Document Outline

; Yoshinaga et al.

1998

). These studies revealed

that human Pb contamination in Japan began in pre-
industrialized Edo era (1603

–1867) probably through the

usage of cosmetics (Hisanaga et al.

; Nakashima et al.

1998

). After Edo era, industrialization started and Pb

emission from industrial activities to the environment
should have affected human body burden of Pb. In fact,
Hirao et al. (

) found excess Pb loading to Tokyo Bay

started in 1900s and that maximum contamination was seen
in 1960

–1970 in the analysis of sediment core samples

covering the period of 1809

–1978. They attributed this

maximum to the use of leaded gasoline by 1970s. It is
noted that Japan phased out leaded regular gasoline by
1975 and achieved 100% share of unleaded gasoline by
1985 (Yoshinaga

2003

), and this was reflected in the

decreasing trend of Pb concentration in the layers of
sediment core corresponding to the period after 1970s.

Responsible editor: Ake Lennart Bergman

M. Matsumoto
Department of Allied Health Sciences, Kitasato University,
Sagamihara, Kanagawa 228-8555, Japan

J. Yoshinaga (

Department of Environmental Studies, University of Tokyo,
Kashiwa, Chiba 277-8563, Japan
e-mail: junyosh@k.u-tokyo.ac.jp

Environ Sci Pollut Res (2010) 17:643

–649

DOI 10.1007/s11356-009-0255-9

On the contrary, Suzuki et al. (

) revealed that Pb

concentration in Japanese women

’s hair that had been cut in

1920

–1929 was the highest (geometric mean, 136 mg/kg)

among the hair samples cut during the period of 1880

–1969

for use as a hair piece (kamoji in Japanese) or for use as a
needle pad filling and had been stored thereafter. Mean Pb
concentration in the historic hair samples cut in 1880

–1919,

1930

–1949, and 1950–1969 was 84, 18, and 20 mg/kg,

respectively (Suzuki et al.

). Since hair Pb level is

known to be influenced not only by ingested Pb that is
transported by blood flow but also by external adsorption
onto the surface of hair, high hair Pb level suggests that Pb
contamination source(s) is present in human proximate
environment, but not necessarily that the contaminating Pb
is taken up by the body.

Thus, the sediment core data and human hair data were

not consistent in that the period of peak Pb contamination
in Japan during the twentieth century was 1960s and 1970s
for general environment and 1920s for human proximate
environment. It is the aim of this study to get insight into
the source of Pb in human proximate environment in the
twentieth century by the analysis of Pb isotope ratios in the
same hair samples as Suzuki et al. analyzed with particular
focus on the ratios in the hair samples of the early twentieth
century. The Pb isotope ratios in the hair samples are to be
compared with those in sediment core samples reported in
Hirao et al. (

2 Materials and methods

2.1 Materials

Historic hair samples of Japanese women analyzed in this
study were obtained from Department of Human Ecology,
the University of Tokyo. These samples had been cut for use
for hair piece or needle pad filling and stored by the subject
herself or by the family member by the time of sampling
carried out by the members of the Department. Date of
haircut was obtained from the subject and/or family member.
Further information on this historic sample and its sampling
is given in Suzuki et al. (

). From the historic hair

sample collection of the department, 40 samples were
selected for this study because there left adequate amount
of sample for Pb isotope analysis. These samples had been
stored after washing with acetone and water according to
International Atomic Energy Agency method (Ryabukhin

1978

). Note that washing method did not alter Pb

concentration in these hair samples (Suzuki et al.

Additional hair samples from contemporary Japanese

were obtained: a composite male hair samples taken in a
barber shop in suburb of Tokyo during 1 month of
December, 2004 and four samples from female subjects

living in the suburb of Tokyo (Chiba and Saitama
Prefecture). The composite male hair sample was prepared
by washing with nonionic detergent (0.4% polyoxyethylene
lauryl ether) and finely cutting with stainless scissors to
make homogeneous composite. Four contemporary hair
samples were not washed. Table

shows the summary of

the hair samples analyzed in this study.

2.2 Methods

A portion of the hair sample (approximately 100 mg) was
accurately weighed in a capped Teflon vessel (N-3, San-ai
Kagaku Ltd., Nagoya, Japan) and digested with 1.5 mL of
ultrapure HNO

(AA-100, Tama Chemical Co. Ltd.,

Kawasaki, Japan) by Teflon double digestion vessel method
(Okamoto and Fuwa

1984

). The digest was made up to 10 g

with purified water to make a stock sample solution.
Procedural blank was included in every batch of sample
digestion and diluted as the hair sample.

Lead concentration in the stock solution was measured

by inductively coupled plasma mass spectrometry (ICPMS;
Agilent 7500ce, Agilent Technology Co. Ltd., Tokyo,
Japan) after appropriate dilution with 0.1 mol/L ultrapure
HNO

. A commercial standard solution (XTC-13, SPEX,

USA) was used for the calibration (Table

Isotope ratios (

207

Pb/

206

Pb and

208

Pb/

206

Pb) were mea-

sured by ICPMS after diluting the stock solution to 5 ng/g of
Pb concentration with 0.1 mol/L HNO

. A certified reference

material (CRM) from the National Institute of Standard and
Technology, USA (NIST SRM 981) was used for the
calibration of ICPMS. Typical within-day precision of isotope
ratio measurement was 0.3% (1

σ) for both of the ratios.

Human hair CRM from the National Institute for Environ-

mental Studies, Japan (NIES CRM No. 13; Yoshinaga et al.

1997

) was digested in every batch of sample digestions and

Pb concentration and isotope ratios were measured for
analytical quality assurance. Table

shows the analytical

result of NIES CRM No. 13 during the period of this study.

Table 1 Hair samples analyzed in this study

Sampling
decade

Total N

Sample type

Hair piece

Needle pad

Unknown/others

1910

1920

1930

1940

1950

1960

2000

Total

644

Environ Sci Pollut Res (2010) 17:643

–649

Reference value for the isotope ratios of this CRM was
provided from the NIES where the ratios measured by ICP
multicollector mass spectrometry with thallium internal
standardization preceded by HNO

digestion with micro-

wave digestion system and solid-phase chelate extraction
(Tanaka et al.

). Good agreement with the certified Pb

concentration and the reference isotope ratios were
obtained in the present study (Table

) indicating

satisfactory accuracy for concentration and isotope ratio
measurements.

3 Results

3.1 Pb concentration in hair samples

The concentration of Pb in historic hair samples analyzed in
this study (n=40) ranged from 2.18 to 313 mg/kg: The
mean concentration was 95.1 mg/kg (standard deviation
(SD), 84.5 mg/kg) and median was 77.0 mg/kg. The Pb
concentration in these samples has already been measured
and included in the published data (Suzuki et al.

). A

close agreement was found between the results Suzuki et al.
obtained and the present measurement results ([Suzuki et al.
result] = 1.06 × [present result]

−3.34, r=0.992, data not

shown) in spite of the fact that there is 25 years interval
between the two measurements, and analytical method is
different (ICP atomic emission spectrometry for Suzuki et
al. and ICPMS for this study). Similar level of Pb was
determined in historical hair samples (1870s

–1920s) of the

USA (Weiss et al.

1972

The concentration of Pb in contemporary hair sample

was lower than that in historic hair sample: The concentra-
tion in male hair composite was 0.79 mg/kg and that in
samples from female subjects (n=4) ranged from 0.99 to
1.37 mg/kg. Median Pb concentration of contemporary hair
was 1.07 mg/kg (n=5).

The concentrations of Pb in historic and contemporary

hair samples are plotted against year of cutting hair in
Fig.

. In this figure, Pb concentration data in one of the

sediment core samples reported by Hirao et al. (

) are

included for comparison of decade-dependent variation
pattern of sediment and hair.

3.2 Isotope ratios of Pb in hair samples

Isotope ratios of Pb in historic and contemporary hair
samples were plotted in Fig.

. Considerable variation in

the isotope ratios was obtained. Mean isotope ratios of Pb
in hair samples divided by decade of hair cutting are plotted
in Fig.

. Bars attached to the marks in this figure denote 1

SD for the two ratios. Note that number of samples of
1930s and 1940s were two (Table

); therefore, SD was not

attached to the corresponding marks. The samples of 1910s
and 1920s formed one cluster, those of 1930

–1950s formed

Table 2 Operational parameters of ICPMS for the measurements of
lead concentration and lead isotopes

Lead concentration

Lead isotope

Instrument

Agilent 7500ce

RF power

1,500 W

Plasma gas flow rate

15 L/min

Carrier gas flow rate

0.9 L/min

Make-up gas flow rate

0.3 L/min

Sample uptake rate

0.35 rps (by peristaltic pump)

Data acquisition

3 point/mass

Dwell time

1 ms/point

Number of scans

100

1,000

Integration time

0.3 s/mass

3 s/mass

Number of measurement

Monitoredm/z

208

206,207,208

Table 3 Analytical results of NIES CRM No. 13 human hair during
this study

Pb concentration

Pb isotope ratio

Certified

Measured
(n=12)

Reference

Measured

Concentration

4.6±0.4

4.9±0.2

–

207

Pb/

206

–

0.8583±0.0005

0.858±0.000

208

Pb/

206

–

2.0984±0.0008

2.101±0.005

Determined by ICP multicollector mass spectrometry with thallium

internal standard (Tanaka et al.

)

100

150

200

250

300

350

1870

1890

1910

1930

1950

1970

1990

2010

Year

Pb concentration (mg/kg)

Fig. 1 Time trend of Pb concentration in Japanese women

’s hair

(closed circle) and sediment core sampled from Tokyo Bay (open
circle; Hirao et al.

Environ Sci Pollut Res (2010) 17:643

–649

645

the second cluster, and those of 1960s and contemporary
formed third cluster. In this figure, isotope ratios of
domestic Pb ore (Sato

1975

) and Broken Hill-type Pb ore

(Brown

1962

), a typical Australian ore, are also plotted as

they are known to be used in Japan in great amount in the
early twentieth century (Hirao et al.

). Generally, hair

isotope ratios decreased from upper right to lower left of the
figure, or the direction from Australian to domestic ore, as
the decade is getting newer. This trend is more clearly
shown in Fig.

in which individual isotope ratio

(

207

Pb/

206

Pb) is plotted against year of hair cutting. The

same pattern was seen when

208

Pb/

206

Pb was used because

of correlation between the two ratios (Fig.

); therefore,

data were not shown. In Fig.

, isotope ratios of Pb in one

of the sediment core samples reported by Hirao et al. (

are included as in Fig.

. Isotope ratio decreased as the year

of cutting hair was more recent. This figure also demon-
strates that decade-dependent isotopic variation seen in
human hair is much greater than that in Tokyo Bay
sediment.

4 Discussion

Use of hair metal concentration as an indicator of intake/
body burden of the metal has been considered of limited
value for most of metals partly because of exogenous
contamination problem. Lead is one of typical metals that
suffer from such exogenous contamination (Suzuki

and hair is considered virtually of no value for assessment
of internal dose. In this study, however, we regard Pb in
hair to represent Pb of both exogenous (e.g., atmospheric
particle adsorption) and endogenous (e.g., drinking water
and food) origins and that present in the proximate
environment of the person regardless of whether the metal
has been taken up by body or not. Presence of a
contaminant in the proximate environment of a person
indicates that the person can take up the contaminant by
some chance.

As shown in Fig.

, time trend in Pb concentration in

hair and Tokyo Bay sediment core during the twentieth
century was different. Around 1910s and 1920s, when hair
concentration was the highest, only a small increase from
background Pb concentration was observed in sediment.
Moreover, the Pb level in the early twentieth century was
followed by decrease for hair whereas with an increase in
the 1970s for sediment. Lead in sediment should have
originated from direct industrial discharge into coastal
water, transport of discharge via the riverine system and
wet and dry deposition of contaminated atmospheric
particles. Therefore, it is generally considered that sediment
Pb level reflects Pb levels in waterway and atmosphere.
The discrepancy in time trend indicates that Pb contamina-

2.100

2.120

2.140

2.160

2.180

2.200

2.220

2.240

0.840

0.860

0.880

0.900

0.920

0.940

0.960

207Pb/206Pb

208Pb/206Pb

Fig. 2 Distribution of Pb isotope ratios in Japanese women

’s hair

samples

2.06

2.08

2.10

2.12

2.14

2.16

2.18

2.20

2.22

2.24

0.82

0.84

0.86

0.88

0.90

0.92

0.94

0.96

0.98

207Pb/206Pb

208Pb/206Pb

1910s & 20s

1930-50s

1960s & 2000s

Australia

(Broken Hill)

Fig. 3 Mean isotope ratios of Pb in Japanese women

’s hair divided by

decade (1910s to contemporary) of haircut. Isotope ratios of Japanese
and Australian ores are from Sato (

1975

) and Brown (

1962

respectively

0.840

0.860

0.880

0.900

0.920

0.940

0.960

1870

1890

1910

1930

1950

1970

1990

2010

Year

207Pb/206Pb

Fig. 4 Time trend of

207

Pb/

206

Pb in Japanese women

’s hair (closed

circle) and sediment core sampled from Tokyo Bay (open circle; Hirao
et al.

)

646

Environ Sci Pollut Res (2010) 17:643

–649

tion of proximate human environment of early twentieth
century, reflected in hair concentration, was unlikely to be
due to general environmental contamination.

The similar situation was seen in the preceding Edo era

(1603

–1867). Lead concentration in human bone was so

high in this era that some of the children of a higher social
class (more contaminated than lower social class people)
were suspected to have died of Pb intoxication (Yoshinaga
et al.

2003

). However, there are no data that indicate Pb

contamination of general environment during the Edo era.
Hirao et al. (

) detected Pb at background levels

(<20 mg/kg) in sediment core corresponding to nineteenth
century. Inano et al. (

2004

) also did not find excess Pb in

sediment core taken from Osaka Castle moat before 1900.
Murozumi et al. (

1982

) found that Pb concentration was

constant around 15 ng/g in tree ring of Cercidiphyllum
japonica in remote area of Japan from 1800 to around
1900, which was followed by increase to 50 ng/g. All these
data indicate that human Pb contamination in Edo era was not
due to environmental contamination but rather to the
contamination of human proximate environment

—in the case

of Edo era, it was believed that face powder containing PbCO

was the cause. The closer similarity with Edo situation
suggests that it was also the case in the early period of the
twentieth century: Pb-containing face powder was prohibited
only in 1930 in Japan due to increased number of Pb
intoxication cases of children (Takahashi

1997

Isotope ratios of Pb in the hair samples of the early

twentieth century were quite different from Japanese
domestic Pb signature and close to the ratios of Australian
ore (Fig.

) which occupied the greatest share of Pb used in

that period of Japan (43%; Hirao et al.

). The

207

Pb/

206

Pb in hair of 1910s and 1920s was plotted against

Pb concentration on individual basis in Fig.

. The more

contaminated with Pb, the more Australian signature was
noted, though isotope ratio levels off around 0.94. Thus, the

source of Pb in proximate environment of the women of the
early twentieth century was estimated to be Australian Pb,
and it is speculated that face powder containing Pb used in
Japan of the early twentieth century was manufactured from
Australian Pb. It must be mentioned, however, that no
supporting information was available on the origin of Pb
used for face powder manufacture of those days. In
preceding Edo era, face powder was manufactured with
domestic Pb because virtually no international trade was
present (Shoguns of Edo era closed Japan against foreign
countries) and isotope ratios of elevated Pb found in human
bone of that era (Yoshinaga et al.

1998

) were consistent

with this fact. Although Hirao et al. (

) detected small

bias in the isotope ratios of Pb in sediment layer
corresponding to the late nineteenth century toward those
of Australian ore, the magnitude of impact of Australian Pb
they found was far much less than that found in this study
as shown in Fig.

. It was thus suggested that industrial

usage of Australian Pb affected much less magnitude to
general environment than did the Pb-containing face
powder to human proximate environment.

Higher isotope ratios of Pb in hair samples of the early

twentieth century steadily decreased thereafter to the lower-
left direction of Fig.

. This may be attributed to phase out

of Pb-containing face powder, which had Australian
signature, and exposure to environmental Pb, in turn, had
been getting predominant among Japanese people. With
regard to environmental Pb of early to mid-twentieth
century, source country of Pb import became diversified
with the increased Pb demand in Japanese industries, and

0.880

0.890

0.900

0.910

0.920

0.930

0.940

0.950

0.960

100

150

200

250

300

350

Pb concentration (mg/kg)

207Pb/206Pb

1910s

1920s

Fig. 5 Relation between Pb concentration and

207

Pb/

206

Pb in

Japanese women

’s hair of 1910s and 1920s

2.080

2.090

2.100

2.110

2.120

2.130

2.140

0.830

0.840

0.850

0.860

0.870

0.880

0.890

207Pb/206Pb

208Pb/206Pb

Food

Rock

Soil(rural)

Atmospheric particle

Soil(urban)

Fig. 6 Comparison of isotope ratios of contemporary Japanese hair
(closed circle, n=5) with environmental samples. Food, Certified
Reference Material

“Typical Japanese Diet”(Tanaka et al.

);

atmospheric particles, mean and standard deviation of 7 cities (Mukai
et al.

1993

); rock, parent rocks of remote region (Murozumi et al.

1982

); soil (rural), soil of remote region (Murozumi et al.

1982

); soil

(urban), surface soil of playgrounds in Tokyo (Takaoka et al.

2006

)

Environ Sci Pollut Res (2010) 17:643

–649

647

thus, relative dependence on Australian Pb, which was one
of the world major ores with the least radiogenic (the
highest

207

Pb/

206

Pb or

208

Pb/

206

Pb) ratios, decreased: This

should have resulted in less Australian signature in
industrially emitted Pb of those days that was reflected in
more radiogenic signature of sediment Pb (Fig.

; Hirao et

al.

). As a result, discrepancy between isotope ratios of

Pb in hair and sediment became smaller in the midcentury
than in the early twentieth century (Fig.

Discrepancy between the isotope ratios of Pb in hair and

in sediment seems to be the minimum in 1960s (Fig.

when environmental Pb contamination level was the
maximum in Japan (e.g., Fig.

). The major source of

environmental Pb contamination of this period should have
been leaded gasoline, which should also have affected
isotope ratios of Pb in sediment. It is thus assumed from the
present hair data and sediment data of Hirao et al. (

that isotope ratios of lead in atmospheric particles of
1960s

–1970s in Japan, which should have been strongly

affected by gasoline Pb, were around 0.86 and 2.12 for

207

Pb/

206

Pb and

208

Pb/

206

Pb, respectively. Isotope ratios of

atmospheric Pb of those days are not known. Mukai et al.
(

1993

) stated that before 1980, leaded gasoline used in

Japan was added with alkyl Pb of American companies
with radiogenic signature (

207

Pb/

206

Pb: around 0.82

–0.85)

and that of UK company with less radiogenic signature
(0.92

–0.95). If this was the case, our estimate of Pb isotope

ratios of the atmospheric particles of 1960s

–1970s (0.86)

was intermediate of those of alkyl Pb of the two source
countries indicating almost equal mixing of the two.

The isotope ratios of Pb in the hair samples of

contemporary Japanese (0.869 and 2.115) were similar to
those in 1960s Japanese (0.869 and 2.117) in spite of the
notion that hair Pb of 1960s Japanese was strongly affected
by leaded gasoline which is no longer present in the
contemporary atmosphere. As has been indicated by the
low Pb concentration in contemporary hair (median,
1.07 mg/kg), it is assumed that there is no particular source
of Pb in the proximate environment of the contemporary
Japanese. The present contemporary result indicates that
mixture of Pb isotopes from various small sources
accidentally coincides with the isotope ratios of leaded
gasoline used by 1970s in Japan. Figure

shows isotope

ratios of Pb in various environmental samples analyzed
after 1980s (Murozumi et al.

; Mukai et al.

Takaoka et al.

; Tanaka et al.

) along with those of

the present contemporary hair samples. Isotope ratios of
hair Pb are not consistent with those of atmosphere or food.
They overlapped those of urban soil (average

207

Pb/

206

and

208

Pb/

206

Pb was 0.863 and 2.106, respectively; Takaoka

et al.

2006

) indicating that adsorption of wind-blown soil

particles onto hair surface was probably the predominant
contributor to the coincidence. Takaoka et al. (