close

Вход

Забыли?

вход по аккаунту

?

The Structure and Interactions of Species in Supersaturated Caustic Aluminate Solutions.

код для вставкиСкачать
Dev. Chem. Eng. Mineral Process., 10(5/6),pp. 553-566, 2002.
The Structure and Interactions of Species in
Supersaturated Caustic Aluminate Solutions
Jonas Addai-Mensah*, Jun Li and Clive A. Prestidge
Ian Wark Research Institute, Universityof South Australia,
M w s o n Lakes, Adelaide 5095, South Australia
The influence of akali metal ions (Na' and
c),
hydrogen isotope (H and 0)and
temperature on apparent hyakodynamic molar volumes, structure and interactions of
species in synthetic, supersaturated caustic aluminate solutions (Bayer liquors) has
been investigated. A strong dependency of structure, size and species interactions
(ion-ion and ion-solvenr) on the three primay variables was revealed. Sodium and
deuterium-based liquors displayed higher viscosities, reflected in greater attractive
interactions, higher activation energies and entropies of viscousjlow, than potassium
and hydrogen-based liquors. The apparent, hydrodynamic molar volumes of the
viscosity modibing species were larger in sodium and hydrogenated than in
potassium and deuterated aluminate liquors, and decreased with increasing
temperature. Ionic strength-independent average molar mass of Al(III)-containing
species close to 100 g mot' and corresponding to monomeric, tetrahydroxo aluminate
(Al(0H)i) species was observed.
Introduction
Gibbsite (y-Al(OH)3) is industrially produced by the Bayer process via precipitation
from seeded, supersaturated sodium aluminate solution (pregnant Bayer liquor).
Spectroscopic studies of fresh Bayer liquors have indicated the predominant existence
* Author for correspondence.
553
J. Addai-Mensah, J. Li and C.A. Prestidge
of hydrated, tetrahydroxo Al(0H)i monomer ions together with dimeric (e.g.
Al20(OH);-) and polymeric (e.g. Ala0(OH),,6-) minor species [ 1-31. Considerable
interactions (e.g. ion pairing, hydrogen bonding, hydration) between species and
solvent, leading to structuring and polycondensation of Al(II1)-containing species,
take place during the precipitation process [2-81. Knowledge of the colloidal species
structure, size and the nature of interactions between themselves and the surrounding
solvent is essential in fostering our understanding of gibbsite particle-solution species
interactions and crystal growth in supersaturated caustic aluminate solutions.
Reported viscosity studies of supersaturated caustic aluminate solutions [6-11, 191
have shown that extensive “structuring” and ionic interactions occur. In recent studies
[l 1, 191, alkali metal ion specific interactions and apparent molar volumes of 91 f 5
and 60 f 0.2 cm3 mar' respectively for Al(II1)-containing species in supersaturated
sodium and potassium solutions at low concentrations ([Al(III)] < 0.37 M], [caustic] =
0.5 M) and temperature range 25-65OC, were reported. Further investigation focusing
on alkali metal ion type, ionic strength, temperature and water isotope effect is
warranted for the full characterization of the species and behaviour of Bayer liquors.
The main aim of the present work was to investigate the nature of ionic species in
fresh, optically clear Bayer liquors, their specific interactions and the influence of
ionic strength, alkali metal ion, hydrogen isotope (H vs D) and temperature (22-75”C)
on the structure of the solution. This was achieved by estimating hydrodynamic molar
volumes of viscosity modifiers and interaction coefficients from viscosity
measurements of sodium and potassium aluminate solutions at different
concentrations and temperatures. In addition, the activation energy and entropy of
viscous flow and molar mass are determined.
Models For Analyses
The Eying’s rate process model [12] relates the solution absolute viscosity (7) to the
activation entropy (ns) and activation energy (EyIs)to initiate flow between molecules
if the solution molar volume (V)and AS are assumed constant, namely:
554
Structure and Interactions of Species in Supersaturated Caustic Aluminate Solutions
where h is Planck's constant; N is the Avogadro number; R is the gas constant; and T
is the absolute temperature (K).
The hydrodynamic, rigid molar volume (V,> and interactions between particulate
matter in a suspension may be characterized by a linearized Einstein-Vand model of
viscosity [14, 151:
(2)
where qo is absolute viscosity of solvent (Al(III)-fiee caustic solution); Q is a
parameter which defines mutual interaction between particles and their Brownian
motion (positive I repulsion and negative
= attraction); and C is the particles molar
concentration.
The weight average molar mass and second virial coefficient of solution species
may be determined by the following Rayleigh light scattering equation [16,171:
H%=YM+2BC
where
7
(3)
is the excess turbidity of solute; M is the average molar mass of solute
species; B is the second virial coefficient reflecting the net attraction (= B negative) or
repulsion (= B positive) between the solute species; C is solute concentration; and H is
an optical parameter dependent upon the wavelength of light.
Experimental Details
Fresh, optically clear, synthetic hydrogenated and deuterated sodium and potassium
aluminate solutions were used in this study. They were prepared fiom AR grade
aluminum metal, NaOH, NaOD, KOH, KOD and D20 (Aldrich Chemicals, USA) and
Milli-Q water (specific conductivity < 0.5 ps cm" and surface tension 72.8 mN m-' at
J. Addai-Mensah,J. Li and C.A.Prestidge
2OOC). Stock solutions containing 4 or 6 M caustic (NaOH or KOH or NaOD or
KOD) and 2.92 or 4.38 M AI(II1) with [caustic]/[Al(III)] = 1.37 were firstly prepared
and diluted with Milli-Q water or D20 or caustic to give solution concentrations in the
range: 1 . 2 5 4 3 8 M Al(II1) at [caustic]/[A1(III)] = 1.37 and 1.25-2.92 M AI(II1) at 4
M caustic or 1.254.38 M at 6 M caustic. Concentrations of dissolved species were
determined by inductively coupled-plasma atomic emission spectroscopy (ICP-AE,
Spectro Analytical Instruments, SIM-SEQ). A Haake CV20 concentric cylinder
rheometer (Haake, Germany) fitted with a stainless steel Mooney-Ewart (45 mm
diameter) concentric cylinder sensor, was used to determine liquor viscosity. Excess
turbidity of liquors was determined using a Brice Phoenix BP3000D light scattering
photometer at operating wavelength of 546 nm.
Results and Discussion
Viscosity of Caustic Aluminate Liquors
The typical absolute (qJ and relative viscosity (q/q0)behaviour of hydrogenated and
deuterated caustic aluminate solutions at different AI(II1) concentrations and
temperatures is shown in Figures 1 and 2. The increase of viscosity with increasing
Al(II1) concentration is evident as expected with the increase less pronounced at Bayer
process temperatures 65 and 75OC than at 22OC, see Figure 1. This result implies that
strong attractive interactions between the Na'- / K
'
- Al(III)-containing and OHsolute
species and solvent (H2O)exist in the solution and that these interactions (e.g.
hydrogen bonding) weaken with increasing temperature. In Figure 1, the viscosity of
sodium aluminate liquor of similar composition and at 20°C reported in the literature
[20] is shown for comparison, indicating a good agreement between the present and
reported data. The effect of alkali metal ion and hydrogen isotope on the relative
viscosity is clearly observed at 75'C in Figure 2. The presence of Na' results in a
'
K
significantly higher viscosity than does ,
and H instead of D results in higher
relative viscosity under similar solution conditions. The observed hydrogen isotope
effect on relative viscosity arises from the fact although no marked difference in
absolute viscosity was observed for deuterated and hydrogenated aluminate solutions,
the viscosity of D20solvent was significantly higher (-1.3 times) than that of H20.
556
Structure and Interactions ofspecies in Supersaturated Caustic Aluminate Solutions
20
I5
L4
B
20 “C, Ikkatai (1%3)
6 10
8
u
.
I
s 5
0
Figure 1. Absolute viscosity of sodium aluminate solutions as afinction of
AI(III) concentration at a constant [NaOH]/[Al(IIiJ] = 1.37,
[NaOH] = 1 - 6 M and at different temperatures..
U
’
1
0
2
3
4
5
Figure 2. Relative viscosity of caustic aluminate solutions as afunction of
Al(IIiJ concentration at a constant [Caustic]/[Al(IIiJ]
=
1.37,
-
[Caustic] = 1 6 M and 75°C.
557
J. Addai-Mensah,J.Li and C.A. Prestidge
2.8
3.0
3.2
3.4
3.6
2.5
2.0
n
.-h
m
1.5
z
1 .o
CI
8m
R
I
0.5
0.0
2.8
3.0
3.2
3.4
3.6
l/T, K-'
Figure 3. Arrheniusplots for caustic aluminate solutions with (Top): Al(l1I)
= 2.92 M,
[Caustic] = 4Mand (Bottom): Al(III) = 4.38 M; [Caustic] = 6M
Activation Energy and Entrupy of ViscousFlow
A typical Arrhenius plot of absolute viscosity and temperature data is exhibited in
Figure 3 on a semi-logarithmic scale according to Equation 1. In all cases, linear
relationships were obtained. This indicates that theoretical treatment of the viscosity
data in the manner of Eying [121 is acceptable. The activation energies and entropies
558
Structure and Interactions of Species in Supersaturated Caustic Aluminate Solutions
of deuteratedhydrogenated sodiudpotassium aluminate liquors with H20/D20
dilution and of sodiudpotassium aluminate liquors with 4 M/6 M caustic dilution are
summarized in Table 1. At constant [Caustic]/[Al(III)] molar ratio (water dilution) the
activation energy of viscous flow increased by -36% when total the NaOH or NaOD
concentration increased from 4 to 6 M for Na' based solutions (see Table 1).
Similarly for K' based solutions, the activation energy of viscous flow increased
by -13% when the total KOH or KOD concentration increased fkom 4 to 6 M. Table
1B shows that at 2.92 M Al(II1) concentration the activation energy for sodium
aluminate liquors increased by 27% when the total NaOH concentration increased
from 4 to 6 M. At constant caustic concentration (e.g. 4 M) however, no such change
in the activation energy was observed upon increasing Al(II1) concentration from 1.25
to 2.92 M. These observations confirm: (i) increasing species interactions at the higher
ionic strength; and (ii) the stronger ion-pairing ability of Na' ions. At a given
concentration, the activation energy for viscous flow was observed to be similar for
the hydrogenated and deuterated liquors, indicating that the net ordering of species
was independent of the type of hydrogen isotope (D or H).
Table IA. Activation energy and entropyfor the viscousflow of caustic aluminate
solutions at a constant ratio of [Caustic]/[Al(lIg]
=
Liquor
1.37.
AS (J mol-'K-')
Caustic = 4 M
Caustic = 6 M Caustic = 4 M
Caustic = 6 M
Al(II1) = 2.92
Al(II1) = 4.38 M
Al(II1) = 2.92
Al(II1) = 4.38
22 f 2
30f2
27f4
46f4
22f 2
30f2
29f4
45 f 6
18f2
20k2
18+ 4
20 4
20+2
24rt2
23 f 4
28 i 4
+
559
J. Addai-Mensah, J. Li and C.A. Prestidge
Table IB. Activation energy and entropyfor viscousflow of caustic aluminate
solutions at constant caustic: [NaOH] or [KOH]= 4 or 6 M.
AS (J mol-'K-')
Liquor
Al(II1) = 1.25 M 2.92 M
M
Al(II1) = 1.25 M
4
20f2
22f2
27 f 4
27f4
6
24f2
28f2
35f4
42f4
4
16f2
18f2
15f4
18f4
6
15 f 2
18f2
11 f 4
17f4
2.92 M
The entropy of activation may be visualized as the extent of disruption of order or
structure in the solutions before flow can take place. The activation entropies also
increased with increasing ionic strength and were significantly greater in sodium than
in potassium aluminate liquors at equivalent concentrations (see Table 1). This
suggests that a greater ordering and interactions of the ions, for instance the formation
of ion pairs (e.g. Na+Al(OH)i or Na+OH-) and hydration, occurs with Na+ as opposed
to K', affecting the surrounding solvent molecules as moving dipoles [12, 131.
Hydrodynamic Molar Volumes of Species
The estimated average hydrodynamic molar volumes of hydrated species in
sodiumlpotassiumaluminate solutions with water or caustic dilution at 22,45, 65 and
75°C are given Tables 2A and 2B. The molar volumes increased significantly with
increasing Al(II1) concentration and were larger for Na+-based species than those
found in K+-based aluminate liquors. It thus follows &om this observation that the
aluminate ions (AI(0H)i or Al(0D)i) undergo significant solvation through Hbonding or possibly form poly Al(II1)-containing species. Na' ion and hydrogen based
560
Structure and Interactions of Species in Supersaturated Caustic Aluminate Solutions
species displayed larger molar volumes than those formed in K' ion and deuterium
based liquors. Over the temperature range 22-75"C,the molar volumes of the caustic
aluminate species followed the sequence: N T ~ I ( O H>) ~NaAl(OD)4 > KAI(OH)4 >
KAI(ODk, and decreased markedly with increasing temperature.
Furthermore, the average, hydrodynamic molar volumes for 4 and 6 M sodium and
potassium aluminate liquors were substantially similar, indicating that the species
present in Bayer liquors are ionic strength independent.
The values of the interaction coefficient (Q) were generally negative, except in a
few cases for potassium aluminate and one instance of sodium aluminate liquors, as
shown in Figures 4 and 5. For both Na' and K' based, water diluted liquors, negative
values of Q were observed over the temperature range 22-75°C (see Figure 4).
Table 2A. The average hydrodvnamic molar volume of species in hydrogenated and
deuterated caustic aluminate solutions: [Caustic]/[Al(IIg]
4.38 M and [Caustic]
=
=
1.37, Al(III)
=
1.25 -
1.7 - 6 M) at different temperatures, estimated by the
Einstein- Vand model (Equation 2).
Hydrodynamic Molar Volume (cm3 mol-')
Liquor
22°C
45°C
65°C
188f 1 1
177 f 10
142 f 8
135 f 7
155 f 9
153 f 8
132 f 8
114f7
145 f 10
131f 9
113 f 7
108 f 7
*
106 f 7
96f6
94f 6
122 10
75°C
J. Addai-Mensah, J. Li and C.A. Prestidge
Table 2B. The average hydrodynamic molar volume of species in caustic aluminate
solutions: [Caustic] = 4 or 6 M and AI(III) = 1.25-2.92 M or 2.50-4.38 M at
diferent temperatures estimated by the Einstein- Vand model (Equation 2).
Average Hydrodynamic Molar Volume (cm3 mol-')
Liquor
M
22OC
NaAI(0H) 4
4
236 f 12
NaAl(0H) 4
6
I w O H )4
KAKOH) 4
450c
65°C
750c
210f 1 1
160 f 8
143 f 7
238 f 12
201 f 1 1
197 f 10
129 f 6
4
105 f 1 1
169f 17
106 f 1 1
90 f 9
6
133 f 13
138 f 14
125 f 13
91f 9
0.1
0.0
a
c,
-0.1
8
1'
0
KAI(0D)t.O NaAl(0Db
p
.
0)
i!
u
-Oa2
-0.3
-0.4
-0.5
20
30
40
70
60
Temperature, OC
50
80
Figure 4. Coeflcient Q calculated using the Einstein-Vand equation (Equation 2) as
a function of temperature for solutions containing: AI(III)
[Caustic]
562
=
1.7 - 6 M with [Caustic]/[Al(III)]
=
1.37.
=
1.25 - 4.38 M and
Structure and Interactions of Species in Supersaturated Caustic Aluminate Solutions
This indicates that net attractive interactions are dominant in the liquors over the
concentration range investigated. At constant caustic concentration (i.e. liquors with
' based
caustic dilution, see Figure 5 ) , the observation of positive Q values for K
liquors and Na+ based liquors at 75°C suggests the presence of repulsive interactions.
This may possibly be due to symmetric interactions of species which lead to long
range coulombic forces becoming important. A subtle variation in the Q with
increasing temperature is observed which suggests that interactions between the ions
are weakly temperature dependent. In this case, the decrease in the molar volume with
increasing temperature is seemingly due to dehydration. Notably, stronger attractive
(or weaker repulsive) species interactions, indicative of the stronger ion pairing, were
observed in sodium as compared with potassium aluminate liquors. Stronger
interactions among species are indicated in deuterated rather than in hydrogenated
solutions.
0.75
0.50
-
0.25
-
e
Z
1
4)
;
u
0NaAl (OH)4,4M
0 NaAl(OH),6M
.I
0
0.00
-
-0.25
1
10
20
30
40
5 0 0 60
Temperature, C
70
80
Figure 5. Coeficient Q calculated using the Einstein-Vand equation (Equation 2) as
a function of temperaturefor solutions containing: [Caustic] = 4 or 6 M and Al(I/I) =
1.50 - 2.92 M or 2.50 - 4.38 M.
563
J. Addai-Mensah, J. Li and C.A. Prestidge
Average Molar Mass of Al(III)-conlainingSpecies in Bayer Liquors
Analysis of HC/z versus solute (Na'Al(0H)i or K'Al(0H)i) concentration C plots
from Equation 3 indicated average molar mass of 130 f 35 and 98 f 41 g mol-',
respectively for Na+Al(OH)i and K'Al(0H)i
ion paired species. Taking the
molecular weights of NaAl(OH)4 (118 g mof') and KAl(OH)4 (134 g mol-') into
account, the predominance of Al(II1)-containing species with an average molar mass
to monomeric of Al(0H); ions (95 g mor') is indicated. Second virial coefficients (B)
values of 0.013 f 0.001 and 0.015 f 0.001 were estimated for the Na' and K' based
liquors. These indicate that weak scatterer-scatterer repulsive interactions are
dominant in these systems. The repulsive interactions are less for Na' than for K'
based liquors. This observation is consistent with the metastable nature of
supersaturated caustic aluminate solutions, i.e. no precipitate formed without seed
addition even after several hours at room temperature, and longer induction times
observed for unseeded Al(OH)3 crystallization in K+based than in Na' based liquors
[18, 191.
Conclusions
Alkali metal ion-mediated solution speciation and structuring have been observed in
this study. The activation energy and entropy of viscous flow are greater in
concentrated sodium aluminate solutions as compared to potassium aluminate liquors.
The specific solute-solute interactions are also stronger in sodium than in potassium
aluminate solutions. The viscosity modifying species in sodium and deuterated
aluminate solutions have larger molar volumes than those in potassium and
hydrogenated aluminate solutions in the Bayer region. These Al(II1)-containing
species, with an average molar mass close to 100 g mol-' provide evidence for the
predominance of monomeric tetrahydroxo aluminate species in Bayer liquors.
Nomenclature
B
Second virial coefficient
C
Solute molar concentration (M)
564
Structure and Interactions of Species in Supersaturated Caustic Aluminate Solutions
E",,
Activation energy of viscous flow
(J mol-')
h
Planck's constant
( 6 . 6 2 6 10-34
~
H
Optical parameter dependent upon
J s)
the wavelength of light
M
Average molar mass of solute species
(kg mot-')
N
Avogadro number
(6.022 x
As
Activation entropy
(J mol-' K-I)
R
T
Universal gas constant
(8.3145 J mol-' K')
Absolute temperature
(K)
V
Molar volume of species
(m3moP)
v
h
Hydrodynamic, rigid molar volume
(m3 mol-')
11
Absolute viscosity of solution
(Pa s)
7
Excess turbidity of solute.
mor')
References
Moolenaar, R.J., Evans, J.C., and McKeever, L.D. 1970. The structure of the aluminate ion in
solutions at high pH. J. Phys. Chem.,74(20), 3629-3636.
2. Zambo, J. 1986. Structure of sodium aluminate liquors; molecular model of the mechanism of their
decomposition.Lighi Meials, 199-215.
3. Watling, H. 1998. Spectroscopy of concentrated sodium aluminate solutions. Applied S'ciroscopy,
52(2), 250-258.
4. Watling, H.R., Fleming, S.D.,Bronswijk W.V., and Rohl, A.L. 1998. Ionic structure in caustic
aluminate solutions and the precipitation of gibbsite.JChem. Soc., Dalton Trans., 391 1-3917.
5 . Sipos, P., V d i , I., May, P.M., and Hefter, G.1997. Formation of NaOHo~a,,~and Na[Al(OH)4]o ion
pairs in concentrated alkaline aluminate solutions. Progress in Coordinaiion and Organomeiallic
Chem., G. Ondrejovic and A. Sirota (eds), Slovak Tech. Univ. Press, Bratislava, 303-308.
6. Eremin, N.I., Volokhov, Y.A., and Mironov, V.E. 1974. Structure and behaviour of aluminate ions in
solution. Russian Chem. Reviews, 43(2), 92-106.
7. Myund, L.A., Sizyakov, V.M., Burkov, M.K., Zakharzhevskaya, V.O., and Borzenko, O.A. 1995.
Aluminate solutions at various temperatures. Russiun J. Applied Chem., 68(12), 1721-1724.
8. Romanov, L.G., Ni, L.P, Pivovarov, S.R., and Bertyakora, L.V. 1974. Nature of aluminate liquors.
Sov. Non-Ferrous Meials Res., 2,273-279.
9. Kuznetsov, S.I., Antipin, L.N., and Vazhenin, S.F. 1957. Nature of the variations of certain properties
of aluminate solutions in the decomposition process. J. Appl. Chem. (USSR)),30,377-381.
10. lkkatai, T., and Okada, N. 1963. Viscosity, specific gravity, and equilibrium concentration of sodium
aluminate solutions. Extractive Metallurgy of Aluminum, Gerard, G. (ed.), Interscience Publishers,
New York, 159-1 75.
11. Li, J., Prestidge, C.A., and Addai-Mensah, J. 2000. Viscosity, density, and refractive index of aqueous
sodium and potassium aluminate solutions.J. Chem. Eng. Dutu, 45(4), 665-67 1.
12. Gladstone, S.,Laidler, K.J., and Eying, H. 1941. Theory of Rate Processes. McGraw-Hill, New
York.
1.
\
"
565
J. Addai-Mensah, J. Li and C.A.Prestidge
13. Miller, M.L., and Doran, M. 1956. Concentrated salt solutions. 11. Viscosity and density of sodium
thiocyanate, sodium perchlorate and sodium iodide. J. Phys. Chem.,60,186-189.
14. Vand, V. 1948. Viscosity of solutions and suspensions, 1. J.Phys. Colloid C k m . , 52,277-299.
15. Thomas, D.G. 1965. Transport characteristics of suspension: VIII. A note on the viscosity of
Newtonian suspensionsof uniform spherical particles. J. Colloid Sci., 20,267-277.
16. Hiemenz, P.C., and Rajagopalan, R. 1986. Chapter 5, Light scattering, in Principles of Colloid and
Surface Chemistry, D. Marcel fed.). Marcel Defier, New York, 223.
17. Kerker, M. 1969. Chapter 9, Scattering by liquids, in The Scattering of Light and Other
Electromagnetic Radiation, E.M. Loebl (ed.), Academic Press, New York,487.
18. Smart, R.St.C., Counter, J., Dawe, I., Addai-Mensah, J., Gerson, A.R., and Ralston, J. 1998.
Nucleation and precipitation mechanisms in precursors to colloidal alumina, in Ceramic Interfaces,
Properties and Applications, Smart, R3t.C. and Nowotny, J. (eds), IOM Communications Ltd., UK,
483.
19. Li, J. 2000. The Influence of Alkali Metal Ion on Gibbsite Crystallization from Synthetic Bayer
Liquors. Ph.D. Dissertation, University of South Australia.
566
Документ
Категория
Без категории
Просмотров
3
Размер файла
522 Кб
Теги
species, solutions, structure, interactions, aluminate, supersaturated, caustic
1/--страниц
Пожаловаться на содержимое документа