After the determination of activity A and number of atoms N of long-lived radionuclides, the half-lives of radionuclides can be calculated according to the relation between radioactivity of nuclides and time A=-λN, where λ=ln 2/T_1/2. As the number of atoms is usually relatively small for long-lived radionuclides, they cannot be measured with ordinary MS because of its strong background interference. AMS has molecular and isobaric background eliminating capabilities, thus it can be employed for the measurements of half-lives of long-lived radionuclides with relatively small number of decay atoms.

Table 1 AMS measurements of major nucleus

Taking ⁷⁹Se for an example, as a fission product of ²³⁵U, one can find that its half-life is very long(10⁵ to 10⁶ a), which is a major issue in the field of landfill waste treatment. Since 1948, the half-life of ⁷⁹Se has been measured repeatedly, and the difference among the measured values are evident(table 2). In 2002, Jiang Songsheng et al^[21] measured the half-life of ⁷⁹Se to be(2.95±0.38)×10⁵ a based on CIAE-AMS system using radiochemical separation+liquid scintillation+AMS method. The data were adopted by the U.S. National Nuclear Data Center and the IAEA Nuclear Data Center, as the reference data for half-life of ⁷⁹Se. Currently, we are improving the AMS measurement method and using absolute measurement method to measure the number of ⁷⁹Se atoms in the test sample, thereby accurately measuring the half-life of ⁷⁹Se. After a series of improvements, we obtained that the half-life of ⁷⁹Se was(2.79±0.16)×10⁵ a, within the error being range of the results obtained in 2002, and with lower relative uncertainty from 12.9% to 5.7%. Thus it is a big step forward in the measurements of half-life of ⁷⁹Se.

Table 2 Measurements of half-life of 79Se[22]

Measurement of small nuclear reaction cross section is an important research direction of experimental nuclear physics. It can reveal the interaction mechanism between incident particles and target nuclei, which is conducive to deepen the understanding of nuclear force and structure, and is the fundamental basis for testing nuclear theory. In addition, nuclear cross section data are also the basis for the application of nuclear science and technology, which are of importance especially to the building and improvement of theoretical models of nuclear reactions, design of fusion reactor, building of nuclear data libraries, as well as nuclear astrophysics.

Due to the extremely high sensitivity(minimum detection limit of 10⁴ atoms)of AMS in measurements of the nuclear reaction products, i.e. nuclides, AMS can achieve the measurements of some small nuclear reaction cross sections. The AMS measurements of nuclear reaction cross sections are based on the formula σ=n/(IN), in which n is the number of atoms produced by reaction and measured by AMS, I is the intensity of incident beam, and N is the number of target nuclei.

The cross sections of ²³⁸U(n,3n)²³⁶U nuclear fission reaction induced by 14 MeV neutrons are important nuclear data. If the emitted neutrons are measured online using direct measurement method, the accuracy will not be high due to the influences from multiple aspects such as prompt and scattered neutrons during the fission process. If the neutrons are measured by neutron activation, the long half-life and low generation level of residual nucleus ²³⁶U(T_1/2=2.34×10⁷ a)will make it very difficult to be measured. If the residual nucleus ²³⁶U is measured by ordinary MS, the ordinary MS will be subject to molecular background interferences(such as ²³⁵UH⁺), and its maximum measurement sensitivity of ²³⁶U is ²³⁶U/²³⁸U≈1×10^-6. In comparison, AMS lifts the restriction of isobaric and molecular background interferences existing in ordinary MS, which is a good option for analyzing the trace of ²³⁶U(²³⁶U/²³⁸U～10^-8≈10^-14). In the actual test, to measure the cross section of 14 MeV neutron induced reaction ²³⁸U(n,3n)²³⁶U, we only need to measure ²³⁶U/²³⁸U, i.e. the ratio of atom numbers between daughter nucleus ²³⁶U and mother nucleus ²³⁸U, as well as neutron fluence.

CIAE-AMS group was the first one in the world to establish the method for AMS measurements of 14 MeV neutron ²³⁸U(n,3n)²³⁶U reaction cross section^[50], demonstrating that the AMS method is the best option for measurements of long-lived nuclide generating fission nuclide reaction cross sections. The method has laid the foundation for successful measurements of significant reaction cross sections of other fission nuclides(Pu,Am). The study has initially given the reaction cross section data of ²³⁸U(n,3n)²³⁶U. The reaction cross sections with neutron energies(14.65±0.40)MeV(0° irradiated samples)and(14.18±0.30)MeV(85° irradiated samples)are(556.7±43.4)mb and(489.3±54.3)mb, respectively.

Some of the nuclear reaction cross sections by AMS method currently in the world are listed in table 3.

Table 3 Measurements of cross section by AMS

Since the 1990s, the experimental measurements of peak-to-valley ratio of long-lived fission products have been promoted because of the proposals of the clean nuclear energy system separation, the transmutation plan and the deepening of research on potential long-term risks in deep geological disposal of high level radioactive waste. As the fission products of uranium and other fissile nuclei, ^121mSn and ¹²⁶Sn are two long-lived radionuclides whose peak-to-valley ratio determination is relatively sensitive. Their half-lives are(43.9±5.0)a and(2.30±0.14)×10⁵ a(National Nuclear Data Center), respectively. With the proceeding of human activities, large amount of artificial ^121mSn and ¹²⁶Sn are introduced into the environment, which results in significantly higher ^121mSn and ¹²⁶Sn contents in some areas. The nuclides ^121mSn and ¹²⁶Sn have become important indicator nuclides for long-term contamination in nuclear engineering environment.

The establishment of highly sensitive measurement method for fission product nuclides ^121mSn and ¹²⁶Sn can lay the foundation for the application of peak-to-valley ratio as well as the research of environmental safety. However, due to long half-life, weak radioactivity and very low content of ^121mSn and ¹²⁶Sn in samples, measurements can hardly be achieved by conventional decay counting because of too large demand for samples. Nor can the ordinary MS achieve high sensitivity measurements due to the limits in measurement sensitivity.

CIAE-AMS group has successfully established a highly sensitive method for AMS measurements of ^121mSn and ¹²⁶Sn based on HI-13 tandem AMS^[51], with the measurement sensitivity better than(4.2±0.1)×10^-10 and(5.2±2.3)×10^-12, respectively.

Both ³H and ¹²⁹I are important nuclides in the field of nuclear energy and nuclear safety application. With the continuous application and development of fission reactors, fusion reactors and nuclear materials, tritium has been gaining increasing attention as one of the important nuclides. Monitoring of tritium content in surrounding air, primary loop water(heavy water)and structural components(zirconium alloy)is an important role. The sensitivities of common tritium monitoring techniques are around 0.1 Bq at present, which limits the progress of research work. AMS measurement technique can improve tritium measurement sensitivity at least by one order of magnitude^[52], which has great development potential in the field of tritium application.

¹²⁹I has a half-life of 1.6×10⁷ a, and a fission yield of 0.8%, which is a long-lived fission nuclide produced by spontaneous fission of ²³⁸U or neutron-induced fission of ²³⁵U. No matter whether it is during reactor leak or spent fuel reprocessing, ¹²⁹I will inevitably release into the environment and accumulate in the environment to cause long-term contamination. At present, AMS is the most powerful means for measuring the trace of ¹²⁹I in the environment, with isotope abundance ratio being reaching 10^-13 to 10^-15. AMS measurement of ¹²⁹I in environmental samples has become an important means of nuclear inspection internationally.

CIAE-AMS group measured ¹²⁹I content in the aerosol samples in Beijing during the Fukushima Nuclear Power Plant(FNPP)accident, and compared it with the ¹³¹I content in the aerosol samples of the same period(Fig.2)^[53]. The results showed that around 2011- 03-26, ¹²⁹I concentration in the air started to be higher than the background level of March 20, indicating that the ¹²⁹I produced in the accident had arrived in Beijing. In comparison, ¹³¹I content in the same sample did not change significantly. Due to the cumulative effect of ¹²⁹I, to 2011- 03-28, ¹²⁹I concentration in the air had been significantly elevated, which was 3 to 4 times higher than the background level, while the ¹³¹I content in the same sample still had no significant change. Therefore, in the early stage of nuclear leak accident, conventional radioactivity measurement approaches for short-lived nuclides often have the shortcomings of low sensitivity and long measurement time. The measurement of ¹²⁹I by AMS has great advantages in terms of nuclear accident warning and detection, because during the early stage of nuclear accident, the amount of ¹²⁹I released is 1 to 2 orders of magnitude higher than that of ¹³¹I. And the data analysis results can be more reasonable and credible because much ¹²⁹I can be obtained in a short time. On-machine measurement time of ¹²⁹I is only 2 to 3 h by AMS, while the on-machine measurement of conventional γ spectrometry takes 20 h or longer. ¹²⁹I samples require relatively short preprocessing time, although ¹³¹I samples do not require preprocessing, too much time needed for sample collection. The total demand for ¹²⁹I samples is relatively simplenss. All of these provide more favorable evidences for the monitoring of early stage nuclear leak.

Fig.2 (Color online)129I and 131I contents in the aerosol samples in Beijing during FNPP accident in 2011[53]