GEOLOGICA CARPATHICA, 52, 3, BRATISLAVA, JUNE 2001
139—146
LOW TEMPERATURE ABIOGENIC SYNTHESIS OF DOLOMITE
JÁN BABČAN
1
and JAROSLAV ŠEVC
2*
1
Department of Geochemistry, Faculty of Science, Comenius University, Mlynská dolina, 842 15 Bratislava, Slovak Republic
2
Geological Institute, Faculty of Science, Comenius University, Mlynská dolina, 842 15 Bratislava, Slovak Republic
(Manuscript received September 8, 2000; accepted in revised form March 15, 2001)
Abstract: Endowed with the theoretical knowledge of carbonate evaporation and the precipitation of CaMg carbon-
ates from aqueous solutions the authors systematically researched the low temperature formation of dolomite-like
carbonates. The results obtained at 40 °C indicate that the formation of rhombohedral CaMg carbonates depends on
the solid matter/water ratio (solidus index Is). The higher the ratio, the more MgCO
3
enters into the carbonate struc-
ture. The results also confirm a gradual transition of these relationships from calcite, through Mg-calcite, Ca-dolomite
and Q-dolomite (Quasi-dolomite) to Mg-dolomite (Ca
44
Mg
56
). The structure of these synthesised carbonates is disor-
dered, much like their natural counterparts in lagoons, sabkhas, soils etc. When studying Ca-dolomite in the labora-
tory it was found to restructure into stable dolomite at 100 °C and into calcite at 180 °C. Synthetized dolomite (Ca
50
Mg
50
)
does not show superstructural X-ray reflections and hence it was denominated as a Quasi-dolomite (Q-D).
Key words: hydrothermal alterations, synthesis, Q-dolomite, dolomite, Ca-dolomite, Mg-calcite.
Introduction
In 1850 Naumann wrote that Arduino was the first to publish
his genetic considerations of Ca-Mg carbonates, later termed
dolomites. He attributed them to:”... replacement of lime-
stone by agents coming from depth”. It is clear that Arduino
considered that metasomatic reactions were the main agent
responsible for the origin of dolomite. But this was not an
isolated opinion. Sometime earlier Heim (1806, cit. in Nau-
mann 1850) also assumed that dolomite was a limestone
metamorphosed through “vaporous explosions (Dampfexplo-
sionen)”.
The presumption that dolomite developed at relatively
high temperatures, was also supported by experimental mod-
elling results. Von Morlot (1847 cit. in Naumann 1850) was
probably the first to confirm the metasomatic origin of dolo-
mite when he reacted limestone with “bitter salt” solution
(MgSO
4
) at 220 °C. Furthermore, Marignac (1847 cit. in
Naumann 1850) is alleged to have made a successful experi-
ment using chloride MgCl
2
instead of sulphate, but at the
lower temperature of 200 °C.
The view that the secondary origin of dolomite was at the
expense of primary CaCO
3
remained practically unchanged
until the 1960’s. The opinion prevailed that all dolomite
formed via metasomatic replacement of calcite or aragonite.
In the light of increased experimental results, views as to the
mechanisms of this process have shifted away from assum-
ing that it occurs at endogenous – high temperatures to
more recently the assumption that it can also take place at
low temperatures. However the experiments ruled out the
possibility that it could occur below 100 °C which once
again contradicts obervations made in nature of mineral asso-
ciations with dolomite.
A breakthrough came with the recent finding of, dolomite-
like minerals in littoral sabkhas, lagoons and salt lakes. Until
then, attempts to produce dolomite under conditions either
corresponding to, or similar to sabkhas were unsuccessful.
The only exceptions were Oppenheimer’s experiments (1964
cit. in Hsü 1967) which succeeded in changing natural Mg-
calcite into dolomite in a synthetic marine environment with
the addition of pepton, ferrous phosphate, a hydrous solution
of peat and quartz. The temperatures during this experiment
ranged between 22 and 25 °C. We note that this genetic mod-
el of dolomite was not generally accepted and most authors
did not refer to it when dealing with the dolomite problem.
Usdowski (1989) was the only experimentor into synthetized
dolomite at temperatures below 100 °C (i.e. 60 and 90 °C).
The genesis of dolomite is a geological problem which has
been referred as the “dolomite question” or “ the dolomite
problem” since the mid 19th century, when Sorby (1856 cit.
in Hardie 1987) found dolomite in unconsolidated sediments.
The key point to this question was formulated by Friedman
& Sanders (1967) and it persisted for a long time in its origi-
nal form: “to reliably explain the origin of dolomite and do-
lomitic rocks one can neither use the comparison with natu-
ral dolomite, nor the experiments”.
It would take a lot of time and space to synthesize only a
fraction of previous reviews on the dolomite problem. Such
reviews appear from time to time in the literature and we
shall mention some studies referring to this problem in the
discussion.
During our studies we recorded evidence of dolomite for-
mation produced through low temperature reactions. The
partial results were interpreted and presented at the XVII In-
ternational Geological Congress in Moscow (Babčan 1984)
and in this paper we present the complete data and other
*Corresponding author: kgee@fns.uniba.sk
140 BABČAN and ŠEVC
details about the low temperature abiogenic synthesis of do-
lomite.
Experiments
Experimental procedures and identification of reaction products
We used varied experimental procedures and therefore de-
tailed descriptions will be given separately for each group.
The analytical and mineral determination methods of the
products were similar.
In most cases we have characterized the reaction environ-
ment before and after the reactions by using the pH values
measured with a Radelkies pH-meter with a reproducibility
of measurements ±0.05 units. The contents of the basic
chemical components were measured only in solutions (che-
latometrically). We did not determine the chemical composi-
tion of solid reaction products because of their varied miner-
al compositions. The contents of Ca, or Mg in the reaction
products were determined by using the content differences in
input materials and in solutions once the experiments were
over. The CaCO
3
and MgCO
3
contents in synthesized dolo-
mite were tentatively determined using the Lumsden’s
(1979) principle, which was extended to include Mg-calcite
and Ca-dolomite.
In agreement with Baron (1960) we consider X-ray diffrac-
tion to be the only suitable method to determine mineral
composition of very fine-grained and heterogeneous carbon-
ate products. X-ray analysis was made on a DRON UM 1 in-
strument using Cu-radiation, Ni-filter, 30—40 kV and an elec-
tric current of 25—30 mA. To ensure the accuracy of the
lattice parameter determinations of the newly formed prod-
ucts, NaCl was used as the internal standard.
Because most natural minerals and synthetic precipitates are
mixtures of various CaMg carbonates, one of the most reliable
methods of determining their CaCO
3
and MgCO
3
content is
X-ray diffractometric analysis and the associated d
10.4
reflec-
tion values. The mol % CaCO
3
content in a rhombohedral
CaMg carbonate is then calculated using Lumsden’s equation
NCaCO
3
= M.d
10.4
+ B
where N = mol % CaCO
3
, d is the (10.4) reflection value in
angström (Å) and M and B are constants whose values are
333.333 and 911.99 respectively. In contrast to Lippmann
(1973), Reeder (1983) reported that the relation between the
lattice parameters and chemical composition is linear only
when the CaCO
3
content ranges between 48 and 57 mol %.
The recent classification of unconsolidated carbonates based
on low magnesia (<4 mol % MgCO
3
) and high magnesia (>5
mol % MgCO
3
) content, is insufficient and too broad. In
agreement with Lippmann’s proposal (1973) we suggest in-
cluding Ca-dolomite as an independent member of the nomen-
clature within a given range of d
10.4
reflexion values. For dolo-
mite, that is Q-dolomite, we outline a field representing the
carbonate compositions ranging within Ca
55—45
Mg
45—55
. For
carbonates with the composition Ca
45—25
Mg
55—75
we employ
the name Mg-dolomite and with the composition Ca
25—5
Mg
75—95
the name Ca-magnesite.
Trigonal CaMg carbonates obtained during our experiments,
and classified using the system proposed by us are mentioned
below (Fig. 1).
Evaporite formation modelling at 50
°C
These experiments were originally aimed at elucidating the
problems of magnesite formation in eastern Slovak gypsum-
anhydrite deposits (Babčan 1980). We ran experiments with
solutions containing 0.4 M CaCl
2
, 0.4 M MgCl
2
, 0.5 M
Na
2
SO
4
and 1 M NaHCO
3
, while the controls were carried out
with the CaCl
2
, or MgCl
2
solutions, separately. The precipita-
tion kinetics were evaluated in relation to the degree of evapo-
ration and to the amounts of Ca, or Mg that entered into reac-
tion products. The procedure included heating all reagents to
50 °C prior to mixing and the addition of NaHCO
3
in its solid
form. The amounts of reacted Ca and Mg were calculated by
using the difference between the amount added to the experi-
ment and the amount remaining in the solution at a given time.
The solid components were separated using filtration and then
submitted to X-ray diffraction analysis, microscopic study and
DTA. The results are listed in Table 1.
In systems containing dissolved CaCl
2
,
the results showed
that gypsum crystallized in a very well ordered form immedi-
ately after the addition of Na
2
SO
4
. X-ray diffraction was sen-
sitive enough to indicate even a small admixture of calcite.
However, after 6 hours the gypsum crystals had totally van-
ished through evaporation. In systems with MgCl
2
the solid
phases did not precipitate unless 25 % of the solution was
evaporated. This process was accompanied by the formation
of nesquehonite. Once the solution gelatinized, hydromagne-
site formed.
Fig. 1. A system proposed to classify metastable rhombohedral CaMg carbonates (d
10.4
spacings values are given in nm
×
10).
3.036 2.888 2.741
l––––––––––––––––––––––––––––––I–––––––––––––––––––––––––––––-I
Calcite Dolomite Magnesite
3.021 2.962 2.903 2.873 2.814 2.756
I–––––––––––I–––––––––––I––––––I–––––––––––I––––––––––-I
Mg-calcite Ca-dolomite Q-dolomite Mg-dolomite Ca-magnesite
LOW TEMPERATURE ABIOGENIC SYNTHESIS OF DOLOMITE 141
In mixed CaCl
2
and MgCl
2
solutions, gypsum co-precipi-
tated with aragonite and as soon as 25 % of the solution was
evaporated, Mg-calcite began to precipitate. A remarkable
feature was that as a progressive amount of Mg reacted, the
d-values of (10.4) reflections in the resulting Mg-calcite
shifted towards the Ca-dolomite field (d
10.4
= 2.970
×
10
—1
nm
to 2.928
×
10
—1
nm).
A study of reactions in the gypsum—MgSO
4
—Na
2
CO
3
systems
at 40
°C
Apart from Cl
—
—SO
4
2—
—CO
3
2—
systems we also investigated
evaporation in other systems, and the results differed consid-
erably. Thus, in a nitrate environment (NaNO
3
was used in-
stead of Na
2
SO
4
) calcite was the main product, while nesque-
honite and brucite were rare. In the presence of NaCl calcite
formed almost alone, while hydromagnesite was sporadic and
magnesite rare. When we used Na
2
B
4
O
7
, calcite was identi-
fied among the products along with an unidentified calcium
borate and hydromagnesite.
Because the above mentioned experiments indicated the
distinct influence of a sulphate environment upon the forma-
tion of Mg-calcite; we carried out a series of experiments us-
ing sulphates – natural gypsum (from Kateřinky near Opa-
va, Czech Republic) and MgSO
4
solution instead of chloride
compounds. In contrast to the mixed chloride-sulphate envi-
ronments, in these environments the Mg-calcite developed
even at 5 °C. The experimental results achieved in the sul-
phate—carbonate systems, were very interesting, because of
their changing pH values.
The basic material was finely powdered gypsum (651 mg)
suspended in 40 ml of 0.1 M MgSO
4
(92.25 mg). One ml of
a 0.75 M Na
2
CO
3
solution with pH 10.69 was added to each
batch of solution. To preserve roughly equal volume and
mass concentrations of MgSO
4
, the volume of water was re-
duced. These systems were then left for 14 days to react at
40 °C. The pH of the resulting suspensions was measured,
the solid components were filtered away and then finally
submitted to X-ray diffraction analysis. We measured the
contents of unreacted Mg in the solution. A summary of the
results is given in Table 2. The sequence of minerals corre-
sponds to visually estimated abundances from X-ray records.
In experiments with high concentrations of Na
2
CO
3
and high
pH, the formation of Q-dolomite was evident.
Thermodynamic presupposition of dolomite formation
Observations of nature clearly show that dolomite associ-
ates with other minerals in the sediments. In the majority of
cases, it is unlikely that this is occurring under elevated tem-
peratures. Dolomite formation should be easily reproduced
in the laboratory but surprisingly there is a lack of experi-
ments confirming this. As stated above, the only low temper-
ature dolomite synthesis achieved at 25 °C was that, report-
ed by Oppenheimer & Master (1965), which, however has,
not been widely accepted.
Apart from natural observations, theoretical calculations
also clearly indicate that it should be easy to produce dolo-
mite in a lab. With this in mind we checked theoretical cal-
culations that confirm this presumption using thermodynam-
CaCl
2
systems
MgCl
2
systems
CaCl
2
+ MgCl
2
systems
% evapo-
ration
% react Ca
products
% react Mg
products
% react Ca
% react Mg
products
0
78.4
G, C
0
0
62.2
9.9
G, A
25
99.5
C, A
13.2
N
96.0
4.5
A, MgC
50
99.9
C, A
83.4
N
89.9
41.9
MgC, A
75
99.9
C, A
83.1
N
85.3
39.9
MgC, A
100
100.0
C, A
87.4
HM,N
99.9
99.8
CaD, A
Table 1: Results of evaporite formation modelling (experiments series No. 742, for more details see text).
A – aragonite, C – calcite, CaD – Ca-dolomite, G – gypsum, HM – hydromagnesite, MgC – Mg-calcite, N – nesquehonite, O – solution without precipitate
exper.
ml
pH in solution
% of reacted
products of
No.
Na
2
CO
3
begin.
end.
Mg
reactions
769 a
0
7.18
7.70
0
G, C
b
1
9.31
8.08
1.51
G, A, C
c
2
9.46
8.09
3.11
G, A, C
d
3
9.52
8.13
0.77
G, A, C
e
4
9.41
8.14
1.91
G, A, C
f
5
9.45
8.14
5.68
C, A, HM
g
6
9.49
8.14
16.51
C, HM, A,evid.of QD
h
7
9.51
8.75
24.16
C, A, HM,evid.of QD
i
8
9.64
8.97
35.75
C, A, HM,evid.of QD
j
9
9.81
8.92
56.86
C, A, HM, N, QD
k
10
9.86
8.96
67.66
C, A, HM, QD
l
11
9.83
9.19
76.38
C, A, HM, QD
m
12
9.90
9.24
86.25
C, A, HM, QD
n
14
10.07
9.82
93.89
C, A, HM, QD
Table 2: Reaction results in systems gypsum—MgSO
4
—Na
2
CO
3
in relation to pH values of the environment (temperature 40 °C, for more
details see text).
Note: Besides of 40
°C we have also experimented at 4, 5, 25 and 100 °C; Abbreviations see Table 2, QD — Quasi-dolomite
142 BABČAN and ŠEVC
ic probability. We calculated equilibrium constant values
(from the Gibbs energies) for 11 reactions postulating either
direct, or indirect syntheses of dolomite (Table 3). The data
used for these thermodynamic calculations were obtained from
the monographic volumes of Garrels & Christ (1965), Nau-
mov et al. (1971) and Me nik (1972).
Three important conclusions can be drawn from these calcu-
lations:
1. The highest equilibrium constant corresponds to reaction
No. 11. We postulate that the input of all components into the
reaction are in ionic form. However, owing to distinct hydrata-
tion properties of both ions, and especially of Mg
2+
the reac-
tion is unrealistic.
2. Most reactions in purely carbonate systems are thermody-
namically improbable, while the systems with sulphate partici-
pation are much more probable.
3. The high thermodynamic probability of systems with sol-
uble neutral molecules (CaCO
3
0
, MgCO
3
0
, MgSO
4
0
) is a re-
markable feature.
In view of these conclusions our experimental work was fo-
cused on creating reaction conditions that would comply with
the criteria stated on pps 140 and 141. A sulphate environment
was established using gypsum and magnesium sulphate (reac-
tions Nos. 8 to 10). To prepare soluble undissociated CaCO
3
0
and MgSO
4
0
molecules, we used our experience from the
evaporite study, especially the preparation of evaporation ex-
periments, when gypsum crystals formed, immediately after
the solutions were mixed and after 6 hours of evaporation the
gypsum was totally replaced by aragonite and Mg-calcite. We
presuppose that dissolved CaCO
3
0
and MgCO
3
0
molecules
form as a transitional product during the formation of crystal-
line CaCO
3
and MgCO
3
.3H
2
O on account of original gypsum
and epsomite.
The first series of experiments conducted using these pre-
conditions brought remarkable results (listed in Table 4). Nev-
ertheless, we assume that they are accidental, because the re-
peated experiments deviated to some extent from the original
ones, especially from those conducted at 40 °C. Many carbon-
ates formed within these systems, but the dolomite reflections
are clearly visible on the X-ray records. The experiments last-
ed 14 days and the repetitions (841 series) took 10 more days.
The results of repeated experiments, made at 40 °C differed,
mainly in the order in which the minerals were represented in
the products.
Experimental results in systems with changeable component
vs. water ratios
In the next series of experiments we changed both the con-
tent of gypsum and the amount of magnesium sulphate and
water. The water appeared to be crucial, because when its
amount in a given system dropped, the d
10.4
reflections of
rhombohedral carbonates shifted toward lower values (more
MgCO
3
, Table 5).
We prepared for testing a mixture of solid basic materials –
gypsum, epsomite (MgSO
4
.7H
2
O) and Na
2
CO
3
in the
amounts corresponding to molal ratios 1:1:2. The mixture was
thoroughly stirred, ground in an agate mortar and constant
amounts were then inserted into glass ampules. Water (as shown
in Table 5) was then added. The ampules were then welded to-
gether and constantly stirred. They were kept for 7 days at a
temperature of 40 °C. After this period the ampules were
opened, and the pH of the suspension and X-ray diffractograms
of the solids were measured. The results are listed in Table 5.
In the next series of experiments we changed the contents
of gypsum and epsomite in the original materials. Calcite
was the only mineral that formed in the presence of pure gyp-
sum at 40 °C in systems similar to those of No. 868 of the ex-
periment series, while hydromagnesite and magnesite formed
in Mg systems.
Number
reaction type
log K
1
2 CaCO
3
+ Mg
2+
→ CaMg(CO
3
)
2
+ Ca
2+
-1.28
2
CaCO
3
+ Mg
2+
+ HCO
3
→ CaMg(CO
3
)
2
+ H
+
-3.25
3
2 CaCO
3
+ MgSO
4
o
→ CaMg(CO
3
)
2
+ CaSO
4
o
-1.08
4
Ca
2+
+ Mg
2+
+ 2 HCO
3
-
→ CaMg(CO
3
)
2
+ 2 H
+
-5.21
5
CaCO
3
+ MgCO
3
→ CaMg(CO
3
)
2
+1.90
6
CaCO
3
o
+ MgCO
3
o
→ CaMg(CO
3
)
2
+8.86
7
2 CaCO
3
o
+ Mg
2+
→ CaMg(CO
3
)
2
+ Ca
2+
+9.04
8
2 CaCO
3
o
+ MgSO
4
→ CaMg(CO
3
)
2
+ CaSO
4
+11.24
9
2 CaCO
3
o
+ MgSO
4
o
+ 2 H
2
O
→ CaMg(CO
3
)
2
+ CaSO
4
.2 H
2
O
+11.42
10
CaSO
4
.2 H
2
O + Mg
2+
+ 2 CO
3
2-
→ CaMg(CO
3
)
2
+ SO
4
2-
+ 2 H
2
O
+10.89
11
Ca
2+
+ Mg
2+
+ 2 CO
3
2-
→ CaMg(CO
3
)
2
+15.45
Table 3: Presumed reaction types leading to formation of dolomite and their equilibrium constants.
Table 4: Experimental results for No. 840 series.
exper.
Temp.
Gypsum
Na
2
CO
3
MgSO
4
H
2
O
pH
products
No.
°C
g
g
g
ml
begin.
end.
a
25
6.15
6.42
1.24
520
9.85
9.14
A,CaD
2.943
b
25
6.15
6.42
2.48
520
9.40
8.85
A, MhK
c
25
6.15
6.42
4.95
520
9.40
8.85
N, A
d
40
6.15
6.42
1.24
520
9.72
9.06
QD
2.894
, HM, A
e
40
6.15
6.42
2.48
520
9.35
8.70
QD
2.877,
A,HM,
f
40
6.15
6.42
4.95
520
9.14
8.60
A, HM, QD
2.894
Note: Abbreviations see Table 2, MhK
— monohydrocalcite CaCO
3
.H
2
O
LOW TEMPERATURE ABIOGENIC SYNTHESIS OF DOLOMITE 143
Hydrothermal alteration of synthetic Ca-dolomite
According to Malinin (1970), calcite has minimal solubility
in water at 150 °C. This information lead us to believe that it
may be a key to the transition problem between unstable
CaMg carbonate and stable minerals. Therefore, we made a
series of tentative experiments with 4 synthetic carbonate mix-
tures of aragonite, Mg-calcite, Ca-dolomite and water em-
placed the glass ampules, sealed and then exposed to elevated
temperatures. After 11 days of exposition at 100 °C and 5 days
of exposition at 180 °C we observed distinct changes in all but
one sample (No. 831e): at 100 °C the Ca-dolomite was re-
placed by Q-dolomite and at 180 °C by calcite, as shows by
the d values of (10.4) reflections in Table 8.
In the experiments in which CaCl
2
was used instead of
gypsum the products were similar to those identified in pre-
vious sulphate—carbonate systems. The used components,
their amounts and reaction results are listed in Table 6.
The experiments corresponding to the results listed in Ta-
bles 5 and 6 confirm that the solid matter/water ratio has a
crucial effect upon the formation of rhombohedral CaMg car-
bonates. Some problems occurred in experiments Nos. 872d
and 872e. Despite being similar, calcite formed in the former
and Mg-dolomite + calcite in the latter. These anomalies oc-
curred in samples with the addition of CaCl
2
in which mini-
mal water was used. This feature developed because the
CaCl
2
, which we used in the reaction contained surplus con-
stitutional water, thus, we had to use a concentrated solution
in which the CaCl
2
content was analytically determined to
exactly measure the amount of water in the reaction mixture.
Experimental results in systems with precise amount of water
in the reaction environment at 40
°C
The results of previous experiments indicated that it is
mainly the amount of water in the environment that influenc-
es the composition of the reaction products. Because not only
the solid substances (CaSO
4
.2H
2
O, MgSO
4
.7H
2
O), but also
solutions (saturated CaCl
2
solution) released water to the re-
actional environment we made calculations for each experi-
ment separately to exactly determine the amounts of water.
The solid matter/water ratio was expressed by solidus index
Is, calculated according to the formula:
solid matter (g)
Is =
×
100
sum water (g)
The results of these experiments are listed in Table 7 and are
shown graphically in Fig. 2. The graph illustrates how the d
values of (10.4) reflections depend on the solidus index of
rhombohedral metastable CaMg carbonates. At the same time
the graph shows how a gradual thickening contributes to the
formation of products from calcite to Mg-dolomite.
Table 5: A review of results No. 868 series.
No of
experim.
ml H
2
O
pH
components
d-values (nm
×10)
of main lines
a
7
8.26
MgC, CaD
2.970, very weak 2.919
b
4
8.18
CaD, MgC
2.922, 2.988
c
2
7.79
CaD, CaD-QD
2.907, 2.943
d
1
undeterm.
CaD, C, HM
2.918, 3.035, 6.27
e
0.5
undeterm.
QD-MgD, MgC, HM
2.870, 2.990, 6.30
Table 6: A review of results No. 872 series.
No. of
CaCl
2
MgSO
4
.7H
2
O
Na
2
CO
3
pH
ml
products d (nm
×10)
experim.
g
g
g
H
2
O
of main lines
a
0.30
0.28
0.24
8.01
6.07
CaD-QD
2.904
b
0.39
0.35
0.30
7.68
3.92
QD
2.896
, MgC
2.993
c
0.48
0.44
0.38
6.83
2.42
QD
2.896
d
0.46
0.41
0.35
undet.
1.11
C
3.044
e
0.46
0.41
0.35
undet.
0.55
MgD
2.854
, C
3.010
Fig. 2. Relationship between d
10.4
reflexions from synthetic rhom-
bohedral CaMg carbonates and solidus index.
144 BABČAN and ŠEVC
Discussion
The results presented in the attached tables, as well as the
diagram of relation solid matter/water clearly indicate that
the entry of magnesium into the structure of rhombohedral
CaMg carbonates not only depends on the substance concen-
tration, but mainly on the water content in the reactional en-
vironment. Provided that a given environment contains
enough Ca, Mg and CO
3
ions, rhombohedral carbonate min-
erals with chemical composition corresponding to calcite
through Mg-calcite, Ca-dolomite, Q-dolomite to Mg-dolo-
mite may form at relatively low temperatures. The formation
of these individual substances at equal incipient concentra-
tions of the mentioned compounds depends on the amount of
available water. The lower the amount of water in relation to
other compounds, the more Mg enters the structure of the
newly formed carbonates. When the solidus index exceeds
65, not only the products chemically corresponding to ideal
Q-dolomite (Ca
50
Mg
50
) are formed, but also the Mg richer
members, such as Mg-dolomite (Ca
45—25
Mg
55—75
).
The mechanism of these processes is not yet fully under-
stood. But a clue to the explanation in agreement with Us-
dowski’s opinion (1994), may be due to the properties of the
magnesium atom, or ion. Obviously, the entry of Mg into the
crystal structures with calcium is prevented by a large num-
ber of water molecules that form a hydration envelope
around the Mg
2+
ion. The values of the heat of hydratation
clearly indicate how big the differences are between the hy-
dration properties of Ca
2+
and Mg
2+
ions. For Ca
2+
it is 1510
kJ.mol
—1
, for Mg
2+
1828 kJ.mol
—1
(Gažo et al. 1974). It
seems probable that both, high salinity and increased temper-
atures, lower the hydration of Mg
2+
ions to enable Mg to en-
ter reactions with Ca, accompanied by the formation of
mixed CaMg structures.
However, there is another important point which we have
already mentioned in relation to thermodynamic calculations
– the formation of nondissociated soluble CaCO
3
0
and
MgCO
3
0
molecules. The transition of a chemical component
from one solid mineral form into another may take place
when transitional members form, such as the above men-
tioned undissociated molecules in the presence of water. As
an example we note the change of gypsum into calcite in our
experiments, when it seems probable that as soon as the gyp-
sum dissolves, the Ca
2+
ions react with available CO
3
2—
ions
to form undissociated CaCO
3
0
molecules, from which first
nuclei and then crystals of calcite form when the concentra-
tion reaches the required level. Should there be MgCO
3
0
molecules that in similar way the mixed CaMg minerals
would form. The possible participation of CaCO
3
0
and
MgCO
3
0
molecules in reactions of carbonate formation are
also cited by Brady et al. (1996) and Pokrovsky (1998).
Several remarkable features applicable to this problem
were recorded during research into the kinetics of CaMg car-
bonate precipitation from water solutions (Babčan et al.
1992). From purely chloride Ca and Mg solutions (with mo-
lal Ca/Mg ratio 3:1) and from mixed chloride-sulphate solu-
tions of a similar composition only calcite precipitated from
two minutes to as much as 1 hour of reaction with CO
3
2—
at
temperatures of 5 °C and 25 °C (the amount of reacted Ca
was 87 to 89 %). After 24 hours the only reaction product
No of experim.
solidus index Is
pH
products
d
10.4
values in nm
×10
868e
106.7
undeterm.
HM, A, MgD-QD, MgC
2.870, 2.990
874b
77.5
undeterm.
QD, MgC
2.881, 2.985
868d
65.4
undeterm.
HM, QD, CaD
2.893, 2.939
874c
52.6
undeterm.
QD
2.899
867d
43.2
7.52
QD, MgC
2.896, 2.976
868c
36.9
7.79
CaD-QD, CaD
2.907, 2.944
868b
19.7
8.18
HM, CaD, MgC
2.922, 2.988
866a
12.7
7.95
A, CaD
2.924, 2.933
874f
9.7
8.30
CaD, MgC
2.936, 2.979
866b
6.2
8.05
CaD, A
2.948
874g
5.6
8.50
CaD, A
2.947
866c
3.1
8.16
CaD-MgC, A
2.960, 2.972
866d
1.6
8.35
MgC, A
2.975
874i
0.9
8.94
MgC, A
2.988
866e
0.4
8.82
MgC, A
2.988
874j
0.2
8.62
MgC, A
3.004
879k
0.04
9.23
C
3.035
Table 7: Experimental results in system with exactly adjusted solid matter/water ratio.
Table 8: The results of hydrothermal alteration of synthetic Ca-dolomite and Mg-calcite.
Exp.
components
d
10.4
products
d
10
.
4
products
d
10.4
No.
of orig.sampl.
nm
×10
100 °C
nm
×10
180 °C
nm
×10
831a
A, CaD
2.94
A, D
2.90
A, C
3.05
831b
A, CaD
2.94
A, D
2.89
A, C
3.03
831c
A, MgC
2.97
A, D
2.90
A, C
3.04
831e
A, MgC
2.96
A, CaD
2.92
A, CaD
2.91
Note: Abbreviations see Table 2
Note: Abbreviations see Table 2
LOW TEMPERATURE ABIOGENIC SYNTHESIS OF DOLOMITE 145
was Mg-calcite (with 8 mol % MgCO
3
), while the original
calcite completely vanished. We cannot assume that Mg en-
tered the solid structure of the original calcite, but we can as-
sume that the original calcite dissolved and then reacted in
the form of CaCO
3
0
with available MgCO
3
0
. The MgCO
3
0
molecules obviously form with a slow reaction time as the
crystals with a carbonate Mg structure first appeared after 10
days. These kinetic experiments enabled us to see that to pro-
duce Mg-calcite the amount of Mg entering its structure de-
pended on temperature. At 25 °C it was on average 8 mol %
and at 50 °C it was already almost 20 mol %, although, these
values indicate that the setting is equilibrated, as suggested
before by Berner (1974 cit. in Drever 1982).
Another important pre-condition of CaMg carbonate for-
mation is the pH value in the environment, which ranged in
our experiments between 7.5 and 10. In an environment with
a high salt content the dissociation of adequate compounds is
reduced to correspond to pH values lower than those listed in
Table 7. In dense systems (Is = 50—100) the pH of the envi-
ronment could not be measured.
Despite finding that some experiments made at 25 °C ap-
peared promising (Table 2), we finally decided to use the tem-
perature of 40 °C (±2 °C) because firstly we expected a faster
rate of reactions and secondly, because of similar processes
taking place in nature at some localities (Persian Gulf etc.).
Most of our experimental results simulated the processes
of CaMg carbonate precipitation from various systems, rang-
ing from brines to strongly concentrated brines such as those
of coastal lagoons (Persian Gulf, Bahamas, Qatar, Southern
Australia etc.), sabkhas (Arabian Peninsula), soils (calc-
crusts, dolocrusts) and crusts covering historical buildings
made of limestone.
Several data referring to this problem in the literature
could not be confirmed by our experimental results. For in-
stance, the assumption of De Boer (1977), Folk & Land
(1975), Friedman & Sanders (1967), Hardie (1987), Kukal
(1986), Leeder (1982), Lippmann (1973), Rosen et al.
(1989), Tucker et al. (1990), Usdowski (1989, 1994), that a
high Mg/Ca ratio is necessary to produce carbonates with a
high content of MgCO
3
component could not be confirmed.
In our experiments we obtained the products with various
contents of CaCO
3
and MgCO
3
(Table 7) using an equal ratio
between starting substances. However, we should note that
the Mg/Ca ratio does not characterize given reaction condi-
tions realistically. The Ca/Mg ratio corresponds to the initial
content of Ca and Mg compounds and it changes immediate-
ly after the addition of CO
3
2—
compounds into the system.
Most of the calcium immediately reacts and the formation of
CaCO
3
changes the real reaction environment to a complete-
ly different environment relative to the theoretical, or origi-
nal environment.
To characterize the low temperature reaction environment
we use the solidus index Is, a value that can easily be calculat-
ed during laboratory experiments. For natural systems an ade-
quate variable may be the value of total salinity, as confirmed,
for instance, by the experiments of Erenburg (1961) and Glov-
er & Sippell (1967) who obtained the richer MgCO
3
products
in an environment with higher concentrations of NaCl during
the synthesis of CaMg carbonates. Kazanskij (1976) reports
that in sediments from the Balchaš Sea the Ca-dolomite (ac-
cording to Kazanskij protodolomite) appeared in depressions
in which the salinity of water was 4.6—5.3 g/kg.
In the highly concentrated coastal lagoons, sabkhas etc.,
no environments with individual Ca
2+,
Mg
2+,
or CO
3
2—
ions
are expected to occur. They are present as undissociated mol-
ecules, molecule associates, hydrates etc. Therefore the cal-
culations postulated by several authors, such as solubility
calculations, equilibrium constants etc., cannot be made. For
instance we do not use values of ionic strengths for the char-
acterization of a medium.
Although, there are indications to explain the nature of this
problem, the question of how the unstable CaMg carbonates
change into stable ones with ordered structure, still remains
unanswered. Some of specialists assume that the mentioned
change is a matter of time (Purser et al. 1994). However,
there is evidence that the unstable Mg-calcite with 20 to 40
mol % MgCO
3
remained unchanged in the Tertiary lime-
stones of the Swiss Jura (Kübler 1958). On the other hand,
another remarkable finding was the well ordered dolomite in
the Coorong Holocene sediments (Rosen et al. 1989). Even
more interesting was the find of stoichiometric transparent
dolomite (besides Ca-dolomite) within the weathering crusts
(patinas) on limestones used in historic buildings (Rodriguer-
navarro et al. 1997)
Remarkable there have also been results from experimen-
tal research into the recrystallization kinetics of synthetic
CaMg carbonates (with 41.7 mol % MgCO
3
, in our terminol-
ogy carbonates at the boundary between Ca-dolomite and Q-
dolomite) at 50—200 °C reported by Malone et al. (1996). In
the seawater-like solution a complete restructuring was not
achieved even at 200 °C (a product developed with a maxi-
mum of 48.6 mol % MgCO
3
). After 336 days some 30 % of
the carbonate recrystallized at a temperature of 50 °C to be-
come dolomite.
In addition we note that repeated X-ray analyses of ran-
domly selected samples, stored in a dry place before the anal-
ysis, did not show any trace of changes even in samples
stored for 18—20 years. It seems that a transport medium (wa-
ter) is inevitably required to recrystallise unstable CaMg car-
bonates. If the unstable CaMg carbonate structures remain
unchanged in the rocks for a long time (as in the case of
Swiss Jurassic limestones), it may be that the original sedi-
ments were quickly lithified. As a result water is prevented
from migration and various exchange reactions, including re-
structuring, could not take place.
Conclusions
On the basis of thermodynamical calculation and systematic
studies of carbonate formation from various environments a do-
lomite-like carbonate was prepared. Synthetized dolomite
(Ca
50
Mg
50
) does not show superstructural X-ray reflections and
hence it was denominated as a Quasi-dolomite (Q-dolomite).
The results obtained at 40 °C indicate that the formation of
rhombohedral CaMg carbonates depends on the solid matter/
146 BABČAN and ŠEVC
water ratio (solidus index Is). The higher the ratio, the more
MgCO
3
enters into the carbonate structure. The results also
confirm a gradual transition of these relationships from cal-
cite, through Mg-calcite, Ca-dolomite and Q-dolomite (Qua-
si-dolomite) to Mg-dolomite (Ca
44
Mg
56
). When studying Ca-
dolomite in the laboratory it was found to restructure into
stable dolomite at 100 °C and into calcite at 180 °C.
References
Babčan J. 1980: Die theoretische und experimentelle Modelierung
des Entstehung von Magnesit in Evaporiten des Westkarpa-
then. Geol. Zbor. Geol. Carpath. 31, 65—74.
Babčan J. 1984: Low temperature synthesis of dolomite. Abstr. 27-
th Internat. Geol. Congress, Moskva, V, 10.
Babčan J., Iró S. & Ševc J. 1992: Experimental study of carbonate
formation kinetics in a Ca-Mg-CO
3
-Cl systems at temperature
of 25 °C. Miner. Slovaca 24, 1—7.
Baron G. 1960: Sur la synthese de la dolomie. Application au phe-
nomene de dolomitisation. Rev. Inst. Franç. Pétrol. Ann. Com-
bust. Liquides 15, 3—68.
Brady P.V., Krumhansl J.L. & Papenguth I.W. 1996: Surface com-
plexation clues to dolomite growth. Geochim. Cosmochim.
Acta 60, 727—731.
De Boer R.B. 1977: Stability of Ca-Mg carbonates. Geochim. Cos-
mochim. Acta 41, 265—270.
Drever J.I. 1982: The geochemistry of natural waters. Prentice-Hall,
Englewood Cliffs, New Jersey, 1—440.
Erenburg G.B. 1961: Artificial mixed carbonates in the CaCO
3
-
MgCO
3
series. Žurn. Strukt. Chimii 2, 178—182.
Folk R.L. & Land L.S. 1975: Mg/Ca ratio and salinity, two controls
over crystallization of dolomite. Bull. Amer. Assoc. Petrol.
Geol. 59, 60—68 (cit. in Hardie 1987).
Friedman G.M. & Sanders J.E 1967: Origin and occurence of dolos-
tones. Carbonate rocks, Part A. Elsevier, Amsterdam, 167—348.
Garrels R.M. & Christ C.L. 1965: Solutions, Minerals, and Equilib-
ria. Harper and Row, New York, 1—450.
Gažo J. and col. 1974: General and inorganic chemistry. 2-nd Edit.,
Alfa Bratislava—SNTL, Praha, 1—808 (in Slovak).
Glover E.D. & Sippell R.F. 1967: Synthesis of magnesium calcite.
Geochim. Cosmochim. Acta 31, 603—613.
Hardie L.A. 1987: Dolomitization: a critical view of some curent
views. J. Sed. Petrology 57, 166—183.
Hsü J.K. 1967: Origin of dolomite in sedimentary sequences. Min-
eral. Dep. 2, 133—138.
Kazanskij Ju.P. 1976: Sedimentology. Izd. Nauka, Novosibirsk, 1—
272 (in Russian).
Kukal Z. 1986: Elements of sedimentology. Academia, Praha, 1—
325 (in Czech).
Kübler B. 1958: Calcites magnesienne d´eau douce dans le Tertiaire
supérieur du Jura neuchatelois. Eclogae Geol. Helv. 51, 676—
685 (Suisse).
Leeder M.R. 1982: Sedimentology, Process and Product. G. Allen
and Univ. London, Boston, Sydney, 1—439.
Lippmann F. 1973: Sedimentary Carbonate Minerals. Springer
Verl., Berlin-Heidelberg-New York, 1—219.
Lumsden D.N. 1979: Discrepancy between thin section and X-ray
estimates of dolomite in limestones. J. Sed. Petrology 49,
429—436.
Malinin S.D. 1970: An application of theory of strong elecrolytes to
the solubulity of carbonates at high-temperature. Geochimia
540—551 (in Russian).
Malone M.J., Baker B.A. & Burns S.J. 1996: Recrystallization of
dolomite – An experimental study from 50—200 degrees.
Geochim. Cosmochim. Acta 60, 2189—2207.
Me nik Ju.P. 1972: Handbook – Thermodynamic constants in the
analysis of conditions for iron ores origin. Naukovaja Dumka,
Kijev, 1—195 (in Russian).
Naumann C.F. 1850: Lehrbuch der Geognosie, Bd. 1. Verl. W. En-
gelmann, Leipzig, 1—1000.
Naumov G.B., Ryženko B.N. & Chodakovskij I.L. 1971: Hand-
book of thermodynamic values. Atomizdat, Moskva, 1—195
(in Russian).
Oppenheimer C.H. & Master I.M. 1965: Transition of silicate and
carbonate crystal structure by photosynthesis and metabolism.
Geol. Soc. Amer. Spec. Pap. 76, 1—125.
Pokrovsky O.S. 1998: Precipitation of calcium and magnesium car-
bonates from homogenous supersaturated solutionns. J. Crystal
Growth 186, 233—239.
Purser B.H., Tucker M.E. & Zenger D.H. 1994: Problems, progress
and future research concerning dolomites and dolomitization.
Dolomites – A volume in honor of Dolomieu. Spec. Publ. No-
21 of the Intern. Assoc. of sedimentologist. Blackwell Sci.
Publ., Oxford etc. 3—20.
Reeder R.J. 1983: Crystallochemistry of trigonal carbonates. In:
Reeder R.J. & Ribbe P.H.(Eds.): Carbonates, mineralogy and
chemistry. Min. Soc. Amer. Rev. Mineralogy 3—61.
Rodriguernavarro C., Sebastian E. & Rodriguergallego M. 1997: An
urban model for dolomite precipitation – authinenic dolomite
on weathering. Sed. Geol. 109, 1—11.
Rosen M.R., Miser D.E., Starcher M.A. & Warren K.J. 1989: For-
mation of dolomite in the Coorong region, South Australia.
Geochim. Cosmochim. Acta 53, 661—669.
Tucker M.E., Wright V.P. & Dicson J.A.D. 1990: Carbonate sedi-
mentology. Blackwell Sci. Publ., Oxford etc., 1—482
Usdowski E. 1989: Synthesis of dolomite and magnesite at 60 °C in
the system Ca-Mg-CO
3
-Cl-H
2
O. Naturwissenschaften 76,
374—375.
Usdowski E. 1994: Synthesis of dolomite and geochemical implica-
tions. Spec. Publ. Int. Assoc. Sed. 21, 345—360.