2.1 Animals

Wistar rats of 2-months old were purchased from Guangxi Medical University, Nanning, China, and fed with standard diet. All animal experiments were performed according to the Guidelines for the Care and Use of Laboratory Animals ^[26]. The mice were divided into the experimental, control and background groups, with 3 mice in each group. The tracer isotope ²⁶Al is the laboratory standard kept from 1996 at the AMS lab of China Institute of Atomic Energy (CIAE) AMS. Decay counting method was used to calibrate the ²⁶Al/²⁷Al ratio. The ²⁶Al/²⁷Al ratio of AlCl₃ solution for injection was 1.5×10⁻⁹, with the ²⁶Al content of 2.5 μg/mL. The solution was prepared by successive dilution of the original standard sample. The three animal groups were treated as follows:

1) Experimental group: The mice are used to inject ²⁶Al for experimental tracing. 30 mice were divided into 10 subgroups (n =3) for 10 time scales. The mice were tail mainline injected with 1 mL of ²⁶AlCl₃ (pH=4.5), and were blood-sampled and euthanized for brain-sampling at 0.5, 2, 4, 8, 16, 48, 72, 96, 120 and 168 h after injection.

2) Control group (n =3): The mice were injected with 1 mL ²⁷AlCl₃ solution, for the behavior observation.

3) Background group (n =3): The mice were injected with 1mL normal saline solution for AMS measurement of the ²⁶Al as background sample.

2.2 Sample preparation

The beam current of ²⁶AlO⁻ is 20–40 times higher than the ²⁶Al⁻, but the number of ²⁶Mg isobar will become too large to identify ²⁶Al of ultra-low level. Besides, the particle transmission of single atom is higher than molecular ions. Therefore, a single atomic negative Al⁻ form is adopted in our AMS measurement. Taking advantage of the electric potential of Mg electron affinity is lower than zero, Mg^- ions cannot be extracted, but the Al^- can, from the AMS multi-cathode source. As the interfere isobar ²⁶Mg is no longer a problem, a simplified preparation process was developed (Fig. 1).

Fig. 1.Full-screen View

The sample preparation procedure.

1) The blood/brain samples were decomposed with 10 mL of concentrated nitric acid at 80°C for 24 h in a Teflon-sealed vessel

2) After cooling, 15 mg of ²⁷AlCl₃ were added in the sample solution as a carrier, and then heat at 140°C for 4 h in a Teflon-sealed vessel inserted in a stainless steel botter.

3) After cooling, the blood samples were centrifuged and Al-contained solution samples will be obtained.

4) Add NaOH solution to the Al-containing sample solution, and carefully adjust the solution to PH7–8. Al(OH)₃ sediment can be formed at pH=7–8, while a very few of Mg(OH)₂ can be formed at this pH value. Therefore, the most of ²⁶Mg, isobar of our interesting isotope ²⁶Al, could be eliminated in this step.

5) Al(OH)₃ sediment was obtained via centrifugation, and then converted to Al₂O₃ at 1000°C for 2h.

In the AMS measurement, each sample was mixed with an equal mass of pure silver powder for improving thermal and electric conductivity, and pressed into a sample holder of a 40-sample NEC MC-SNICS ion source.

2.3 Analysis of ²⁶Al by accelerator mass spectrometry

The number of ²⁶Al was measured with the AMS at CIAE (Fig.2). Since its installation in 1989, ²⁶Al, ³⁶Cl, ⁴¹Ca, ⁵⁵Fe, ⁶⁴Gu, ⁷⁹Se, ⁹⁹Tc, ¹²⁹I,¹⁵¹Sm, ¹⁸²Hf, ²³⁶U and ²³⁷Np have been measured for applications in biomedicine, nuclear physics, nuclear astrophysics and environmental science ^[11,27-36]. Fig. 2 shows schematically the AMS system based on an HI-13 Tandem accelerator. This work was carried out on AMS Beamline 1. The AMS Beamline 2 was for AMS research of medium-heavy mass nuclides..

Fig. 2.Full-screen View

Schematic diagram of the accelerator mass spectrometer based on the HI-13 tandem accelerator at CIAE.

Single atomic ions of Al⁻ from the ion source are selected by the low energy magnetic analysis system for injection into HI-13Tandem accelerator, typically with 40 nA beam current at the lower energy Faraday cup (LEFC). The Al⁻ ions are pre-accelerated to 110 keV in the low energy system, and to 7 MeV by the first column of the tandem accelerator. The negative ions are stripped by a 3 μg/cm² thick carbon foil, and the position ions are further accelerated in the second column by the terminal voltage. A 90° double-focusing analyzing magnet was used to select ions ²⁶Al⁷⁺, with energy of 56.1 MeV. According to theoretical simulation of the charge stripping, the charge state 7+ at 7 MV has a large fraction of ~34%. The ²⁶Al⁷⁺ ion beams are transported to the AMS Beamline 1 and detection system, with a switching magnet and 15° electrostatic deflector. The analyzer magnet, switching magnet, electrostatic analyzer and other beam optical elements are adjusted to obtain optimal ²⁶Al⁷⁺ beam line.

Since no interference of ²⁶Mg isobar, a Si(Au) surface barrier detector is used to record the number and energy of the ²⁶Al⁷⁺. The ²⁷Al beam current is record by the LEFC in low energy system, and the ²⁶Al/²⁷Al isotope ratios can be obtained. In this study, each sample was measured at least three times. All isotope ratios were normalized by ²⁶Al/²⁷Al standards sample.

3.1 ²⁶Al-AMS based on extracting Al^- ions

Typically, the beam current of ²⁷Al⁻ is ~40 nA at LEFC. The ions extracting is stable (Fig.3), which is crucial for measurement precision. Typical current is lower than that at UTTAC and SUERC^[37,38], because of mainly the low electron affinity of Al, and the not-so-good ion source and low energy system, caused by a low vacuum problem at that time. However, this did not have much effect on this work.

Fig. 3.Full-screen View

Beam currents of ²⁷Al^−- of the commercial Al₂O₃ sample and biological Al₂O₃ sample prepared at our AMS lab..

The preparation of biological sample is a key step in the tracing experiment. As shown Fig. 3, beam current of the prepared Al₂O₃ sample is over 40 nA, and it is larger than that of commercial sample. The overall particle transmission between the low energy injection system and the detection system is ~3%. Fig .4(a) is the 56.1 MeV ²⁶Al⁷⁺ ion spectra of the background sample prepared with the same sample preparation procedure. It demonstrates that there were no interference isotopes in our measurement. The combination of our sample preparation procedure and extraction of single atomic negative Al⁻ from the ion source, the ²⁶Mg isobar is highly suppressed. Fig. 4(b) is spectra of the blood sample, and it can be sure that the peak at 56.1 MeV is the counts of ²⁶Al⁷⁺ ions. The simplicity preparation procedure can satisfy the biological tracing requirement.

Fig. 4.Full-screen View

Spectra of 56.1 MeV ²⁶Al⁷⁺ ions, detected by a Si(Au) surface barrier detector.

We used the blank sample (commercial Al₂O₃ and our background sample) to test the AMS system. The sensitivity is good, with the ²⁶Al/²⁷Al ratio of lower than 5×10⁻¹⁵.

3.2 Al biological behavior in blood

In this work to establish the ²⁶Al-AMS method for biological tracing, the ²⁶Al/²⁷Al ratio, instead of the ²⁶Al concentration, is used. The ²⁶Al concentration can be got from ²⁶Al = (²⁶Al/²⁷Al) ratioײ⁷Al concentration, which can be measured by spectrophotometry.

The data of ²⁶Al/²⁷Al ratios in blood are shown in Fig.5. Because of limit in beam time of the tandem accelerator, we measured the blood samples of six time points (0.5, 2, 8, 16, 48, 168 h). The data could be fitted with a nonlinear function:

Fig. 5Full-screen View

Biological behavior of ²⁶Al/²⁷Al in blood. The data, were fitted by a nonlinear function and a two-component function (t_b,≈8 h).

The decrease index is similar with the result of Priest et al.^[3], who found that the Al concentration (A_Al) in the blood after injection varied in a function of time as A_Al = 0.206 t ^−1.250.

We also found, as shown in Fig. 5, that the data also can be well fitted by a two-component function of time,

where the t_b≈8 h is break time, and the exponents −(0.457±0.121) and −(2.921±0.195) denote respectively the time before and after t_b,, corresponding to the slow and fast decrease phase.

3.3 Al biological behavior in brain

Brain samples of five time points (2,16,48,72 and 168 h) were measured. The ²⁶Al/²⁷Al results are shown in Fig. 6. It can be seen that the brain ²⁶Al/²⁷Al ratio went up quickly in 16 h after the Al injection. It retained stable in the first week, with a nonlinear function of ²⁶Al / ²⁷Al =1.41×10⁻¹³ t ^{−(0.142±0.136)} .

Fig. 6Full-screen View

Biological behavior of ²⁶Al/²⁷Al in brain. The data were fitted by a nonlinear function and a two-component function (t_b,≈12 h).

Also, the data could be well fitted by a two-component function of time, as shown in Fig.6,

where the t_b≈12 h is the break time. The Al content in brain increased before t_b with an increase index of 1.620±0.125,corresponding to the slow decrease phase of the blood sample (Fig.5). After the Al came into the brain, it would decrease slowly, with a decrease index of −(0.142±0.136). If the Al intake continues, the Al accumulated in brain would cause many kinds of diseases.