Photon from neutron-proton bremsstrahlung in p+Pb reactions is examined as a potential probe of the neutron-skin thickness in different centralities and at different proton incident energies. It is shown that the best choice of reaction environment is about 140MeV for the incident proton and the 95%-100% centrality for the reaction system since the incident proton mainly interacts with neutrons inside the skin of the target and thus leads to different photon production to maximal extent. Moreover, considering two main uncertainties from both photon production probability and nucleon-nucleon cross section in the reaction, I propose to use the ratio of photon production from two reactions to measure the neutron-skin thickness because of its cancellation effects on these uncertainties simultaneously, but the preserved about 13%-15% sensitivities on the varied neutron-skin thickness from 0.1 to 0.3fm within the current experimental uncertainty range of the neutron-skin size in 208 Pb. The neutron-skin of nuclei is a fundamental physical quantity in nuclear physics, and has received considerable attention due to its importance in determining the structure of neutron-rich nuclei in nuclear physics and the property of neutron-rich matter in astrophysics. To determine the neutron-skin thickness of nuclei, one should known the proton density distribution and neutron density distribution, and then determines the corresponding neutron-skin thickness by calculating the rootmean-square (rms) radius difference between proton and neutron. Presently, the proton rms radius can be determined precisely, typically with an error of 0.02fm or better for many nuclei [1][2][3]; the neutron rms radius is much less well-known although many efforts have been devoted to probing the neutron density distribution by theoretical and experimental methods such as the nucleon elastic scattering [4][5][6][7], the inelastic excitation of the giant dipole and spin-dipole resonances [8,9], the pygmy dipole resonance [10,11] and experiments in exotic atoms [12][13][14][15][16][17]. This is because almost all of these probes are hadronic ones and need model assumptions to deal with the strong force introducing possible systematic uncertainties even if some of them reach small errors [18]. In this situation, the Parity Radius Experiment (PREX-I) at the Jefferson Laboratory (J-Lab) [19] has been performed to measure the neutron-skin thickness of 208 Pb using parity violating e-Pb scattering, the measured value of 0.33
+0.16−0.18 fm in 208 Pb obviously differs from previous value of 0.11 ± 0.06 fm of 208 Pb from π + -Pb scattering [20] albeit largely overlapping with each other within error bars. However, the obtained results from PREX-I experiment suffer from large uncertain- * Email address: wei.gaofeng@foxmail.com ties although the PREX-I experiment aims to a modelindependent measurement of the neutron-skin thickness of 208 Pb. It is interesting, however, to note that the neutron-skin for 208 Pb as thick as 0.33+0.16 fm reported by the PREX-I experimen...