未知化学**弹药的定性识别在犀护社会安全方面是十分重要的, 可指导化学**的分类处理。 瞬发伽马射线中子活化分析(PGNAA)技术利用分析活化产生的伽马射线能谱可以实现对物质中元素的无损, 快速检测, 在化学**识别中具有独特的优势。 因此, 本研究基于PGNAA技术进行了化学**弹药类型识别装置的设计, 同时使用逻辑树判别方法对化学**样品进行定性分析。 首先, 基于高纯锗(HPGe)探测器与Cf-252中子源, 使用蒙特卡罗MCNP程序对装置结构进行设计优化, 主要包括中子源容器尺寸、 伽马屏蔽体厚度以及探测器相对位置等。 为了最大化样品活化产生的特征伽马射线, 需要提高样品位置处的热中子通量, 采用聚乙烯作为慢化体, 模拟结果显示聚乙烯厚度达到6 cm, 宽度达到12 cm时, 样品中热中子通量达到较高水平。 为了降低周围材料活化噪声的干扰, 选择铅作为屏蔽结构, 模拟显示铅屏蔽厚度达到5 cm时, 可满足屏蔽要求。 同时, 探测器与样品之间的距离也会影响对伽马射线的探测, 最终模拟确定探测器与样品之间的距离为28 cm时, 特征信号计数最高。 根据优化结果搭建测量装置, 使用分析纯试剂根据真实化学**元素含量配制化学**模拟样品, 通过对5种化学**模拟样品的测量获得伽马能谱。 对能谱中的特征峰处理过程中, 基于特征峰对元素进行分析, 针对计数统计性较好的元素(如H, Cl, S)的特征峰, 使用高斯及多项式拟合的方式对特征峰处的高能量康普顿平台进行扣除, 获得特征伽马射线的全能峰信息。 而对统计性较差的元素特征峰(如N元素的10.829 MeV), 采用能量区间加和法, 对该能量下的全能峰至单逃逸峰之间的计数求和, 进而可确定该元素在样品中的存在情况, 最后利用建立的逻辑树判别方法根据元素存在信息对样品类型进行判别。 实验结果表明, 利用该优化的装置可以获得5种模拟样品的能谱, 结合能谱分析方法可以得到化学**模拟样品中的H, Cl, S和N等元素的存在信息, 最后使用逻辑树判别方法可以对化学**样品种类进行判别。

Identifying the unknown chemical weapons is an important work for maintaining social security, and it can guide the chemical weapons destruction. Prompt gamma-ray neutron activation analysis (PGNAA) technology has the advantages of being non-destructive and rapid. In this research, the device was designed for identifying the chemical weapons based on PGNAA technology, and logic-tree-based discrimination method was employed to conduct qualitative analysis on samples.Firstly, with the high purity germanium (HPGe) detector and Cf-252 neutron source as the core instruments, the structures of device were optimized using the Monte Carlo MCNP code, including neutron source moderator, the thickness of the shielding body, and the relative position of the detector. To maximize the characteristic gamma ray generated by sample activation, it is necessary to increase the thermal neutron flux in the sample. In this research, polyethylene is used as moderator to increase the scattering of neutrons before the sample, so that more neutrons are thermalized. The simulation results showed that the thermal neutron flux in the sample reaches a high level when the polyethylene has a thickness of 6 cm and a width of 12 cm. In order to reduce the interference of the surrounding material activation noise, lead is selected as the shielding structure. And the simulation showed that it can meet the shielding requirement when the lead shielding thickness reaches 5 cm. At the same time, the distance between the detector and the sample also affects the detection of gamma rays. The final simulation determines that the distance between the detector and the sample is 28 cm, and the characteristic signal count is the highest.According to the optimization results, the experimental device was set up. And the simulated samples of chemical weapon were prepared according to the actual elements by using the analytical pure reagent, and the gamma spectrum was obtained by measuring the five samples. In the process of analyzing the characteristic peaks in the spectrum, the elements were analyzed based on the characteristic peaks of the elements. The elements with well statistic were analyzed by using gauss and polynomial fitting (such as H, Cl, S). The high-energy Compton platform at the characteristic peak was deducted to obtain the full peak information. For the element characteristic peak with poor statistic (such as 10.829 MeV of the N element), the energy interval summation method was used to sum the counts between the full peaks and the single escape energy peaks, so we can get the information of the elements in the sample. Finally, the logic-tree-based discrimination method was used for sample identification. The analysis results showed that the information of H, Cl, S, N and other elements in the simulated samples of chemical weapons can be obtained by using the spectrum fitting method. And the type of simulated samples of chemical weapon can be identified by combining logic-tree-based discrimination method.