I 
GEOL. CROAT. SOil 27 - 32 4 Figs. 1 Tab. ZAGREB 1997 

Strontium Dependence of the Lattice Constants of Barites from 
the Kresevo Area in Central Bosnia (Bosnia and Herzegovina) 

Dragutin SLOVENEC', Dubravko SIFTAR " Milko ,JAKSIC2 and Ivan ,JURKOVIC' 

Key words: Strontiulll bearing barites, Latlice con­
stants, Inhomogeneous st rontium distribution in 
barite, XRD determination of Sf in barite, Kresevo, 
Central Bosn ia. 

Abstract 
The jan ice constants of a sy nthetic BaS04 and fourtee n natural 

b'lri[c samples from lhe Kresevo area (Centra! Bosnia) were deter­
mined by X-ray powder diffraction. The natural samp les contained 
from 0.3- [7 .6 mole % 51"S04' and negligible concentrations of Ca and 
Pb «0.03 and 0 .02 wt. %, res pec tiv ely ), thu s rep resen ting almost 
pure (l3a,Sr)S04 solid so lut ions. 

The diffraoion lines of moslnaturaJ samp les were broadened in 
rc!~l1ion 10 those of synthetic I3aS04 • Thi s results from obvious com­
positional heterogeneity, e.g. the heterogeneous distribution of stron­
tium, as is clcarly observed from Sr variation pal(erns across a surface 
of the barite obtained by proton microprobe analysis . Nevertheless, 
the present values of unit-cell parameters arc in agreemen! with those 
of synthetic (13a ,S r)SO_1 so lid solu tion s documented ill the literature. 

It has been proven Ihal the int erplanar spacing drol may be used 
for dete rm ination o f thc Sr co nt cn t of natural barites; the absolute 
error v;lI·ies from ±O.3 mole % SrS04 (for homogcncous samples hav­
ing th e difrracti0l11ine 004 clearly resolved in spcctral doublet com­
ponc nts) up \0 ±O.5 llloic % SrS04 (for the samples with compositio­
nal hcterogeneity having a broadened and rciative poorly defined dif­
fraction linc CXl4) . 

1. INTRODUCTION 

Experiment al studies of the system BaS04 - SrS04 

showed that a comp lete series of solid solution exis ts 
between the orthorombi e rorms of BaS04 and SrS04 , 

i. e. between barite and celest ite (e.g. BOSTROM et a1. , 
1967). However, the natural intermediary members 
between 10 and 90 mole % SrS04 are very rarely obser­
ved. 

Hydrothermal barites from the Kresevo area in Cen­
tral Bosnia contain from 0.1 - 17 .6 mole % SrS04 (i.e. 
from 0.1 to 14.4 wI. % SrS04) as reported by SIFT AR 
(1988) who chemically analysed 33 samples. The wide 

I racllily of Mining, Geology and Petrol eum Engineeri ng, University 
of Zagreb, Pi crottijeva 6, HR-l0CXlO Zagreb, Croatia. 

2 Ruder 13oskovic In stitut e, 13ijenicka ces ta 54. HR- I 0000 Zagreb, 
Croatia. 

Sr concentration range, as well as a relatively large 
number of available samples provided an excellent 
opportunity to examine how the unit ce ll parameters 
depend on Sr content in natural barites. For the purpose 
of the present examination 14 barite samples were used 
(7 samples analysed in SIFT AR, 1988, and 7 new sam­
ples, also analyzed by SIFTAR in this study). In addi­
tion , the unit-cell parameters of a synthetic BaS04 were 
determined which was prepared by sedimentation from 
analytically pure chemicals (H2S04 and BaCl,- H20 , 
with less than 0.05 wt. % Sr). Synthetic BaS04 was 
heated at 900°C for 3 hours in order to obtain larger 
crystallites. The treated sample did not show any mea­
surable broadening of X-ray diffraction lines due to the 
small crystallite size. 

In this paper the presen t data as we ll as the cited 
data of lattice constants and the Miller indices are prc­
sented on the basis or the space group Pnma (62). 

2. BASIC GEOLOGICAL DATA 

The barite bearing Krescvo region is a part of the 
Central Bosnian are Mountains. It is an area about 22 
km long and 2-4 km wide. The oldest rocks are pre­
Devonian metamorphic rocks which grade into 
metasandstones (JOVANOVIC ct aI., 1977). Devonian 
rossiliferous layered or bedded dolomites, limestones or 
marbles overlie this metamorphic complex. Magmatic 
rocks are rcprcsented by Upper Palaeozoic rhyolites and 
melarhyolites which either form sil ls in the metamor­
phic complex, or extrusions on the vergc 01" the Devon­
ian carbonate rocks. Upper Permian con tin en tal and 
lagoonal deposits which unconfo rmably overli e older 
Palaeozoic rocks pass continously into Lower and Mid­
dle Triassic deposits. The vein and metasomatic barite 
deposits of the Kresevo area arc located almost exclu­
sively in the Devonian carbonatc rocks, and as such rep­
resent the largest barite accumuiation in the Palaeozoic 
rocks of the Dinarides. Thc main gangue mineral is 
barite (over 95 wt. %), locally with rhombohedral cal­
ci te, octahedral lluorite and siderite. The main ore min­
eral is mercurian tetrahedrite. Barite contains average 
4.19 wi. % SrS04 (an average or 56 samples). Its aver­
age su lphur isotope composition from 40 analyses is 
+9.48%0 (JURKOVIC, 1987; JURKOVIC et aI., 1995). 



2R 

3. METHODOLOGY 

The following methods were used: X-ray powder 

diffraction (XRO), optical emission spec troscopy 

(OES), chemical gravime tric analysis and proton micro­
probe, part icle induced X-ray emission (PIXE) analysis. 

XRD patterns were produced using a Philips dif­
fractometer, with a proportional coun ter and graphite 

m onoch romator, using CuKa radiation. Uniquely 
indexed diffraction lines 410,32 1,303 and 004, all in 

the region from (-) = 22° to (-) = 26°, were lIsed in calcu­
lation or the parameters of the orthorombic unit-cell. 
The angular scale of the diffractometer was corrected 
by lI sing quartz as the internal standard and its diffrac­
tion line 112 at e ~ 25.070° for CuKa, radiation (the 

corresponding interplanar spacing being d ~ 1.8178 A; 
BORG & SMITH, 1969). The homogenized mixture of 

barite and quartz contained ===15 wI. % quart z. Operat­
ing condi tions were a divergence slit and rece iving slit 
o r 10 ancl 0.2 111111, respectively; scan rate or 1/2 0 per 
minute; chart speed of 10 mOl/minute; a full scale of 

I x 103 cps; and a time constant of J see. Three patterns 
were produced for each sample. The position of the 
quartz diffraction line 112 as well as the barite diffrac­
ti on line 004 for the CuKa 1 radiation (several barite 

samples have the diffraction line 004 resolved in the 
spectral double t components) were measured at 4/5 of 
their he ighl. The posi tions of the non-resolved barite 

diffracti on line 004 (most of the samples) as well as the 

positions of the barite diffraction lines 410,321 and 
303 for the CuKa radiation were meas ured at 1/2 of 
their height. 

Mole% SrS04 

Sample Locality Determined Derived 
No. by DES from dOO4 

BaSO,-synth. <0.1 0.0+0.3 
1 Martinovac 0.3±0.1 O.4±0.3 
2 Tocak 1.4±0.1 1.6±O.4 

3 Radesa 1.6±O.1 0.9±O.4 
4 Cvjetnjak 2.9±0.2 2.5±0.3 
5 Ravni Vaganac 3.5±0.2 4.2±O.5 

6 Tocak 5.4±0.4 6. 1±0.4 
7 Martinovac 7.0±0.4 6.7±0.5 
8 Gusta Suma 8.1±0.5 8.6±O.4 
9 Gusta Suma 8.6±O.5 8.3±0.4 
10 Dubrave-Dugi Dol 8.7±0.5 8.1 ±0.3 
11 Donjr Martrnovac 9.6±0.5 8.8 ±0.5 
12 Brjele Jame 12.8±O.7 12.4±0.4 
13 Bijele Stijene 15.7±O.8 15.5±O.4 
14 Kolovoje 17.6±0.9 18.3±0.3 

Gcologkl Croatica 50/1 

Barite samples prepared for chemical analysis were 
proven free of XRD detectable impurities. Any negligi ­
ble calcite and dolomite admixtures were removed by 

acid treatment. 
T he purity control of' the selected barite samples by 

OES showed numerous Ba lines in all samples, several 
strong or weak Sr lines depending on the s trontium con­
tent , and some weak lines of ea and Pb in all samples. 
The chemical analysis determined all the samples as 
su lphates. Due to the purity of the samples the quantita­
tive cstimation of barium and sulphate was not neces­

sary. 
The strontium content was dctermined by OES, 

using barium as an internal standard and measuring the 
intensi ty ratios of spectral lines in selected line pairs, 
espccially Srn 338.07 nm - Ban 277.14 nm and Sri 
335.l3 nm - Bal 335.68 nm (S IFfAR, 1975). For the 

purpose of check ing the analytical data from SIFf AR 
(1988) and [or new samples, four selected barite sam­
ples with s trontium contents highe r than 7 mole % were 
analyzed aner dilution with equal amounts of pure bari­
till} sulphate containing less than 0.1 wt. % of strontium 
sulphate. The analyses wcre performed by OES, and by 
a chemical gravimc tric method involving the precipita­
tion of barium as bariulll chromate from a homoge­
neous solution, with subsequent gravimetric determina­
tion of Sr as stront ium sulphate (following the method 
of GORDON et ai. , 1959). Thi s chemical mcthod is 
vcry time-consuming and not suitable [or routine analy­
sis, howeve r the results arc somewhat better than by 
OES. Every samplc in the present study was analyscd 
three times by OES and the above mentioned samples 
also by the chcmical method. 

d""IA alA blA ciA 

1.7890(2) 8.879(2) 5.456(1 ) 7.156(1) 
1.7887(2) 8.879(1 ) 5.454(2) 7.155(1) 
1.7878(3) 8.878(3) 5.455(2) 7.151 (1) 
1.7883(3) 8.875(3) 5.453(1 ) 7.153(1 ) 
1.7871 (2) 8.886(2) 5.455(1 ) 7.148(1 ) 
1.7858(4) 8.856(2) 5.451 (2) 7.143(2) 
1.7843(3) 8.849(1 ) 5.451 (2) 7. t 37(1) 

1.7838(4) 8.842(2) 5.453(2) 7.135(2) 
1.7823(3) 8.831(1) 5.446(2) 7.129(1) 
1.7826(3) 8.839(3) 5.443(2) 7.130(1 ) 
1.7827(2) 8.834(1 ) 5.452(2) 7.131 (1) 
1.7822(4) 8.825(2) 5.446(1 ) 7.129(2) 
1.7794(3) 8.803(1 ) 5.448(1 ) 7.1 t8(1) 
1.7770(3) 8.795(3) 5.443(2) 7.108(1 ) 
1.7748(2) 8.765(2) 5.446(1 ) 7.099(1 ) 

Table I. Locality, slrontilllll content, imerp lanar spacing d(}1)4 and la llice constants of natural barites from the Kresevo area and the synthetic 
BaSO.;' Samples 1.3.4,7. II and 14 correspond 10 the samples K 2c, K 12. K 19, K 2<1, K 3 and K 26b (SIFTAR, 1988). respectively. Sam­
ple 9 corresponds on ly !O the fine-grained pari of barite sam pl e K R (S IFTAR, 198R). The SI" contcnt of the samples 8, 11, 13 and 14 was 
also delennined by the chemical £l"llvim c!ric method which gave the fotlowin£ resu lt s (mole % SrS0 4 ): 8.0tO.3, 9A±OA, IS.6tOA, 
17.S ±O.S, respectively. 



Slovcncc. Silbr. J:lksic & Jurkovic: Strontium Dcpendence ofthc ~lIIicc Constants of Bariles ... 29 

" !2 
<:> v 
<:> 

j 
OJ ..., 

J '" ..., 
~ 
'" Q 

b 

c 

IxlO Jcps 

26 25 24 23 22 
8/', CuK" 

Fig. I XRD pa1!erns of lhe sampl es, in the reg ion from 22" 10 26° 
(0): a - syn lhel ic 13aSO~; b - samp le 14; C - sample 7. The angular 
position of lhe quartz diffraction li ne 11 2 (used fo r the angular 
scale c<ll ibration) is show n by lhe arrow . 

Calcium and lead conte nts were determ ined by OES 
comparing the in tensities of ea and Pb li nes with those 
in the s tandards containing know n amou nts o f lhese 

clements. 
Two bari te samples were analysed lIs ing the foclised 

proton mi c robeam and PJXE spectroscopy. The mea­
surements were made using a 3MeY proton beam from 

the Tandcm Van de Graarf accele rator in the Ruder 
Boskovi c In s titute. The nuc lea r microp robe fac ilit y 
desc ri bed e lsewhere (lAKSIC e t aI., 1993; RYAN & 
GR IFrIN, 1993) foc used the beam to a spatial resolu­
tion oj' 5 pill with a beam current of 100 pA. The beam 
was scanned linearly across the sample surface. C harac­

teris tic X- rays were de tec ted using the semiconductor 
Si( Li ) detec tor. From the total PJXE spectrum , counts 
be longing to a particular X-ray line were recorded 
usi ng thc mu ltipa rame te r data acqu is ition sys tem. The 
pos it ion or the beam at the time of an event is a lso 

recorded. Data were the n displayed as a dis tribution of 
X- ray line in tens it y across the particular line, for each 
or the lines observed ill spectrum. In thi s a rt ic le only 

the s tron tium La line resul ts were present ed due to the 
lowe r sampling depth (5 pm) w hen compared to the 
strontiulll Ka line. 

4. RESULTS AND DISCUSSION 

Locality, Sr-contcnt, inte rplanar spacing d(YJ.) ancllat­
ti ce constants or fourteen natural bar ites as weI! as the 

data fo r synthet ic BaS04 arc g ivcn in Table 1. The nat­
ura l samples contained from 0.3-17.6 mo le % SrS04 • 

The fract ions o f Ca and Pb, which can a lso replace Ba 
in the c rysta l s truc lUre, were negl ig ible; the samples 
contained less than 0.03 wt. % Ca and less than 0.02 wt. 
% Pb, thus representing almos t pure (Ba,Sr)SO<) solid 
soluti ons. 

Diffrac ti on lines of some natural samples were 
almost as sharp as those of the sy nthe tic BaSO" (Fig. 
la , b). However, diffraction lines oj" other samples were 
not so well-defined, and the lines or samplc 7 were sub­
stantially broadened in rclation to synthet ic BaS04 (Fig. 
Ic). The widths at half maximum or the peak inte nsity 

were measured for s tron g diffraction lines 002 at 8 = 
12.5° and 2 11 at 8 = 14.5° of all the samples, and com­
pared with those of synthet ic BaSO,) . It was found that 
the w idths of synthetic BaSO" and samples I , 8, 10, 13 
and 14 were s imilar. The widt hs of diffraction lines o f 
sample 7 were the biggest (about 30% bigger than those 
of synthetic BaS04), and the widths o f d iffraction lines 
of othe r sampl es were up to 10% bigger than those of 
sy nthe ti c BaS04 • The observed broadeni ng m ay be 
attributed to composi ti onal heterogene ity (e .g . to the 
inhomogeneous distribution or Sr). H yd rothe rmal 
barites may even reveal mi c rohc terogene ity (e.g. 
LEEDER et a I., 1983). Conseque ntly, c rys tallites or 
the ir parts, in a given XRD powdered sample may have 
different concentrations of Sr, and thus have s lightl y 
different lattice constants caus ing the broadening of dif­
frac tion lines. 

Profiles of Sr de termined by PIXE along li ncar tra­
verses (Fig. 2) show cons iderable composi tio na l 
c hanges in sample 7 (the broadening o f difrrac tion lines 
is about 30%), and lower observed he tcrogeneity in 
sample 9 (the broade ning of diffraction lines is about 
6%). A polished surface of the fine-grained barite sam ­
pic 9 (bari te g rains are smalle r than 0.6 111m in diame­

te r) was profiled by PIXE a long an arbitrary linear tra­
verse. The cleavage parti c le of tabul a r barit e 7 was 
a nalysed; profiles shown in Fi g. 2 were take n on the 
poli shed cleavage plane (001) alon g the s traight line 
whic h bisec ts a sha rp angle (=78°) between c leavage 
planes 1210 ), e.g. along [010]. Profiles of Sr were a lso 
taken on the cleavage plane (2 10) along the s trai ght line 
perpcndicular to [00 I J. These profiles show a relatively 
homogeneous d istr ibution o f SI'. The obscrved features 
o f sam pl e 7 indicate the zonatio n w ith di scontinuity 
planes paralle l to 12!o I. 

The un it -cell parame te rs of a g iven sample wcre 
obtained as follows. Firstl y, the parameter c was deter­
mined from the interplanar spac ing dQ04. Nex t the para­
meter a was found from the inlerpl anar spac ing d3fJJ and 
the already obtaincd parameter c. The parameter b was 
then calc ulated using the previously determined val ues 
or a and c and the interplanar spacings dJ 10 (b , was cal­
c ulated) and dm (b2 was calc ul ated); b (th e value in 
Tab le I) = (h, + 2"2)/3. 

The dependence of the unit -ce ll parameters or the 
Kresevo barites on the s trontiulll con te nt is s hown in 



10 Geologia CrO:llica 50/1 

4~-----~ 4~-----~ 4~-----~ 

0 

0 200 400 600 800 

Distonce I flm 

4 

3 Wlii~1~~~1\\! ~ 
~2 
L 
Vl 

1 

3 

::2 
~ 
;: 
Vl 

1 

o 200 400 600 800 

Distoncefllm 

4~~----~ 

~: - ~~~¥~~y 
;: 
Vl 

3 

~2 
L 
Vl. 

1 

o 200 400 600 800 

Distancefllm 

Fig. 2 Distribution of strontium concentration across lhe surrace of barite 

0 o sample 7 (upper profiles) and sample 9 (lower profiles). On a given sam-

0 200 400 600 800 o 200 400 600 800 pIc the profiles were taken a long straight line, and the distances between 
Distance/pm Oistance/)lIT1 individual profiles were 3 111m. 

Fig. 3, which also shows the values of unit-cell parame­
ters of the natural samples found by WIESER (1982), 
and of synthetic solid solution (Ba,Sr)S04 found by 
BOSTROM et al. (1967) and by BURKHARD (1973), 
as well as the values presented by GOLDISH (1989). It 
can be observed that there are considerable differences 
bet ween the present values and those of the natural 
samples obtained by WIESER (1982). The present val­
ues arc in agreement with those of BOSTROM et al. 
(1967), and especially with those of GOLDlSH (1989), 
who synthesized (Ba,Sr)S04 crystals with 25, 50 and 75 
mole % SrS04 . From the unil~cell parameters of these 
crystals and those of BaS04 and SrS04 (g iven on PDF 
ca rd 24-1035 and PDF ca rd 5-593, respect ively), and 
taking into account deviations of unit -cell parameters 
from linear relations (deviations from Vegard's law), 
GOLDISH (1989) derived unit-cell parameters for 
intermediate compositions (between 0. 1 and 0.9 mole 
fraction) at intervals of 0.1 mole fraction . 

However, Fig. 3 shows that the present va lues of the 
unit-cell parameters arc rather different from those of 

all. 
0 

8.90 
blA \ 

\ 
\ 

\ 

" • \0 
8.85 \ 5.50 

\ 0 
\ 0 ---• 0 --~ 
~o ---0 

'0, 0 0 
OJ' 8.80 . \ 545 0'. \ 

\ 
\ 

• Q 
\ 

8.75 5.40 
0 10 20 0 10 20 

Mole %5r50, Mole %5r504 

synthetic crystals obtained by BURKHARD (1973). 
For example the difference between the c parameters 
for pure BaS04 is 0.015 A, or 0.21 %; i.e. it is greater 
than the experiment,,, error. BURKHARD (1973) deter­

mined the parameters using single crystals, the parame­
ter b from precession diffraction patterns, and parame­
ters a and c from Weissenberg diffraction patterns, fol ­

lowing the method proposed by CHRIST (1956). As a 
hi gh accuracy was c laimed for CHRIST's (1956) 
method, the parameter c was in the present work deter­
mined by two addi tiona l procedures described in the 
fo llowing paragraph. 

Sample 1 was very close to pure BaS04, containing 
only 0.3 mole % SrSO, (Table I). A single crystal of 
this sample, having the planes [DOl) parallcl to its sur­

face (approx. 7x5 mm 2
), was used to detect a series of 

sharp diffraction lines 001 at high Bragg angles (8)50°) 
for the rad iations CuKa p Ka2 and Kf31 by means of the 
counter diffractometer wit hout the monochromator. 

From the observed Bragg angles approximate values of 
the c parameter were fou nd. In order to minim ize sys-

ciA 

7.20 

7. 10 

705 
0 

0 

, 
''Q , 

' 00 ,0 
,0 

'u , , 
"-

• 0, 

10 20 
Mole %5rS04 

Fig. 3 Depcndcncc of the 
unit -cell parametcrs {/, /; 
and c on Sr content in so­
lid solutions (Ba,Sr)SO.j : 
empty circlcs - natural 
samples (WIESER, 1982); 
cmpty squares and dashed 
lines - synthetic crystals 
(BURKHARD, 1973); full 
lines - synthctic crystals 
(BOSTROM ct aI. , 1967); 
crosses - thc values pre­
sented by GOLDtSH 
(1989); full circles - nat ­
ural samplcs (from Illis 
study) . 



Slovcm;c, Siftar, Jaksic & Jurkovic: Strontiuml)epenclcncc of Ihe Lattice Constanls of Baritcs .. . J I 

1.785 
• 

1.780 

1775 

o 5 10 15 20 
Mole 'Io5r504 

Fi g. 4 The values of interplanar spacing (/001 vs. composition: full cir­
c les - synthetic BaS04 and natural samples (from this study): 
crosses - the values derived from the data presented by GOLD­
[SI-[ (1989). The straight linc is filled to the values derived from 
thc data presellted by GOLDlSH (1989). 

temmic errors, a graphical extrapolation toward e:::: 90° 

was applied using the function cose·ctge. A value of 
7.155 A. was obtained for parameter c, being ~qual to 
the value given in Table 1, but derived from the inter­
planar spacing dOO4 ' Using the angular separation of 
neighbouring diffraction lines 001 and following the 
method proposed by POPOVIC (1973), in which most 
of systematic errors are minimized, a value of 7. 154 A. 
was obtained from four pairs of diffraction lines having 
e> SOO. Besides, it should be stressed that the present 
values of synthetic BaS04 (Table 1) are in very good 
agreement with those for BaSO" given on the PDF card 
24- 1035. 

There are several methods or Sr determination by 
XRD in barites based on measurement of particular 
interplanar spacings. For instance, the d2 11 spacing was 
recomillended by GOLDlSH (1989) and the d]02 spac­
ing by OSACAR et al. (1991). The value of the inter­
planar spacing dOQ..t (quartz was used as internal standard 
for the angular scale calibration) tested on the samples 
from Krcsevo area also appears satisfactory in the 
determination of Sr in natural barites . 

On the PDF card 24 - 1035 the relative intensity or 
the diffraction line 004 is only 4% , and this line is 
therefore considered as unusable. However, barite has 
exccllcnt cleavage along 100 I j, and the powdered sam­
ple in the specimen holder has a high degree of prefer­
entially oriented grains. As a result the intensity of the 

diffraction line on the pattern obtained by the diffrac­
tometer will be several magnitudes higher than that 
according to PDF data, especially if the surface of the 
sample is well "polished". The XRD patterns of three 
samples prepared in this way are shown in Fig. 1. In the 
range from 0 to ::=20 mole % SrS04 (the range investi­
gated in this study) the angular distance in e between 
the diffraction line 004 of barite and the 112 diffraction 
line of quartz is smaller than 0.8°, which practically 
excludes systematic intstrumental errors. Besides, the 
112 diffraction line of quartz as well as the diffraction 
line of relatively homogeneous barite are resolved in a 
spcctral doublet which allows a high accuracy of dif­
fraction angle e measurement. Quartz is a very suitable 
mineral and often occurcs in barite ore samples. 

The dependencc of the interplanar spacing d OO4 of 
barites from the Kresevo area on strontium content is 
shown in Fig. 4, which also shows the correspond ing 
value of synthetic BaSO" (this study), as well as the 
values derived from the data presented by GOLDISH 
(1989). Between Goldish's values we can fit a straight 
line which deviates about ±O.OOOOS A, corresponding to 
about 0.06 mole % SrS04 • The value of synthetic 
BaS04 as well as the values of natural samples plot 
very ncar to this straight line showing (a) a relatively 
good agreement of OES and XRD data, and (b) that the 
diffraction line 004 may be used for detcrmination of 
the Sf content of barites. 

An approximate concentration of Sr in the range 
[rom 0 to ~ 20 mole % SrSO" may by determined from 
the diagram on Fig. 4, or using the equation: d()(J4(A) :::: 
1.78903 - 0.07782 x S (S = Illole fraction SrSO,). 

A homogeneous sample has a diffraction line 004 
clearly resolved in the spectral doublet components, 
and the estimated error of d OO4 is ±0.0002 A, corre­
sponding to an absolute error of about ±0.3 mole % 
SrS04 in Sr determination, according to Fig. 4. Howev­
er, the 004 dirfraction line or most samples from the 
Bosnian deposits is broadened and not so well-defined. 
In these cases the spacing value is measured with an 
error not exceeding ±0.0004 A. . In this case the absolute 
error is ::; ±O.S mole % SrS04 , i.e. from ±OA to ±O.S 
mole % SrSO, Cfable l). 

The values of the Sr content of the Kre.sevo barites 
derived from the XRD data arc compared with thc OES 
data in Table I. 

The described method of XRD determination or Sr 
is similar to that proposed by GOLDISI-I (1989) who 
recommended using diffraction line 211. This very 
sharp diffraction line for pure BaS04 is at e::= 1404° , 
while the diffraction line 004 is at e ~25 . 5°. We 
observed that with the same measurement of accuracy 
of the diffraction angle e the absolute error in mole % 
SrS04 is about two times higher when the diffraction 
line 211 is used, compared to the results obtained by the 
usc of the dirrraction line 004. 

OSACAR et a1. (1991) proposed the strong dillrac­
tion line 102 of barite and halite (Nael) as the internal 
standard (its difTraction line II J), for XRD determina-



32 

l ion of Sr in Ihe range from 0 10 J 2 mo le % SrS04 • 

Accord ing 10 Osaca!" c t aI. , thc absolutc erro r is ±O. 5 
mole % SrS04 in Sr determination. The recording time 
for one sample is 10 mi nutes . The advan tage of this 
me thod is the recovery of the ori ginal sample aft er the 
analys is by dissolving the Nael in water. The main dis­
advantage of this met hod is that the strongest line of 
quartz is part ial ly superimposed with the 102 di ffrac­
tion li nc of barite. 

5. CONC LUSION 

The proposed method of S r determ inat ion in barites 
may be applied as rout ine analys is s ince it provides 
rapid detenni nat ion for a set of samples. In the cases of 
considerable sa mple heterogenei ty (sec sample 7) the 
average concentrati ons of Sr obt ai ned by X RD and 
chem ical ana lysis arc comparable. The method aids in 
the tes ting of poss ible rough errors obtained by diffcr­
ent chem ical analyses. Moreover, the comparison of the 
w idt hs of corres ponding d iffracti on lines for a g iven 
sample and pu re BaS04 may give informat ion about the 
re lat ive inhomogene ity of Sr distribution in the sample. 

Different pan s of even very small hyd roth erm al 
barite fragments may have very differe nt Sr contents. 
Due to th is it is oj" essential importance that the same 
homogenized sample is used for different types of 
ana lyses. In contrast, d iscrepancies can occur in the 
results of complementary analyses. 

4. REFERENCES 

BORG. l. Y. & SM ITH, D.K. (1969): Calculatcd X-ray 
powdcr patterns fo r s ilicate mi nerals.- Geological 
Society of Amcrica, Boulder, Colorado, 896 p. 

BOSTROM, K. , FRAZER, J. & BLANKENBURG, J. 
(1967): Subsolidu s phase relations and lattice con­
stant s in the sys tcm BaS04-S rS04-PbS04.- Arki v 
Mineral. Geol. , 4, 477-485. 

BU RKHARD, A. (1973): Optische und rbntgenogra­
ph ischc Untersuchungen am System 3aSO'I-SrS04 
(Ba ryt·Coeles tin).- Schweiz. Min. Pctr. Mitt. , 53, 
185- 197. 

CHR IST, c.L. (1956): Precision dete rmination of lat­
ticc constants of s ingle c rystals us ing the conven­
tional Weissenbcrg camcra.- Amer. Miner., 4 1, 569-
580. 

GOLDIS H, E. ( 1989): X·ray diffrac tion analysis of bar· 
ium-stron tiu m sulfate (bari te-ce lestite) so li d sol u­
tions.· Pow. Dirr. , 4, 214·2 16. 

Geologia Croatica 5011 

GORDON, L., SALUTSKY, M.L. & WILLARD, H.H. 
( 1959): The precipi tat ion from homogeneous solu­
ti on.- Wiley, New York, 187 p. 

JAKS IC, M. , KUKEC, L. & YA LKOYIC, Y. ( 1993): 
T he Zag reb nuclear microprobe faei lity. - Nucl. 
Instr. and Meth ., B77, 49·5 1. 

JOY ANOYIC, R., MOJ ICEY IC, M., TOKIC, S. & 
ROK IC, L j. (1977): Osnov na geoloska karta 
I: 100.000. Tumac za list Sarajevo K34· 1 (Geo logy 
oj" Sarajevo sheet).- Inst itu\ za geoloska ist razivanja 
Sarajevo i Zavod za inzenjersku gcologiju i hidroge­
ologij u Gradev inskog rak ulteta Sarajevo (1970· 
1974). Say. geol. zavod, Beograd, 52 p. 

JURKOY IC, l. (1987 ): Barite depos its on Mou nt 
Meduv rsje sou th and south-eas t of the town of 
Kresevo, Bosni a.· Geol. vjesnik. 40,3 13-336. 

JURKOYIC, I. . PALI NKAS, L., SLOYENEC, D., 
SIFTAR, S. & PEZDlC, J. (1995): Paragenes is, 
geochemistry and crysta ll ochem ist ry of the ba ri te 
de posits in the Bosnian Sch ist Mounta ins. - EUG-8, 
April 1995, Strasbourg, TERRA Abstracts, 7, 331. 
Strasbourg. 

LEEDER, 0 ., BAUM, H. & HUNG ER . H.-J. (1983): 
Heteroge ner S tro ntiu l11c inbau in B aryl.- Z. geo l. 
Wiss., II . 1137- 11 40. 

OSACAR. M.C., BESTEIRO, J. & GONZALES , J. 
( 1991): X·ray diffraction analysis of strontiu m in 
barites.· Pow. Diff. . 6, 70-73. 

POPOY IC, S. (1973): Un it -ce ll dimension measu re­
ments from pairs of X-ray diffrac tion lines.- 1. App \. 
Crys\. , 6,122- 128. 

RYAN, C.G. & GRIFFIN, W. L. (1993): The nuclea r 
microprobe as a tool in geology and mineral explo­
ra tion.- Nuel. Instr. and Meth., B77. 38 1-389. 

SIFTAR, D. (1975): Determi nation of stront ium in 
barite by spectrochemica l meth od .- Zborni k radova 
Rud arsko-geol osko·na ftn og fakuitcta, 45-49. Zag­
reb. 

S IFTAR , D. ( 1988): The chemical characteris ti cs of 
bari te from some Bosnian deposit s.- Rudarsko­
metalurski zbornik, 35, 75-89, LjUbljana. 

WIESER, T. (1982): Barites and cclcstoba ritcs in th c 
fl ysch or the Polish Carpath ians. - Archi wull1 Miner­
alogicZllc, XXXYlll , 13-27. 

Manuscript received November 11, 1996. 
Revised manuscript accepted April 28, 1997 .