z1}^{A1}X + {}_{Z2}^{A2}Y + m_0^1n, \end{eqnarray}$$(2) where the neutrons|${}_0^1n$|emitted from the pulsed neutron source react with 235U and release |${}_{Z1}^{A1}X \,.\, {}_{Z2}^{A2}Y \, .\, m_0^1n $| . X, Y is fission fragments, medium quality unstable and radioactive nuclides, which undergo beta decay, and to form stable isotopes; |$m_0^1n$| is the prompt neutron generated after excitation of 235U by a neutron source. This study mainly analyses the logging response law of prompt neutrons released at the point of fission. Neutron logging is used in this study to resolve the neutron transport problem in uranium ore formation. We constructed the characteristic numerical environment by simulating a pulsed neutron source in the centre of a borehole embedded in a uranium formation. The bedrock media around the well is divided into two materials. The dike structure is embedded in the matrix surrounding rock, which is predominantly composed of quartz, the other material is the surrounding rock matrix and its main component is calcite. The fracture structure is embedded in the bedrock media and is most dense in the innermost layer (Fig. 1). The chemical composition of the numerical model is shown in Table 1. Figure 1. Open in new tabDownload slide Diagram of the borehole environment. 1 – Surrounding rock matrix; 2 – Quartz vein; 3 – Wellhole and 4 – Fracture. Figure 1. Open in new tabDownload slide Diagram of the borehole environment. 1 – Surrounding rock matrix; 2 – Quartz vein; 3 – Wellhole and 4 – Fracture. Table 1. Chemical composition of the numerical model. Component name . Component proportion (per cent) . SiO2 41.25 CaO 33.67 CO2 23.24 Fe2O3 1.49 H2O 0.21 MgO 0.14 Component name . Component proportion (per cent) . SiO2 41.25 CaO 33.67 CO2 23.24 Fe2O3 1.49 H2O 0.21 MgO 0.14 Open in new tab Table 1. Chemical composition of the numerical model. Component name . Component proportion (per cent) . SiO2 41.25 CaO 33.67 CO2 23.24 Fe2O3 1.49 H2O 0.21 MgO 0.14 Component name . Component proportion (per cent) . SiO2 41.25 CaO 33.67 CO2 23.24 Fe2O3 1.49 H2O 0.21 MgO 0.14 Open in new tab Uranium ore occurs in the fissure structure (Fig. 1). The aspect ratio of fracture structures is 0.01 and the porosity is 0.005. We set the pulse neutron source at the centre of the vertical shaft. The fluid in the hole is mud. The normal direction of the main body of all fracture structures is the same as that of the shaft, which varies from 0° to 90°. The simulation calculation of neutron transport and reaction process is mainly divided into three steps: (1) determine the sequence of neutron state parameters; (2) determine the neutron transport process and (3) record the results of neutron transport and reaction. Step 1 determines the neutron's state parameters, and obtains the sequence of state parameters: $$\begin{eqnarray} {S_m} = \left( {{r_m},{E_m},{\Omega _m},{t_m},{W_m}} \right), \end{eqnarray}$$(3) where |${r_m}$| is the mth collision point; |${E_m}$| is the energy after the mth collision; |${\Omega _m}$| is the direction of motion after the mth collision; |${t_m}$| is the time of the mth collision and |${W_m}$| is the weight after the mth collision. The source neutron enters the formation and reacts with the atomic nucleus of the formation material multiple times. Eventually, the energy runs out and stops transporting and the neutron is absorbed by the nucleus of the formation. The simulation calculation of this process is based on the simulation of the neutron free path lf: $$\begin{eqnarray} {l_f} = - \frac{{\ln \cdot \xi }}{{{\Sigma _t}}}, \end{eqnarray}$$(4) where |${\Sigma _t}$|is the macroscopic total cross section, the measurement of the probability of an interaction between the neutron and the nucleus of formation material, and|$\xi $|is random numbers which obey uniform distribution in the interval (0, 1). The location of the next collision point |${r_m}$| is |${r_m} = {r_{m - 1}} + {l_f}$|⁠. Step 2 determines the type of atomic nucleus of the formation material that reacts with neutrons. As there are so many materials in the formation, including uranium, determining what nuclei neutrons react with is a major part of the simulation. First, n kinds of nuclei are set in the formation, and then the total reaction cross sections of the reaction with neutrons and the reaction cross sections of each nucleus are obtained (⁠|${\Sigma _1}$|⁠, |${\Sigma _2}$|⁠, …, |${\Sigma _n}$|⁠). In this study, the collision nuclides are determined by direct sampling of discrete random variables. The probability that the neutron reacts with the n nuclide is |${P_n} = {\Sigma _n}/{\Sigma _t}$|⁠. Sampling rules for determining the types of reaction nuclides: $$\begin{eqnarray} {\rm{Neutron}}\,{\rm{reaction \ nuclide}} = \left\{ \begin{array}{@{}l@{}} {\rm{First\,nuclide }}\,0 \le \xi < {P_1}\\ {\mathop{\rm Second\,nuclide}\nolimits} {\rm{ }}\,{P_1} \le \xi < {P_1} + {P_2}\\ {\rm{ }} \cdots {\rm{ }} \cdots {\rm{ }}\\ {n{\rm th}}\,{\rm{nuclide }}\,\Sigma _1^{n - 1}{P_i} \le \xi \le 1. \end{array} \right. \end{eqnarray}$$(5) This step further determines the type of reaction between neutrons and the nuclei of formation materials. Neutrons may react with the nucleus of the formation in a variety of ways (such as elastic scattering, inelastic scattering, trapping reaction and fission). The probability of each reaction is as follows: |${\sigma _{el}},{\rm{ }}{\sigma _{in}},{\rm{ }}{\sigma _a},{\rm{ }}{\sigma _f}$|⁠. The total reaction probability is |${\sigma _t}$|⁠. Response types were determined by a discrete random variable sampling method: $$\begin{eqnarray} {\mathop{\rm Reaction\,}\nolimits} {\rm{type }} = \left\{ \begin{array}{@{}l@{}} {\rm{Elastic\, }}{\rm{scattering }}\left( {0 \le \xi < \frac{{{\sigma _{el}}}}{{{\sigma _t}}}} \right)\\ {\rm{Inelastic\,scattering }}\left( {\frac{{{\sigma _{el}}}}{{{\sigma _t}}} \le \xi < \frac{{{\sigma _{el}} + {\sigma _{in}}}}{{{\sigma _t}}}} \right)\\ {\rm{ }} \cdots {\rm{ }} \cdots \\ {\rm{Fission }}\left( {1 - \frac{{{\sigma _f}}}{{{\sigma _t}}} \le \xi \le \frac{{{\sigma _{el}}}}{{{\sigma _t}}}} \right). \end{array} \right. \end{eqnarray}$$(6) Step 3. records the results. If the neutron energy is exhausted and absorbed by the nucleus of the formation, the number of neutrons is not recorded; otherwise, the number of neutrons is recorded. Because the calculated object in this study is uranium ore, the content of which in the strata is low, a more precise statistical estimation method should be adopted in the recording stage to reduce error and improve the accuracy of the recording. The statistical estimation method can make more use of the history of neutrons. When a neutron penetrates the medium directly without collision, or penetrates the medium after multiple collisions, these events are incompatible, and the recording function is: $$\begin{eqnarray} {f_m} = \prod\limits_{l = 1}^m {\frac{{{\sum _s}\left( {{Z_l},{E_{l - 1}}} \right)}}{{{\sum _t}\left( {{Z_l},{E_{l - 1}}} \right)}}} \bullet \eta \left( {0 \le {Z_l} \le a} \right) \bullet f\left( {{Z_m},{E_m},\cos {\alpha _m}} \right). \end{eqnarray}$$(7) Then, an unbiased estimate P of the penetration probability is: $$\begin{eqnarray} P = \sum\limits_{m = l}^\infty {} \prod\limits_{l = 1}^m {\frac{{{\sum _s}\left( {{Z_l},{E_{l - 1}}} \right)}}{{{\sum _t}\left( {{Z_l},{E_{l - 1}}} \right)}}} \bullet \eta \left( {0 \le {Z_l} \le a} \right) \bullet f\left( {{Z_m},{E_m},\cos {\alpha _m}} \right). \end{eqnarray}$$(8) This formula shows that particles contribute (⁠|$\prod\limits_{l = 1}^m {\frac{{{\sum _s}( {{Z_l},{E_{l - 1}}} )}}{{{\sum _t}( {{Z_l},{E_{l - 1}}} )}}} \bullet f( {{Z_m},{E_m},\cos {\alpha _m}} )$|⁠) as long as they collide within the system. If the particles get reflected or escape from the system, particle history ends. The total penetration probability composed of these penetrating processes can be represented by P, |$P = \sum\limits_{m = 0}^\infty {{P_m}} $|⁠; here, |${P_m}$| is the probability that a neutron will pass through the medium after mth collisions. The contribution of neutrons to the penetration probability is not limited to the last collision by statistical estimation. The statistical estimation method deals with the transport length of particles analytically and makes more use of the neutron motion history. Because this method has more statistical properties than other methods, it is called statistical estimation method. For any state of the neutron (⁠|$Z,E,\cos \alpha $|⁠), the contribution to the penetration shield is |$\exp \{ { - \sum\nolimits_t {( E ) \cdot \frac{{a - Z}}{{\cos a}}} } \}( {\cos a > 0} )$|⁠, transport length|$l = - \ln \xi /{\Sigma _t}( E )$|⁠, when |$l > \frac{{a - Z}}{{\cos a}}( {\cos a > 0} )$|⁠, neutron penetrates shielding and contributes while: $$\begin{eqnarray} \begin{array}{@{}l@{}} P\left( {l > \frac{{a - Z}}{{\cos a}}} \right) = P\left( { - \frac{1}{{\sum\nolimits_t {\left( E \right)} }}\ln \xi > \frac{{a - Z}}{{\cos a}}} \right)\\ = P\left( {\xi \le \exp \left\{ { - \sum\nolimits_t {\left( E \right) \cdot \frac{{a - Z}}{{\cos a}}} } \right\}} \right)\\ = \exp \left\{ { - \sum\nolimits_t {\left( E \right) \cdot \frac{{a - Z}}{{\cos a}}} } \right\}. \end{array} \end{eqnarray}$$(9) In the direct recording method under the condition of statistical estimation method, the history of neutrons, namely the contribution after the mth collision, should be firstly simulated: $$\begin{eqnarray} P_m^{\prime} = \left\{ \begin{array}{@{}l@{}} \exp \left\{ { - \sum\nolimits_t {\left( {{E_m}} \right)\frac{{a - {Z_m}}}{{\cos {\alpha _m}}}} } \right\}{\rm{ }}\cos {\alpha _m} > 0\\ 0\,\,{\rm{ others}}, \end{array} \right. \end{eqnarray}$$(10) where a is the thickness of the medium, Zm is the distance from the point where the collision occurred to the emission source, and |${\alpha _m}$| is the included angle between the direction of motion of the particle and the radial direction. When m = 0, |$P_0^{\prime}$| is the contribution of the neutron from the source to penetrate the shield without collision. Through the above calculation, the probability of neutron recording can be determined, and then: The contribution of the neutron to the detector is as follows: $$\begin{eqnarray} {\eta _n} = \left\{ \begin{array}{@{}l@{}} 0 \, ( {\rm Capturing \, reactions \, occur \, between} \\ \quad {\rm the \, neutron \, and \, nuclide })\\ 1 \, ( {\rm There \, is \, no \, capturing \, reaction} \\ \quad {\rm between\, the \, neutron \, and \,nuclide} ) .\end{array} \right. \end{eqnarray}$$(11) According to the formula, the neutrons are divided into two types: uncaptured neutrons continue to be tracked and calculated, while captured neutrons are no longer tracked and calculated. To calculate the distribution of thermal neutron density in uranium deposits, the probability size (macroscopic cross-section) of the reaction is determined as follows (Zhang et al. 2018). The distribution of thermal neurons is calculated by the dynamic diffusion equation: $$\begin{eqnarray} \frac{{{\rm d}n}}{{{\rm d}t}} = S + {D_0}{\nabla ^2}n - \sum {vn}, \end{eqnarray}$$(12) where n is the thermal neutron density; v is neutron velocity (cm s-1); S is the density of the thermal neutron source; |${D_0}$| is the neutron diffusion coefficient (relative to the thermal neutron density) and |$\Sigma $| is the thermal neutron capture cross-section. Neutrons emitted from the source can be decelerated from fast neutrons to thermal neutron after 1 μs. The density distribution of thermal neutrons can be obtained by numerical solution of a dynamic diffusion equation. In this study, thermal neutron counting is the superposition of borehole counting and formation counting: $$\begin{eqnarray} N\left( t \right) = {N_{10*e}}^{{{{ - t}}/ \!{{{\tau _1}}}}} + {N_{20*e}}^{{{{ - t}}/ \!{{{\tau _2}}}}}, \end{eqnarray}$$(13) where |${N_{10}}$| and |${N_{20}}$| represent the thermal neutron number at the initial time in the borehole and formation, respectively; |${\tau _1}$| and |${\tau _2}$| represent the thermal neutron lifetime of the borehole and formation, respectively. The macroscopic capture cross-section (⁠|$\Sigma $|⁠) is related to the thermal neutron lifetime as follows: |$\tau = \frac{1}{{v\Sigma }}$|⁠, in which |$v = 2.1978 \times {10^5}$| cm s−1, so the form can be written as:|$\tau = \frac{{4550}}{\Sigma }$|⁠. 3 LOGGING RESPONSE CHARACTERISTICS OF URANIUM DEPOSIT To analyse the influence factors of fracture parameters on neutron transport state in uranium ore environment, we set the range of fracture angle variation as 0°–90°. Table 2 shows the borehole background parameters. Table 2. Parameters of borehole model and pulsed neutron source. Parameter name . Parameter values . Initial time of neutron pulse 0 s Initial energy of neutron pulse 14.1 MeV Pulse width 5 μs Radius of the borehole 0.1 m Parameter name . Parameter values . Initial time of neutron pulse 0 s Initial energy of neutron pulse 14.1 MeV Pulse width 5 μs Radius of the borehole 0.1 m Open in new tab Table 2. Parameters of borehole model and pulsed neutron source. Parameter name . Parameter values . Initial time of neutron pulse 0 s Initial energy of neutron pulse 14.1 MeV Pulse width 5 μs Radius of the borehole 0.1 m Parameter name . Parameter values . Initial time of neutron pulse 0 s Initial energy of neutron pulse 14.1 MeV Pulse width 5 μs Radius of the borehole 0.1 m Open in new tab By changing fracture parameters, the logging response of fracture type uranium ore under corresponding conditions is obtained. These include: thermal neutron density distribution diagram, angle-neutron sensitivity diagram, number density-neutron energy counting spectrum and number density-neutron time relation spectrum. Figs 2–5 show the thermal neutron density distributions for the fracture type uranium deposit. The fracture angle increased gradually from small to large (0°–90°), and it was found that thermal neutron density distribution presented different states in the fractured uranium ore environment at different angles. Figure 2. Open in new tabDownload slide Distribution of thermal neutron density in horizontally angled fractured uranium deposits. (a) Fracture number density 0.1; (b) fracture number density 0.3 and (c) fracture number density 0.4. Figure 2. Open in new tabDownload slide Distribution of thermal neutron density in horizontally angled fractured uranium deposits. (a) Fracture number density 0.1; (b) fracture number density 0.3 and (c) fracture number density 0.4. Figure 3. Open in new tabDownload slide Distribution of thermal neutron density in medium-low angled fractured uranium deposits. (a) Fracture number density 0.1; (b) fracture number density 0.3 (c) fracture number density 0.4. Figure 3. Open in new tabDownload slide Distribution of thermal neutron density in medium-low angled fractured uranium deposits. (a) Fracture number density 0.1; (b) fracture number density 0.3 (c) fracture number density 0.4. Figure 4. Open in new tabDownload slide Distribution of thermal neutron density in medium-high angled fractured uranium deposits. (a) Fracture number density 0.1; (b) fracture number density 0.3 and (c) fracture number density 0.4. Figure 4. Open in new tabDownload slide Distribution of thermal neutron density in medium-high angled fractured uranium deposits. (a) Fracture number density 0.1; (b) fracture number density 0.3 and (c) fracture number density 0.4. Figure 5. Open in new tabDownload slide Distribution of thermal neutron density in vertical angled fractured uranium deposits. (a) Fracture number density 0.1; (b) fracture number density 0.3 and (c) fracture number density 0.4. Figure 5. Open in new tabDownload slide Distribution of thermal neutron density in vertical angled fractured uranium deposits. (a) Fracture number density 0.1; (b) fracture number density 0.3 and (c) fracture number density 0.4. Fig. 2 shows that the distribution of thermal neutron density in the approximate horizontal (0°–29°) fracture environment is significantly different when the number density is 0.1, 0.3 and 0.4, respectively: in the range with small number density (0.1–0.2), the peak value of thermal neutron density distribution is the lowest, and the number of peak value is small. The distribution structure is single and flat, so the thermal neutron has low sensitivity to uranium environment under low number density. When the fracture number density is high (0.3–0.4), the number of peaks in the thermal neutron density distribution increased and some of the peaks increased, and the overall distribution peak tip became sharper. This indicates that thermal neutron distribution is more sensitive to uranium ore environment under the condition of high density of fractures, resulting in sharp peaks in local areas and significant rise and fall of density distribution. As Fig. 3 shows, in the medium-low angle (30°–59°) fractured uranium ore environment, with the increase of fracture number density, the thermal neutron distribution still conforms to the trend change from flat and simple distribution to dispersion and significant fluctuation difference. However, the variation of the environment is slightly smaller than that of similar horizontal fractures. That is, when the fracture number density is high (0.4), the thermal neutron distribution sensitivity is slightly lower than that of the approximate horizontal fracture environment in the medium-low angle fracture type uranium ore environment. Nevertheless, when the fracture number density is 0.4, the peaks of thermal neutron density distribution decreases more compared with other fracture environments. This indicates that the thermal neutron is more easily absorbed by the uranium ore environment under a high density and medium-low angle fracture environment. In the medium-high angle (60°–79°) fractured uranium ore environment, with the increase of fracture number density, the sensitivity of thermal neutron density distribution increases, similar to the trend of the first two angle environments; however, overall, the sensitivity is significantly improved compared with the first two fracture environments. In the low-number density environment (0.1–0.2), the distribution of thermal neutron density changed from the original flat and smooth state to a rough and uneven state, with a peak rising trend. As the fracture number density increased (0.3–0.4), the difference degree of peak value became increasingly obvious. The number increases with the increase in the peak value. Thus, overall, the sensitivity of thermal neutron distribution in this fracture angle environment improved the most. As Fig. 5 shows, in the uranium ore environment with approximate vertical (80°–90°) fracture, the distribution difference of thermal neutron density increases gradually with the increase of fracture number density; however, compared with the low-angle fracture environment, the sensitivity of thermal neutron density distribution increases more, especially in the low-number density (0.2) fracture environment. To compare the thermal neutron density distribution in the fractured uranium ore environment from different angles more intuitively, we used |$Sen$| variable to measure the sensitivity of thermal neutron density distribution to different environments (Zhang et al. 2018): $$\begin{eqnarray} Sen = \frac{{{{\left( {a + b} \right)}_i}}}{{\sum\limits_1^n {{{\left( {a + b} \right)}_i}} }} \cdot 100\,{\rm per\,cent} + \frac{{{A_i}}}{{\sum\limits_1^n {{A_i}} }} \cdot 100\,{\rm per\,cent} {\rm{, 1}} \le i \le n, \end{eqnarray}$$(14) $$\begin{eqnarray} {A_i} = {A_{i1}} + {A_{i2}} + \cdots + {A_{ia}},{\rm{ 1}} \le a, \end{eqnarray}$$(15) where a is peak number of the thermal neutron density distribution in each set of data (each fracture angle and each fracture number density corresponds to a set of data); b is the minimum number of the thermal neutron density distribution in each set of data; |${A_{ia}}$| is the single peak value; |${A_i}$| is the sum of all peak values in each set of data and n is the number of data groups. The sensitivity can be used to more comprehensively and prominently measure the thermal neutron distribution state under certain fracture conditions. In practical applications, it can reflect the characteristic state of thermal neutrons under certain conditions, thus providing distinct discrimination information for the occurrence state of the fracture uranium ore with a specific angle and number density. As seen in Fig. 6, the sensitivity of thermal neutron distribution tends to increase with the increase of the fracture angle. However, in low angle environment, the sensitivity is generally low, especially in low number density, where the sensitivity is close to 0, and the growth trend is slow. When the fracture angle increased to a medium-high angle, the sensitivity increased rapidly. In the environment of 0.2 number density, the sensitivity increased the most, increasing by approximately eight times. Figure 6. Open in new tabDownload slide Thermal neutron density distribution sensitivity for different fracture angles. Figure 6. Open in new tabDownload slide Thermal neutron density distribution sensitivity for different fracture angles. Overall, the fracture density has the greatest influence on the sensitivity of thermal neutron distribution. The higher the number density, the higher the sensitivity, which is positively correlated with the inhomogeneity of uranium ore distribution in the strata. At the same time, the higher the fracture angle is, the greater the sensitivity will be, especially in the medium-high angle environment, where the sensitivity will produce a substantial increase, which indicates that thermal neutrons are more sensitive to the high-angle fracture type uranium ore environment. Fig. 7 shows the thermal neutron energy spectrum, that is after the neutron source sends neutrons to the uranium ore formation for a period of time, the neutrons react with the nuclei of different substances in the formation, and the neutron energy produces different attenuation. Finally, the spectrum of the number of neutrons in different energy ranges is recorded. Fig. 7 shows the neutron energy spectrum under the condition that the density of fracture number varies from 0.1 to 0.4 and the angle varies from 0° to 90°. Figure 7. Open in new tabDownload slide Neutron energy spectrum of uranium ore at different fracture angles and number densities. (a) Approximate horizontal angle; (b) medium-low angles; (c) medium-high angles and (d) approximate vertical angle. Figure 7. Open in new tabDownload slide Neutron energy spectrum of uranium ore at different fracture angles and number densities. (a) Approximate horizontal angle; (b) medium-low angles; (c) medium-high angles and (d) approximate vertical angle. Fig. 7 illustrates that, with the increase of the angle, the number of energy spectrometers in the environment with the fracture number density of 0.1 decreases slowly, with a small decrease rate (12.68 per cent). In the energy range, the energy region with the largest count is |$0 - 0.04 \times {10^{ - 3}}\,{\rm{MeV}}$|⁠, after which the count drops sharply. Only at approximately vertical angles did the count show a slight tendency to diffuse to higher energy regions. Under the condition of fracture number density 0.2, the counting peak decreases slowly in the approximate horizontal and low-angle fracture environment, and the distribution and variation rules are similar to the numerical density 0.1 environment. Under the condition of changing from medium to high angle to approximately vertical angle, the energy peak decreases significantly (35.06 per cent), and the counting energy zone is dispersed, indicating that the fracture uranium environment at a higher angle has a large absorption of neutron. Under the condition of 0.3 fracture number density, the overall count is less than 0.1–0.2 and decreases slightly as the fracture angle increases. In the case of fracture number density 0.4, in the medium-low angled fractured uranium deposits, the energy count is not in the interval of |$0 - 0.04 \times {10^{ - 3}}{\rm{MeV}}$|⁠; however, it moves to the high-energy region, and the energy peak is scattered in the interval. Especially in the 40°–80° fracture environment, the maximum concentration of energy counts is in the |$0.6 \times {10^{ - 3}}\,{\rm{MeV}}$| energy region. Thus, in general, under different number density conditions, the neutron energy counting peak positions in uranium strata are different. With the increase of number density, the neutron energy counting peak tends to move to the high-energy region. As the number density increases, the count decreases as can be seen in Fig. 8. However, in the number density range of 0.1–0.4, the count decreases with the increase of the angle; while at 70°, the count peak spectra have an inflection point. The energy count peak attenuation is larger in the fracture environment with a high number density (0.4) at high angles (70°–90°). Figure 8. Open in new tabDownload slide Energy count peak spectra at different fracture angles and number densities. Figure 8. Open in new tabDownload slide Energy count peak spectra at different fracture angles and number densities. The characteristic law of neutron count change can be obtained through time spectrum. In different stratigraphic environments, time spectrum will show different characteristics. Fig. 9 shows that in terms of time spectrum, neutron count is almost independent of angle with the increase of time in uranium ore environment with number density 0.1. However, as the density of fracture number increases, the influence of fracture angle on the peak value of time spectrum count increases gradually (in the case of 0.2 number density, the count decreases by 27.27 per cent with the increase of the angle; When the number density is 0.4, the count decreases by 52.94 per cent with the increase of the angle). That is, as the angle increases, the count decreases more as the number density increases. Even under the condition of 0.2–0.4 number density, the dispersion peak in high-angle environment appears. However, in the near-vertical environment, the total count decreases and the dispersion peak is not obvious because the neutron is absorbed by the nucleus of the formation to a large extent. Figure 9. Open in new tabDownload slide Thermal neutron counting time spectrum at different fracture angles and number densities. (a) Approximate horizontal angle; (b) medium-low angles; (c) medium-high angles and (d) approximate vertical angle. Figure 9. Open in new tabDownload slide Thermal neutron counting time spectrum at different fracture angles and number densities. (a) Approximate horizontal angle; (b) medium-low angles; (c) medium-high angles and (d) approximate vertical angle. 4 VERIFICATION OF MODEL VALIDITY In order to verify the validity of this fracture model, we collected actual neutron log data to study fractured uranium deposits and compared the field data with the simulated data. The uranium ore area used for field data collection in this paper was rich in fracture structures which are mainly in the NW and NNE directions according to the distribution direction. The uranium ore bodies in this area are mainly distributed in a fissure shape with a single ore body generally extending tens to hundreds of meters, and a maximum strike length of up to 2 km. Among them, the NNE group is large in scale and extends far. It runs longitudinally through the west of the mining area extending up to 2 km, and an average of 2–8 m wide. The NNE group is the main ore-controlling structure in the area with a strike NNE 8°–32°, dip NWW and dip angle of 40°–60°. In the footplate of the above faulted structures there exists a series of small fissure groups with a NW strike approximately 320°, incline to the NE, and the dip angle is generally 60°–80°. The degree of fracture development is directly related to the wealth of uranium mineralization. The fracture density is over 12, the best mineralization, and the dense fracture zone is the main ore-bearing structure. The petrochemicals in the mining area are rich in silicon and gangue minerals mainly include quartz, chalcedony, hydromica, carbonate and fluorite. The main ore bodies are buried shallowly and the majority of the buried depth is 0–150 m. A small number of ore bodies are buried deep at 420 m. Under these conditions, we reviewed the thermal neutron density distribution at a certain time point after the pulse for the uranium deposit. Then, we compared the results with the simulation results to obtain Figs 10 and 11. Figure 10. Open in new tabDownload slide Thermal neutron density distribution for actual log data: (a) medium-low angle and (b) medium-high angle. Figure 10. Open in new tabDownload slide Thermal neutron density distribution for actual log data: (a) medium-low angle and (b) medium-high angle. Figure 11. Open in new tabDownload slide Thermal neutron density distribution for simulated data: (a) medium-low angle and (b) medium-high angle. Figure 11. Open in new tabDownload slide Thermal neutron density distribution for simulated data: (a) medium-low angle and (b) medium-high angle. In actual conditions, there are many interferences and limiting factors, so these need to be considered in the comparison results. First, It is impossible for the fracture angles to be strictly consistent under natural conditions, which can only ensure that most of the fracture groups participating in the comparison have the same fracture angles. Secondly, The presence of fluid in the well, well pore structure itself, large pore structure in the surrounding rock medium, and unsaturated fluid in the fracture and pore structure will have certain interferences on detection results (such as the capture degree of thermal neutron), thereby resulting in a decrease in the counting rate. However, we can make up for this by increasing the source count. Thirdly, the density of the fracture number in the actual uranium ore environment is higher than that in the digital simulation environment, and the distribution is uneven. Therefore, compared with the simulation results in the ideal model condition, the fast neutron deceleration time and capture time are shorter and the transport distance is shorter in actual conditions. The results showed a concentrated and uneven distribution of thermal neutrons detected near the well hole, with almost no counting in the far area away from the well hole (i.e. the lower area in the figure shows large areas of dark blue, with no peaks and no counting). However, from the viewpoint of the statistical law, basic conformance to the requirements of the experimental comparison of the selected stratum environment, consistent with the main characteristics of the numerical model (the fracture structure exists in a large number of strata in groups, the fracture angle is basically the same, the composition of surrounding rock media is similar to the numerical model, and the influence of the pore structure is small), can be used for validation purposes. Figs 10 and 11 show the thermal neutron density distributions for different fracture uranium deposit environments. Fig. 10 shows the thermal neutron density distribution for the fractured uranium deposit based on the actual log data. Fig. 11 was created using the simulated data. By comparing Figs 10 and 11, it can be seen that the differences between the count peaks for the uranium deposit in the natural environment fracture group are pronounced. The count peak value is lower in the medium-low angle uranium deposit environment than in the medium-high angle environment. According to the plan density distribution, the thermal neutron diffusion area in the medium-high angle uranium deposit is larger, which indicates that the thermal neutron absorption rate of the medium-high angle uranium deposit is lower than that of the medium-low angle uranium deposit environment. Since the actual log data results agree well with the simulated result, the simulation is significant. 5 CONCLUSIONS Based on the field uranium mining environment, the relationship between fracture parameters (angle and number density) and logging responses of uranium ore is obtained by pulsed neutron logging method. The following findings were made. The sensitivity of thermal neutron distribution is positively correlated with fracture angle, especially in high-angle and high number density environments. Among them, number density has the most significant effect on sensitivity. In other words, high-angle and high-porosity fracture type uranium strata have the greatest influence on thermal neutrons. Under the condition of high porosity, the neutron energy spectrum displays a dispersion peak and moves towards a high energy region. Moreover, the energy decay rate is the highest in the high-angle fracture environment. It is noteworthy that in a high number density environment, medium-low angle uranium ore has the strongest thermal neutron absorption in a low energy range. In terms of the neutron time count spectrum, the count decay is greatest in the high- angle fracture environment. Moreover, the higher the density of the fracture number, the more susceptible the counting is to the influence of angle factor. The calculated results can be used to analyse the complex fracture structure in the field uranium exploration, which in turn can help judge and predict the occurrence state of fractured uranium deposits. The findings of this study can be used to identify fractured uranium deposits in different occurrence states. This is rare in the field of neutron logging. However, as our findings are based on the theory of numerical model, this implies certain restrictions, for example low saturation fluid or high rock porosity will affect the results of the analysis and lead to an increase in error. This restriction needs to be addressed in subsequent studies. In terms of future research, this study will contribute to the analysis of delayed neutron logging method and uranium ore quantification, and deepen the study of the influence law of a well diameter. ACKNOWLEDGEMENTS The authors are grateful to the anonymous reviewers for their suggestions and valuable comments. The study was supported by the National Natural Science Foundation of China (41761090, 11465002) and Natural Science Foundation of Jiangxi Province (20181BAB213018, 20171ACB2021). The data used to support the findings of the article titled “Simulation of multi-angle fractured uranium deposits by neutron logging” are available from the corresponding author upon request. REFERENCES Abdel-Fattah Mohamed I. , 2015 . Impact of depositional environment on petrophysical reservoir characteristics in Obaiyed field, Western Desert, Egypt , Arab. J. Geosci. , 8 ( 11 ), 9301 – 9314 . 10.1007/s12517-015-1913-5 Google Scholar OpenURL Placeholder Text WorldCat Crossref Ajayi O. , Torres-Verdín C, Preeg W.E., 2015a . Fast numerical simulation of logging-while-drilling gamma-ray spectroscopy measurements , Geophysics , 80 ( 5 ), D501 – D523 . 10.1190/geo2014-0584.1 Google Scholar OpenURL Placeholder Text WorldCat Crossref Ajayi O. , Torres-Verdín C, Preeg W.E., 2015b . Inversion-based interpretation of logging-while-drilling gamma-ray spectroscopy measurements , Geophysics , 81 ( 1 ), D9 – D34 . 10.1190/geo2015-0293.1 Google Scholar OpenURL Placeholder Text WorldCat Crossref Animesh M. , Mohanty W.K., Sharma S.P., Arkoprovo B., Joydip S., Bhatt A.K., 2015 . Geophysical signatures of uranium mineralization and its subsurface validation at Beldih, Purulia District, West Bengal, India: a case study , Geophys. Prospect , 63 ( 3 ), 713 – 726 . Google Scholar OpenURL Placeholder Text WorldCat Asfahani J. , 2017 . Porosity and hydraulic conductivity estimation of the basaltic aquifer in southern syria by using nuclear and electrical well logging techniques , Acta. Geophys. , 65 ( 4 ), 1 – 11 . 10.1007/s11600-017-0056-3 Google Scholar OpenURL Placeholder Text WorldCat Crossref Bilal E. , Petrescu L., 2017 . Speciation of natural uranium and thorium in surface waters around a uranium mine (Bistrita Mts. Romania) , Geochim. cosmochim. Acta. , 70 ( 18 ), A486 – A486 . Google Scholar OpenURL Placeholder Text WorldCat Bogdanovich B.Y. , Kozlovskii K.I., Nesterovich A.V., Shikanov A.E., II'Inskii A.V., Urmanov E.G., 2017 . Possibilities of increasing logging efficiency with simultaneous analysis of neutron and gamma fields formed in oil wells , At. Energ ., 122 ( 4 ), 1 – 7 . 10.1007/s10512-017-0266-4 Google Scholar OpenURL Placeholder Text WorldCat Crossref Chen S. , Hattori K., Grunsky E.C., 2018 . Identification of sand stones above blind uranium deposits using multivariate statistical assessment of compositional data, Athabasca Basin, Canada , J. Geochem. Explor. ., 188 , 229 – 239 . 10.1016/j.gexplo.2018.01.026 Google Scholar OpenURL Placeholder Text WorldCat Crossref Dance T. , Paterson L., 2016 . Observations of carbon dioxide saturation distribution and residual trapping using core analysis and repeat pulsed-neutron logging at the CO2CRC Otway site , Int. J. Greenhouse Gas Control ., 47 , 210 – 220 . 10.1016/j.ijggc.2016.01.042 Google Scholar OpenURL Placeholder Text WorldCat Crossref Hajnal Z. , Takacs E., Pandit B., Annesley I.R., 2015 . Uranium mineralization indicators from seismic and well log data in the shea creek area at the southern margin of the carswell impact structure, Athabasca Basin, Canada , Geophys. Prospect. , 63 ( 4 ), 861 – 880 . 10.1111/1365-2478.12274 Google Scholar OpenURL Placeholder Text WorldCat Crossref Hudson J.A. , 1980 . Overall properties of a cracked solid , Math. Proc. Cambridge Philos. Soc. , 88 ( 2 ), 371 – 384 . 10.1017/S0305004100057674 Google Scholar OpenURL Placeholder Text WorldCat Crossref Hudson J.A. , 1991 . Overall properties of heterogeneous material , Geophys. J. Int. , 107 ( 3 ), 505 – 511 . 10.1111/j.1365-246X.1991.tb01411.x Google Scholar OpenURL Placeholder Text WorldCat Crossref Hudson J.A. , Liu E., Crampin S., 1996 . The mechanical properties of materials with interconnected cracks and pores , Geophys. J. Int. , 124 ( 1 ), 105 – 112 . 10.1111/j.1365-246X.1996.tb06355.x Google Scholar OpenURL Placeholder Text WorldCat Crossref Ortega E. , Torres-Verdín C, Preeg W.E., 2015 . 2D Inversion-based interpretation of logging-while-drilling thermal-neutron and gamma-ray time decays , Geophysics , 80 ( 3 ), D247 – D263 . 10.1190/geo2014-0336.1 Google Scholar OpenURL Placeholder Text WorldCat Crossref Ortega E. , Luycx M., Torres-Verdín C, Preeg W.E., 2017 . Joint petrophysical interpretation of multidetector nuclear measurements in the presence of invasion, Shoulder-Bed, and Well-deviation effects , Geophysics , 82 ( 1 ), D13 – D30 . 10.1190/geo2016-0037.1 Google Scholar OpenURL Placeholder Text WorldCat Crossref Tian L. , Zhang F., Wang X., Ti Y., Su B., Lu M., Chen R., 2018 . Logging method for formation water salinity determination by PGNAA technology , J. Radioanal. Nucl. Chem. , 318 ( 2 ), 1143 – 1149 . 10.1007/s10967-018-6215-4 Google Scholar OpenURL Placeholder Text WorldCat Crossref Tün M. , Pekkan E., Özel O., Guney Y., 2016 . An investigation into the bedrock depth in the Eskisehir Quaternary Basin (Turkey) using the microtremor method , Geophys. J. Int. , 207 ( 1 ), 589 – 607 . 10.1093/gji/ggw294 Google Scholar OpenURL Placeholder Text WorldCat Crossref Yang Z.C. , Jiang T., Xu X.X., Lin J., Jia H.Q., Ma R.Q., Chen Z.Z., 2016 . Research on correlation detection reference signal based on the reconstructed excitation signal for electromagnetic seismic vibrator , Chin. J. Geophys. , 59 ( 01 ), 318 – 329 . Google Scholar OpenURL Placeholder Text WorldCat Zhang X.A. , Wang Z.W., 2015 . Distinguishing oil and water layers by interpreting acoustic logging data with changing well diameters , Geophys. Prospect. , 63 ( 3 ), 669 – 679 . 10.1111/1365-2478.12220 Google Scholar OpenURL Placeholder Text WorldCat Crossref Zhang X.A. , Wang Z.W., Yang Z.C., 2017 . Distinguishing oil and water layers in a porous cracked medium by interpreting acoustic logging data on the basis of Hudson theory , J. Earth Sci. , 28 ( 3 ), 500 – 506 . 10.1007/s12583-017-0620-4 Google Scholar OpenURL Placeholder Text WorldCat Crossref Zhang X.A. , Yang Z.C., Tang B., Wang R.B., Wei X., 2018 . Distinguishing oil and water layers in a cracked porous medium using pulsed neutron logging data based on Hudson's crack theory , Geophys. J. Int. , 213 , 1345 – 1359 . 10.1093/gji/ggy065 Google Scholar OpenURL Placeholder Text WorldCat Crossref © The Author(s) 2020. Published by Oxford University Press on behalf of The Royal Astronomical Society. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Simulation of multi-angle fractured uranium deposits by neutron logging
JF - Geophysical Journal International
DO - 10.1093/gji/ggaa436
DA - 2020-10-15
UR - https://www.deepdyve.com/lp/oxford-university-press/simulation-of-multi-angle-fractured-uranium-deposits-by-neutron-c6dC6EO4OS
SP - 2027
EP - 2037
VL - 223
IS - 3
DP - DeepDyve
ER -