P r e p a r a t i o n method o f barium s u l f a t e from m e r c u r i c s u l f i d e f o r s u l f u r i s o t o p e r a t i o measurement
Maya Kawano
1， Ak i h o T s u t s u i
2， T a k e s h i Min
ami1，2*I
Graduate School0 /
Science and Engineering Research， and 2 Department0 /
L砕 Sciences. School0 /
Science and Engineering， Kinki Universiη3‑4‑1 Kowakae，
Higashi・osaka，Osaka 577‑8502， JAPAN (Received December 1，2014)
Abstract
The sulfur isotope ratio in mercuric sulfide called vermilion has been measured for identification of the original mines of vermilion used in ancient buria1 mounds and antiquities. Barium sulfate is prepared丘omvermilion for measurement of the sulfur isotope ratio. However， as vermilion is composed of sulfur and mercury， one issue is the prevention of contamination of barium sulfate by rnercury. Mercury is a toxic metal and pollutes the piping system of analytical ins甘uments.The aim of the present study was to establish a method of sulfur isotope analysis sampling
出
atdoes not include mercury. Mercuric sulfide was dissolved in reverse aqua regia， and sulfur ion was oxidized to sulfate ion with the addition of bromine. When we observed whether mercuηr was adsorbed by either cation‑or anion‑exchange resins， mercury completely adhered to neither resin. Barium chloride was then added to the solution without using an ion‑exchange resin. No mercury was detected in the precipitate of barium sulfate， and the recoveηratio was over 90 % when the ratios of water for the reverse aqua regia before adding barium chloride were more than double. We concluded血atthis method for precipitating of barium sulfate by the addition of bariurn ch10ride is simple and useful for removing mercu庁・Keywords: vermilion， su凶JIisotope， reverse aqua regia， barium ch1oride， barium sulfate， merc町y，ion‑exchange resin
1. Introduction
Vermilion was used as a precious red‑color pigment in the ancient world. Its vivid red color made it one of the most valuable symbols of power [1， 2]. In ancient Japan企omthe middle of血e Yayoi period to the Kofun pe吋od，升om the fourth century B.C. to the sixth cen加ryA.D.， vermilion was used to decorate the physical remains and the inner surfaces of many burial rnounds. Large amounts of vermilion ‑more than
*
Corresponding author10 kg ‑were observed in the burial mounds of powerful people. There are more than 100 cinnabar mines in Japan [3]， and vermilion col1ected
企
om cinnabar ore was used for painting of earthenware in the last part of Johmon period characterized by the hunting‑gathering lifestyle. In Yayoi period， which was based on a wet‑rice agnc叫同relifestyle [4]， vermilion was used for decorating the physical rernains and the inner surfaces of burial facilities‑49‑
in funerary ceremonies [5]. This custom was transmitted企omChina around the middle of Yayoi period， and vermilion was at first imported企omChina along with the custom. However， as people knew that vermilion could be collected in Japan， it is thought that Japanese vermilion instead of Chinese vermilion came to be used for the custom. Around this time the first unified dynasty， the ancient Yamato dynasty， was bom [6， 7]. It is known企omthe old records that there were several mines collecting cinnabar ores around the center of the ancient Yamato dynasty [8]. Therefore， to identi命 the original source of vermilion used in Yayoi and Kofun period would provide information on historical shifts of power.
For this pu中ose，we have been measuring the isotope ratio of sulfur， a component of vermilion [6， 7， 9]. Ishihara et al. [10] reported that the sulfur isotope ratio varied regionally rather than by sulfide species or type or by the commodity of the ore deposits. In addition， we previously reported that the ratios of sulfur isotopes of cinnabar ores collected from Chinese mines showed positive values [11， 12]， while the ratios of cinnabar ores collected from Japanese mines showed negative values [3]. Therefore， it is reasonable to measure the sulfur isotope ratio of vermilion in order to identi布 the original mine of that vermilion.
However， as vermilion is composed of sulfur and mercury， it is important to remove mercury from the assay system. Mercury is a toxic metal and pollutes the piping systems of analytical instruments. We did not consider in detail how much mercury was contained in the measurement samples. The aim of the present study was thus to establish a
50
method of sulfur isotope analysis that does not include mercury 合om vermilion.
2. Materials and Methods
Extra‑pure mercuric sulfide was purchased from Nacalai Tesque Co. Ltd. (Kyoto， Japan). Nitric acid and hydrochloric acid for ultratrace analysis， Dowex 50Wx8， 100‑200 mesh， for strongly acidic cation‑exchange resin， Dowex 1x8， 200‑400 mesh， for strongly basic anion‑exchange resin were 企om Wako Pure Chemicals Ltd. (Osaka， Japan). The other chemicals were ultra‑pure reagents purchased from Wako Pure Chemicals.
Removal 01 mercuη 斤om the preparation solution using cation‑ or
anion‑exchange resin
About 10 mg of mercuric sulfide was added to a test tube， and after two ml of reverse aqua regia (RAR) were added， the tube was heated at 95 oC for 2 ho町 s. After one drop of bromine was added to the cooling solution， the solution was heated at 95 oC for 2 hours again. Four ml of ultra‑pure water， Direct‑Q (Nippon Millipore K.K.， Tokyo， Japan)， was added to the solution after cooling. The solution was applied to the column with 10 ml of cation‑or anion‑exchange resin， and the pass‑through solution was collected. Mercury content was measured with mercury measurement equipment (HG‑450‑5C30， Hiranuma Sangyo Co. Ltd， Ibaraki， Japan).
Comparison
0 1
recovery method from mercuric sulfide to barium sulfiα:teMercuric sulfide was heated with 2 ml of RAR at 95 oC for 2 hours， and after adding one drop of bromine， the solution was heated at 95 oC for 2 hours again.
After cooling， 2 or 4 rnl of ultrapure water was added， and then the condition for precipitation of bariurn sulfate was cornpared as follows without ion‑exchange procedure: (a) A丘町 the solution was heated at 65 oC， one rnl of 11' ..ιBaCh was added to the solution for the production of bariurn sulfate and the solution was heated at 65 oC for one hour to rnak:e the precipitate; (b) After the solution was heated at 65 oC， one rnl of 1M‑BaCh was added to the solution and the solution was heated at 65 OC ovemight to rnake the precipitate; (c) One rnl of 1M‑BaCh was added to the solution at roorn ternperature and the solution was left at roorn ternperature ovemight. The precipitate was washed with ultrapure water 5 tirnes and dried at 65 oC for the rneasurernent of世y weight.
To exarnine the e百ectof the ratio of water with RAR for precipitating bariurn sulfate， the volurnes of RAR and water were changed as follows: (1) 3 rnl of RAR and 3 rnl of ultrapure water; (2) 4 rnl of RAR and 2 rnl of ul仕apurewater; (3) 1 rnl of RAR and 5 rnl of ultrapure water.
Stα，tistical analysis
Results are expressed as rnean士 SD (n=3). Statistical evaluation ofthe results was performed by one‑way ANOVA analysis. Significance was established at p<0.05 and p<0.01 using Sche旺ピs rnultiple cornparison procedure.
3. Results
Table 1 shows the e百ectof adsorption of rnercury using cation‑ or anion‑exchange resin. Mercuric sulfide was dissolved in RAR， and after the addition of brornine we thought that sulfur would becorne sulfate ion.
Table 1. Adsorption effect by ion‑exchange resin of mercury Before passage: Solution before applyin呂toion‑exchange resin
After passage: P国Sー出roughsolution. a且町applyingto ion‑exchange resm
cation‑exchange reSlll
anion‑ex.change
I
15.06:t 1.61I
0.01 :t 0.01r巴Slll
Total Hg contents in solution (ng)
Generally， as sulfate ion is adsorbed to an anion‑exchange resin， the rnercury content in the pass‑through solution using the anion‑exchange resin cannot be rneasured. However， in the present study， the effects of both anion‑ and cation‑exchange resins were checked to exarnine whether rnercu巧r lOn was rernoved企ornthe pass‑through solution. When the solution with rnercury and sulfate ions was applied to the colurnn filled with cation‑exchange resins， the total rnercu巧rcontent in the pass‑through solution was 4.31 ng， and the recovery ratio was 36.6 %. About 63.4 % of rnercury in the starting solution was adsorbed by the resin. In contrast， when rnercuric sulfide was dissolved in the sarne way as the above and applied to the colurnn fil1ed with anion‑exchange resins， the total rnercury content in the pass‑through solution was 0.01 ng， and the recoveηr ratio was 0.07 %. Alrnost all of the rnercury was rernoved企ornthe pass‑through solution. Although we thought that rnercury was adsorbed to the anion‑exchange resin， sulfate ion is also adsorbed to the resin. Therefore， the anion‑exchange resin cannot be used for collection of sulfate ion.
Next， bariurn chloride solution was added and a precipitate of bariurn sulfate was rnade. Mercury sulfide was dissolved in 2 rnl of RAR， and after one drop of brornine was added， 2 rnl of ultrapure water was added to the solution. After changing the precipitation
11ム Fh u
shown in Fig. 3; ~弓len 5 ml of ultrapure water to 1 ml of RAR was used， the recovery ratio of barium sulfate was
100 %. However， the recovery ratios of barium sulfate were 37.6 % and 16.0 %， respectively， when 3 ml of ultrapure water for 3 ml of RAR and 2 ml of
γl
O
H
﹁
1 1
⁝
ii l
i
‑
‑ lil i‑
‑d l
n u
0 0 0 0 0 0 2 0 8 6 4 2 1 1 ( 式 ) 0 豊富
55
￡
02
EP
Eo
oE
ultrapure water for 4 ml of RAR were used. No mercury was detected in any
戸士、 120 E )
者。1曲
司m
E 80 コ
h
.~ :: 60
0
言0 40 E ~ 20 0 8
RAR 3 ml + water 3 ml RAR 4 ml + water 2 ml RAR 1 ml + water 5 ml Fig. 3 Change ofrecovery ratio ofbarium sulfate with change ofthe ratio ofRAR田dwater
The solution w田heated65・Cfor one hOUf for making the precipit田Iofbarium sulfale
"‑p<O.Ol
precipitate (data not shown). 4. Discussion
We previously established a method to distinguish whether the original source of vermilion was Japan or China by measuring the sulfur isotope ratio in vermilion [6]. Initially we collected sulfur dioxide gas produced 企om vermilion according to the procedure described by Yanagisawa and Sakai [13]. However， more than 100 mg of vermilion was necessa巧Tfor this method [6]， while the amounts of vermilion that can be collected企omburial mounds are often less than 100 mg. Next， we directly introduced vermilion into an element analyzer， and sulfur dioxide gas directly introduced into a spectrometer for isotope analysis Analysis was possible approximately 10 mg of vermilion this method [7]. However， polluted the pipetting systems instrumen .t That is why
Fig. 1 Recovery回tioof barium sulfate by ratio of RAR叩dwater(l:l) RT: room tempera担 問**p<O.OI
conditions， the results are shown in Fig. 1. When the solution was maintained at 65 oC overnight after adding barium chloride， the recovery ratio of barium sulfate was 65.4 %. In contrast， the recovery ratios of barium sulfate were 7.5 % and 10.6 %， respectively， when the solution was maintained at room temperature overnight and at 65 oC for 1 hour. In addition， mercury was detected in any precipitate (data shown).
Figure 2 shows the result when 2 ml of RAR and 4 ml of ultrapure water were used. The barium sulfate precipitate was 100 % recovered by leaving at 65 oC overnight and the barium sulfate precipitate was 100 % recovered under all conditions， i.e.， 65 oC for 1 hour， at room temperature overnight as well as at 65 oC overnight. In addition， mercury was not detected in any precipitate (data not shown).
not not
was mass [7]. with by mercury of the further
65・C，1 hour 65・C，overnlght RT，overnight
Fig. 2 Recoverγratio ofbarium sulfate by ratio ofRAR and water (1 :2)
The recovery ratios of barium sulfate for different ratios of RAR and ultrapure water and leaving at 65 oC for 1 hour are
っ ︐
U
Fh u
~ 120
)
】ω C.E 100
ロ悶
~ 80
￡国0 60
。
~ 40
b u
d 20 u ω
』
lmprovement was necessary.
It is k:nown that mercury has both monovalent and divalent cations. The use of a cation‑exchange resin was thus considered as a method to remove mercury. However， as shown in Table 1， the cation‑exchange resin could not remove mercury仕omthe solution. In contrast， almost ‑but not quite ‑all of mercury could be removed 企om the solution by using anion‑exchange resin. Shirakashi et al. [14] reported a divalent ion of mercury in solution existed as an anionic mercury complex such as [HgX3
r
and [Hg~t. Especially when bromine ion is present， like in the present study， and the solution pH is low， mercury can easily adhere to an anion‑exchange resin. Furthermore， when the solution pH is low， cation‑exchange resin has positive ion characteristics， and it is thought that anion mercury complex adheres to the positive ion characteristics. In the present study， the use of anion‑exchange resin was difficult， because both sulfate ion and anionic mercury complex would be adsorbed by the anion‑exchange resin.How can mercury be removed企om the final sample for sulfur isotope analysis? We observed that mercury was not detected in the precipitate of barium sulfate obtained 合omthe pass‑through solution of the cation‑exchange resin column with barium chloride solution (data not shown). However， the recovery ratio of the barium sulfate produced in this way was not stable. Therefore， we examined how the recovery ratio of barium sulfate could be increased. Generally， the solubility of barium sulfate decreases with the addition of dilute sulfuric acid to yield a precipitate of barium sulfate. However， we also wanted to avoid contamination of sulfur
企om dilute sulfuric acid for measurement of the original sulfur isotope ratio. It is k:nown that the precipitate of barium sulfate has high adsorption characteristics， and it is necessary to remove cations， especially
仕ivalentcations，企omthe solution before barium sulfate is formed. However， mercury was not detected in the precipitate of barium sulfate in the present study. It is probable that mercury ion became an anion mercury complex and was not adsorbed into the precipitate. In addition， when making a large precipitate of barium sulfate， it is necessary to make the precipitate丘oma dilute solution and to keep the solution warm. From the results shown in Figs. 1， 2， and 3， a high recovery ratio of barium sulfate precipitate was obtained when ulむapurewater was added at more than 2 times volume of RAR and when the precipitate forming solution was maintained for more than one hour at 65 oC. In our previous study using reference materials such as IAEA‑S2 and IAEA‑S3， isotopic fractionation occurred when starting amount of sample was less than 10 mg [15]. Therefore， 10 mg or more samples would be necessary to obtain reliable data for measurement of the sul白risotope ratio.
In conclusion， for the measurement of sulfur isotope ratio， mercuric sulfide was dissolved in RAR and bromine was needed to form sulfuric ion. Then， after adding ultrapure water at more than 2 volumes of RAR， 1M barium chloride was added to the solution to yield a precipitate of barium sulfide. Mercury was not detected in the precipitate.百le above procedure is a very effective method for making a sulfur isotope ratio analysis sample企omvermilion.
円︿U
F同U
Acknowledgement
This work was supported by JSPS
KAKENHI Grant Number 26242016.
References
[1] Caley， E.R. (1982) Joumal ofChemical Education， 5，419‑424.
[2] Martin‑Gil， ，.J Martin‑Gil， F.J.， Delibes‑de‑Castro， G.， Zap瓜ero‑Magdaleno，P.， Sarabia‑Herrero， F.J. (1995) Experientia， 51， 759‑761.
[3] Minami， T.， Bunno， M.， Imazu， S. (2008) Annual Reports by Research Institute for Science and Technology， 20， 41‑48 (in Japanese).
[4] Imamura， K. (1996) New perspectives on insular East Asia. University of Hawaii Press.
[5] Minami， T.， Kawano， M.， Furukawa， N.， Takahashi， K.， Takeuchi， A.， Imazu， S. (2013) Chikyukagaku (Geochemistry)， 47， 237‑243 (in Japanese).
[6] Minami， T.， Imai， A.， Bunno， M.， Kawakami， K.， Imazu， S. (2005) Geoarchaeology， 20， 79‑84.
[7] Kawano， M.， Takeuchi， A.， Takahashi. K.， Imazu， S.， Minami， T. (2014) ISIJ Intemational， 54， 1155‑1158.
[8] Oku， Y. (1981) The Archaeological Joumal， 438， 17‑18 (in Japanese).
[9] Minami， T.， Iwahashi， T.， Ohba， T.， Ito， S.， Shibasaki， A.， Sakamoto， T.， Imazu， S.， Watanabe， S. (2011) Archaeology and Natural Science， 62， 65‑72 (in Japanese). [10] Ishihara， S.， Sasaki， A.， Minagawa， T.， B田mo，M.， Shishido， A.， Tanaka， R. (2000)
Bulletin ofthe Geological Survey of Japan， 51，283‑297.
[11] Minami， T.， Imai， A.， Bunno， M.， Tomita， K.， Hisa， Y.， Okayama， M.， Yang， Z.， Imazu， S. (2004) Archaeology and Natural Science， 46， 67‑74 (in Japanese). [12] Minami，工， Yang， Z.， Bunno， M.， Shimazaki， H. (2009) Archaeology and Natural
Science， 58， 25‑29 (in Japanese).
[13] Yanagisawa， ， .FSakai， H. (1983) Analytical Chemistry， 55， 985‑987.
[14] Shirakashi， T.， Kakii， K.， Kuriyama， M. (1987) Nippon Kagaku Kaishi， (8)， 1609‑1614.
[15] Kawano， M.， Tokuda， M.， Imazu， S.， Minami， T. (2014) Joumal of Applied and Laser Spectroscopy， 1， 37‑40.
A
斗ム
F同υ
