STUDIES OF RADIOAC'IIVE COMPOUNDS:
V-SODDYITEl
D. H. GonuAN,2UniaersityoJ Toronto,Toronl,o,
Canado.
ABSTRACT
New observations are made on the rare uranium silicate mineral soddyite. The formula
5uo3'2sio2'6Hzo
is established as the most plausible. New crystaliographic data are
recorded: orthorhombic, a:8.32, b:ll.2I,
c:18.7I A; space group Fdd.d,.powder data
are presented which are in agreement with cell dimensions. soddyite is optically negative,
2Y:84", a:1.650,8:1.635, t:1.772,X:c,Y:b,Z:a,pleochroic,
X c o l o r l e s s ,p v e r y
pale yellow, Z pale yellow-greenl dispersion negligible, in disagreement with previous obobservations of r)u strong.
Soddyite is a rare hydrous uranium silicate mineral, having an extremely limited geographical distribution. It was first described by
Schoep (1922) from Kasolo, Belgian Congo. In subsequent papers,
Schoep(1923a,b; 1927,1930) gave more detaileddescriptions,and nothing more on it appeared in the mineralogical literature until Gevers,
Partridge & Joubert (1937) reported its occurrence in Namaqualand,
Africa. Namaqualand material was not available to the author. Five
specimensof soddyite from Kasolo served as material for this study.
Puysrcer Pnoprnrrns
Soddyite occurs as idiomorphic crystals often in groups or clusters,
which are sometimes divergent. It also occurs as flat blades or fibres of
the cross-fibre type in fissuresl also massively intergrown with curite.
Larger crystals up to a millimeter in length are generally mosaical, and
usually totally opaque, or have at the most a mere selvageof transparent
material' smaller crystals, which are usually tabular are translucent
and sometimes quite transparent except for a cloudiness towards the
centre. Zoning of the transparent and opaque material, as shown in
Fig. 1, is common.
The color is greenish-yellowfor the cross-fibretype, canary yellow for
the opaque material and amber yellow for the transparent crystals. It is
readily distinguished from the orange-red of curite and the slightly
brownish hue of kasolite. The streak is pale yellow. The transparent servages on opaque crystals have a vitreous to adamantine lustre. The
opaque varieties are sub-vitreousto dull. Massively it is earthy. The hardnessis between 3 and 4 on Mohs' scale.The density of a clean cluster of
crystals weighingapproximately 18 mg. gave 4.70i0.01 as measuredon
I Extracted from a thesis for the
degree of Doctor of philosophy in the university
Toronto.
2 Holding a National Research Council
of Canada Studentship.
386
of
STUDIES OF RADIOACTM
COMPOLINDS: V-'SODDYI1.E
387
elongated in [001] showing zoning of transparent and
Frc. 1. Soddyite. A-Crystal
elongated in 1110] showing murkiness toward the centre'
opaque material. B-Crystal
Both photomicrographs taken with plane polarized light, X112'
the Berman balance. This compares well with Schoep'sdetermination of
4.627 by the pycnometer method. Two cleavageswere noted: (001) perfect, (111) good. No visible fluorescencewas observed from soddyite
under long or short wave ultraviolet excitation.
t"'
Frc2'soddvi
tion(\ : 1'5418A);
:?J:Jl;;'ll1?TH*[:H"i :I,*:L'Ja
X-nev Cnvsrar,looRAPHY
showed a large c period with a strong pseudo-periodof *c' When the Z
factor was subsequently calculated as 3, a tentative explanation of the
388
D. H, GORMAN
Tesrn 1. Sooovrrn-SUo3. 2SiOr.6HrO:X-nav powopn Dare
Orthorhombic,
Fddd;a:8.32, b:11.21, c:lB.7lA: Z:3
d
I
(cu) (meas)
hkl
d)
(calc)
Id
(Cu) (meas)
|
6.28 111 6.28
4 7 9 022 4 . 8 1
467
+ . . ' ) / 1oo+
[1 1 3 4 . 5 6
3 . 8 0 202 3 . 8 0
9
3
10
9
3. 3 6
I rst
J.J/
l22o 3 . 3 4
1 (Cu) d (meas)
(calc)
2.99
2.80
2. 7 1
2.49
234
2.28
2.21
7
5
8
8
1
2
300
2.82
2.73
2.49
2.33
2.28
133
040
224
206
008
226
,
z.zz
Jl
1 (Cu) d (meas)
3
3
r.77t
1.710
1. 3 8 6
2
4
2
1.681
1.654
1606
1.344
1.228
t
.r.JJ/
r.IJJ
3
i
4
1.524
1.s02
1.412
t.142
1.121
r . Jo.)
I.ZIJ
I
a
hhl
I
5
3
5
5
6
d
(calc)
a
(cu) (meas)
I'kl
2.10
205
1.990
1.918
333
153
119
422
.rrJ
1.867 I
]t246
i (Cu) d (meas)
1(Cu)
2.ll
2.04
1 .985
1.9r7
1 867
1.864
d (meas)
1. 0 8 8
1.046
1.022
0.999
0.984
0.946
0 .9 3 1
+
+
2
+
+
0.903
08e2
0.875
0.861
0.833
2
1
0819
0.803
0.9r2
+
0.7e6
pseudo-periodpresenteditself. rt is conjecturedthat there is a threefold
disposition of the heavy uranium atoms or uranium groupings along the
c axis.
The a and 6 lengths were obtained from excellent weissenberg photographs, which also showed clearly orthorhombic symmetry. The space
group proved to be the rather unusual Fd.dd. The extinction conditions
imposed by such a spacegroup are (hkl) ptesent only with lx, h, l, all odcl
or all even; (001), (0&0), (h00), (hk\), (\kl), (h\l) present only with
all indices even and half the sum even. rn Table 2 the single crystal data
are Dresented.
Tenr,B 2. Sooovrrn: X-ney Sruclr Cnvsrl.l. Der,t
8 . 3 2A + . 0 1
1 1. 2 1
18.71
orthorhombic2/m 2/m 2/ m
a i b : c : 0 . 7 4 21i i 7 . 6 6 9
po:Qo:rn:2 249: l. 669: I
space group Fild.d
z:3
A rotation photograph was also taken about [110], the elongation direction of some crystals. The period determined was i.00 A, which is in
good agreement with 6.99 A calculated from o and b. As a further check
STaDIES OF RADIOACTM COMPOUNDS:V-SODDYITE
389
on cell dimensionsand spacegroup the first 17 Iines of the powder photograph were indexed and their spacings calculated. The excellent agreement of measuredand calculatedspacingsmay be seenin Table 1'
CRYSTALLoGRAPHY
MonpuorocrcAl
Schoep has noted in all his papers that soddyite crystals were not
amenable to accurate goniometric work. Nevertheless he was able to
determine the orthorhombic symmetry and an axial ratio oI a:b:c
:0.7959: 1 : 1.6685.This is only in fair agreementwith the axial ratio obtained in this study by r-ray work. Undoubtedly this discrepancy is due
to the poor quality of the crystals which both Schoep and the present
writer have observed.It is evident from the variation in measuredangles
for the form (111) as shown in Table 3 that an axial ratio calculatedfrom
goniometric measurementscannot be as accurate as that calculated from
excellent tr-ray measurements.
Tesr-B 3. Soonvttr:
Moesunnl
eNo Cer,culerrn
Two Crncln Axcr-ns
, r ( 1 1 1- ,)r ( 1 1 1 )
Schoep (1923a)
Schoep (1923b)
Schoep (1930)
Gorman
107"
52" 02',
51" 29',
53'07' (meas)
53'25' (calc)
106' 14' (meas)
106'50' (calc)
Individual crystals of soddyite invariably have an acute dipyramid
form. They show a changein habit from an elongationin [001],through
equidimensional,to a platy type elongated in the [110] direction, but
neverthelessdipyramidal. It appearsas if opposite facesof the dipyramid
have been compressedand the crystal attenuated in [110]' It is this
platy, elongated habit, thought by Schoep (1922) to show a prism form,
lh"t hur led George(1949) and other observersto describethis elongation
as [001].The transition in habit is clearly shown in Fig. 3.
The dominant form is the dipyramid (111) which on some crystals is
terminated by the basal pinacoid (001). Other faces with d values the
same as that of (111) and p values varying ftom74" 25' to 79" 04'were
observed,although their goniometric signalswere poor. These faceswhen
plotted on the gnomonic projection could not be given simple indices,
and the writer hesitates to impute any indices, regardlessof complexity
to faces giving such poor signals, especially when aberration caused by
striations had to be considered.The crystals are heavily striated parallel
to the trace of the basal pinacoid (001). These striations are due in part
390
D. E. GORMAN
to an oscillation between dipyramids with varying p angles,but in the
m a i n t o o s c i l l a t i o nb e t w e e n( 1 1 1 )a n d ( 1 1 1 ) .
The platy crystals lessfrequently exhibit (111) but rather are formed
by dipyramids with p valuesgreater than that of (111). of thesep values
74" 45'occurs most frequently. when the lessplaty types possessthese
possibly vicinal faces, barrel shapeclcrystals result from the continuous
changein p angles.
Frc. 3. soddvite.Showing
habitchange
fromelongation
in [001]to [110];andforms:
p(lll),7 (dipyramid
c(001),
with.t thesameas111,andp varyingfrom 24"25,
to 79"04,).
Schoep(1930) recordsthe forms (113) and (114),neither of which appeared on the score of crystals measured in this study; nor was any indication of such faces seen when myriads of crystals were observed under
the binocular microscooe.
served here. It was dfficult to distinguish between a crystal and a cleavage fragment since the cleavageplanes are equivalent to the prominent
crystal faces,and very few incipient cracks are present.
STUDIES OF RADIOACTM COMPOUNDS:V-SODDYITE
39r
Oprrcer PnopBnrrBs
Some diffi.culty was encountered in determining the optical constants
of soddyite becauseof the murky, pebbly appearanceof the more abundant material. This may account for the discrepancybetween the indices
of refraction originally given by Schoep (1923a) and those recorded by
more recent workers. However, there is good agreementamong the recent workers, and ProfessorSchoepdoes indicate in his own hand on a
copy of his original article in the University of Toronto Library that his
indices are within wide limits. Some of the opticai data as determined by
various observersare given in Table 4.
Tesln
4. Solovrrn:
A Colrpenrsox ol Solrc Orrrc.qr Dlra
Sign
d
Kasolo
Schoep (1923a)
Larsen &
Berman (1934)
George (1949)
Gorman (1952)
Namaqualand
Gevers, etc. (1937)
I .645
1.662
Large
r.7r
t.715
r.7t2
Neg.
Neg.
Neg.
Near 90o
Large
84'(calc)
1.699
N"g.
Near 70'
The correct optical orientation X : c, V :b, Z : a followsfrom the new
morphological setting. Most of the material available was weakly pleochroic: X colorless,Y very pale yellow, Z pale green-yellow.There was a
stronger pleochroism in some of the cross-fibretype. In Becke line tests
no appreciabledispersionwas observed,in contrast with the other authors
listed in Table 4, who report strong dispersionr) z.
CnBursrny AND MoDE oF OccURRENcE
Table 5 showsthree analysesmade by Schoepon soddyite from Kasolo.
Also given are the average of the three analysesaccording to Schoep,the
theoreticalcompositionscalculatedfor two suggestedchemicalformulae
along with the molecular ratios derived therefrom. Schoep chose 12UOs
. 5 S i O z . 1 4 H r Oa s m o r e l i k e l y t h a n 5 U O 3 . 2 S i O r . 6 S i O(za d v o c a t e db y
Wherry, 1922) since the weight percentages calculated for it are in
slightly better agreementwith his averageanalysis.
The weight of the cell contents was calculated from the ceil dimensions
and the measured density. This figure, with the atomic proportions derived from analysis 3 in Table 5led to 3[5UOr'2SiO2'6HrO]as the cell
contents most compatible with the chemical data. The derivation of the
cell contents is presented in Table 6. The calculated Z lactor of 3 is ap-
392
D. H. C,ORMAN
Tesln 5. Cnnurc.q.r,Dere oN Soooyrm (After Schoep)
UOa
SiOz
HrO
8s.53
7.86
85.13 85.79 85.87
7.88
7.6I
7.87
6.16
6.26
86.10
7.50
6.30
86.25
7.30
6.50
0.300 0.300
0.131 0.125
0.350 0.350
99.56 100.00 99.90 100.05
1, 2, 3. Soddyite, Kasolo; anal. Schoep (1930). 4. Schoep'saverage of analyses1, 2 and 3.
5, 6. Schoep's calculated weight percentagesfor 12UO: 5SiOz. 14II:O and 5UO3.2SiOr.
6H2O respectively. 7,8. Schoep's molecular proportions for 5 and 6 respectively.
parently related to the pseudo-period of $ which was previously noted.
The calculated specific gravity of 4.75 compares well with the measured
value of 4.70.
Wildish (1930) has reported the Uo content of soddyite as 43.98/s determined by the emanation method, and 44/6 by the gravimetric method
The material analysed was certainly not soddyite, but quite possibly
kasolite, which has a Uo content of roughly 40/p. These two minerals
are similar in color and not easily distinguished in the hand specimen,
by those not familiar with their crystal morphology.
As noted at the beginning of this paper, soddyite is both rare and
limited in occurrence.It has been reported from only two localities:
Kasolo, in the Belgian Congo of Africa, and the Norrabees pegmatite
area south of the Orange River, Namaqualand, Africa. At Kasolo it
forms a massive intimate mixture with curite. The cross-fibreor crossbladed type always occurs as fissure fillings in the soddyite-curite complex. Euhedral crystals and crystal clusters are found associated with
needlesof sklodowskite on curite, the soddyite seemingly formed after
T,c.sr,n6. Ar,IA,rvsrseno Crrr- CorrnNr ol Sonovrrr,
UOa
SiOr
HzO
8 5 .7 9
7. 6 r
U
Si
6.16
9 9 .5 6
14.94
6.29
5 7. 0 0
15
6
o
14.88
6.26
56.79
HrO
1 8 .5 2
18.60
18
96.45
96.83
I Analysis by Schoep on Kasolo material.
2 Atoms per unit cell. 3. Atoms per unit cell
b a s e do n O : 5 7 . 4 . I d e a l c e l l c o n t e n t f o r 3 [ 5 U O s . 2 S i O r . 6 H r O ] .
STUDIES OF RADIOACTM
COMPOUNDS: V-SODDYITE
393
the curite and before the sklodowskite. Kasolite, the lead bearing hydrous uranium silicate is another associatedmineral.
Schoephas pointed out that there seemsto be an intimate association
of lead and nonlead speciesat Kasolo, for examplecurite-soddyite, curitebecquerelite, and kasolite-soddyite, the lead bearing species having
formed first as an alteration of uraninite. The paragenesisat the Nicholson Mine, Lake Athabaska, Canada appears analogous, although no
soddyite has as yet been found there.
The Namaqualand soddyite as reported by Gevers, Partridge and
quattz,
Joubert (1937) occurs in pegmatites, as a bladed incrustation on
malachite.
with
associated
Socldite was the original name given to this mineral by Schoep in
honor of ProfessorF. Soddy of Oxford University. Billiet (1926) changed
it to soddyite, the changebeing approved by Schoep,who used this more
euphonioussynonym in all his later publications.
Acrrqowr-BoGEMENTS
I wish to record my thanks to Dr. V. B. Meen and Dr. G. Switzer for
the loan of specimens of this rare mineral; to Dr. E. W. Nuffield for
criticisms;and to the National Research Council of Canada, my patron'
for allowing the publication of this work.
Rrnmrrcas
de immersie
Blrrrnr, Y. (1926): Over de bepaling van becquerelite, janthiniet,-van
method van Becke, Natmt'r. Tijd'. Ant.'7,ll2.
Gooncr, D. (1949): Mineralogy of uranium and thorium bearing minerals, U'S'A'E'C',
R.M.O.,563, 195.
GBvnns, T., P,lntnocr, F., & Jounonr, G. (1937): The pegmatite area south of Orange
172'
River, Namaqualand., Mem. Geol'.Surt. South Africa,3l,
LensnN, E. & BnnueN, H. (1934): The microscopic determination of the non-opaque
minerals, [/.S.G.S. Bull. 848' 183.
Scroer, A. (1922): La soddite, nouveau min6ral radioactif, compt. Rend.. Acad. sci'.
Paris, l7O, 1066.
properties optiques,
-1t923a): Sur les lormes des cristeaux de soddite et sur leur
Bull. Soc. Belge.Geol'.Brur.,23r83.
-(1g23b): Les min6raux uranifBre du congo Belge, ertroit ilu Bull,. soc. Belge. GeoI.
Brur.,33, l8l.
-(1927): Kristallografische mededeelingen, kristallen van kasoliet, soddyiet en
brochantiet, Natuur. Tijd. Ant.,9, 1927.
(1930): Les min6raux du gite uranifbre du Katanga, Ann- Musee Congo Bel'ge,
ser.1,1,fasc.2,26.
Turrox, A. (1922): Crystdlography and, praclical mystal measurement,Yol. l. London.
Wnrnnv, E. (1922): Soddite (abstract), Am. Mi'neral',7 , 179.
Wrrotsn, J. (1930): Origin of protactinium, f our. Am. Chem. Soc.,5l' 163'