Trace Element Analysis of Sulphur in a Japanese

TRACE ELEMENT ANALYSIS OF SULPHUR IN A JAPANESE SWORD
N. Mori1, S. Kuypers2, T. Tanaka3, K. Nakajima1 and T. Kimura3
Jeol Ltd.1-2 Musashino, 3 Akishima, JP-196-8558 Tokyo, Japan
2Jeol BV　Planet II, bldg. B, Leuvensesteenweg 542, BE-1930 Zaventem, Belgium
3National Institute for Materials Science (NIMS), Advanced Nano-Characterization Center,
Advanced Surface Chemical Analysis Group
1-2-1 Sengen, JP-305-0047 Tsukuba, Ibaraki, Japan
Cross Section of Samurai Sword
The Japanese sword or Nihonto is a piece of art, a wonder of metallurgical
craftsmanship and – even today – a fascinating object for materials research.
A traditional Japanese sword is made from tama-hagane (literally: jewel
steel). Tama-hagane is steel with a carbon content around 1 % and has the
chemical composition best suited for edged steels. Its composition makes it
exceedingly malleable. The amount of other chemical impurities present is
extremely low. Tama-hagane is obtained by a traditional method called tatara.
Tatara iron manufacturing has been developed by the master sword makers for
more than 1000 years.　The sword was cut, and analyzed.
Edge
Mine
JXA-8230
A
B
Ferrite+Perlite
Perlite
BEI Image 10kV 5000 X
soft
5 μm
hard
Mass%
traditional steel
low carbon steel
Martensite
1.0
C
O
0.5
Si
0.0
0
Al
1
2
3 4
5
6
mm
Line analysis of C 10kV, 10nA
JXA-8230 stage mappings at 20 kV, 500 nA over a large area (6mm x 30mm).
We investigated a sword that was preserved in an old Japanese village headman house. One of the characteristics of a Japanese sword is that is consists of two
different steels: a harder outer jacket of steel wrapped around a softer core of steel. A specimen including both steels was produced as cross-section perpendicular
to the length of the sword. Stage mappings was performed over a large cross-sectional area of 6 mm x 30 mm with EPMA (JXA-8230). The outer skin of the
blade showed few impurities other than carbon. A striped pattern of Si and Al due to oxide inclusions such as SiO2 and Al2O3 particles was clearly observed. In
fact, tama-hagane manufacturing with the tatara process from iron sand contains, though small, certain amounts of SiO2 and Al2O3 particles. Furthermore, the
manufacturing process of the Japanese sword consists of many times of folding, hammering and re-welding of the billet of steel, a process which produces a striped
distribution of SiO2 and Al2O3 particles. The observation of the outer skin of the blade agreed very well with those made by traditional sword makers.
Low carbon analysis technique
35000
LNT is effective in the low carbon analysis
y = 24470x + 11269
Mass%
Mass%
30000
Stainless Steel Mapping
LD E2H
Intensity
25000
20000
L D E2H
L D E2
S TE
15000
LD E2
LNT
y = 7816.4x + 3893.1
10000
Mass%
0.1% Difference
Object lens
STE
5000
D etection lim it
(10kV,1uA ,10sec)
L D E 2H :12ppm
L D E 2:23ppm
S T E :49ppm
y = 1970x + 1016.9
0
0
0.2
0.4
0.6
m ass%
0.8
1
High sensitivity multi-layered analyzing crystals
Cold fin
specimen
LNT traps the contamination
by the cold fin.
C
C
LNT ON
LNT OFF
LNT and TMP are effective in the Contamination decrease.
S　Kα
A
230ppm
B
230ppm
50ppm
S
Acc.V. - 20kV,
50ppm
Probe Current - 500nA
Analysis Time - 20s
Mn
JXA-8230 stage mappings
at 20 kV, 500 nA over a large area (6mm x 30mm).
Quantitative analysis　of Steel in Samurai Sword
Element A ( mass% ) B ( mass% ) Detection limit ( massppm )
P
0.0159
0.0171
21
Ca
0.1034
96
Mn
0.3516
60
Si
0.1365
0.1325
18
Al
0.0054
0.1090
18
Mg
0.0166
15
S
0.0232
0.0053
21
Cu
0.0155
0.0597
60
Cr
0.0245
45
C
0.3095
0.6054
21
Fe
99.1179
98.9510
Difference
Total
100.0000
100.0000
The inner core of the sword, was found to contain lager amounts of Mn
and S than those in tama-hagane.
Quantitative analysis of S with a spot size of 50 µm indicated that the
inner core of the sword contained 230 ppm whereas the outer skin of the
blade contained 50 ppm. To obtain these trace amount of S, the S-Kα
spectrum was measured to confirm that no discernible overlap due to
other elements was present in the background. The two standard
samples of ICP-MS: JSS003-1 S: 4 ppm and SS001-3 S: 2.2 ppm, were
analyzed with a spot size 30 µm. Only 2 ppm difference in these two
standard samples could be detected with this EPMA so that the measured
value of the outer skin of the blade was considered to be highly reliable.
Acc.V. - 20kV, Probe Current - 500nA, Analysis Time - 20s
z Detection Limits in Qualitative and Quantitative Analysis (Point Analysis)
10
Heavy
Elements
Detection limit (%)
Light
Elements
3 B
CDL =
C0
( P − B ) i ・t
1
EDS
Qual:all
0.1
WDS
Qual:trace, Quant:normal
0.01
EDS
0.001
0
10
20
30
40
50
Atomic number
Conceptual diagram of detection limit
in typical qualitative and quantitative
analysis conditions
5 μm
BEI
O
5 μm
CDL：Detection limit (mass%)
P：X-ray count rate of spectral peak (cps/μA)
B：X-ray count rate of background (cps/μA)
i：Current (μA), t: Time (sec)
C0：Concentration of the element of interest in
the standard (mass%)
Steel std. for ICP-MS
JSS003-1 4ppm
JSS001-3 2.2ppm
9 ppm
7 ppm
Name
Probe current Measurement time Remarks
Qual: all
100nA
5 minutes *2
Qualitative analysis condition for all elements
Qual: trace
100nA
1 minute
Qualitative analysis condition for trace elements
1 minute *3
Normal quantitative analysis condition
Quant: normal 50nA
Quant:trace
Low Sulphur in Steels
Quant: trace
500nA
1 minute *3
Quantitative analysis condition for trace elements
EDS
*1
~ 5 minutes *4
Spectral collection by EDS
5 ppm contamination
at polishing
Acc.V. 20kV, Current 2 μA
*1 Adjusted to optimize the detection limit.
*2 Measurement time is set depending on the configuration of X -ray spectrometers and analyzing crystals in the
instrument used for measurement.
*3 Measurement time is set to 1minute, including the time to move t he analyzing crystals. (Peak: 20sec, BG
±: 10sec)
*4 Real time.
Si
5 μm
5 μm
BEI
O
5 μm
Si
5 μm
S
5 μm
Al
5 μm
Mn
5 μm
Ca
5 μm
MnS
Α
S
5 μm
Al
5 μm
Mn
5 μm
Ca
5 μm
Β
Acc.V.　15ｋV， Current 50nA
Sulphur inner core was found in high magnification mapping as MnS particles with size of few tenths of micrometres. MnS inclusions have never
been observed in tama-hagane. They are rather a fingerprint of modern iron making, which started at the end of the Edo period in the middle of
nineteenth century. This sword was found to be produced relatively recently about100 years old. This was classified as a contemporary work and
brief method.