Environmental Geochemistry and Health 1993 15(2/3) page 93
Identification of Mn-oxide minerals in some soils from Devon, UK, and their varying capacity to adsorb Co and Cu Henrietta M. Lidiard*, Joy E. Rae and Andrew Parker Postgraduate Research Institute for Sedimentology, University of Reading, Whiteknights, PO Box 227, Reading, RG6 2AB, England Abstract A number of Mn-oxide minerals in soils from a farm in North Devon have been tentatively identified using a combination of advanced analytical techniques: scanning electron microscopy (SEM), scanning electron microprobe (SEMP), X-ray diffraction (XRD) and bulk chemical analysis by wet digestion followed by inductively-coupled plasma spectrometry (ICP). The minerals lithiophorite and hollandite are thought to occur throughout the study area although there is considerable geographical variation in the proportions of minerals present. Birnessite, vernadite, romanechite, todorokite and cryptomelane may also be present, although in smaller amounts. The use of SEMP, together with a simple sorption experiment, has allowed a study of the extent of uptake of Co and Cu by different Mn-oxide minerals. Lithiophorite appears to take up Co and Cu more effectively than hollandite within a pH range of 4-6.
Introduction
Manganese oxides are present in soils as nodules, coatings on peds and grains, or as a disseminated black/grey pigment. They are invariably impure and contaminated with clay minerals and organic matter, and often contain significant amounts of Fe (Gilkes and McKenzie, 1988). Mn oxides have proved difficult to study for several reasons, not least their impurity, but also because of their generally low abundance and the extremely poor sensitivity of X-ray diffraction to the minerals. Complications al.so arise from their small crystal size and non-stoichiometric composition. Previous research into Mn oxides has included the study of heavy metal sorption (e.g. Tewari et al., 1972; Loganathan and Burau, 1973; Murray, 1975; Burns, 1976; Means et al., 1978; McKenzie, 1967, 1970, 1972, 1979,1980, 1981 and 1983) and the selectivity order for sorption is well known: Pb > Cu > Mn > Co > Zn > Ni (McKenzie, 1980). Although to some extent pH-dependent, it is widely thought that the adsorption of metal cations occurs strongly in all the cryptocrystalline oxides, regardless of mineral type (McKenzie, 1989). The study reported here casts doubt on the latter
assertion. The two main objectives of the study are: (i) to illustrate the importance of using a number of complimentary analytical techniques for the identification of soil Mn-oxide minerals, and (ii) to determine the role of soil Mn-mineralogy in the concentration of trace elements. The two trace elements, Co and Cu, were specifically selected for investigation since both e l e m e n t s are i m p o r t a n t in agriculture, (Cu deficiency in cro 3s and Co and Cu disorders in
IAWALE~ ENGLAf'
N
,& . 0 5 10 k m | i i i i I , , , , ~
/
Scale
m
$imonsbath •
Station Fleming
nslaple Heastey
Mill Q
QNorth Moiton
*Present address: Department of Environmental Science, Phoenix Building, University of Bradford, West Yorkshire BD7 1DR England
Ipso~th Molton
Figure 1 Geograph'ical location of the study area.
94
Identification of Mn oxide minerals
the area (Mn, Fe and Cu), and field observations of soils suggest that the occurrence of Mn oxides is widespread. Soil float, either within soil profiles or from the surface of ploughed fields, is frequently covered with black sub-metallic encrusting minerals, and the appearance of black soil mottles is common. The geology and details of the mineralization of the study area are outlined in Figure 2. Bedrock is Devonian in age with sediments forming part of a clastic sequence, deposited in a series of marine transgressions onto the edge of the E-W trending coastline of the Old Red Sandstone continent (Goldring etal., 1967). The mineralisation is thought to have occurred as a result of hydrothermal activity during the Upper Carboniferous (Edmonds etal., 1985).It consists of l o w - t e m p e r a t u r e impersistent vein systems running sub-parallel to the strike of the country rocks in an east-west
grazing livestock are important and widespread problems in the UK and elsewhere: Thornton and Webb, 1980). A f a r m area c o n t a i n i n g Mn mineralization was selected as the study site in order that the relatively high abundance of Mn might i n c r e a s e the c h a n c e s of the i s o l a t i o n and identification of a number of different Mn-oxide minerals. In addition, the study site affords the opportunity to relate the uptake of Co and Cu by Mn oxides to Co and Cu deficiency diseases in livestock, since there are livestock deficiency problems related to both Co and Cu (and I and Se) in the area.
StudyArea The study area is a farm on Exmoor in the mineralised region of Heasley Mill, North Devon, UK (Figure 1). There is extensive mineralisation in
i
-~~?B~r~
._,._
- "
N
~,
:..."-. :.. _
I ~ -
~
-
-
~
~-~----S~--
v
:.:.;.........;....:::i. . . . . i! ii.i. Common
:.-; _
-
_
~--+
~
-~0
-
'- +
.__ . . . . . . ".~ : : . . . ' . .
-~'~".......
. . ? . ~ % j o ~ . . . . . . . . . . _._ ~BroyV~i,ey~y~cT, oe--~--.----~.'."
~
~oo,o~,~-~:#;.2 --__- ~ ~\-_-_-~ ~ - -
--'--~-s=j_ -
-- - -
- ~
" --
~
P t
Fe "" " "
"
~
"1" "
"
• " " " • ". " " " ." .,~'-
"
" • " :
•
•
X'/7././Y7/.A/Y,4.//¢,," ,'(77/./YT"/Z/¢/~.4/.~.
•
.'. / " " ~ ' ~ 2 / ~ , ! / ' / : " " " ' ~ [
L.
~°
"
.,..
•
.
".
•
"
"
. "
-
" ". ""
:.......... :. ~Peat
'IDEvON,AN
....
~
" " I SUCCESSION /~7.4f F_xrenI or • " It._ ~ ~ study farm • " " I I PiltOn Shales --2 ~ ^ . ,,i Fault . • " I-oVl~aggybst. ,, • . . . . i'=Z"-I UpcottSlates • Mineral vein
. .
" Fe "
.'IKEY"
"." " " "
• • #/~/2~-)L • • .~"r'ZI. • • " " /(./4/7,X~X)~" ~,~-,/¢//;,,I .... ;//~./z/'./',4./"/'7",Z'/.,/,,#d_zi ,
•
: i.ee: :..".
: .~........~,
•"
z%
" Cu Fe
~" : ' ' H e ' a s l e y ' M i l l ' ' ' ' '
• "~P'"
.~'.. " . . : :. ,ooo~T. , ., . .. •
'
• ...e~...
.,~. : : . ~ ~ - . - . •
. : : - ...
- _ - _ - ~ . . 0 . ' . - . . . . . . - . . ' .
- - - - -- - - ¢ ~ e 2 ~ " "
-
-.=__~_
"
" °,"
"
~
"i
~Morte 0 I
Slates I
.,~ DipangJdirect. 2 km
I
Figure 2 Outline of geology and mineralisation in the study area.
I
H. M. Lidiard, J. E. Rae and A. Parker
95
cattle are moved northwards to graze the reclaimed moorland pastures, where they remain until the following November.
Methodology
MOLTON
/
~
)
KEY: ~9-5-...~ '~-/'JContours in m. A Triangulation point
~ ~
Boundary of farm land SCALE : I : 50,000
Figure 3 Four-fold geographic division of the study farm. direction and the minerals haematite, wad and malachite are among those most commonly found. The study farm comprises four main tracts of land (Figure 3), ranging from altitudes of 200 m above O.D. to more than 500 m above O.D. Barton Pits (grid ref. SS63/73.272 132) and Great Coombeshead (grid ref. 272 130) are the two field sets in the southern region of the field area; they are relatively low-lying and sheltered. Towards the north, the field sets of Five Barrows (grid ref. 273 136) and Fyldon Common (grid ref. 275 135) are much more exposed, situated on the southern margins of E x m o o r uplands. The farm relies predominantly on stock cattle and sheep rearing for its income, and the farmers adhere to the traditional pattern of transhumance found in Exmoor farms. The livestock are housed during the winter months, or are grazed in the sheltered and lower-lying p a s t u r e s a r o u n d B a r t o n Pits and G r e a t Coombeshead. During this time the animals' diet consists mostly of silage and cereals produced on the farm. In May, all the sheep and many of the
Soil samples were collected by auger from the top 10 cm of each field in a c c o r d a n c e with the r e c o m m e n d a t i o n s of the British Ministry of Agriculture, Fisheries and Food (MAFF/ADAS, 1979). A minimum of 25 auger borings were collected in each field, in a ' W ' shape to avoid sampling bias. All trees, electricity pylons and areas of unusual appearance, e.g. tracks or heaps, were avoided. The composite samples were homogenised and carefully subsampled. In addition, a soil profile was sampled from Five Barrows (grid ref. 734 369) by digging a pit and removing samples at 5 cm intervals with a plastic scoop. Soils were oven-dried at 50.1 25-1- 50-0 20.1 - 25.0
15.1 - 2 0 , 0 10.1-15-0
/I Fault
Identification of Mn-oxide minerals in some soils from Devon, UK, and their varying capacity to adsorb Co and Cu Henrietta M. Lidiard*, Joy E. Rae and Andrew Parker Postgraduate Research Institute for Sedimentology, University of Reading, Whiteknights, PO Box 227, Reading, RG6 2AB, England Abstract A number of Mn-oxide minerals in soils from a farm in North Devon have been tentatively identified using a combination of advanced analytical techniques: scanning electron microscopy (SEM), scanning electron microprobe (SEMP), X-ray diffraction (XRD) and bulk chemical analysis by wet digestion followed by inductively-coupled plasma spectrometry (ICP). The minerals lithiophorite and hollandite are thought to occur throughout the study area although there is considerable geographical variation in the proportions of minerals present. Birnessite, vernadite, romanechite, todorokite and cryptomelane may also be present, although in smaller amounts. The use of SEMP, together with a simple sorption experiment, has allowed a study of the extent of uptake of Co and Cu by different Mn-oxide minerals. Lithiophorite appears to take up Co and Cu more effectively than hollandite within a pH range of 4-6.
Introduction
Manganese oxides are present in soils as nodules, coatings on peds and grains, or as a disseminated black/grey pigment. They are invariably impure and contaminated with clay minerals and organic matter, and often contain significant amounts of Fe (Gilkes and McKenzie, 1988). Mn oxides have proved difficult to study for several reasons, not least their impurity, but also because of their generally low abundance and the extremely poor sensitivity of X-ray diffraction to the minerals. Complications al.so arise from their small crystal size and non-stoichiometric composition. Previous research into Mn oxides has included the study of heavy metal sorption (e.g. Tewari et al., 1972; Loganathan and Burau, 1973; Murray, 1975; Burns, 1976; Means et al., 1978; McKenzie, 1967, 1970, 1972, 1979,1980, 1981 and 1983) and the selectivity order for sorption is well known: Pb > Cu > Mn > Co > Zn > Ni (McKenzie, 1980). Although to some extent pH-dependent, it is widely thought that the adsorption of metal cations occurs strongly in all the cryptocrystalline oxides, regardless of mineral type (McKenzie, 1989). The study reported here casts doubt on the latter
assertion. The two main objectives of the study are: (i) to illustrate the importance of using a number of complimentary analytical techniques for the identification of soil Mn-oxide minerals, and (ii) to determine the role of soil Mn-mineralogy in the concentration of trace elements. The two trace elements, Co and Cu, were specifically selected for investigation since both e l e m e n t s are i m p o r t a n t in agriculture, (Cu deficiency in cro 3s and Co and Cu disorders in
IAWALE~ ENGLAf'
N
,& . 0 5 10 k m | i i i i I , , , , ~
/
Scale
m
$imonsbath •
Station Fleming
nslaple Heastey
Mill Q
QNorth Moiton
*Present address: Department of Environmental Science, Phoenix Building, University of Bradford, West Yorkshire BD7 1DR England
Ipso~th Molton
Figure 1 Geograph'ical location of the study area.
94
Identification of Mn oxide minerals
the area (Mn, Fe and Cu), and field observations of soils suggest that the occurrence of Mn oxides is widespread. Soil float, either within soil profiles or from the surface of ploughed fields, is frequently covered with black sub-metallic encrusting minerals, and the appearance of black soil mottles is common. The geology and details of the mineralization of the study area are outlined in Figure 2. Bedrock is Devonian in age with sediments forming part of a clastic sequence, deposited in a series of marine transgressions onto the edge of the E-W trending coastline of the Old Red Sandstone continent (Goldring etal., 1967). The mineralisation is thought to have occurred as a result of hydrothermal activity during the Upper Carboniferous (Edmonds etal., 1985).It consists of l o w - t e m p e r a t u r e impersistent vein systems running sub-parallel to the strike of the country rocks in an east-west
grazing livestock are important and widespread problems in the UK and elsewhere: Thornton and Webb, 1980). A f a r m area c o n t a i n i n g Mn mineralization was selected as the study site in order that the relatively high abundance of Mn might i n c r e a s e the c h a n c e s of the i s o l a t i o n and identification of a number of different Mn-oxide minerals. In addition, the study site affords the opportunity to relate the uptake of Co and Cu by Mn oxides to Co and Cu deficiency diseases in livestock, since there are livestock deficiency problems related to both Co and Cu (and I and Se) in the area.
StudyArea The study area is a farm on Exmoor in the mineralised region of Heasley Mill, North Devon, UK (Figure 1). There is extensive mineralisation in
i
-~~?B~r~
._,._
- "
N
~,
:..."-. :.. _
I ~ -
~
-
-
~
~-~----S~--
v
:.:.;.........;....:::i. . . . . i! ii.i. Common
:.-; _
-
_
~--+
~
-~0
-
'- +
.__ . . . . . . ".~ : : . . . ' . .
-~'~".......
. . ? . ~ % j o ~ . . . . . . . . . . _._ ~BroyV~i,ey~y~cT, oe--~--.----~.'."
~
~oo,o~,~-~:#;.2 --__- ~ ~\-_-_-~ ~ - -
--'--~-s=j_ -
-- - -
- ~
" --
~
P t
Fe "" " "
"
~
"1" "
"
• " " " • ". " " " ." .,~'-
"
" • " :
•
•
X'/7././Y7/.A/Y,4.//¢,," ,'(77/./YT"/Z/¢/~.4/.~.
•
.'. / " " ~ ' ~ 2 / ~ , ! / ' / : " " " ' ~ [
L.
~°
"
.,..
•
.
".
•
"
"
. "
-
" ". ""
:.......... :. ~Peat
'IDEvON,AN
....
~
" " I SUCCESSION /~7.4f F_xrenI or • " It._ ~ ~ study farm • " " I I PiltOn Shales --2 ~ ^ . ,,i Fault . • " I-oVl~aggybst. ,, • . . . . i'=Z"-I UpcottSlates • Mineral vein
. .
" Fe "
.'IKEY"
"." " " "
• • #/~/2~-)L • • .~"r'ZI. • • " " /(./4/7,X~X)~" ~,~-,/¢//;,,I .... ;//~./z/'./',4./"/'7",Z'/.,/,,#d_zi ,
•
: i.ee: :..".
: .~........~,
•"
z%
" Cu Fe
~" : ' ' H e ' a s l e y ' M i l l ' ' ' ' '
• "~P'"
.~'.. " . . : :. ,ooo~T. , ., . .. •
'
• ...e~...
.,~. : : . ~ ~ - . - . •
. : : - ...
- _ - _ - ~ . . 0 . ' . - . . . . . . - . . ' .
- - - - -- - - ¢ ~ e 2 ~ " "
-
-.=__~_
"
" °,"
"
~
"i
~Morte 0 I
Slates I
.,~ DipangJdirect. 2 km
I
Figure 2 Outline of geology and mineralisation in the study area.
I
H. M. Lidiard, J. E. Rae and A. Parker
95
cattle are moved northwards to graze the reclaimed moorland pastures, where they remain until the following November.
Methodology
MOLTON
/
~
)
KEY: ~9-5-...~ '~-/'JContours in m. A Triangulation point
~ ~
Boundary of farm land SCALE : I : 50,000
Figure 3 Four-fold geographic division of the study farm. direction and the minerals haematite, wad and malachite are among those most commonly found. The study farm comprises four main tracts of land (Figure 3), ranging from altitudes of 200 m above O.D. to more than 500 m above O.D. Barton Pits (grid ref. SS63/73.272 132) and Great Coombeshead (grid ref. 272 130) are the two field sets in the southern region of the field area; they are relatively low-lying and sheltered. Towards the north, the field sets of Five Barrows (grid ref. 273 136) and Fyldon Common (grid ref. 275 135) are much more exposed, situated on the southern margins of E x m o o r uplands. The farm relies predominantly on stock cattle and sheep rearing for its income, and the farmers adhere to the traditional pattern of transhumance found in Exmoor farms. The livestock are housed during the winter months, or are grazed in the sheltered and lower-lying p a s t u r e s a r o u n d B a r t o n Pits and G r e a t Coombeshead. During this time the animals' diet consists mostly of silage and cereals produced on the farm. In May, all the sheep and many of the
Soil samples were collected by auger from the top 10 cm of each field in a c c o r d a n c e with the r e c o m m e n d a t i o n s of the British Ministry of Agriculture, Fisheries and Food (MAFF/ADAS, 1979). A minimum of 25 auger borings were collected in each field, in a ' W ' shape to avoid sampling bias. All trees, electricity pylons and areas of unusual appearance, e.g. tracks or heaps, were avoided. The composite samples were homogenised and carefully subsampled. In addition, a soil profile was sampled from Five Barrows (grid ref. 734 369) by digging a pit and removing samples at 5 cm intervals with a plastic scoop. Soils were oven-dried at 50.1 25-1- 50-0 20.1 - 25.0
15.1 - 2 0 , 0 10.1-15-0
/I Fault
