The measurements of neutron capture cross sections of neutron-rich nuclei are challenging but essential for understanding nucleosynthesis and stellar evolution processes in the explosive burning scenario. In the quest of r-process abundances, according to the neutrino-driven-wind model, light neutron-rich unstable nuclei may play a significant role as seed nuclei that influence the abundance pattern. Hence, experimental data for neutron capture cross sections of neutron-rich nuclei are needed. Coulomb dissociation of radioactive ion beams at intermediate energy is a powerful indirect method for inferring capture cross section. As a test case for validation of the indirect method, the neutron capture cross section (n, γ ) for 14 C was inferred from the Coulomb dissociation of 15 C at intermediate energy (600A MeV). A comparison between different theoretical approaches and experimental results for the reaction is discussed. We report for the first time experimental reaction cross sections of 28 Na(n, γ ) 29 Na, 29 Na(n, γ ) 30 Na, 32 Mg(n, γ ) 33 Mg, and 34 Al(n, γ ) 35 Al. The reaction cross sections were inferred indirectly through Coulomb dissociation of 29,30 Na, 33 Mg, and 35 Al at incident projectile energies around 400-430 A MeV using the FRS-LAND setup at GSI, Darmstadt. The neutron capture cross sections were obtained from the photoabsorption cross sections with the aid of the detailed balance theorem. The reaction rates for the neutron-rich Na, Mg, Al nuclei at typical r-process temperatures were obtained from the measured (n, γ ) capture cross sections. The measured neutron capture reaction rates of the neutron-rich nuclei, 28 Na, 29 Na, and 34 Al are significantly lower than those predicted by the Hauser-Feshbach decay model. A similar trend was observed earlier for 17 C and 19 N but in the case of 14 C(n, γ ) 15 C the trend is opposite. The situation is more complicated when the ground state has a multi-particle-hole configuration. For 32 Mg, the measured cross section is about 40-90 % higher than the Hauser-Feshbach prediction.