Content uploaded by Alfonso Mucci
Author content
All content in this area was uploaded by Alfonso Mucci on Jan 02, 2018
Content may be subject to copyright.
Georhimicn er Cosmwhimica Am Vol. 44. pp. 85 to 94
0 Pergamon Press Ltd. 1980. Printed in Great Britain
The solubility of calcite and aragonite in seawater
of 35x0 salinity at 25°C and atmospheric pressure
JOHN W. MORSE, ALFONSO Muccr and FRANK J. MILLERO
Division of Marine and Atmospheric Chemistry, Rosenstiel School of Marine and
Atmospheric Science, University of Miami, 4600 Rickenbacker Causeway, Miami, FL 33149, U.S.A.
(Received 25 April 1979; accepted in reoisedJorm 10 September 1979)
Abstract-The solubilities of synthetic, natural and biogenic aragonite and calcite, in natural seawater of
357;,,, salinity at 25°C and 1 atm pressure, were measured using a closed system technique. Equilibration
times ranged up to several months. The apparent solubility constant determined for c&cite of
4.39( k5.20) x lo-’ moI* kgmZ is in good agreement with other recent solubility measurements and is
constant after 5 days equilibration. When we measured aragonite solubility we observed that it de-
creased with increasing time of equilibration. The value of 6.65( kO.12) x lo-’ mol* kg-‘, determined
for equilibration times in excess of 2 months, is significantly less than that found in other recent
measurements, which employed equilibration times of only a few hours to days. No statistically signifi-
cant difference was found among the synthetic, natural and biogenic material. Solid to solution ratio,
contamination of aragonite with up to 10 wt% calcite and recycling of the aragonite made no statisti~lly
significant difference in solubility when long equilibration times were used.
Measured apparent solubility constants of aragonite and calcite are respectively 22( +3)“/, and
20( f 2)% less than apparent solubility constants calculated from thermodynamic equilibrium constants
and seawater total activity coefficients. These large differences in measured and calculated apparent
solubility constants may be the result of the formation of surface layers of lower solubility than the bulk
solid.
INTRODUCI’ION
THE STUDY of the chemical behavior of calcium carbo-
nate in seawater has long been of major interest to
investigators attempting to understand the oceanic
carbon dioxide system, and the accumulation and dia-
genesis of carbonate minerals in marine sediments.
Although the existence of many complex relationships
among physical, chemical and biological factors has
been clearly demonstrated, a major difficulty has per-
sisted in precisely determining a firm reference point
to which these factors can be clearly related. This
reference point is the solubility of the two most im-
portant calcium carbonate phases in the marine en-
vironment; calcite and aragonite. Until the solubility
problem is resolved, the importance and influence of
other factors will remain ambiguous.
Ideally a classical the~~ynamic calculation
should prove sufficient, where the solubility product is
simply equal to the thermodynamic equilibrium con-
stant.
K, = ~a~+-i=ii- (1)
Ka = =%z+=%~- (2)
K is the thermodynamic equilibrium constant for cal-
cite or aragonite as denoted by the subscript c or a, la
and H indicate the equilibrium activities of calcium or
carbonate for calcite and aragonite, respectively, and
the activity of the solid phase is defined to be unity.
Histori~lly, the major problem has been relating ion
concentrations in seawater to their thermodynamic
activities, particularly when the effects of variable
temperature, pressure and salinity are considered.
Consequently, the problem of determining the solu-
bility of pure calcite or aragonite in seawater is basi-
cally one of knowing the activity of calcium and car-
bonate in seawater under a specified set of conditions.
To avoid the problems associated with the precise
determination of activity coefficients, marine chemists
(e.g. F’YTKOWICZ, 1969) have frequently used apparent
constants based on total ion concentrations. Ideally,
for pure calcite and aragonite:
K: = ii&+ii&- (3)
K: = $ax+=%+- (4)
Where K’ is the apparent solubility constant and m is
the total (free plus complexed) ion concentration in
moi per kg of seawater in equilibrium with the solid
phase. The apparent and thermodynamic constants
should be related in the following ways:
Kf = KC (3
YCd + Ycoi -
K; = K, (6)
YCaJ + Yco: -
Where y is the total ion activity coefficient which
includes the effects of complexing.
The major difficulty with using apparent constants
for the solubility of solid phases in complex electro-
lyte solutions such as seawater is the necessary
assumption of ideal reversibility. This assumption is
simply that the stoichiometry of the dissolution and
precipitation reactions which occur during equilibra-
tion are the same as the bulk solid composition,
85
X6 J. W. MORSE. A. Mucci and F. J. MUERO
There is abundant evidence (e.g. BERNER, 1975) that
this assumption is false for the precipitation of calcite
and aragonite from seawater. Consequently, the
resulting apparent solubilities may not be representi-
tative of the pure bulk solids, but rather of impure
surface phases.
We have undertaken this investigation to determine
the effects of experimental variables on the apparent
solubilities of calcite and aragonite in seawater, and in
an effort to resolve the differences in other recently
determined values for the apparent solubility con-
stants of calcite and aragonite in seawater (MAcIN-
TYKE, 1965: INC~LE rt ul., 1973; BERNER, 1976; PLATH,
1979). This investigation differs from previous investi-
gations in the variety of materials studied, the use of
long equilibration times and its examination of such
factors as solid to solution ratios, calcite contami-
nation of aragonite and recycling of solids. It has been
restricted to determining calcite and aragonite solubi-
lities at 25°C and 1 atm total pressure in filtered
natural seawater of 35”;,,, salinity.
COMP,~RtSON OF RESULTS OF
DIFFERENT INVESTIGATORS
There have been numerous measurements of the
apparent solubility constants of calcite and aragonite
in seawater. The most notable measurements in recent
years of calcite or aragonite solubility in seawater are
those of MACINTYRE (1965), INGLE et ui. (1973), INC~LE
(1975) BERNER (1976) and PLATH (1979). Earlier work
has been reviewed by MAC~NTYRE (1965), EDMOND
(1970) and INCLE et ai. (1973). Relatively pure (greater
than 99?; CaCO,) natural and synthetic, calcites and
aragonites were used in these measurements. Solubili-
ties were reported in terms of the total ion molat or
molar product of calcium and carbonate ions. All of
the measurements, with the exception of INGLE (1975)
were restricted to 1 atm pressure. The following com-
parison of results is restricted to those obtained at
25°C and 1 atm pressure in seawater of approximately
35“,,,, salinity. In order to compare the results, it is
necessary to compute all the solubilities using the
same carbonic acid system apparent constants and
concentration units. The constants used in all calcula-
tions here are those of MEHRRACH et (II. (1973) as refit
by MILLERO (1979).
MACINTYRE (1965) measured caicite and aragonite
solubilities in seawater using both reaction flask and
agitated closed system methods. A major problem in
interpreting his data has resulted from his use of a
partial pressure of carbon dioxide (Pro,) equal to
I atm. This produced equilibrated seawater solutions
having signi~~ntly greater alkalinity and calcium
concentrations than normal seawater. The effect was
equivalent to a change in salinity of approximately
2”,,,,. To correct for this, MACINTYRE (1965) diluted his
starting seawater solutions so that the final salinity
would be approximateiy 35”<,,,. BEN-YAAKOV and
G~LDHABER (1973) noted that the altered composition
of the seawater had a significant effect on the carbonic
acid dissociation constants and solid apparent solu-
bility constants. MACINTYRE’S (1965) experimental
data has been used here to recalculate the apparent
soiubility constants of calcite and aragonite in sea-
water. This was done by using the K; value of MEHR-
BACH et cd. (1973) for seawater of lx”,,,, chlorinity at
25”C, modified by the method of BEN-YAAKOV and
GQLDHABER (1973) for the change in composition due
to increased calcium and alkalinity. The same modiii-
cation technique was used on the resulting K;. and
K; values. The values of k’:. and h’:, are
4.3X(&0.26) x 10~7mo12kg~’ and 7.12(&0.31) x
IO-’ mo12 kg- ‘, respectively. where the value in par-
entheses is the standard deviation.
INC;LE et uf. (1973) determined the solubility of cai-
cite in synthetic seawater using the saturometry
method (WEYL, 1961). The average value of K: is
4.60( f0.10) x lo-’ ’ mol’ kgu2. This value is 5.3”,,
larger than the value of Ki calculated from the data of
MACINTVRE (1965).
BERNER (1976) determined the solubility of arago-
nite in seawater using an open reaction flask main-
tained at a constant P,ol. The average of Ki is
8.21(+0.25) x lo-’ mol’ kg--*, which is 15.3:,;;
greater than the value calculated from the data of
MAC.INTYRE (1965).
PLATH (1979) determined calcite and aragonite
solubilities using a s~turometer technique similar to
that of INGLE et cd. (1973). Only one run of 8-12 hr
duration using reagent grade calcite was made from
supersaturation and undersaturation at 25°C in sea-
water of 34.57”,,,, salinity the result was
Ki = 4.70(+0.10) x IO-’ mol” kgw2. The value at
35”<,*,
calculated using PLATH’S (1979) temperature sali-
nity equation is 4.62 x lo- ’ mol’ kg-*, which is 5.5”~~
greater than the value determined by MACINTYRE
(1965) and 0.4”; greater than the value determined by
INC~LE er (11. (1973). The solubilities of Bermuda oolites
and synthetic aragonite contaminated with an unde-
termined amount of calcite at 25°C in seawater of
32.6?; salinity were 8.69( +0.49) x lo-’ mol’ kg--’
and 5.00( 10.32) x IO--’ mol’ kg- ‘, respectively.
PLATH (1979) recommended rejection of the synthetic
aragonite data due to its contamination with calcite.
Although not specified, the equilibration time is pre-
sumed to be the same as that for calcite (X- 12 hr).
PL.ATH (1979) found that the oolite aragonite solu-
bility was 2.05 times greater than calcite at the same
temperature and salinity. IJsing this ratio at 25 C and
35”,,,, salinity results in a calculated aragonite solu-
bility of 9.46( 50.49) x IO ’ mo12 kg -‘, which is 33”,,
greater than the value determined by MA~INTYRE
(1965) and 15.2’?;, greater than the value determined
by BERNER (1976).
Calculation of K.k md K:fiorn K, ctnd K,
The thermodynamic equilibrium constants for cal-
cite and aragonite of BERNER (1976), and a value of
0.21( &O.OCtS estimated) for the total activity coeffi-
Solubility of calcite and aragonite in seawater 87
cient of calcium (MUERO, 1974) and 0.030(+0.002
estimated) for the total activity coefficient of carbo-
nate (PYTKOWICZ, 1973, in seawater of 35% at 25°C
and 1 atm pressure, were used to calculate the theor-
etical values of K: and K: according to equations 5
and 6. The resulting values are K; = 5.65(+0.48) x
lo-’ mol’ kg-* and K: = 8.40( kO.86) x lo-’ mol’
kg-‘.
MATERIALS AND METHODS
Mallinckrodt brand reagent grade calcium carbonate
was used for all experiments in which synthetic calcite was
studied. The 125420 urn size oortion of foraminifera tests
from a deep sea core’collected from 2247 m depth on the
Ontong Java Plateau by Berger, was also used for calcite
solubility measurements.
Aragonite was synthesized by the method of WRAY and
DANIELS (1957) as modified by KATZ er ul. (1972) at a
temperature of 70°C. X-ray diffraction spectra and SEM
photomicrographs did not indicate the presence of any
vaterite and less than I wt% calcite. A sample of the
ground stalagtitic aragonite used in previous- aragonite
solubility experiments (BERNER, 1976) was furnished bv
Berner. Pterbpod tests in the greater than 125 pm size
range from a core collected at 1935 m depth at Joides site
411 in the Atlantic Ocean were also used in the aragonite
solubility measurements.
Natural seawater was used in all of the solubility
determinations. Gulf Stream near-surface seawater of ap-
proximately 36.5% salinity was collected 5-7 km off the
coast of Miami, filtered through a glass fiber filter and a
0.4 pm Nuclepore filter. The filtered seawater was diluted
to 35”‘{,,, and stored for at least 3 months in a glass carboy.
After this period of time and prior to its use, the water was
refiltered through a 0.4pm Nuclepore filter. It contained
undetectable amounts of reactive phosphate (co.1 pg
PO4 - P/l). The method of STRICKLAND and PARFQNS
(1972) was used to determine the reactive phosphate con-
centrations. This procedure was used to minimize biologi-
cal activity during the long term equilibration experiments.
Solubility measurements were carried out in a closed
system. Sealed bottles containing CaCO, suspensions were
stirred on a rotating table mounted in a constant tempera-
ture bath. Equilibration was followed from both under-
saturated and supersaturated solutions.
Natural seawater diluted to 35&,,, was used without
further alteration for the supersaturated (approximately 5
times with respect to calcite) solution, and was acidified
with 1 N hydrochloric acid to reduce the saturation state
to approximately 30% with respect to calcite for the under-
saturated solution. Air from outside the laboratory, satu-
rated with water, was equilibrated with the solutions prior
to adding the solids. The Pco2 of the solutions was moni-
tored by measuring pH and total alkalinity. The solutions
were equilibrated with air until the calculated Pco, was
360 + IOppm. Brown polypropylene bottles capable of
holding 70ml of solution were used as reaction vessels.
Known quantities of calcite or aragonite were added to
these bottles first. They were then filled with either the
undersaturated or supersaturated seawater solution, which
was allowed to overflow slightly before being sealed with
polypropylene caps. The bottles were transferred to a PVC
rotating table, capable of holding 42 bottles, which was
immersed in a constant temperature bath maintained at
25 k O.l”C by a temperature controlled circulating system.
The table was rotated at a rate of approximately 2 rpm, to
constantly stir the solutions.
After various equilibration periods, the bottles were
taken from the rotating table and quickly placed in a water
bath also maintained at 25°C. The calcium carbonate was
allowed to settle before the pH of the solution was
measured. The bottle was uncapped, and a combination
electrode, fitted with a piece of ‘Tygon’ tubing and ‘Para-
film’ to form an air tight seal, was immediately inserted in
the neck of the bottle. After measurement of pH the solu-
tion was drawn from the bottle using a 5Occ syringe and
filtered through a 0.4nm Nuclepore syringe filter. Total
alkalinity and calcium were then determined on the filtered
solution.
The pH of the initial and equilibrated solutions were
measured using a Leeds and Northrup combination glass
electrode No. 117184. The electrode was calibrated by
using a set of three buffers (6.862, 7.410, 9.18) calibrated on
the NBS scale at 25°C. The buffers and equilibrated solu-
tions were kept in the constant temperature bath at 25°C
prior to and during the measurements. The electrode was
recalibrated after every four pH measurements. Electrode
drift was generally less than 0.003 pH units.
The total alkalinity was determined on 10 ml of solution
using the gran titration technique described by GIESKES
(1974). The precision was +0.3x. The boric acid contribu-
tion to total alkalinity was subtracted from total alkalinity
to obtain the carbonate alkalinity.
Calcium was determined by EGTA titration using the
method described by GIESKES (1974) which is based on a
technique devised by TSUNOGAI et al. (1968). The titrant
was standardized using standard seawater. The calcium
concentration could also be estimated by the change in,
carbonate alkalinity between the initial and equilibrated
solution, assuming a two to one relationship, ACa = i_AAc.
The measured and calculated values of Ca’+ in solution
agreed to within 0.5% or better.
The effects of opening up the closed system to insert the
electrode and syringe filter the solution, on pH, alkalinity
and calcium concentration were not detectable within our
analytical precision.
RESULTS
The analytic results and calculated apparent solu-
bility products of synthetic calcite and a mixed assem-
blage of pelagic foraminifera tests in natural seawater
are presented in Table 1. The average value of K: for
synthetic calcite exposed to seawater, for periods in
excess of 4 weeks, is 4.36( kO.20) x lo-’ mol’ kg- ‘.
This is in good agreement with the average values of
K: obtained for synthetic calcite exposed to seawater
for 5-14 days of 4.39(f0.27) x lo-’ mol’kg-* and
the mixed assemblage of pelagic foraminifera tests of
4.39( kO.20) x lo-’ mol* kg-*.
Analytic results and calculated apparent solubility
products for synthetic aragonite, ground stalagtitic
aragonite and a mixed assemblage of pelagic ptero-
pod tests, in natural seawater, are presented in Table 2.
The apparent solubility product determined for
synthetic aragonite during exposures of 1. 2 and 4
weeks is 6.97(+0.20) x lo-’ mol’ kg-*. There are no
statistically significant differences in the values of K:
for these different periods of exposure to seawater.
However, a lower apparent solubility product, of
6.65(&0.12) x lo-’ mol* kg-‘, is found for exposure
periods in excess of 2 months. The average K: values
obtained for the ground stalagtitic aragonite and
mixed assemblage of pelagic pteropod tests, during
exposures in excess of 50 days, of
6.59 x lo-‘mol* kg-* and 6.73 x lo-’ mol* kg-*,
are in excellent agreement with the long term expo-
sure K: of synthetic aragonite.
Table 1. Calcite apparent solubility constants in seawater
* A K'
c
Type oE Equilibration mCa c
Approach Calcite (days) (mole kgv1x103) (equiv.kg
-1
x103) PH (mole2 kge2x107)
und R 5 10.38 1.045 7.699 3.91
und R 5 10.47 1.077 7.706 4.13
und R 5 10.45 1.027 7.726 4.10
sup R 5 10.28 1.715 7.560 4.71
SUP R 5 10.34 1.726 7.539 4.55
SUP R 5 10.31 1.717 7.558 4.71
und R 7 10.29 0.665 7.935 4.04
und R 7 10.29 0.659 7.935 4.00
SUP R 7 10.00 1.760 7.565 4.75
S'-'Q R 7 10.01 1.766 7.565 4.77
und R 11 10.47 1.134 7.718 4.46
und R 11 10.45 1.154 7.711 4.46
und R 11 10.43 1.188 7.695 4 43
SUP R 11 10.15 1.745 7.550 4 63
sup R 11 10.26 1.856 7.497 4 43
S"P R 11 10.24 1.769 7.508 4 32
und R 14 10.33 0.740 7.895 4 15
und R 14 10.33 0.757 7.870 4 04
SUP R 14 10.00 1.745 7.545 4 51
E
C
z
6
e
‘1?
L
5
E
6
TABLE 1. (Continued)
Approach Type of Equilibration “Ca* *c R’c
Calcite (days) (mole kg-lx 103) (equiv. kg-’ x 103) PH (mole2 kge2x 107>
aup
und
und
sue
aup
und
und
auP
auP
und
und
aup
und
sup
und
und
und
auP
aup
aup
R 14 9.98 1.719 7.555 4.53
R 28 10.41 0.920 7.788 4.17
R 28 10.40 0.900 7.792 4.11
R 28 10.02 1.794 7.530 4.50
R 28 9.96 1.674 7.582 4.67
R 62 10.35 0.779 7.862 4.09
R 62 10.44 0.981 7.769 4.28
R 62 10.04 1.831 7.512 4.42
R 62 10.01 1.774 7.520 4.35
R 108 10.71 1.545 7.575 4.57
R 108 10.50 1.431 7.594 4.32
R 108 10.36 1.832 7.520 4.64
R 123 10.64 1.529 7.553 4.28
R 123 10.20 1.704 7.547 4.51
F 50 10.47 1.204 7.660 4.18
F 50 10.35 0.958 7.778 4.23
F 50 10.32 0.888 7.819 4.26
F 50 10.23 2.232 7.399 4.20
F 50 10.12 2.009 7.501 4.77
F 50 10.17 2.107 7.462 4.62
Und = approach to equilibrium for undersaturation; sup = approach to equilibrium from supersaturation; R = Mallinckdodt brand
reagent grade calcite; F = pelagic foraminifera tests in the 125 to 420 pm size range.
8
Table 2. Aragonite apparent solubility constants in seawater
Time of +I-
% *c K'a
Type of Equilibration
Approach Aragonite (days) -1 3
(mole kg -1
x 10 ) (equiv.kg x 103) PH -2
(mole' kg x 107)
"nd s 7 10.33 1.224 7.910 7.10
und S 7 10.35 1.331 7.893 7.45
S"P s 7 9.95 1.916 7.675 6.52
S"P s 7 9.90 1.914 7.691 6.71
und S 7 10.33 0.691 8.209 7.17
und S 7 10.33 0.700 8.192 7.04
S"P S 7 10.01 1.760 7.748 7.05
S"P
und
und
SUP
S"P
und
und
S"P
SUP
und
und
SUP
SUP
und
S
S
S
S
S
S
S
S
S
S
S
S
S
S
I
14
14
14
14
14
14
14
14
28
28
28
28
62
10.01 1.761 7.745 7.01
10.30 1.261 7.883 6.89
10.31 1.265 7.905 7.24
9.84 1.868 7.721 6.95
10.33 2.087 7.654 7.06
10.34 0.730 8.170 7.05
10.34 0.715 8.165 6.85
10.01 1.764 7.738 6.92
9.98 1.705 7.740 6.70
10.37 0.792 8.122 7.01
10.37 0.796 8.121 7.03
10.11 1.950 7.684 6.88
10.04 1.809 7.722 6.88
10.38 0.816 8.086 6.74
TABLE 2. (Continued)
Type of
Approach Aragonite
und S
sup S
sup S
und S
SW S
Time of
Equilibration
(days) -_
62
62
62
87
87
%3++ -1 3
hole kg x10 )
10.38
LO,04
10.04
9.91
9.64
K’
Ac a
(equiv.kg% lCl3, PW (lnole2 kg-*x 107)
0.821 8.077 6.67
1.822 7.698 6.58
1.815 7.708 6.70
1.490 7.810 6.74
2.010 7.685 6.78
und S 102 10.38 1.578 7.751 6.60
SUP S 102 9.81 2,062 7.638 6.40
und P 51. 10.71 2.026 7.629 6.73
SW P 51 10.43 2.690 7.500 6.52
und Bl 51 10.61 1.748 7.712 6.88
SUP BI 51 10.22 2.217 7.609 6.73
und B2 51 IO.55 1.803 7,648 6.15
SUP B2 51 10.06 2.314 7.620 7.08
und P 62 10.64 1.360 7.769 6.05
sue P 62 la.11 1.964 7.730 7.65
und BL 62 10.46 0.993 7.952 6.34
sup Bl 62 10.13 1.997 7.645 6.50
und B2 62 10.39 0.851 7.999 5.93
SUP B2 62 19.07 1.878 7.719 7.12
Und = appraach to equilibrium from undersaturation; sup = approach to equilibrium from supersaturation; 5 = synthetic aragonite;
BI and B2 =i
size. 2 different samples of ground stalagtitic aragonite provided by Berner; P = pelagic pteropod tests greater than 125 pm in
92 J. W. MORSE, A. MUCCI and F. J. MILLERO
Table 3. Examination of ex~rimental factors which may influence apparent solubility constants
Solid to Solution Ratio
Solid Solid to
Solution Ratio
K'
sP Number of
-2
(mole' kg x 107) Determinations
Calcite 1:70 4.36(+ 0.17) 8
1:140 4.39(* 0.24) 6
Aragonite 1:70 6.63(i 0.17)
.l:LOO 6.61(lt 0.29)
1:140 6.67(r 0.07)
1:230 6.84(+ 0.13)
1:300 6.57(+ 0.62)
1:350 6.73(+ 0.67)
Calcite Contamination of Aragonite
Weight %
Calcite
2 6.66(+ 0.18)
10 6.65(4 0.23)
Recycling Aragonite
K:
-2
(mole* kg x 10') Number of
Determinations
6.65(+ 0.55) 10
Number of
Determinations
10
10
The results of solubjlity measurements to determine
the effects of solid to solution ratio, contamination of
aragonite with calcite, and previous equilibration of
aragonite on apparent solubility are summarized in
Table 3. None of these factors has a statistically
significant influence on solubility over the ranges
studied.
DlSCUSSION
The results of this study and all other recent
measurements of the apparent solubility of calcite in
seawater at 25°C. 1 atmosphere total pressure and
35:;,, salinity agree to within 5% or better (see
Table 4). Considering the different calcite sources, ex-
perimental techniques, equilibration times, and solid
to solution ratios used by the various investigators,
this good agreement is indicative that the apparent
soiubility of calcite in seawater is well established.
The apparent solubiiity of calcite in seawater is
22(*3)% less than that calculated for pure calcite in
seawater. Direct observations, using depth profiling
scanning Auger spectroscopy, by MORSE er al. (1979)
indicate that the apparent solubility of calcite in sea-
Table 4. Summary of calcite and aragonite apparent solubility constants at 25°C and 35’:,, salinity
Kc’ Ka’
2
(mole kg
-2 X 10’) (mole' kg
-2 x 107) Ka’
K'
c
MacIntyre (1965)
Ingle et al. (1973)
Berner (1976)
Plath (1979)
This study
Calculated
4.38(% 0.26) 7.12(t 0.31) 1.63(+ 0.17)
4.60(% 0.10)
8.21(+ 0.25)
4.62(+ 0.10) 9.46(? 0.49) 2.05(+ 0.15)
4.36(? 0.20) 6.65(? 0.12) 1.53(+ 0.10)
5.65(t 0.48) 8.4O(Lt 0.86) 1.49(+ 0.28)
Solubility of calcite and aragonite in seawater 93
IO
F;______,_ --
N- 9L
‘2 I
I
N
-t
E B f I
h 8 -1 I
I I
x I
-Y - \,< ___I -- -
7- Y&
I ---__ --w-w-- $-
61 1 1 1 I 1 I I I 4 I I
0 20 40 60 80 100 120
Time (days)
Fig. I. Apparent solubility of aragonite in seawater versus equilibration time. Vertical bars represent
calculated uncertainty in Kg. P = PLATH (1979); B = BERNER (1976); M = MACINTYRE (1965) as calcu-
lated in this paper; X = Results of this study; T = Theoretical value calculated from the thermodynamic
equilibrium constant for aragonite and total activity coefficients in seawater as described in text.
water may be established by the formation of a Mg-
calcite surface phase containing between 2 and
6 mol% Mg.
Determinations of the apparent volubility of arago-
nite in seawater made by different investigators are in
poor agreement. At present it is not possible to estab-
lish the reasons for the differing results. One major
variable is the time of equilibration. A consistent
trend in the results is a decrease in measured apparent
solubility with increasing time of equilibration (see
Fig. 1). The results of this study and MACINTYRE
(1965) are statistically indistinguishable for the same
period of equilibration. The results of BERNER (1976)
and PLATH (1979), which are based on relatively short
equilibration times, agree within the uncertainty of
10% with the calculated apparent solubility of arago-
nite in seawater. The results of this study for equili-
bration times in excess of 2 months are 20( + 2)% less
than the calculated apparent solubility of aragonite in
seawater.
These observations indicate that the solubility be-
havior of aragonite in seawater is complex. The slow
inversion of the aragonite surface to a Mg-calcite or
formation of an aragonitic phase of greater stability
than pure aragonite may be responsible for this be-
havior. It is, however, important to note that once the
aragonite surface is conditioned by extended exposure
to seawater, that dissolution cannot start until sea-
water becomes undersaturated with respect to the sur-
face phase.
CONCLUSIONS
The apparent solubility constant of calcite in sea-
water determined in this study is in good agreement
with other recent measurements (MACINTYRE, 1965;
INGLE et al., 1973; PLATH, 1979). Time of equilibration
is important for the determination of aragonite solu-
bility in seawater. Values for the apparent solubility
constant of aragonite determined for equilibration
times in excess of 2 months, in this study, are signifi-
cantly less than those determined on other recent
investigations (MACINTYRE, 1965; BERNER, 1975;
PLATH, 1979), where equilibration times of only a few
days were used. Synthetic, natural and biogenic cal-
cium carbonates of greater than 99% purity have the
same apparent solubility constants within a statistical
uncertainty of approximately 5%. Solid to solution
ratio, contamination of aragonite with up to lOwt%
calcite and recycling of aragonite have no statistically
significant influence on solubility when long equili-
bration times are used.
The measured apparent solubility constants of cal-
cite and aragonite are respectively 22( *3)% and
20( f 2)% less than apparent solubility constants cal-
culated from thermodynamic equilibrium constants
and total ion activity coefficients in seawater. A poss-
ible explanation is that the apparent solubilities of
both calcite and aragonite in seawater are determined
by the formation of a surface layer of different com-
position and lower solubility than the bulk solid. If
true, this has serious implications for the use of
apparent solubility constants in understanding the
relationship between water chemistry and calcium
carbonate behavior in the marine environment.
Acknowledgements-We thank Dr R. A. BERNER for stimu-
lating discussions, Dr R. M. PYTKOWICZ for his comments,
SARA ~OTOLONGO for analytical assistance and CATHERINE
TUTTLE for aid in preparing this paper. Support provided
by NSF Marine Chemistry Program grant OCE78-18072.
REFERENCES
BEN-YAAKOV S. and GOLDHABER M. B. (1973) The
influence of seawater composition on the apparent con-
94 J. W. MORSE, A. Muccr and F. J. MILLERO
stants of the carbonate system. Deep-Seu Res. 20, 87-99.
BERNER R. A. (1975) The role of magnesium in the crystal
growth of calcite and aragonite from seawater. Geochim.
Cosmochim. Acta 39,489-504.
BERNER R. A. (1976) The solubility of calcite and aragonite
in seawater at atmospheric pressure and 34.5”:,,, salinity.
Am. J. Sci. 216, 713-730.
EDMOND J. M. (1970) The Carbonic Acid System in Sea-
water. 174 pp. Thesis, University of California, San
Diego.
GIESKES J. M. (1974) Alkalinity, pH, Mg, Ca, Si, PO1 and
NH4. Interstitial water studies, Leg IS. Init. Repts. Deep-
Seu Drilling Project 15, 63-79.
INCLE S. E. (1975) Solubility of calcite in the ocean. Mar.
Chem. 3, 301-319.
I~XLF. S. E., CULBERSON C. H., HAWLEY J. E. and PYTKO-
WICZ R. M. (1973) The solubility of calcite in seawater at
atmospheric pressure and 35”:,,, salinity. Mar. Chem. 1,
295-307.
KATZ A.. SASS E., STARINSKY A. and HOLLAND H. D. (1972)
Strontium behavior in the aragonite-calcite transforma-
tion: an experimental study at 40-98°C. Geochim. Cos-
mochim. Acta 36, 481496.
MA~INTYRE W. G. (1965) The Temperature Variation of the
Solubility Product of Calcium Curbonate in Sea Water.
153 pp. Manuscript, Rep. Ser. No. 200, Fish. Res. Board
Canada.
MEHRRACH C., C~JLBERSON C. H., HAWLEY J. E. and
PYTKOWICZ R. M. (1973) Measurements of the apparent
dissociation constants of carbonic acid in seawater
at atmospheric pressure. Limnol. Oceunoyr. 18. 897.
907.
M~LLERO F. J. (1974) The physical chemistry of seawater.
Annual Review Earth Planet. Sci. Lett. 2, 101-150.
MILLERO F. J. (1979) The thermodynamics of the carbonate
system in seawater. Geochim. Cosmochim. Acta 43,
1651-1661.
MORSE J. W., Muccr A., WALWR L. M. and KAMINSKY
M. S. (1979) Magnesium interaction with the surface of
calcite in seawater. Science 205, 904905.
PLATH D. C. (1979) The Solubility of CaCOx in Seawater
und rhe Determination of Activity Coeflcients in Electro-
/Tte Solutions. 92 pp. M.S. Thesis, Oregon State Univer-
slty, Corvallis.
PYTKOWICZ R. M. (1969) Use of apparent equilibrium con-
stants in chemical oceanography, geochemistry and bio-
chemistry. Geochem. J. 3, 181-184.
PYTKOWI~Z R. M. (1975) Activity coefficients of bicarbo-
nates and carbonates in seawater. Limnol. Oceunogr. 20.
971.-975.
STRICKLAND J. D. H. and PARSONS T. R. (1972) A Pructical
Handbook of Seawater Analysis, pp. 49-52. Fish. Res.
Board Canada.
TSUNO(;AI S., NISHIMURA M. and NAKAYA S. (1968) Com-
plexometric titration of calcium in the presence of larger
amounts of magnesium. Talanta 15, 385-390.
WEYL P. K. (1961) The carbonate saturometer. J. Geol. 69,
33-44.
WRAY J. L. and DANIELS F. (1957) Precipitation of calcite
and aragonite. J. Am. Chem. Sot. 79, 2031-2034.
