Keywords: 
• 热中子探测 /
• 微通道板 /
• 蒙特-卡罗模拟 /
• 优化计算

Abstract: The traditional digital imaging of neutron radiography is based on neutron scintillation screen cooperated with charge coupled device (CCD) camera, whose spatial resolution and neutron detection e?ciency are contradictory. Neutron detection method based on microchannel plates (MCP) could solve the problem appearing in traditional method. It could supply high spatial resolution, high neutron detection e?ciency and high time resolution. It is of benefit to high-resolution neutron radiography and neutron energy choice imaging. Tremsin et al. [Tremsin A S, Feller W B, Downing R G, Mildner D R 2004 U. S. Government Work not Protected by U. S. Copyright p340] calculated the detection e?ciency of thermal neutron sensitivity MCP in 2004. Then his team fabricated a prototype of neutron detection system based on MCP and carried out the neutron imaging experiments on several neutron sources. The experimental results show that spatial resolution is nearly 15 μm and neutron detection e?ciency for cold neutron is more than 70%. In China, Yang Y G et al. [15] from Tsinghua University developed a neutron detection system based on MCP, and preliminary neutron experimental results indicate that spatial resolution is about 200 μm. In order to find the optimal structure of MCP, in this paper we calculate the detection e?ciency of thermal neutron sensitive MCP doped (or coated) by boron and gadolinium with Monte-Carlo method. The neutron detection e?ciency P is determined by three terms P1, P2 and P3, which are related by P =P1 × P2 × P3. Here, P1 is the possibility that the neutrons are absorbed by MCP solid parts, P2 is the possibility that the secondary particle escapes into MCP channel and generates an electron avalanche, and P3 is the possibility that the electron avalanch is recorded by readout system. Theoretical analysis indicates that more solid parts of MCP can make P1 higher and increase the di?culty for secondary particle to escape, and make P2 lower. There may be an optimal geometry to make the total P maximal. This paper gives the calculation method of P1 and P2, and approximates P 3 to 1. The calculation results show that the neutron detection e?ciency depends on channel diameter (or coated thickness) and material, but not on the structure of MCP. When the thickness of MCP is 0.4 mm, the pixel of MCP is 15 μm, and the neutron sensitivity material is 10B2O3, the optimal thermal neutron detection e?ciency is more than 40% with a channel diameter of 8.0 μm for the doped MCP, and it is nearly 60% with a coated thickness of 1.5 μm for the coated MCP. With the same geometry parameters and the neutron sensitive material such as natural Gd2O3, the optimal thermal neutron detection e?ciency is more than 30%with a channel diameter of 9.0 μm for the doped MCP, and it is more than 50%with a coated thickness of 0.5 μm for the coated MCP.