Radiostrontium in soil, grass, milk and bone in U.K.; 1956 results

J.NuclearEnetgy.1957.Vol.6,pp.22to40. PergamonPxss Ltd..London

RADIOSTRONTIUM IN SOIL, GRASS, MILK AND BONE IN U.K. ; 1956 RESULTS F. J. BRYANT,A. C. CHAMBERLAIN, A. MORGANand G. S. SPICER Chemistry ancj Health Physics Divisions, Atomic Energy Research Establishment, Didcot, Berks (Received

13 August

Harwell,

1957)

Abstract-The results of Y3r analysis of soil, grass and sheep bone from twelve stations in England and Wales are given. The ?ir in the top 4 in. of undisturbed soil in July 1956 ranged from 1.9 to 10.0 me/km*, depending on the rainfall. The @%ractivity of herbage and of sheep bone showed a wider range, samples from acid hill soils being relatively more active. Milk from Somerset had a median activity of 4.4 ,upc @OSr/gCa in 1956, compared with 4.1 in 1955. Human-bone specimens obtained in 1956 showed Y3r activity depending on age. The average level in children under 5 was 0.7 ppc ‘%r/g Ca and the average bone dose 2 mrad/year. 1. INTRODUCTION

THE fission products formed in nuclear explosions are carried round the world in the upper air and fall to earth in rain (EISENBUDand HARLEY, 1953, 1955, 1956; STEWART et al., 1955, 1956; LIBBY, 1956). The fission product of greatest biological hazard is gOSr.* The routes by which gOSr enters the human body are shown diagrammatically in Fig. 1. The work described here is concerned with the gOSrcontamination of agricultural produce, and with the resulting trace contamination of human bones.

FIG. I.--Entry

of %r

to human

bone.

As strontium and calcium are chemically related the amount of g”Sr in biological materials is usually expressed in terms of the specific activity relative to calcium. The strontium unit,? or S.U. is defined as: Strontium unit = 1V2 c 9oSr/g calcium. * Medical Research Council (1956) Hazards 7 The unit was referred

to Man of Nuclear andAllied to in the past as the “Sunshine Unit.” 22

Radiation para. 236. H.M.S.O.

Radiostrontium

in soil, grass, milk and bone in U.K.;

1956 results

23

2. METHODS

The analytical methods used have been described in detail in an A.E.R.E. report et al., 1956). After addition of carrier strontium, sample material is treated to bring the alkaline earths into solution. Radiostrontium (8gSr and ?Sr) with the strontium carrier is separated from calcium as nitrate in strong nitric acid solutions. Ferric hydroxide and barium chromate scavengers are included to remove contaminating activities. The separated strontium is stored with yttrium carrier for at least 14 days and the yttrium precipitated as the hydroxide, converted to oxalate, mounted and counted in a suitable low background counter. The 9oY decay curve

(BRYANT

$-

‘O

1 0

20

IO 120

40 60 00 100 hr fo?%,doys for*‘Sr

FIG. 2.-goY

and 89Srdecay.

*

is followed and the activity of the %r deduced. Strontium is also precipitated as carbonate, mounted and the soY allowed to re-equilibrate. The sample is counted and as the activity is then due to 8sSr + sOSr+ soY, the 8sSr activity can be deduced by difference. The count may be repeated at weekly intervals and the 54-day decay of the *sSr observed. Typical decay curves from a sample of vegetation ash are shown in Fig. 2. In this instance the count rate of both soY and 8gSr was reasonably high, and an endwindow counter with background of 7 counts/min was used. When the specific activity of the original sample is low, or only a limited weight is available, very much lower counting rates are found. End-window counters with a background of O-5 counts/min (BRYANT et al. 1956), are then used, but a 20 g sample of human bone gives a maximum s0Sr activity of only a few disintegrations per minute and the 9oY measurement is often impracticable. The total strontium count is therefore reported for human bone. In instances where the activity has been sufficient for the “Y determination, the s”Sr and total strontium counts have been compared

24

F. J. BRYANT, A. C. CHAMBERLAIN, A. MORGAN and

G. S.

SPICER

and the sgSr contribution has been found to be small, as would be expected from the slow turnover of strontium in human bone (Appendix 3). The analytical methods usecl at A.E.R.E. are generally similar to those developed at the University of Chicago by LIBBY and his associates (LIBBY, 1956; MARTELL, 1956) and at the Health and Safety Laboratory, U.S.A.E.C., New York by HARLEY et al. (1956). Several intercomparison samples have been exchanged between different laboratories (HARLEY,1956) and have shown good agreement. There is a variation of practice in the treatment of soils, which is partly due to different objectives. The amount of calcium in the soil which is removed by ion exchange processes, such as ammonium acetate leaching or the electrodialysis method used in the U.S., is variable as between one soil and another and depends also on the 1000

loo-

0.1

1 WC/g

FIG. 3.-Comparison

of ‘Yh/Ca

10 ca by NH&

100

1000

ratios in soil by HCI and NH,Ac

extraction.

conditions of the extraction. The gOSrin the soil appears to be mostly exchangeable. Consequently the gOSr/Ca ratios or S.U. values in soils are usually higher when determined by ion exchange’ methods than when more complete extraction is made by hydrochloric acid leaching or fusion with sodium carbonate. The difference is most marked with calcareous soils. The “Sr/C!a ratio in the roots of plants growing in the soil is more nearly represented by the exchangeable than by the total strontium and calcium analysis, but it is very difficult to reproduce the extraction by the plant by any simple chemical procedure (BOWENand DYMOND,1955, 1956). Since the “exchangeable calcium” in the soil is an imprecise conception it was decided to make extraction with 6M hydrochloric acid the standard procedure. Previous work (BRYANTet al., 1956) had shown that this method extracts as much sOSrfrom the soil as the more heroic method of fusion with sodium carbonate, and it is shown below that measurements of the total 90Sr fall-out by HCl extraction of soil agree well with the estimated cumulative fall-out in rain. Extraction with ammonium acetate at pH 7 as well as with HCl has also been done on a selection of soils, aliquots of dried, ground and mixed soil being taken for the two methods. The details are given by BRYANTet al. (1956) for the 1955 soils and in Appendix 1 of this paper for the 1956 soils. In Fig. 3 the results in S.U. by

Radiostrontiumin soil, grass, milk and bone in U.K.; 1956results

25

the two methods are compared. On calcareous soils, which correspond to low S.U. values, there is considerable divergence between the HCl and NH,Ac results, the latter tending to give higher S.U. ratios. On acid soils there does not seem to be any systematic difference in the S.U. results by the two methods. There is also some variation of practice in the measurement of the g0Sr and Ca content of vegetation ash. Since the stable Sr content is low, Sr carrier is added before extraction with acid and the yield of Sr estimated. The gOSrresult is then corrected for yield, on the assumption that the stable and radioactive isotopes are extracted equally. This procedure is impossible for Ca, and in the method as described by BRYANTet al. (1956) and used at Woolwich, an independent extraction of the ash with HF/HClO, is made for calcium determination. In the Health Physics Division an alternative method has been used in which both Ca and gOSrare determined in the HCIO,/HNO, extract, but it became clear from intercomparison of results (Appendix 2.1) that this extractant does not remove all the Ca from the ash. A third method is being developed in which both gOSrand Ca are determined in the HF/HClO, extract, and details of this method will be published later. 3. SAMPLING

OF

SOIL,

GRASS

AND

SHEEP

BONE

The analytical methods and counting techniques are lengthy and complicated so that the number of samples that can be analysed in a year is limited. The sampling procedure had to be designed with the utmost economy of numbers, but it was decided nevertheless to include in the survey a wide range of climate and soil conditions. Soil and grass samples were confined to permanent grassland, so as to avoid the alterations in the soil activity profile brought about by cultivation. A list of sampling stations in use in 1956 is given in Table 1. Stations A to G are farms carrying sheep on permanent pasture, A to E being hill farms and F and G lowland farms. One half-acre plot was selected for sampling on each station, except Station A, where two such plots were chosen. The plots were intended to be typical but not fully representative of the grazing, which on hill farms varies greatly from place to place. Soil cores were taken in July, 1956, from 10 or 12 points on each plot to depth 4 in. and a square yard of herbage surrounding each soil core was cut as closely as possible with shears. The sites had been grazed up to a few weeks before sampling, and the growth was therefore fresh. The soil and herbage subsamples were bulked and mixed before analysis. Leg bones from a sheep between 12- and 15-months-old, which had grazed pasture similar to and including the sampling plot for at least several months previously, were taken at each station. Yearlings were stipulated to ensure recent but mature bone growth. To supplement the sheep stations, soil and grass were taken at five auxiliary stations. One of these (A3) is in the Cwmystwyth valley, at Pwllpeiran, about 5 miles from the stations Al and A2. The sampling area at A3 is in an upland valley, and is typical of the pastures to which the sheep are brought down in bad weather. The other four stations (H to K) are on former airfields, all within 15 miles of Hat-well, and having similar meteorological but dissimilar soil status. At these auxiliary stations, small plots of about 20 yd2 were fenced off. The soil cores and the herbage were taken from within these enclosed plots. At each of stations H to K in July,

26

G. S. SPICER

F. J. BRYANT, A. C. CHAMBERLAIN, A. MORGAN and

Ref.

TABLE 1 .-SAMPLING

T 41titude

1

Locality

(ft)

L

-

Rainfall 1954-6 (in/year)

STATIONS Soil

-

Samples

Total Ca (g/kg)

PH

Type

Al A2 A3 B

Cwrnystwyth, Cardigan Cwmystwyth, Cardigan Cwmystwyth, Cardigan Vymwy, Montgomery

1200 1100 800 1100

60 60 60 62

4.3 4.5 4.9 5.4

0.14 0.17 1.0 1.8

Peat on shale (Free draining Peat on shale (Free draining:

c

Talgarth,

1050

36

6.2

2.7

D

Princetown,

Devon

1300

81

5.6

5.8

E

Rookhope,

Durham

1600

42

3.6

0.4

F

Norwich,

85

26

7.5

4.7

G

Boxworth,

157

22

68

14.6

H

Aldermaston,

250

25

6.0

1.6

I

Culham,

180

22

6.6

3.0

J

Grove,

250

25

7.1

39

K

Chilton,

400

24

8.0

156

Free draining soil on sandstone Sandy peat on granite Peaty sandy loam with podsol layer Sandy loam with gravel Dark brown loam with chalk Sandy soil with humus Sandy soil on lower greensand Heavy gault clay Calcareous clay with flint

Brecon

Norfolk Cambs.

Berks.

Oxon.

Berks. Berks.

-

-

i

-

-/

Soil, grass, sheep bone Soil, grass Soil, grass, sheep bone Soil, sheep bone Soil, grass, sheep bone Soil, grass, sheep bone Soil, grass, sheep bone Soil, grass, sheep bone Soil, grass Soil, grass

Soil, grass Soil, grass

i

1956, three cores were taken to depths 12 in., and divided into three horizons of 4 in. each to test the penetration with depth. At A3 one single square yard soil sample only was taken in May, but when the sampling was repeated in November twelve cores were taken. At the five auxiliary stations, repeated samples of grass were taken at intervals from May to September, 1956. Samples were of two types: (a) Accumulated growth taken from previously untouched plots. (b) Fresh grass which had grown since the last cutting. 4.

RESULTS

ON

SOIL,

GRASS

AND

SHEEP

BONE

4.1. wSr in soil A list of the results of gOSranalysis of soils taken in 1956 is given in Appendix 1, and a summary of the results by HCl extraction is given in columns 3 to 5 of Table 2. In Fig. 4 the QOSractivity* by HCl extraction per unit area of soil to depth 4 in. is compared with the annual rainfall at the same stations in the years 1954-6. The line on the graph shows the fall-out of YSr which would have occurred if it is proportional to rainfall, using as reference point the cumulative total in rain of 5.6 mc/km2 to July, 1956, measured at Milford Haven, where the annual rainfall is 38 in. (STEWART et al., 1956, extended by further measurements). There is a rather large scatter in the results from the stations Al, A2 and A3, but otherwise there is ??The unit ,u,oc/ma is used in Table 2 and Appendix 1. The unit me/km*, equivalent is used in Fig. 4, because this unit is a commonly used measure of fall-out generally.

to 1000,u~c/ma

Radiostrontium TABLE

in soil, grass, milk and bone in U.K.

2.--J0Sr IN

SOIL, GRASS AND

Station

SHEEP BONE

I

Grass

Soil (HCl extraction)

27

; 1956 results

Bone

PH (g Ca/kg)

1 Cu,dm”) )b-v/g W

CupcW) /Q-w/k) ! (,wc/g Caii

Cuwlg W

I

Al A2 A3 B c D E F G H

4.3 4.5 4.9 5.4 6.2 5.6 36 7.5 6.8 6.0

I

6.6

J

7.1

K

8.0

60 90

4900 5700

800 760

9600

z7 4.7 14.6 1.6

6600 3300 10,000 5000 2900 3400 2600

130 59 14 28 220 5.2 2.6 19

3.0

2500

8.0

39

1900

0.66

156

2200

0.15

0.14 0.17 I.1 1.8 2.7

I

2100 1400 230 1000

I5

~

145 50 64 20 (a) (b) (a) (b) (a) (b)

( i:;

-

2100 790

450 250 150 68 100 89 100 72 190 210

::

1.3 24 2.3 18 2.5 ‘i.8

~

~

116 123 125 625 64 41

1 I

160 41 24 53 71 13 8.7

:: 33 26 32 26 33 28

(a) Accumulated growth. (6) Fresh growth.

0

10 20 30

40

50 60

Rainfall 1955-1957 FIG. 4.-Correlation

70 80 I in/year

of soil %r and rainfall.

good agreement between the fall-out of gOSrbelieved to have occurred in rain, and that found in the top 4 in. of soil. This agreement shows that gOSrremains substantially in the top soil and available to HCl extraction for periods of the order of years. This lack of penetration has previously been reported from the U.S. (LIBBY, 1956), and is shown also in the results from the 4-8 in. and 8-12 in. horizons from stations H to K, given in Appendix 1 and previously reported (BOOKER et al., 1957). The soil results are also expressed as the specific activity of gOSrwith respect to Ca (S.U.) in the extractant. The S.U. content of the soil is a function of the extractant used as well as the soil, especially for calcareous soils. On acid soils different methods of extraction appear to give approximately the same S.U. results (Fig. 3 and BRYANT _ et al., 1956).

. 28

F. J. BRYANT, A. C. CHAMBERLAIN, A. MORGAN and G. S. SPICER

A more serious objection to the S.U. as a unit of gOSrin soil is that it takes no account of the vertical distribution of WSr which may be very non-uniform even within the top 4 in. This is shown in the results from the two stations (D and E) where there was a sufficiently discrete layer of matt between vegetation and soil for a separate sample to be taken. The matt was found to contain about half the total activity in ppc/m2, and the S.U. ratio in it was 44 times that of the soil beneath. Equally sharp variation with depth may occur on other soils. 4.2. g”Sr in grass The detailed results from the sheep stations are given in Appendix 2.1, and-from the other stations in Ap.pendices 2.2 to 2.6. A summary is given in Table 2, in which the figures quoted for stations A3 and H to K are the averages of the results on the sequential samples taken between May and September, 1956. At stations H to K samples of accumulated growth and of new growth were taken separately, and the results on the two types are given in the rows labelled (a) and (b) respectively for each station. The results are expressed in activity per m2 of ground, per kg dry weight and per g Ca in the vegetation. The gOSrin ppc/rn2 in vegetation may be compared with the corresponding figure for soil in the fourth column of Table 2, and the vegetation activity expressed as a percentage of that in the soil. This percentage varies from 2.2 at station B to 0.05 at station H (new growth). The g”Sr expressed in p,uc/m2 in the new growth samples at stations H to K was only about one-sixth of that in the nearby samples of accumulated growth. This was largely due to the difference in the weight of vegetation. Per unit dry weight the accumulated growth had 17 per cent more gOSrthan the new growth and per g calcium 28 per cent more. Only the latter difference is significant on a t test, and both are much smaller than the divergence between the ,upc/m2 on adjacent new and old growth. This suggests that ,upc/kg or ,u,uc/g Ca is the best unit for expressing the activity of vegetation when the conditions of growth are uncontrolled. In Fig. 5 the ,upc/g Ca (S.U.) in vegetation at all stations in mid-1956 is correlated with the calcium content of the top 4 in. of soil. In making the comparison the variation in total fall-out of gOSr&PC/m2 in soil) has been allowed for by normalizing the results of the various stations to a nominal fall-out of 5000 p&m2. Thus for station B, at which the ?Sr in soil was 6600 p+/m2, the vegetation result given in Table 2 has been multiplied by the factor 5000/6600 before insertion in Fig. 5. The object of normalizing the results in this tiay is to avoid spurious correlation due to the association of high rainfall both with high total fall-out and with low soil calcium and pH. The normalized S.U. ratio in grass does not show any correlation with soil calcium when the latter is 1 g/kg dry weight or over, as determined by HCl extraction. Thus the normalized S.U. in grass from Pwllpeiran (A3) and from Chilton (K) are about the ‘same, though the soil at the former station has 1 g Ca/kg and pH 4.9 whereas the latter has 156 g Ca/kg and pH 8-O. The normalized S.U. values in vegetation on the uncultivated, acid and very low calcium soils Al, A2 and E are higher by a factor of 10 to 60 than the values for normal soils. ROMNEY et al. (1957) grew crops in pots containing 7 different soils to which

Radiostrontium

in soil, grass, milk and bone in U.K.;

1956 results

29

g”Sr had been added. The highest Yjr levels were found in plants grown on acidic soils low in calcium and the lowest levels in the alkaline calcareous soils. Plants grown on the Sassafras soil, which had a pH of 4.6 and less than 0.1 g exchangeable Ca per kg, took up about 10 times as much gOSrper unit dry weight as plants grown A, A2

A3

E

HB

Cl

FD

G

K I x Gross

J

fJjlOooc’~ P

0

i

Sheep bone

I

FIG. 5.-Correlation of OOSrin grass and sheep bone with soil calcium. (Normalized to 5000 ,utcc/me in soil)

on alkaline calcareous soils. These experiments were under controlled laboratory conditions, using gOSradded uniformly to homogenized soil, and without the complications introduced by foliar uptake, the profile of gOSrin uncultivated soils, and the presence of peaty matt between vegetation and soil. TABLE

3.-89Sr/8CSr

RATIOS IN GRASS, RAIN AND

SOIL AT CHILTON

89Sr/soSr ratio Date Rain

Grass

5 October 1956 3 December 1956 1 January 1957 1 February 1957 ’ 1 March 1957

I I

10 13 12

j

Soil

19 17

I

l5

~

2.3 2.3 2.1

4.3. sgSr in grass In the autumn of 1956 a series of measurements was made of the 8sSr/g0Sr ratio in grass at Chilton, near Harwell. Grass samples (accumulated growth) were taken at monthly intervals and rain was also collected during the month previous to each grass sample. The sgSr and gOSractivities were measured, and the sgSr/gOSr ratio at the time of sampling worked out. These results are shown in Table 3. The sgSr/gOSr ratio in the soil was also estimated, by numerical integration of the monthly fall-out of sQSr and 90Sr.* * For this purpose results on the *@Srand 9oSr content of rain at Milford Haven due to OSMOND (1957) were used. This was necessary because the Chilton results did not extend far enough back. Over the period during which the ssSr/soSr ratios from both stations were available, there was good agreement between them.

30

F. J. BRYANT, A. C. CHAMBERLAIN,A. MORGAN

and G. S. SPICER

The *gSr/gOSr ratio in rain decreased from 40 : 1 to 15 : 1 during the period of these measurements, but the calculated soil ratio declined only from 2.4 : 1 to 2.1 : 1, because of the effect of build-up and radioactive decay. The sgSr/90Sr ratio in grass was on each occasion intermediate between that of rain and of soil. 4.4. goSr in sheep bone The gOSrto calcium ratio in the bones of yearling sheep at stations A to G are shown in Table 2, and the normalized S.U. values are plotted against soil calcium in Fig. 5. Independent estimations by different laboratories on mixed ash samples show reasonable agreement, but there ‘is an unknown variation between animals from the same flock. The range of results for sheep from different areas in 1956 is as follows*

Lowland sheep Hill sheep

I

No.

Range (S.U.)

Median (S.U.)

I

7.8-15.6 24-160

14 57

6

I

The hills grazed by sheep in Britain are generally areas of high rainfall, the soil is uncultivated, peaty and of low-calcium status, and there is low yield of vegetation. Any or all these factors may tend to enhance the uptake of gOSr. The lowland farms generally have the opposite conditions. It is not possible to deduce from the present results the relative importance of these factors, but it is clear, from Fig. 5, that variation in the total fall-out of gOSris not the sole cause. The bones of hill sheep contain more gOSrper unit fall-out than those of lowland sheep. TABLE 4.-Sr/Ca RATIOS IN GRASS AND SHEEP BONE

-

Stable Sr @g/g Ca) .

OOSr&x/g

Ca)

??

Station

Average

Bone

Grass

O.R.

Bone

730 470 470 650 520 930

5000 1500 1700 2800 2500 3400

0.15 0.31 0.28 0.23 0.21 0.27

160 41 53 71 12.8 8.7

0.24

Grass

O.R.

1750 123 125 625 64 41

0.09 0.33 0.42 0.14 0.20 0.21

,

0.23

4.5. Ratio of strontium to calcium in herbage and sheep bone In Table 4 the Sr/Ca and gOSr/Ca ratios in grass and sheep bone at stations A to G are compared. The estimations of stable Sr were made by the Spectrographic Section, Chemistry Division, (Woolwich Out-station) using methods which will be reported separately. The discrimination against Sr in passage from grass to bone is shown with both stable and radioactive Sr. Following COMAR et al, (1956, * Some additional bones from animals killed in the early part of 1956, reported by BRYANTetat.(1956) have been included.

Radiostrontium in soil, grass, milk and bone in U.K.; 1956results

31

1957), the “Observed Ratio” (O.R.) is defined as: O.R. bone/grass =

Sr/Ca in bone Sr/Ca in grass

The O.R. for the sheep-bone/grass comparison varies between O-15 and O-31 for stable Sr, and between 0.09 and 0.42 for g”Sr. The mean of the O.R.‘s at the various stations is 0.24 (stable) and 0.23 (radioactive). These values are in excellent agreement with results reported elsewhere with various animal species (COMAR et al., 1957).

1955

1954 FIG.

1956

1957

6.-$OSr in Somersetmilk.

5. RADIOSTRONTIUM IN MILK A series of samples of spray-dried skimmed milk from a factory at Frome, Somerset* have been analysed, some in New York (by the kindness of Dr. J. H. HARLEY) and some at Woolwich and Harwell. The results are shown in Fig. 6 together with the cumulative 90Sr fall-out in rain at Milford Haven, reported by STEWART et al. (1956). A comparison of the milk activity with the total fall-out shows that, whereas the latter has increased fairly steadily for the past 3 years, the former rose quickly from about 2 to 5 S.U. in the spring of 1955 and has thereafter stayed fairly constant. The median of thirteen 1956 samples is 4.4 S.U., compared with 4.1 S.U. in 1955. The general trend is consistent with the theory that the milk activity is determined partly by .the cumulative fall-out and partly by the rate of fall-out. In October and again in December, 1956, samples of dried milk from various parts of Britain were obtained, and 8gSr and gOSrdeterminations made, with results shown in Table 5. There is a tendency in both sampling periods for the g%r activity to be higher in milk from the north and west of the British Isles than in that from the south and east. This is probably an effect of rainfall amounts. The activity of just over 10 S.U. found in the October sample from Londonderry, Northern Ireland compares with maxima of 10 S.U. or slightly more, reported from British Columbia? and from North Dakota (HARLEY et al., 1956). The 8gSr/g0Sr ratio in the October milk samples varies from 12 (Yorkshire) to 24 (Somerset), which compares with a ratio of 22 found in Chilton grass (Table 3). There is a marked reduction in 8sSr activity from the October to the December * The location was wrongly referred to as Yeovil by BRYANTel nl. (1956). t Atomic Energy of Canada Limited (1957) Report CRC 689, Chalk River, Ont., Canada.

F. 3. BRYANT, A. C. CHAMBERLAIN,A. MORGAN and G. S. SPICER

32

1956 sampling, due mainly to the change over from fresh grass eaten by the cows in the open autumn of 1956 to hay, silage and other stored foods in winter time. If the cows were eating in December, hay or silage cut the previous June, the *gSr content would have experienced a radioactive decay of three half-lives; reducing it by a factor 8, during the storage period. The stable Sr/Ca ratios in the December series of samples were measured spectrographically, and show a range from 200 to 410 pg Sr/g Ca. TABLE5.-REGIONAL COMPARISON OF Y3r AND ssSr IN DRIEDMILK

Sr (r&g Cal

Areas and date of dried manufacture Carmarthen Carmarthen Yorkshire Yorkshire Cumberland Cumberland Antrim Antrim Londonderry Londonderry Somerset

17 October 1956 29 December 1956 16 October 1956 27 December 1956 19 October 1956 25 December 1956 19 October 1956 28 December 1956 17 October 1956 27 December 1956 26 October 1956 28 December 1956

T

200 240 410 270 280 230

IMPORTANCE

OF MILK

AND

Ratio *%r/%r

Y3;

-

(wclg Ca) 190 30 53 19 100 5, 150 24 220 22 110 25

8.0 7.2 4.3 3.9 6.5 5.6 6.9 7.0 10.3 6.2 4.6 5.5 -

6. RELATIVE

-

?Sr @w/g Ca)

~ : j j

/ j I

23 4.2 12 4.8 15 04 22 3.4 21 3.6 24 4.5

OTHER

SOURCES

OF %r

Milk is the main source of calcium for growing children in Britain but it does not necessarily follow that it is the main source of strontium or of radiostrontium. Strontium-90 reaches the earth in air and rain, neither of which contains appreciable calcium, and it is therefore possible for gOSr to enter the human body by routes different from those of calcium. Some possible alternatives are considered in turn. Inhalation. STEWART et al. (1956) give the mean ?Sr concentration in air at ground level in the years 1952-1955 as 4 x lWs pc/cm3. A person breathing at the “standard man” rate of 20 m3/day would have inhaled 10 ,L+Cin the 4 years. Taking the fraction transferred to bone as O-22,* the resulting body burden would be 0.002 S.U. The same calculation would apply to a child, except insofar as the breathing rate of an active child may be greater in proportion to its body weight. Inhalation cannot therefore be a major factor in determining the body burden of gOSr. Drinking water. The mean WSr content of rain in 1952-5 was 1.7 ,u,uc/l. (STEWART et al., 1956). In 1956 it was about 2.5 ,,~yc/l. A litre of liquid milk contains about 1 g Ca so the figures quoted above for the S.U. content of milk can also be read as ,u,uc/l. The gOSrcontent of milk is thus about twice that of rain-water, volume for volume. Few people in Britain drink raw rain-water. The “Sr content of river and lake water is less than that of rain-water by at least a factor 10 (LIBBY, 1956). New York tap water shows l/30 the activity of local rain. A similar ratio has been found in an * I.C.R.P. (1955) Recommendation of the International Commission on Radiological Protection.

Brit. J. Radiol. NO. 6.

Radiostrontium

in soil, grass, milk and bone in U.K.;

1956,results

33

isolated London sample (OSMOND,1957). It follows that milk is likely to be a more important source than drinking water if the relative consumptions are at all comparable. Cereals and vegetables. If the plants eaten by man have S.U. levels equal to those found in grass they are potentially an important source of sOSr, since they enter into diet without the discrimination against strontium which occurs in the production of milk by cows. Adequate data are lacking, but there are indications that the gOSr levels in cereals and vegetables in Britain are considerably lower than in grass.* The reasons for this include the effects of cultivation and the protection afforded l-6

5

p

1.5 1.4 l-3 1.2 1-l 1.0 O-9 0.6 o-7 D6 0.5 04 0.3 0.2 0.1 0

2468

10

12

14

16 20 40

Age FIG. 7.-Y4r

60 years

in human bone in 1956.

against foliar uptake by the outer leaves and husks of vegetables and cereals, and by washing and other preparatory processes. The addition of mineral calcium will depress the S.U. ratio in flour. The relative importance of milk and other sources of gOSrmay change with time, as the importance of foliar uptake and of the effect of cultivation will lessen as the cumulative fall-out increases and the gOSrbecomes more evenly distributed in the top soil. 7. RADIOSTRONTIUM IN HUMAN BONES A list of human bone specimens to the end of 1956 analysed at Woolwich is given in Appendix 3, and the radiostrontium results? on 34 femora and 3 tibiae among them are shown in Fig. 7. The specimens came mostly from south east England and the Midlands, but there were a few from the west and north west, and these included the two showing the highest radiostrontium activity, namely, 1.55 S.U. in a 2-year-old Plymouth child and 1.3 S.U. in a 3-month-old from Carlis1e.S There is a considerable scatter in the results on bones from infants under 1 year, but thereafter there is a fall off in the g”Sr/Ca ratio with age, as previously reported by KULP et al. (1957). The average gOSractivity of femora in various age groups is given in Table 8 at the end of this paper. * HIYAMA (1957) has reported that brown rice samples in Japan gave 12 S.U. in July, 1956, and 104 S.U. in November, 1956. There is some doubt whether the latter estimate is of 90Sr or of WSr and *@Sr. Since the 89/90 ratio in grasp towards the end of 1956 was about 20 : 1, the distinction is an important one. t As stated in the paragraph on methods, the practice with human bones has been to measure the total radiostrontium activity and attribute it to O%r. On specimens for which 8oY has been measured independently, the ‘?Sr activity deduced has been within 15 per cent of that estimated by the total strontium method. : A provisional result of 2.3 S.U. has been obtained in a l-year-old child in 1957. 3

F. J. BRYANT,A. C. CHAMBERLAIN,A. MORGAN and G. S.

34

As the minimum

weight of bone for a reasonably

SPICER

accurate s0Sr estimation is

50 g it was not possible to make a detailed study of distribution

within the bone,

but two femora from older children were each divided into four portions which were analysed separately as shown in Table 6. The pattern is the same in both specimens, with a maximum in the new sub-epiphyseal bone and a minimum in the old bone in in the centre of the shaft. TABLE6.-DISTRIBLJTION OF “Sr IN HUMANFEMORA (,U,uC/gCa) HB38

Ref.

I

HB 57

.I

Age (years)

9

Distal epiphyseal plate Distal sub-epiphyseal bone Centre of shaft Proximal sub-epiphyseal bone

0.17 0.20 0.11 0.20

0.25 0.34 0.22 0.27

Stable Sr measurements have made been on a number of human bones, and the results are given in Appendix

3. The average at all ages in the present series is 320 ,ug

Sr/g Ca, which is not significantly different from the average of 270 rug Sr/g Ca found in milk.

Since, however, the stable Sr content of foods not derived from milk is about

10 times greater than this, it is by no means certain that milk is the main source of stable Sr to human bone. TABLET.--DOSERATEFROM“Sr

AND

NATURAL

Source of radiation

RADIOACTTVITY

TO

BONE

Dose rate (mremlyear)

Natural radiation

external sources radium in bone

82 39

total

121

Wr in children under 5 in 1956 median level (0.70 S.U.) maximum level (1.55 S.U.)

8.

DOSE

RATE

FROM

1

2 4

RADIOSTRONTIUM

In Table 7 the average dose rate to bone from the gOSrburden in children in 1956 is compared with that due to natural radioactivity

(SPIERS, 1956).

The highest R’?Sractivity recorded in 1956 gives an average dose to bone which is one-tenth of that due to the natural radium, when allowance is made for the relative biological efficiency of the a-rays, **and one-thirtieth of the total natural dose to bone from internal and external sources. * R.B.E. of cc-rays is 10 (ICRP, 1955, see footnote p. 32).

Radiostrontium

9. MAXIMUM

in soil, grass, milk and bone in U.K.; PERMISSIBLE

BODY

35

1956 results

BURDEN

OF

$OSr

The maximum permissible body burden for occupational workers is 1 PC*, which is approximately equivalent to 1000 S.U. The Medical Research Council Committee (1956) have proposed a limit of one-tenth of the occupational body burden for the general population, and state also that “immediate consideration would be required if the concentration in human bones showed signs of rising greatly beyond onehundredth of that corresponding to the maximum permissible occupational level.” The highest gOSractivity in human bone found in our series to the end of 1956 is 1.55 S.U. which is l/60 of the maximum permissible for the general population, TABLE8.-90Sr

IN BIOLOGICAL

I Material

No. of samples

MATERIALS IN 1956 90Sr activity (S.U.) I ,~~~~_ _ _~_

Max.

Min.

Median

91 11 24 7.8 2.9 3.9

130 37 57 13.7 4.4 6.7

. ~~ ._________ Grass (acid hill soils) Grass (normal soils) Sheep bones (hills) Sheep bones (lowland) Milk (Somerset) Milk (other areas)

I I

9 61 6 7 13 10

Human bones (femora and tibiae)

I

2100 77 170 15.6 5.7 103

I

O-5 years

25

5-20 years >20 years

10 2

:

1.55

0.15

0.70

I

0.38 0.13

0.15 0.06

0.26 -

and l/6 of the level above which “immediate consideration would be required.” It has been suggested that a dose rate 10 times the human occupational level would be acceptable for animals (CHAMBERLAIN, LOUTIT,MARTIN,and SCOTTRUSSELL 1956). The gOSr maximum permissible level for sheep would then be 10,000 S.U. The highest levels recorded in Britain are at Cwmystwyth, when 183 S.U. were found in October, 1955 (BRYANTet al., 1956) and 160 S.U. in July, 1956 (Table 2). 10. SUMMARY

OF

1956 RESULTS

.

The range of 1956 results on the various materials sampled and the median values are given in Table 8. Acknowledgements-We

are indebted to Mr. K. H. JONESand to all the members of the National Agricultural Advisory Service of the Ministry of Agriculture, Fisheries and Food who obtained samples for us. Dr. RICE WILLIAMS,in particular, gave very valuable advice and practical help in obtaining the Welsh samples. We are indebted to Dr. M. BODIAN,Dr. J. S. FAULDS,Dr. R. H. MOLE, Dr. A. H. CAMERON and Dr. C. A. JONES who went to great trouble to provide the human bone specimens, and to Dr. J. F. LOUTIT, Dr. E. E. POCHINand Dr. R. SCOTT RUSSELL for helpful discussions. * Seefootnoteto ICRP on p. 32.

36

F. J. BRYANT, A. C. CHAMBERLAIN, A. MORGAN and G. S. SPICER APPENDIX List df Soil Samples-(Soil

1

samples to death 4 in. excevt where otherwise stated)

Sample

Soil

I

Station Date

Ref.

Al

A31

7 May 1956

Al Al

A48 A46

9 July 1956 20 November 1956

Al

Mean

A2

A42

%r

I

Method kg/ma g Call%

1

50

10 12

-37 45

irz,uc/g _

0.13 0.11 0.09 0.19 0.20

Ca ppc/ma

520 330 1440 680 800

3400 1800 4800 5700 7200

HCl NH4Ac HCl HCI HCl

_9 July 1956

10

44

0.14

44

0.17

800

4900

HCl

760

5700

HCl

110 110 98

9700 8900 4900

HCl HCl NH*Ac --

150

10,000

HCl

59

6600

HCl

14

3300

HCI

19 88 28

5900 4100 10,000

HCl HCl HCl

120 560 220

2200 3000 5200

HCl HCl HCl

5.2

2900

HCl

2.6

3400

__ __

A3

A30

7 May 1956

1

79

1.11 1.04 0.65

A3

A41

20 November 1956

12

66

0.97

B

A43

6 July 1956

10

62

1.8

C-

A23

D

A40

_._ __ 20 March 1956

10

86

2.7

18 July 1956

10

57 5.4 62

5.5 8.7 5.8

59 5.3 64

0.31 1.0 0,37

(matt) (total)

__

_. E

A41 (matt) (total)

10 July 1956

12

F

A33

4 kly

16

G

A32

3 July 1956

H

A36

31 July 1956

_. 1956

12 3

4-8” 8-12”

I

A31

A35

31 July 1956

3

A38 4-8” 8-12”

90

14.6

86

1.6 1.4 1.5 1.5

104 128 131

27 July 1956

3

75 87 107

4-8” 8-12”

K

4.7

107 128

4-8” 8-12”

J

119

27 July 1956

3

96 132 158

__.

3.0 2.3 2.7 3.5 39 I.7 17 9.5 156 6.6 185 204

_.

__

__

-

HCl -

19 17 <0.8 <0.3

2600 2000 <150 <50

HCI. NH,Ac HCl HCl

8.0 8.5 0.6 <0.3

2500 2000 200 <150

HCl NH,Ac HCl HCl

0.66 2.2 0.18 0.22

1900 1300 270 220

HCl NH,Ac HCl HCl

0.15 3,4
2200 2200 t150 <150

HCl NH,Ac HCl HCl

Radiostrontium

in soil, grass, milk and bone in U.K.; APPENDIX

37

1956 results

2

List of Grass Samples 2.1, saSr in grass on sheep farms

Ref.

Dry wt. (g/m?

Station

Lab.

DSl

15 1.1 2.0 1.6 8.2 7.7 17.3 15.6 1.5 1.0 7.1 6.3

069 045

B

065

D

044

E

038 037

P G

4 July 1956 3 July 1956

45 60

1900 2200 1400 1400 134 112 140 110 650 600 64 41

1500 1700 2800 2500 -

2.2. s0Sr in grass at Pwllpeiran, Cwmystwyth (Station A3) Ref.

Station

032 039 057 068

A3 A3 A3 A3

1

Date

!

18 June 3 July 8 August 17 July

1956 1956 1956 1956

/

Average

1

Dry wt. Wm2)

(

71

1

2.0

15

)

229

/

116

2.3. s0Sr in grass at Aldermaston (Station H) Ref.

Date

027/Z 2713 2714 2715

29 29 29 29

1956 1956 1956 1956

,

MCXI

29 May 1956

I

03515 D35/1-4 D41/5 041/l-4 048 D55/5 055/l-4 Mean

May May May May

19June

1

(a)

1.7

153

207 1.5 1.3 15 189 1.0 27 1.6 Samples lost in analysis 144 19

1956

g;

10 July 1956 31 July 1956

Es

23 August

1956

z:;

23 August

1956

(a) (b)

10.7

I

I

23.8 1.5 11.9 1.9 37 0.6

68 115 100 63 70

’

_I

1

40

;: 63

260 34

2.4. ‘OS’rin grass at Culham (Station I) Ref.

024/l 24/Z 2414 2415

--Mean

8 8 8 8 /

May May May May

1956 1956 1956 1956

/

8 May 1956

‘_-l

29 May 1956 29 May 19Jun.e 19June 10 July 10 July 31 July 31 July MeatI

1956 1956 1956 1956 1956 1956

(0)

I

::

147 3:: 3::

::: :;

-

148

(a) (b)

~ ~

1:: 23

(a) (b)

~

215 26

(a) Accumulated growth. (b) New growth since cutting 3 weeks previously.

1 1

4.6

;:; 2.9 42 3.3 4.3 2.4 1.9

1 ) ~

‘--I

I

28

186

146

‘;,3

135 140 78 46 145 110 53 60

39

24 1.3 44 41 5.6 1.4

;

36 27 :: :z 32

F. J. BRYANT, A.

38 2.5.

C.

CHAMBERLAIN,

MORGAN and

A.

G.

S. SPICER

90Sr in grass at Grove (Station J)

Ref.

8 May

1956 1956 8 May 1956 8 May 1956 8 May 1956

023/l 23/Z 23/3 2314 23/5

46 37 5.2 2.7

8 May

8 May 1956 D28/5 028/l-4 03315 033/l-4 04215 042/l-4

:

29 May 1956 29 May 1956 19June 19June1956 10 July 1956

1

64

’

1.8

4.0

2.8 3.2

18.4 2.5

; I

63

~

___ 35

(b)

31 10 July 1956

049

(a)

Lost

:; 175 39

Meat3

105 72

I

:‘6

200

/

40

2.6. sOSringrass at Chilton (Station K)

D25ll

9 May 1956

/‘s

9 May 1956 9 May 1956 9 May 1956

Mean

03015 030/l-4 03615 036/l-4 040/S D47jl-4 05415 054/l-4

I

/

30 May 1956 30 Mav 1956 2OJunk1956 20 June 1956 11 July 1956 11 July 1956

I

(a) (b)

i

I

18

i

;; 22

;:: 5.2 9.2 7.1 13.5 9.0 11.5

153 24

I 2

27 48

186 206

1:: 31 120

zi;

5.1

145 130 115 130 170 240 240 210 260 320

(a) :;

I

26 24 21 4.4 32 7.1 29 2.9 26 7.1

177 19 185

(b)

1

88

(01

Cb)

1 August Au&t 1956 9 September 1956

(a)

3.7 3.8

II!~~~~‘t!~I~~! i 1g 1;‘I ii 12

(a) Accumulated growth. (b) New growth since cutting 3 weeks previously.

Radiostrontium

in soil, grass, milk and bone in U.K.;

APPENDIX

1956 results

39

3

List of Human Bone Samples T

Ref. 1

1955 HBl 2 3 4 5 6 I 8 9 10

Date

October 1955 October 1955 October 1955 October 1955 October 1955 December 1955 November 1955 December 1955 November 1955 December 1955

1Age at death

12 years 21 years 40 years 38 years 6 years 1 year 21 years 31 years 23 years 10 years

Bone

Ribs Ribs Ribs Ribs Ribs Ribs Ribs Ribs

Sr

District

(pglg Cal ! (n) j (b)

Swindon Swindon Swindon Reading Swindon Birmingham Reading Oxford Reading Birmingham

! ,

~ YSr piuc/gCa

330 380 350

-

-

0.2 0.15 0.05 0.05 I 0.15 1.2 ~ 13.2 D.06 0.16 0.50 0.57

I-

1956 11 12 13 14 15 16 17

18 19 20 21 22 23 24 25 26 21 28 29 30 31 32 33 34 35 36 37 38 39 40 41

January January January Januray January January January

1956 1956 1956 1956 1956 1956 1956

January 1956

~ January 1956 / February 1956 ~ February 1956 January 1956 December 1955 December 1955 / December 1955 February 1956 February 1956 January 1956 February 1956 February 1956 February 1956 May 1956 April 1956 June 1956 May 1956 April 1956 May 1956 May 1956 June 1956 July 1956 July 1956

14 years 1 year 34 years 164 years 50 years 65 years Still-born

Ribs

~ Ribs ~ Ribs

Tibia Tibia Tibia I Sternum and ’ femur 64 years Tibia 67 years Tibia l* years Femur 3 months Femur 33 years Ribs 18 years Ribs 40 years Ribs 34 years Ribs 8 months Ribs 20 years Ribs 16 years Ribs 2 years Femur 11 years Femur 94 years 1 Femur 1 months/ Femur 1 2 Femur l* years Femur 2$ years Femur 8 years Femur 12 years Femur 14 years Femur 2 days ~ Femur 7 days Femur 12 days Femur

Birmingham Birmingham Dudley Carlisle Carlisle Carlisle Carlisle

Carlisle Carlisle Carlisle Carlisle Swindon Swindon Swindon Swindon Birmingham Birmingham Birmingham Birmingham Birmingham Birmingham Herts. Sussex Surrey London Middlesex London Surrey London Kent / London

i

-

I 160 250 -

450 380 450 -

~

250 150

530 320 300 420 190

.

0.76 1.1 1.05 0.25 0.06 0.13 0.45

1.1

got analysed \lot analysed 0.8 I 1.3 I 0.1 0.2 0.07 0.5 / 0.2 0.2 ’ 0.55 0.15 0.24 0.5 0.15 / 0.9 0.8 0.27 0.24 0.17 0.45 0.15 0.35 ’

F. J. BRYANT, A. C. CHAMBERLAIN,A. MORGAN and G. S. SPICER

40

Ref.

Date

4ge at death

-

Bone

Sr (:flg/g Ca)

District

42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59

July 1956 July 1956 July 1956 June 1956 June 1956 June 1956 July 1953 July 1956 July 1956 July 1956 June 1956 December December December December December December December

3 days 1 month 2 months 5+ months 6 months 6 months 6 months 2 years 2& years 3& years I+ years 2* years 3 years 9 years 2 years 9 years 7 years 4 months

1956 1956 1956 1956 1956 1956 1956

WSr activity:

Femur Femur Femur Femur Femur Femur Femur Femur Femur Femur Femur Femur Femur Femur Femur Femur Femur Femur

London London Sussex London Bucks. London Blackpool Middlesex Essex Sussex Essex Surrey Surrey Sussex Plymouth Berks. Norwich Cambridge

-

240 -

T;%r

ppc/g Ca

(a) 0.8 0.4 0.55 1.1 0.75 1.2 1.1 0.75 0.70 064 0.38 054 0.72 032 1.55 0.27 0.35 0.28

(b)

1.0 1.1

1.35

-

(a) estimated by total radiostrontium count. (6) estimated by s0Y count. REFERENCES

BANKERD. V., BRYANTF. J., CHAMBERLAIN A. C., MORGANA. and SPICER G. S. (1957) A.E.R.E. HP/R. 2182. BOWENH. J. M. and DYMONDJ. A. (1955) Proc. Roy. Sot. B 144, 355. BOWENH. J. M. and DYMONDJ. A. (1956) J. Exp. Bet. 7,264. BRYANT F. J., CHAMBERLAIN A. C., MORGANA. and SPICER G. S. (1956) A.E.R.E. HP/R. 2056. H.M.S.O. CHAMBERLAIN A. C., LOUT~~J. F., MARTIN R. P. and Scorr RUSSELLR. (1956) Proceedings ofthe International Conference on the Peaceful Uses of Atomic Energy Vol. 13. United Nations, New York. COMAR C. L. and WASSERMANR. H. (1956) Radioisotopes in the Study of Mineral Metabolism in Progress in Nuclear Energy Series I, Vol. 6, Pergamon Press, London. COMAR C. L., SCOTTRUSSELLR. and WASSERMAN R. H. (1957) Science In press. EISENBUDM. and HARLEYJ. H. (1953) Science 17, 141. EISENBUDM. and HARLEY5. H. (1955) Science 121, 677. EISENBUDM. and HARLEYJ. H. (1956) Science 124,251. HARLEYJ. H., HARDY E. P., WELFORDG. A., WHITNEYI. B. and EISENBUDM. (1956) U.S.A.E.C. report N. Y.O.O. 4761, Washington. HIYAMAY. (1957) Radiological Data in Japan, Government of Japan. KULP J. L., ECKELMANNW. R. and SCHULERTA. R. (1957) Science 125,219. LIBBYW. F. (1956) Proc. Nat. Acad. Sci. Wash. 42, 365. LIBBYW. F. (1957) Address to American Physical Society. Issued by U.S.A.E.C. as press release of 26 April 1957. MARTELLE. A. (1956) The Chicago Sunshine Method University of Chicago, Chicago, Ill. GSMONDR. G. (1957) Personal communication. ROMNEYE. M., NEEL J. W., NISHITAH., OLAF~~NJ. H. and LARXIN K. H. (1957) Soil Sci. 83, 169. SPIERSF. W. (1956) &it. J. Radiol. XXIX, 409. STEWARTN. G., CREAKSR. N. and FISHERE. M. R. (1955) A.E.R.E. HP/R. 1701. STEWARTN. G., CREAKSR. N. and FISHERE. M. R. (1956) A.E.R.E. HP/R. 2017.