Chemical Geology, 78 (1989) 315-324 315

Elsevier Science Publishers B.V., Amsterdam - - Printed in The Netherlands

Experimental concentration-time curves for the iron(ll)

sulphide precipitation process in aqueous solutions and
their interpretation*

DAVID RICKARD
Department o[ Geology, University of Wales, Cardif[ CF1 3 YE, Wales (Great Britain)
(Accepted for publication August 24, 1989)

Abstract

Rickard, D., 1989. Experimental concentration-time curves for the iron (II) sulphide precipitation process in aqueous
solutions and their interpretation. In: J. Schott and A.C. Lasaga (Editors), Kinetic Geochemistry. Chem. Geol., 78:
315-324.

The first precipitate formed through the reaction between aqueous Fe (II) salts and dissolved sulphide at ambient
temperatures and pH < 9, appears to be a highly disordered gel approaching the composition Fe (HS) 2 on a water-free
basis. After ~ 0.4 s this precipitate loses sulphide and amorphous FeS begins to appear. The mackinawite structure
begins to develop after several hours.
The rate of formation of the initial precipitate can be approximated by a pseudo first-order reaction, directly de-
pendent on total sulphide concentration and with an apparent pseudo first-order rate constant of 48 _+9 s - 1. Dissolved
Fe concentration does not appear to be rate limiting.
The estimated solubility of the initial phase is variable but consistently one to two orders of magnitude greater than
the measured solubilities for amorphous FeS. In natural systems this may lead to variable Fe (II) solubilities in sul-
phidic environments. This initial material may play a more central role in iron sulphide reaction pathways than either
mackinawite or amorphous FeS.

1. I n t r o d u c t i o n found that it transformed to mackinawite in a

stirred aqueous H2S solution after a few days at
The reaction between aqueous sulphide and room temperature. The chemical composition
dissolved Fe (II) salts is familiar to most chem- of the precipitate is reported to vary from Feo.sTS
ists and geochemists. It leads to the rapid pre- to FeSI.1 (Berner, 1967; Rickard, 1969; Swee-
cipitation of a black iron (II) sulphide which is ney and Kaplan 1973). W.A. Kornicker (in
not easily defined. Berner (1967) found it to be Morse et al., 1987 ) thought that its properties
X-ray amorphous and more soluble than mack- changed on drying, suggesting it may be a hy-
inawite. Rickard (1969) found distinctive drate. Berner (1967) and Rickard (1969) sug-
mackinawite X-ray diffraction XRD peaks after gested that H2S was absorbed. In general, the
1 hr. W.A. Kornicker (in Morse et al., 1987) material is extremely sensitive to oxidation, es-
pecially when dry, and small amounts of oxi-
*Paper presented at the International Congress of Geo-
dation during preparation and analysis may
chemistry and Cosmochemistry, Paris, France, August 29- c a u s e s o m e of the observed variations.
September 2, 1988. It is not clear why this material should form

0009-2541/89/$03.50 © 1989 Elsevier Science Publishers B.V.

316 D. RICKARD

in the first place. It is evidently metastable with ues as the sample is taken. Therefore, contin-
respect to crystalline mackinawite, which is ap- uous measurement systems must be employed.
patently unstable in the Fe-S system at low I used silver-silver sulphide electrodes in this
temperatures relative to pyrite and members of system, since they gave direct information about
the pyrrhotite group (Berner, 1967; Rickard, dissolved sulphide concentration. However, this
1969; L.A. Taylor 1970; Morse et al., 1987). Its meant that solutions had to be buffered, limit-
formation is an illustration of Ostwald's (1897) ing the possibility of product identification. In
step rule which states that, if a reaction can re- fact, the buffering of this acid-producing reac-
sult in several products, it is the least stable tion limits the dissolved Fe(II) concentration
phase that forms first. The fundamental causes that can be used to a maximum of~ 10-3 M at
of this rule have been ascribed to energy con- the geochemically interesting pH range of 7-8.
siderations by Ostwald himself, entropy flow by This in turn produced such small quantities of
van Santen (1984), and steric effects by Rick- extremely fine-grained iron (II) sulphide prod-
ard (1969) and Morse and Casey (1988). uct that it could not be filtered and analysed
The formation of black iron(II) sulphide is directly. On the other hand, the steady-state
an important reaction in sedimentary geo- system meant that there was no problem with
chemistry, ore formation and even nuclear re- electrode response times, and various parts of
actor design (cf. P. Taylor, 1980). It forms as the reaction could be examined in detail, even
an apparent precursor to more stable iron sul- though > 90% of the reaction was generally
phides and thus determines Fe solubility in completed within 250 ms.
those environments. This paper reports the re-
sults of a first experimental investigation of 2.1. T-tube
concentration-time curves for the reaction be-
tween aqueous sulphide and dissolved Fe (II) The T-tube is shown diagrammatically in Fig.
salts to form this material. 1. It is constructed from thick walled glass tub-
ing with a 1.8-ram inner diameter. The mixing
2. M e t h o d s chamber at the junction of the T has parallel
inlets and a perpendicular outflow. Solutions are
The reaction, even at low total dissolved pumped through the arms of the T-tube with a
Fe (II) and sulphide concentrations, is very fast. peristaltic pump. Tests showed that streaming
Its time-based behaviour can only be accessed potentials did not interfere with the results and
by developing a steady-state system preceded maximum flow rates could be used. Calcula-
by a rapid mix. These design requirements are tions of flow rate via volume of solution col-
satisfied by a T-tube, where Fe (II) salts and lected show that mixing is complete in < 10 ms
aqueous sulphide are brought along separate - hence the term "millisecond mix" for the mix-
arms to a mixing chamber (cf. Rickard, 1989). ing chamber. The first electrode port is situated
The resultant product solution flows out of the at 0.13 s at maximum flow rate; subsequent
mixing chamber at a known rate so that its electrode ports are at 0.26, 0.40, 0.53, 0.60 and
characteristics at any time can be investigated 0.81 s. It is believed that these times can be de-
at specific distances down the tube. The T-tube creased by up to two orders of magnitude with
approach (Fig. 1) has certain limitations. The a more powerful pump, and reproducible re-
tube itself must be of small enough bore to al- sults still obtained.
low steady-state conditions to be achieved Electrodes were made of Ag wire bent into
without using very large quantities of reagents, the tube and cemented in place using a resin.
Solutions cannot easily be tapped and sampled The Ag wire was soldered to Cu leads. Junction
at various points, because the reaction contin- potential problems were overcome by standard-
EXPERIMENTAL CONCENTRATION-TIME CURVESFOR THE Fe (II)-SULFIDEPRECIPITATION 317

REACTANT IN

TIME

ELECTRODE PORTS

BRIDGE

Q REFERENCE
REACTANT IN ELECTRODE

Fig. 1. Schematic representation of the T-tube system.

isation (see below). The electrode leads were 2.3. Experimental

connected through an Orion ® electrode switch
to an Orion ® EA920 ion meter. Temperature was not controlled in this ex-
perimental series, which were aimed at discov-
2.2. Reagents ering if a rate could be determined. Tempera-
ture was measured for each run and was
Sulphide solutions were made by dissolving 18 + 2 oC in the fume cupboard.
Na2S in O2-free distilled water and diluting to Each run was preceded by electrode clean-
the required concentration. During reaction sing in distilled H20 and dilute HC1 followed by
runs the sulphide solutions were kept under ni- standardisation of the electrodes at the re-
trogen. Fe solutions were made up with iron (II) quired pH with sulphide solutions of known
ammonium sulphate, which is relatively resis- concentrations. Since the T-tube is in a steady
tant to oxidation, and O2-free distilled water, state, there is no problem in electrode response
The sulphide solutions were buffered with times. Solutions could be pumped through until
standard buffers. Solution mixing is rapid but the millivolt readings were constant.
Fe precipitation does not occur in sulphide-free Experimental runs were made by pumping
runs. At the concentrations used in this exper- Fe (II) and sulphide solutions through the T-
imentation, iron(II) hydroxide does not pre- tube at equal rates via the multichannel peri-
cipitate until > pH 8.4 and even then it is a slow staltic pump. Runs usually took several min-
process relative to the sulphide reaction. More utes to complete. Duplicate runs were made in
problems were encountered with the buffer so- all cases.
lutions. At pH 7-9 the buffers were essentially The net overall reaction leads to the signifi-
phosphate solutions with a reactant concentra- cant production of hydrogen ion:
tion of ~ 0.028 M, and 0.013 M after mixing.
However, sulphide-free runs revealed no phos- Fe 2+ + H S - ~ F e S + H +
phate precipitate until several minutes of re-
action. Above pH 9, carbonate buffers were and
used, but no carbonates were detected in pre-
cipitates. At these elevated pH-values, the for- Fe2+ ÷ H2 S ~ F e S + 2H +
mation of alkali iron sulphides was a greater The pH of the end solution was measured to
problem, ensure that no significant pH changes occurred
318 D. R I C K A R D

along the tube ( < 0.01 p H units). Because of relative. In fact, it was probably generally far
protonation reactions, dissolved sulphide could better than this. The nature of the concentra-
only be calculated if the p H was constant along t i o n - t i m e curves obtained were surprisingly
the tube. The addition of concentrated buffer complex and led to repeated experimental runs
solutions made it impossible to define chemi- to confirm them.
cally the precipitate in these runs (see p. 322 ).
A new experimental procedure using micro-pH 3. R e s u l t s a n d a n a l y s i s
electrodes in conjunction with silver-silver sul-
phide electrodes is planned, which will over- A typical plot of dissolved sulphide concen-
come this buffer problem, tration vs. time is shown in Fig. 2. The curve is
complex and involves two major reaction re-
gimes. Regime 1 consists of a rapid initial re-
2.4. Precision and accuracy action which typically results in the removal of
up to 99% of total dissolved sulphide from so-
The range of the experiments was limited by lution. Regime 2 involves an initial increase in
the Fe and sulphide concentration range giving dissolved sulphide concentration followed by a
linear electrode response, mass balance, p H slow, steady decrease in dissolved sulphide con-
control and flow rates, centrations. A steady-state final dissolved sul-
The electrodes showed linear response from phide concentration is not observed in this
at least 10 -1 down to 10 -7 M total dissolved experimentation.
sulphide. Repeated measurements showed one
standard deviation of < 1 mV over this range,
Log vs. T i m e
which is rather less than ___1% reproducibility. -2
Since the m e a s u r e m e n t made was removal of s, : 000~M 0 pH78
Fe i = 0.001 iv/ A pH 7
sulphide, a dissolved F e ( I I ) concentration ap- -3
proximately equal to the sulphide was em-
REGIME 1 REGIME 2
ployed. Trial runs showed that above 10 -3 M
total dissolved sulphide reaction rate was too -4
fast to be measured. Below 10 -4 M total dis- o~
solved sulphide, the reaction could not be fol- - -5 o [] [] [] []
lowed to > 99% completion. The experimenta- ~ ~ A
tion was made in the region 10-4-10 -3 M total -6
dissolved sulphide and total dissolved Fe (II).
The electrode method of concentration mea-
-7
surement has unusual properties because it is o, 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
T i m e (s)
essentially on a logarithmic scale. This means
that the last 1% of reaction could be measured Fig. 2. Log total dissolved sulphide, S (M) vs. time (s) for
a s accurately as the first 99%. However, the rate runs with initial total sulphide, Si, and initial total dis-
was generally so fast that 99% of reaction was solved iron, Fei, concentrations of 0.001 M, at pH 7.0 and
7.8. The rapid initial decrease in total dissolved sulphide
achieved within 0.26 s. reaches a constant value in <400 ms (regime 1 ) and this
Although considerable reproducibility was is followed by a small increase in S before a final decrease
achieved, overall accuracy was affected by ( a ) is observed (regime2). The concentration-time curves are
sulphide concentration; (b) p H control; and (c) interpreted a s representing two concurrent reactions, o n e
resulting in the initial precipitation of an iron sulphide phase
electrode response. The experimental analyti- (dominating regime 1 ) and one causing the re-reaction of
cal accuracy was estimated to be within _+10% this initial phase with the solution.
EXPERIMENTAL CONCENTRATION-TIME CURVES FOR THE Fe (II)-SULFIDE PRECIPITATION 319

3.1. R e g i m e I TABLE 1
R a w d a t a from experimental runs
T h e c o n c e n t r a t i o n - t i m e c u r v e in r e g i m e I is
a n e x p o n e n t i a l d e c r e a s e in d i s s o l v e d s u l p h i d e Run pH Fe, Si k' R~ HSv S~
No. (104M) (104M) {s -1) (103Ms) (104M) (107M)
c o n c e n t r a t i o n f r o m t h e initial v a l u e (Si) to a
7-33 7 10 10 51 51 5.0 42
v a l u e w h e r e t h e r a t e t e n d s to zero (Ss). F o r a 7-34 10 1 43 4.3 0.5 5.1
s i m p l e p s e u d o f i r s t - o r d e r r e a c t i o n in which: 7-x4 5 1 2s 2.s 6.9
7-44 1 1 36 3.6 0.5
-dS/dt ~= [S] 7.2-33 7.2 10 10 46 46 6.1 31

w h e r e S is t h e t o t a l d i s s o l v e d s u l p h i d e c o n c e n - 7.2-3x 10 5 55 28 3.1 16
7.2-34 10 1 45 4.5 0.6 5.0
t r a t i o n in m o l e s p e r litre ( M ) a n d t is t i m e in 7.2-x4 5 1 36 3.6
seconds, we c a n write: 7.2-44 1 1 30 3.0 0.6 26
74-33 7.4 IO 10 71 71 7.8 5.2
-dS/dt=k' [S] 74-44 1 I 45 4.5 0.8 13

w h e r e k ' is t h e p s e u d o f i r s t - o r d e r r a t e c o n s t a n t 75-3x 7.5 10 5 34 17 4 4.9

7.5-34 10 1 57 5.7 0.8 1
in s - 1. F r o m t h e r e a c t i o n p r o g r e s s c u r v e we ob- 7.5-x4 5 1 37 3.7 0.8
thin the boundary condition that, where, 7.6-33 7.6 10 10 55 49 8 60

-dS/dt-*O, S-*S~ 7.6-3X 10 5 53 27 4 25

7.6-34 10 1 47 4.7 0.8 1

that is: 7.8-33 7.8 10 10 43 43 9 93

7.8-3X 10 5 44 22 4.5 32
- d S / d t = k' S - k ' Ss (1) 7.8-34 10 1 44 4.4 8.9 1.3
8 33 8 10 10 52 52 9 17
T h e g e n e r a l s o l u t i o n for t h i s e q u a t i o n is: 8-3x 10 5 60 30 4.5 25
S=S~ +c exp (-k't) 8-34 10 1 44 4.4 0.9 5
8-X4 5 1 44 4.4
8-44 1 1 57 5.7 0.9 14
w h e r e c is t h e c o n s t a n t o f i n t e g r a t i o n . F r o m t h e
second boundary condition where 8.2-33 8.2 10 10 71 71 10 10
8.2-34 10 1 48 4.8 1
t--) 0, S --, S i 8.2-44 1 1 48 4.8 1 8
8.5-33 8.5 10 10 36 36 10 57
we obtain: 8.5-44 1 1 38 3.8 1 27

9-33 9 10 10 49 49 10 8
S = S ~ + (Si - S s ) e x p ( - k ' t ) (2) 9-44 1 1 53 5.3 1 3
Values for k ' for 30 m i n . b e t w e e n p H 7 a n d Blanksindicatewherethe datumwas not applicableor unmeasurable.
10, Si-- 1 0 - 4 - 1 0 -3 M a n d t o t a l d i s s o l v e d iron, Thefirst figurein the run No. indicatesthe pH, the next the pFeland
the t h i r d the psi. X refers to a p X of 3.5, i refers to initial concentra-
Fei = 10-4_ 10-3 M are t a b u l a t e d in T a b l e I. T h e tions of total dissolvedsulphide (S), total dissolvediron (Fe), and
m e a n v a l u e o f k' is 48 s - 1 w i t h a s t a n d a r d de- bisulphideHS-. R is the rate and k' the apparentpseudofirst-order
v i a t i o n of + 9 S - ~. C o n s i d e r i n g t h e e x p o n e n t i a l rate constant.
n a t u r e of t h e r a t e d a t a t h i s is a r e l a t i v e l y con-
s i s t e n t result. I t i n d i c a t e s t h a t t h e r a t e c a n b e a n d a p l o t of In [ S i - S ) / ( S - Ss) ] vs. t s h o u l d
a p p r o x i m a t e l y d e s c r i b e d as a s i m p l e f u n c t i o n of be a s t r a i g h t line. H o w e v e r , t h i s r e l a t i o n s h i p is
total dissolved sulphide concentration, valid u p to ~ 75% of r e a c t i o n ; a b o v e t h i s it is
difficult to e v a l u a t e t h e l i n e a r i t y or o t h e r w i s e
3.1.1. I n i t i a l rate data. R e a r r a n g i n g , eq. 1 we o f t h e plot. S i n c e t h e d a t a p o i n t s m e a s u r e d in
obtain: r e g i m e 1 are a t 0.13 a n d 0.26 s, t h e n t h e p l o t
In [ (Si - S~ ) / (S - S~) ] -- e x p ( k ' t) m e r e l y c o n f i r m s t h e a p p r o x i m a t e v a l i d i t y o f eq.
1. H o w e v e r , k ' , t h e p s e u d o f i r s t - o r d e r r a t e con-
320 D. RICKARD

Colculated initiol rote vs. pH

stant, includes the effects of total dissolved Fe, 2
pH and sulphide speciation on the rate. From
eq. 1: [] [] [] []

-dS/dt=k' (S-Ss) v
where ~ v
%
t-*0, S--~S~
and ~ ~ ~

- d S / d t = k' (Si - Ss) 3s i= 0.001,

w Sl : 0.0005 M
Since, in this experimentation, Si is always ~s, ooooo~M
>>Ss: -s -7
-dS/dt~k'Si LogH
Fig. 3. Log initial rate ( M s) vs. log H + for runs with vary-
In other words, the initial rate, Ri, can be cal- ing initial dissolvedsulphide (Si) concentrations.The ini-
culated. This initial rate concept is useful in the tial rates are independent of p H between p H 7 and 8.
analysis of reaction progress curves since at
t = 0, the concentration of the reactants can be catcu,ot,d initialrotev s . initiolHSconcentration
considered to equal their known initial values, so :
By keeping all reactants c o n s t a n t apart from 70 ]I
one, the order of reaction with respect to t h a t 60
component can be determined (cf. Rickard, I
1975). The thesis t h a t the variation in k' is _ s0 l
caused by the effect of other components can ~ 40
be tested. _~ I

A plot of Ri vs. p H shows no significant vari- ~ 30 r~ ~ t

ation, suggesting t h a t the rate is independent 20 T
of pH (Fig. 3). Since we can write:
~0
C
S=H2S+HS-
0
and o 0.0002 0.0004 0.0006 0.0008 o.oo,0
HS i (M)

H 2 S ~- HS - + H+ ( K = 10 - 7 ) Fig. 4. Initial rate vs. initial H S - concentration suggesting

t h a t the rate-controlling reaction involves the bisulphide
S = H S ( 1 + [ H + ] / 1 0 - 7) ion specifically.
bisulphide concentration is related to the total
dissolved sulphide concentration through a first-order rate constant of a factor of 2, in the
function of pH. The function: experimental range. However, since:
( 1 + [H + ]/10 -7) H2S~HS- +H +

is substantial in the experimental range from is fast relative to the rate-determining step, any
pH 7 to 8, varying from 2 at p H 7 to just above dependence on sulphide speciation would only
u n i t y above p H 8. This means that, if the rate- be observed in systems where the total availa-
determining reaction was dependent on sul- bility of sulphide is not rate limiting.
phide speciation (e.g., bisulphide concentra- Fig. 4 is a plot of calculated initial rate vs.
tion), we should expect a variation in the pseudo bisulphide concentration for runs where the
EXPERIMENTAL CONCENTRATION-TIME CURVESFOR THE Fe (II) -SULFIDE PRECIPITATION 321

initial S/Fe ratios were 1 0 - 4 : 1 0 -35 and ,00 s.... s,

1 0 - 4 . 10-4; t h a t is, w h e r e t o t a l dissolved Fe is °6, p"
pH 72
76 ~, :0o01M
equal or less t h a n t o t a l dissolved sulphide. In 8o ~ pH78

e a c h case t h e r e is a w e a k p H t r e n d . I n c r e a s i n g
pH means that HS- becomes more abundant 60
relative to H2S. T h i s t r e n d t o w a r d s i n c r e a s i n g ~ 40 []
initial rate w i t h i n c r e a s i n g p H t h e r e f o r e sug- ~ v
gests t h e possibility of a b i s u l p h i d e d e p e n d e n c e 20
o n t h e rate. U n f o r t u n a t e l y , t h e e x p e r i m e n t a - []
t i o n has b e e n c o n s t r a i n e d in e x t e n d i n g t h e case 0 ~ ~ ~ ~ ; ~°;o
of limiting t o t a l dissolved sulphide b y b u f f e r i n g ~, ~'0' M~
capabilities.
N o m a r k e d or s y s t e m a t i c v a r i a t i o n has b e e n ,00 ~.... P"
o b s e r v e d o f t h e r a t e w i t h t o t a l dissolved Fe con- 90l ~
El si
Si :0 00°5,
: 0001 M [3
80 ~ S~ = 00001 M
c e n t r a t i o n . I f t h e r e a c t i o n h a d a f i r s t - o r d e r Fe ~e~:0001
c o n c e n t r a t i o n this s h o u l d be a p p a r e n t in t h e 70
r u n s w h e n t h e initial dissolved Fe a n d sulphide ~ 80 []
c o n c e n t r a t i o n s were equal. H o w e v e r , it a p p e a r s ~ 60 []
in t h e s e e x p e r i m e n t a l r u n s t h a t t h e dissolved ~40T
Fe c o n c e n t r a t i o n is n o t r a t e limiting. 30f
20 ~7
~7
10
(b)
3.2. Regime2 °7.0 '
7.1 ~
72 '
73 '
74 7:~ ~6 7;7 ~ 7~ 80
pH

T h e i n t e r f a c e b e t w e e n regimes I a n d 2 in t h e Fig. 5. Steady-state concentration at end of regime 1, S., of

reaction p r o g r e s s curves is delineated by a zone total dissolved sulphide v.s.: (a) initial total dissolved sup
w h e r e t h e r a t e t e n d s to zero. In t h e kinetic the- phide concentration, S~; and (b) pH.
ory o f t h e r m o d y n a m i c equilibrium, this zone
CalcuLatedinitial rate vs Ss
m i g h t be i n t e r p r e t e d as a s p a c e - t i m e zone where 100
t h e p r e c i p i t a t i n g p h a s e is in e q u i l i b r i u m w i t h []
t h e solution. H o w e v e r , plots o f t h e t o t a l dis- 80
solved sulphide c o n c e n t r a t i o n (S~) a t this in-
t e r f a c e shows t h a t it varies w i t h b o t h initial dis- ~
solved sulphide c o n c e n t r a t i o n a n d p H (Fig. 5 ). ~ ~0 ~~
Unless the reaction can somehow remember ~ []
how it s t a r t e d out, t h e i n t e r p r e t a t i o n o f S~ as a n 20 [] ~ i
e q u i l i b r i u m c o n c e n t r a t i o n is precluded. T h i s ~ [] ~
sulphide c o n c e n t r a t i o n is a kinetic s t e a d y - s t a t e 0 ~ [] ~'0 ,'0 ~0 ~0
f u n c t i o n , d i r e c t l y p r o p o r t i o n a l to t h e calcu- ~ ( '°~ ~ )

lated initial r a t e (Fig. 6). T h e first b o u n d a r y Fig. 6. Initial rate, R~ vs. steady-state sulphide concentra-
c o n d i t i o n o f t h e r a t e e q u a t i o n was d e f i n e d as: tion at the end of regime 1, S~. T h e crude dependence of S~
on R i suggests a kinetic origin for S~ as a steady-state con-
S - . S~ where - dS/dt-* 0 centration resulting from concurrent reactions.
T h e kinetic n a t u r e o f this s t e a d y - s t a t e sul-
p h i d e c o n c e n t r a t i o n is f u r t h e r s h o w n o n t h e re- t h a t this increase was a n e x p e r i m e n t a l a r t e f a c t ,
a c t i o n progress c u r v e s b y t h e increase in dis- c a u s e d b y e l e c t r o d e m i s b e h a v i o u r in t h e T - t u b e
solved sulphide in regime 2. I originally t h o u g h t runs. H o w e v e r , it could be s h o w n t h a t : ( a ) t h e
322 D. RICKARD

increase was repeatable; (b) it could be made can be approximated by the relationship:
to occur at different electrodes by varying flow
rates; and (c) it could be demonstrated at the -dS/dt=k' (S-Ss)
same reaction time using a single electrode and where k' is the pseudo first-order rate constant
varying flow rates for the reactions. The in- of 4 8 + 9 s-l; and Ss is the steady-state dis-
crease in dissolved sulphide concentration solved sulphide concentration. Other features
amounts to up to 1% of the sulphide precipi- of the reaction are the independence of rate on
tated. The amount dissolved is proportional to initial total dissolved Fe concentration and the
the steady-state sulphide concentration, which weak inverse [H + ] dependence where total
is in turn directly dependent on the initial rate. sulphide is limiting. This latter observation
The final section of the reaction progress suggests that bisulphide might be the reactant
curves is a slight decrease in total dissolved sul- in the rate-determining reaction step; this is not
phide. No final steady-state value has been distinguishable from total dissolved sulphide
reached in these experimental runs. where sulphide is readily available because the
dissociation of H2S is fast relative to the rate-
3.3. Nature of the product determining step.
The absence of any significant dependence of
It has not been possible to define the nature the reaction rate on dissolved Fe concentration
of the product of these experimental runs di- implies that dissolved Fe 2+ is involved in an
rectly. Attempts to filter directly from the mil- earlier, faster process which does not involve
lisecond mix through a 0.4-/~m Millipore ® filter precipitation. This would be consistent with:
were unsuccessful. The precipitate passed
through the filter. Attempts to coagulate the Fe 2+ + H S - -, (FeSH) (aq)
q- (fast)
particles with commercial coagulants did not and
succeed: the precipitates were too dilute and
dispersed for the coagulation to succeed. Con- (FellS) + +HS--~Fe(SH)2(~) (slow)
tinuous filtration of the product for > 10 min.
gave a greenish precipitate of an iron phos- as the initial precipitation process.
phate, which block up the filter pores suffi- The existence of Fe (SH) 2 as the first precip-
ciently for a black iron sulphide precipitate to itated ferrous sulphide was first suggested by
accumulate. Analysis of this product by optical Baas Becking (1956). Subsequently, P. Taylor
spectrophotometer, direct X-ray fluorescence (1980) suggested the formation of FeSH + as an
(XRF) and atomic absorption spectrometry intermediary on stereochemical arguments. In
(AAS) gave a substantial phosphate con- time FeSH + then polymerises to form the Fe2S2
centration, groups characteristic of mackinawite.
XRD analysis gave no results, even on the In fact, it seems unlikely if this initial precip-
bulk precipitate. The precipitate is X-ray itated phase is mackinawite. Its rate depen-
amorphous. This, together with the ease with dence on bisulphide suggests an Fe (SH)2 com-
which it penetrated the Millipore ® filter sug- position. Furthermore, it undergoes a small but
gests a particle size of < 200 ~,. significant redistribution of sulphide some tens
of milliseconds after formation. It is possible
4. D i s c u s s i o n that this is a re-reaction with solution, but it
appears more likely that sulphide is being re-
The precipitation reaction of iron(II) sul- leased during a phase transformation, for
phide from aqueous solution at ~ 2 0 ° C is a example:
complex process. The main part of the reaction Fe (SH)2 -~FeS + H 2 S
EXPERIMENTAL CONCENTRATION-TIME CURVES FOR THE Fe(II)-SULFIDE PRECIPITATION 323

It is probable that this reaction is proceeding discussed in some detail by Rickard and SjS-
continuously throughout the reaction progress, berg (1983) for calcite dissolution kinetics and
but is only observed when the rate of removal derives essentially from the reaction affinity of
of dissolved sulphide from solution is less than irreversible thermodynamics. The reaction af-
the production of sulphide through this reac- finity can be defined as the negative sum of the
tion. The apparent kinetic equilibrium ob- chemical potentials of the reactants each mul-
tained of ~ 10-6 M total dissolved sulphide tiplied by its stoichiometric coefficient. The ob-
would suggest a solubilityproductoftheprecip- served relationship between rate and ( S - S s )
itated FeS of ~ 10-12. This compares well with is consistent with the reaction rate being a lin-
a solubility product of 10-11-10 -1° for amor- ear function of chemical affinity.
phous FeS and 10-12-10-11 for well-crystalline However, in the case of iron (II) sulphide
mackinawite, as calculated by Morse et al. precipitation, Ss is not an equilibrium value. It
(1987) from measurements by Berner (1967). is related to the rate at which the phase forms.
Since the surface free energy of the precipitate In other words, it appears that a non-ordered
must make a substantial contribution to its ef- ferrous sulphide phase, which includes Fe 2+ and
fective solubility, it would appear that, on the H S - but also H20 and probably H2S is dumped
1 : 1, Fe: S stoichiometry the precipitated phase from solution. It subsequently undergoes re-or-
would be more stable (less soluble ) than either ganisation in the solid state by rejection of H2S
well-crystalline mackinawite or amorphous FeS. (and He0) to form amorphous FeS, the mack-
The conundrum can be solved if a 1:2 Fe:S inawite precursor.
stoichiometry is assumed. In this case, the sol- The concept of a "dumped" iron(II) sul-
ubility product is ~ 10-9-10 -s, a magnitude phide, with a disorder approaching the disorder
greater than amorphous FeS. Such a stoichi- within the precipitating solution, as the first
ometry is satisfied by Fe (SH) 2. formed phase is very close to Ostwald's original
explanation for his rule. That is, the first formed
4.1. Kinetic implications of the process phase will be formed by the reaction having the
smallest, and not the largest free energy change.
The rate equation (eq. 1) shows that the pre- The relationship between order (as measured
cipitation rate of ferrous sulphide depends on a by internal entropy, S), free energy (G) and
function (S-S~), where S is the total dissolved chemical affinity (A) is familiar from classical
sulphide concentrate at any time and S~ is a thermodynamics:
steady-state sulphide concentration where the diS= - (AG/T).d~
rate tends to zero. This steady-state concentra-
tion has been shown not to be related to the where ~ is the extent of reaction. Subsequent
thermodynamic solubility of the precipitated change in the material is directed by the elec-
phase. It is the point in the reaction where R1, tronic nature of the components of the mate-
the rate of the reaction: rial, principally Fe(II) and sulphide. The sta-
tistical chances of hitting upon even an ordered
( FeSH ) + + S H - --. Fe (SH) 2 mackinawite cell directly from solution must be
equals the rate, R2, of the reaction: negligible.

Fe (SH) 2 -" FeS + H2 S 5. C o n c l u s i o n s

However, the function ( S - Ss) is analogous to
the £2function (the difference between solution The initial precipitate from the reaction be-
concentration and equilibrium concentration), tween dissolved Fe (II) and aqueous sulphide
observed in several other processes. It has been at ambient temperatures forms at a finite rate.
324 D. RICKARD

It appears to possess a bulk composition ap- sence of oxidizing agents is mackinawite.

proaching Fe(SH)2, but also includes varying Mackinawite is itself a material of variable
amounts of water and hydrogen sulphide. In composition, often with substantial contents of
fact, by analogy with iron oxyhydroxides, it other metals. The role ofmackinawite or amor-
probably also incorporates other dissolved ma- phous FeS as a precursor material for other sul-
terial as well. In natural systems, this "dump- phides has been discussed. However, it may well
ing" of a disordered, indefinite sulphide phase be that this initial precipitated Fe (SH)2 plays
probably leads to extensive scavenging of trace a more central role in iron sulphide reaction
metals from solution. These trace elements may pathways than either mackinawite or amor-
be expelled during the transformation of the phous FeS.
initial "Fe (SH)2" to amorphous FeS.
The solubility of this initial precipitate is
variable, and dependent on the rate of its for- References
mation. If the Fe: S stoichiometry is ~ 1" 2, the
Fe concentration product is probably ~ 10-9 at
Baas Becking, L.G.M., 1956. Biological processes in the es-
pH = 7-8, or one or two magnitudes more than tuarine environment, VI. State of iron in the estuarine
measured values for amorphous FeS in the same environment and iron sulfides. K. Ned. Akad. Wet. Proc.
pH range. In natural systems, this indefinite Ser. B., 59: 181-189.
solubility will lead to variable measured dis- Berner, R.A., 1967. Thermodynamic stability of sedimen-
tary iron sulfides. Am. J. Sci., 265: 773-785.
solved Fe (II) concentrations in sulphidic e n - Morse, J.W. and Casey, W.H., 1988. Ostwald processes and
vironments. The solubility appears to coincide mineral paragenesis in sediments. Am. J. Sci., 288: 537-
with dissolved sulphide concentrations, not 560. ~
necessarily as a result of sulphide complexa- Morse, J.W., Millero, F.J., Cornwell, J.C. and Rickard, D.
1987. The chemistry of the hydrogen sulfide and iron
tion, but merely for kinetic reasons. The parti- sulfide systems in natural waters. Earth Sci. Rev., 24:
cle size of the initial precipitate is extremely 1-42.
small. Walton {1967) notes that for precipita- Ostwald, W.Z., 1987. Studien fiber die Bildung und Um-
tion reactions in high supersaturation the ini- wandlung fester Khrper, 1. Abhandlung: fJbers~ttigung
und Oberkaltung. Z. Phys. Chem., 22: 289-330.
tial nuclei can be smaller than a unit cell. In Rickard, D., 1969. The chemistry of iron sulphide forma-
relatively dilute solutions, these nuclei do not tion at low temperatures. Stockholm Contrib. Geol., 20:
coagulate rapidly and may be difficult to sepa- 67-95.
rate from solution for analysis. Rickard, D., 1975. Kinetics and mechanism of pyrite for-
mation at low temperatures. Am. J. Sci., 275: 636-652.
After some tens of milliseconds, the precipi- Rickard, D., 1989. An apparatus for the study of fast pre-
tation reaction becomes slow enough that a sec- cipitation reactions.Mineral.Mad. (in press).
ond reaction, involving loss of sulphide from the Rickard, D. and Sj~berg E.L., 1983. Mixed kinetic control
initial precipitate, can be observed. The rejec- of calcite dissolution rates. Am. J. Sci., 283: 815-830.
tion of sulphide and, probably, water reflects Sweeney, R.E. and Kaplan, I.R., 1973. Pyrite framboid for-
mation: laboratory synthesis and marine sediments.
the beginning of ordering of the precipi- Econ. Geol.,68: 618-634.
tate: essentially crystal growth. It is interesting Taylor, L.A., 1970. Low temperature phase relations in the
to note that in the absence of coagulation, crys- iron-sulfur system. Carnegie Inst. Washington Yearb.,
tal growth must initially be reactioning with the 69: 318-322.
Taylor, P., 1980. The stereochemistry of iron sulfides: a
solution itself, as suggested by W.A. Kornicker structuralrationalefor the crystallization ofsomemeta-
(in Morse et al., 1987 ). stable phases from aqueous solutions. Am. Mineral., 65:
Several investigations {e.g., Rickard, 1969; P. 1025-1029.
Taylor, 1980; W.A. Kornicker in Morse et al., Van Santen, R.A., 1984. The Ostwald step rule. J. Phys.
Chem., 88: 5768-5769.
1 9 8 7 ) suggest that the first crystalline product Walton, A.G., 1967. The Formation and Properties of Pre-
of this process in sulphide solutions in the ab- cipitates. Wiley-Interscience, New York, N.Y., 232 pp.