**5. Light elements analysis**

**4. Hair analysis**

34 Forensic Analysis - From Death to Justice

arsenic of H's hairs were quite doubtful.

the spectral data was deleted [44].

using 100 h of the KEK-PF beam time for hundreds of hair shafts.

H's hair was analyzed by St. Marianna University School of Medicine, using AAS (atomic absorption spectrometry) after NaOH digestion and hydride generation technique. They found 90 ppb As3+ in her hair. The method was the same as reported in Ref. [41]. St. Marianna University knew that As3+ was not appropriately analyzed when using the method of Ref. [41], and thus they had not analyzed As3+ from 1984 to 1997 [42]. However they analyzed As3+ in 1998 in the forensic analysis of H's hairs and concluded that 90 ppb As3+ was exogenously attached to her hair. They used an old atomic absorption spectrometer, which was made in 1970s, i.e. too old, using paper and a pen-recorder, and measured the peak height with a rule. In early 1990s, the chemical state analysis of arsenic had been performed by an ion chroma‐ tography (IC)/ICP-MS or an HPLC/MS [43] and the analytical instruments had been compu‐ terized. The advancements of these analytical instruments were due to the zenith of the semiconductor industry [44]. Therefore St. Marianna University obtained As3+ concentration using an old non-computerized AAS machine, where chemical state of arsenic compounds changed depending on pH. Therefore the forensic analysis results on the chemical state of

Many of H's hairs were also measured at BL-4A of KEK-PF (Photon Factory at High Energy Accelerator Research Organization, a synchrotron radiation facility in Tsukuba), and found an arsenic particle on a hair. The synchrotron radiation beam size was 4 or 1 mm width along with the hair shaft. It is still not clear how many hair shafts were measured and how many particles were found. It is said that arsenic particle was found on only one or two hair shaft(s),

At SPring-8, the same hair was measured but arsenic signal was not detected and testified that

**Figure 5.** Photo of measuring the paper cup (1) at the beamline BL-4A of KEK-PF, taken from the legal document of the

Tokyo University of Science [39]. The document is a public document, not a copyrighted matter.

The results of light elements analysis are summarized in **Table 2**. This table is the results of the National Research Institute of Police Science. M's green can (2) was pure arsenic oxide imported from China. M's milk can (3) was directly taken from the green can (2). If **Table 2** is compared with **Table 1**, the concentrations of the impurity light elements are mostly inversely related to the arsenic concentration. That is to say, when As wt% was less, then Al and/or Ca concentrations were higher, for e.g., M's "Heavy" (4), M's brown tupper (5), and T's milk can (6). However, paper cup (1) was different. Sodium and iron concentrations were higher, but arsenic concentration was also high. If saline water like the sea water about one litter was poured into another green 50-kg can, and then dried, sodium concentration could be explained, but the sea water should take the mass ratio Na:Mg:Ca=100:12:4, which was significantly different from the ratio in **Table 2**. Based on the discussion at Section 3 and the present section, the paper cup arsenic oxide powder (1) was taken out from another green can imported from the same industry, but this green can was once exposed to saline water containing Na, Mg, and Ca, when mining, smelted, shipping, or in use.

It is known that arsenic green 50-kg cans were imported from China twice in a year for total 10 years. The M's green can was one of the 60 cans imported at the same time by a ship, known by the shipping mark on the can. At the top, 10 or 15 cans were sold in Wakayama city in a month, and consequently at least more than one hundred cans were sold in Wakayama city before the arsenic curry incidence.

Starch was found for several arsenic oxide powder evidences using infrared (IR) analyses twice and iodine-starch tests twice. M's brown tupper (5) and T's milk can (6) results were contra‐ dicted as shown in **Table 2**. The paper cup (1) did not contain starch, and if the sentence was correct, starch powder mixed in the arsenic disappeared when taken by the paper cup.

Barium is not a light element but was found in several arsenic oxide evidences. Barium was an impurity element in SiO2 for M's brown tupper (5), because barium was not water soluble. Barium was an impurity element of Ca for T's milk can (6), because it was water soluble; calcium was due to the cement. However, the barium in paper cup (1) has not been analyzed whether water soluble or not. Based on these fact, Osaka Electro-Communication University and Hiroshima University, who performed forensic analysis according to the order of judge in 2001, concluded that paper cup arsenic oxide (1) was identical to either M's brown tupper (5) or T's milk can with the probability of 80% [47]. But this was wrong, because the concen‐ tration of arsenic was higher for paper cup than those of (5) or (6); also because of the discussion related to the hexagon radar chart in Section 3.
