Yan 1995

The Scienceof the Total Environment 172(1995) 95-118

ELSEVIER

Is the glass phase dissolution rate always a limiting factor in the leaching processes of combustion residues? J. Yan”, I. Neretnieks Department

of Chemical

Engineering

and Technology,

Royal Institute

of Technology,

S-l 00 44 Stockholm,

Sweden

Received 6 September 1994;accepted25 January 1995

Abstract The kinetic characteristics of waste glass phases in some coal fly ashes are simulated by the hydration free energy approach of multiple-component glass. A geochemical model is used to calculate the dissolution processes of waste glass in a low flow-through system for coal fly ashes. The contribution of waste glass dissolution for the leaching of solid waste is discussed on the basis of dissolution mechanisms and kinetics of glass phases, reaction time and disposal condition. It has been found that the dissolution rate of a glass phase is not always a limiting factor in the

leaching processes of coal fly ashes. Additional retention mechanisms may retard the release of toxic elements but only under some limited conditions. The disposal environment together with kinetic properties of waste glass phases need to be accounted, for the dissolution of the glass phase. Glass dissolution under some, not improbable, circumstances may considerably contribute to the long-term release of toxic metals. Keywords: Waste glass phase; Solid waste; Coal fly ash; Dissolution mechanism and kinetics; Geochemical modeling; Release of toxic metals

1. Introduction

The glass phase is one of the most important parts of solid wastes, especially combustion, incineration or smelting residues such as fossil fuel combustion wastes, municipal solid waste (MSW) ash and slags that may be a main constituent of industrial and municipal solid wastes (Gut&rez et al., 1993; Kirby and Rimstidt, 1993; Yan and

* Corresponding author.

Neretnieks, 1993b). According to current estimates, the production of combustion residues is 70 million tons annually for coal ash, 4 million tons annually for municipal waste combustion (MWC), and 1 million tons annually for wastewater sludge incineration (WSI) ashes in the U.S. (Theis and Gardner, 1990). Although recycling or reuse of these waste materials is possible in some instances, most of them are usually disposed of in landfills. Therefore, one of the major environmental concerns in the disposal of the waste residues is their leaching characteristics. More

0048-9697/95/$09.50 0 1995 Elsevier ScienceBV. All rights reserved. SSDI

0048-9697(95)04727-I

96

J. Yan, I. Neretnieks

/The

Science of the Total Environment

information is required to permit a thorough assessment of leachability for this kind of solid waste with the ultimate objective of defining environmentally sound waste management practices. In general, many solid wastes are heterogeneous materials. This heterogeneity mainly depends on distribution of mineral elements in the solid wastes (Yan and Neretnieks, 1993b). There are close relationships between the dissolution behavior of a species and its distribution. The relationships determine the dissolution mechanisms of a species in certain conditions. This is the basis for connecting the intrinsic physical and chemical properties of the wastes with their dissolution characteristics. During the dissolution processes of solid wastes, some fractions (slats or soluble oxides) of solid waste are easily leached in a short period of time, some fractions (acid-soluble components) may be dissolved under certain conditions and the others (glass phases and some magnetic fractions) will take a relatively long period of time to be altered and dissolved (Yan and Neretnieks, 1994b). Dissolution behavior of the species varies due to the differences in dissolution mechanisms and rates of species that exist in various fractions of solid waste. There are two dominant features of glass phases present in the solid wastes. One is their high percentage of wastes. For example, the glass phase constitutes 60 to 90% of most coal fly ashes (Roy et al., 1985). Some MWC ashes have been found to contain a glass phase (Kirby and Rimistidt, 1993; Eighmy et al., 1994) as have some commercial slags (Eberendu and Daugherty, 1984). The other is that more toxic elements, heavy metals, for example, may be enriched in the glass phase as compared to other phases of solid waste (Hulett et al., 1980; EPRI, 1981, 1983; Norton et al., 1986; Rai et al., 1987). In combustion residues, some toxic elements contained in the glass phase may be as high as over 90% (Yan and Neretnieks, 1994b). The glass phase is generally considered as a host of these trace elements in solid waste (Norton et al., 1986; Hemming and Berry, 1988). Many toxic elements are incorporated into the glass phase by becoming part of the random three-dimensional glass network. Dissolution of the silicate matrix is a necessary precursor to the

172 (1995) 95-118

leaching of the toxic elements, or does at least occur simultaneously with leaching of other components. Therefore, the degree of dissolution of the glass phase may be a reasonable measure of the potential for release of the toxic elements from the solid wastes. Glass phases should be considered metastable solids. In thermodynamics they have a higher free energy than an assemblage of crystals of the same bulk composition. Most glass phases exhibit a tendency to react with aqueous solutions to form more stable assemblages of hydrated phases (White, 1988). With respect to the kinetics, on the other hand, the glass phases have a stability that works against re-crystallization and further reactions with aqueous solutions. For solid waste in which the glass phase is a main component, the durability of the glass phase must account for the thermodynamics and kinetics, particularly for long-term leachability of the solid waste. It is not very clear, however, how the glass phase dissolution contributes to the leachability of the solid waste. There are several reasons for knowing little about the properties of the glass phases in leaching processes of the solid wastes. Owing to the relative stability of the glass phase, long-term observations are needed to understand the leachability of the glass phase. This time factor is difficult to simulate by present laboratory leaching tests if the kinetics of the glass phase has to be considered. In laboratory leaching tests, an aggressive leachant or a high ratio of the solution volume to solid volume (or weight) is often used to accelerate the leaching progresses, but this does not provide a correct relationship between the results of the short-term leaching tests and long-term leaching behaviors of a solid waste. In this kind of leaching test, the kinetic factors of the solid phase dissolution are easily ignored. Besides the limits of current laboratory leaching tests for dissolution of the glass phase, few longterm field observations of leaching characteristics of the solid wastes are available, though a fouryear natural weathering test (Andrade et al., 1990) and a seven-year large-scale lysimeter test (Hjelmar, 1990) for coal fly ash have been reported. On the other hand, the complexity of dissolution and alteration of glass phase also

J. Yan, I. Neretniekx

/The

Science of the Total Environment

makes it difficult to distinguish the dissolution or alteration (or changing from metastable phase into stable phase) characteristics of glass phase itself in the leaching process of solid waste. The dissolution or alteration of the glass phase often appears with formation of secondary phases. This makes the alteration processes of glass phase very difficult to be directly observed from the changes of the solution chemistry by current laboratory test methods and field investigations. Because of the limited understanding of the dissolution behavior of waste glass, the glass phases in solid waste are usually considered as an inert phase (van der Sloot et al., 1985; Kosson et al., 1991). According to this assumption, the dissolution rate of waste glass is so low that there is little or no contribution for release of toxic elements during leaching of solid waste. In fact, whether the dissolution rate of the glass phase is always a limiting factor in leaching processes of solid wastes is still an open question. In this paper, we try to take into account dissolution kinetics of the waste glass phase, the time factor and possible conditions of waste disposal in order to understand the contribution of waste glass for dissolution of solid wastes. 2. Dissolution glass phase

mechanisms

and kinetics of waste

The current theoretical and experimental studies of glass corrosion mainly involve fields including commercial glasses, non-crystalline solid materials, natural glasses and nuclear waste glasses (Adams, 1988). In recent decades, most of the fundamental work focuses on nuclear waste glasses. The field of rock/water interaction, especially the reaction of silicate minerals with aqueous solutions, also provides essential understanding of the glass dissolution and surface chemistry. Since most waste glass phases belong to the multi-component non-crystalline silicate, this similarity in chemical composition and structure provides a basis for studying their reaction mechanisms and kinetics in an aqueous solution by using the basic theory and experimental technology of glass corrosion. All of these promote the development of unifying concepts of the reac-

97

172 (1995) 95-118

tion mechanisms and prediction of long-term chemical durability, in particular for multi-component waste glasses. 2.1. Dissolution

mechanisms

of waste glass phase

Regarding waste disposal, the primary concern is the reactivity of the glass phase with water or other aqueous solutions, and the potential for release of toxic elements. The interaction of a multi-component waste glass phase with an aqueous solution is a complicated process. The dissolution of multi-component silicate glass phases, primarily involve selective leaching (Doremus, 1975; McGrail et al., 1984; Lee and Clark, 1986; Drad et al., 1988); pH drift (Barkatt et al., 1986; Hench, 1988); matrix dissolution (Adams, 1988; Advocat et al., 1991); formation of a reaction layer (gel layer) (Ewing and Jercinovic, 1987; Abrajano et al., 1989; Cunnane et al., 1993); precipitation of insoluble species dissolved from the glass in or on the gel layer (Michaux et al., 1992); and surface alteration and formation of secondary phase (Ewing and Jercinovic, 1987; Abrajano et al., 1990). Since glass is a metastable phase it may reach a metastable saturation with an aqueous solution as a development of the surface reaction process. It cannot be in thermodynamic equilibrium with an aqueous solution, Therefore, glass will continue to react with an aqueous medium, although the reaction rate may be low (White, 1986). The glass phase dissolution phenomena and mechanisms have been discussed in detail in our early work (Yan and Neretnieks, 1993a; Yan and Neretnieks, 1994a). Regarding dissolution kinetics of waste glasses, the most important mechanisms may be the selective leaching, matrix dissolution, phase alteration and formation of secondary phase in or on the interfaces between solution and glass surface. 2.2. Dissolution

kinetics of waste glass phase

Several rate-controlling processes are involved in dissolution of the complex waste glasses. In general, these fundamental processes may be grouped into mass transport, surface reaction and solubility limit of the secondary phase. For most silicate glasses, dissolution in an aqueous solution exhibits two limiting stages (Nogues et al., 1985;

98

I. Yan, I. Neretnieks / The Science of the Total Environment 172 (1995) 95-118

Bunker et al., 1988). At short dissolution times, a mobile element (for example, alkali metals or boron) release is consistent with a transport (diffusion) controlled process. After the initial transient period and for long-term dissolution, the species release may be consistent with a rate limiting step involving the reactions occurring at the interface between the reacted glass phase and the bulk glass before the concentration of the species reaches a saturation limit. Many studies provide evidence that the surface dissolution reaction controls the overall dissolution rate of complex silicate glasses including natural glasses and waste glasses (Chick and Pederson, 1984; Conradt et al., 1985; Grambow, 1985; Crovisier et al., 1987; Lutze, 1988; Mouche and Vernaz, 1988; Advocat et al., 1990; Bourcier et al., 1990; Knauss et al., 1990; Bourcier, 1991; Cunnane et al., 1993). In most situations the surface layer of the reacted glass phase exhibits little or no effect of a diffision barrier. There is, however, evidence that the surface layers may be important under some circumstances (Grambow et al., 1992). In a closed system or for a sufficiently long reaction time, the rate controlling process of complex silicate glasses may be complicated by precipitation of secondary phases from the solution or by in situ altering of the gel layer. The composition and property changes of surface layers, particularly the alteration of metastable phases, may affect transport processes or thermodynamic equilibrium limits, and thus the overall dissolution rate of glass phases. An approach assuming surface reaction control is widely used to model the dissolution processes of waste glasses. This approach is developed on the basis of the theory of water/mineral interface reaction. On the basis of the transition state theory and surface complex chemistry, a general rate law (Lasaga, 1981, 1984; Aagaard and Helgeson, 1982; Helgeson et al., 1984; Murphy and Helgeson, 1987) is proposed to account for the silicate hydrolysis at constant pressure and temperature:

where r is the rate of reaction, n the number of moles of reactant mineral in the system, t the time, s the effective surface area, k the rate constant, a, the activity of the ith species in the system, ii,,j the reaction coefficient of the ith species other than the activated complex in the jth activation reaction forming the activated complex, A the chemical affinity for overall hydrolysis reaction, R the gas constant, T the absolute temperature, u the ratio of the rate of decomposition of the activated complex to that of the overall reaction. According to the general hydrolysis theory of silicate mineral as mentioned above and reaction kinetics of silica with water (Rimstidt and Barnes, 1980), Grambow developed a model that accounts for dissolution behaviors of waste glasses and experimental observations. This model has been widely applied to modeling of dissolution of waste and natural glasses under various environmental conditions (Freude et al., 1985; Grambow, 1985, 1987; Grambow et al., 1985a, 1985b, 1986, 1987; Croviser et al., 1985; Grambow and Strachan, 1988; Iseghem and Grambow, 1988; Zwicky et al., 1989). Grambow’s model assumes that the surface reaction (matrix dissolution) controls the overall dissolution process and considers the influences of chemical affinity and solubility effects on the dissolution rate. This model also considers the diffusion effect of silicic acid through the alteration layers on the glass surface and an empirical long-term dissolution rate. A general rate equation for matrix dissolution of glass is shown as follows: rm = k+(l-exp(g))

where r, is the rate of glass matrix dissolution, the forward rate constant, and A* the affinity of the rate limiting reaction.

k,

where IAP”is the activity product and K*is the stability constant for the rate-limiting reaction. In the most simple case for glass dissolution, the

99

J. Yan, I. Neretnieks / The Science of the Total Environment 172 (199.5) 95-118

rate limiting SiO, + 2H,O

Therefore, the dissolution kinetics of a waste glass is described by four major parameters:

reaction is considered as: -+ H,SiO,

(4) . . . 0

Then:

where aH,sio 4 is the activity of silicic acid in the solution. Considering the influence of silica transport through the surface layers on dissolution rate, the transport rate rt can be given by a modification of Fick’s first law: rt = g(aS -ah> + rain

(6)

where D is the diffusion coefficient for silica, L the thickness of the transport barrier (For a growing surface layer, L is calculated from the amount of glass reacted at each reaction step and for the Nerst’s film, a fixed value for the film thickness may be used (Grambow, 1987)), a, and a,, are the activities of silicic acid at the glass surface and in bulk solution, respectively, and rfin the final rate of reaction, The last term accounts for the experimental observation that the dissolution of glass still continues even under saturated conditions. In this case, rt = rfin. At a steady state, i.e., rt = r, and an,sio = a, ,-k+~l-~~=r(U~-li,i,,;

(7)

the the the the the

forward rate constant k, stability constant for silica control K* final dissolution rate yfin diffusion coefficient of silicic acid through surface 1ayersD

3. Prediction of the kinetic properties glass phases

of waste

According to the theories of dissolution mechanisms and kinetics of waste glass, the dissolution behaviors of a waste glass phase can be determined by its kinetic properties and environmental conditions. The main kinetic parameters of waste glasses can be obtained from experimental measurements or theoretical prediction and estimation. So far, many dissolution kinetic data are available for nuclear waste glasses and some natural glasses but little is known about other waste glasses. Because the glass phases in conventional solid wastes such as combustion residues usually are mixed with other constituents, it is difficult to separate the glass phases from the solid waste for measurement of their kinetic properties even when the composition, structure and fractions of the glass phases have been determined by various methods. Reasonable theoretical prediction and estimation may, however, provide a way to study the dissolution kinetics of the waste glass phases. 3.1. A thermodynamic approach for prediction of chemical durability of wasteglassphases

then abg + (K” - ab> + rfi, a, =

(8) ;K*

+ k,

and g(K*

- ab> + rfi,

rm=K+

(9) ;K*

+ k,

A thermodynamic approach has been successfully proposed for prediction of the chemical durability of various glasses. This approach is based on the hydration theory of glass (Paul, 1977, 1990). The glass phase is assumed to be a mechanical mixture of orthosilicate and oxide components. The long-term chemical resistance of a glass is mainly determined by the thermodynamic activity and stability of its component oxides in an aqueous solution. The thermodynamic stability of a glass can be considered to be the

J. Yan, I. Neretnieks /The Science of the Total Environment 172 (1995) 95-118

100

stability of its component oxides, which in turn is a function of the activity of the particular oxide in glass and the equilibrium constants of hydration, ionization and complexation. This approach assumes that the calculated free energy of hydration of the glass is the chemical driving force for glass alteration or dissolution. The chemical potential, pi, of component i in the glass is the free energy of formation of the component. The chemical potential of the hydrated form of component i is assumed to be the free energy of formation of the hydrated species. Pi,glass

(10)

= AGf,i,glass

k,hydrated

(11)

= AGf,i,hydrated

The chemical driving force for glass hydration, A lFthydratj,,n,is given by: n A phydration (12) = C ( f4,glass - &hydrated) i=l

A ~hydration

=

f:

AGi,hydration

(13)

i=l

The calculated free energy of hydration of the glass is related to the free energies of hydration of its component oxides by the following equation: AGglass,hydration

=

i i=l

(AGi,hydrationoXi)

glass and the logarithm of the normalized mass loss of component i (usually SiO,), Li, under a standard leaching testing condition have been determined: AGglass,hydration

= c

l

Log

Li

(15)

where C is the constant of proportionality. This empirical relationship connects the intrinsic chemical durability of a glass phase with its composition. There is experimental evidence that a constant and maximal dissolution rate may be obtained at infinite dilution or at a high flow rate and constant temperature and pH (Lutze, 1988). The rate is only dependent on glass composition and structure. It can therefore be considered to be one of the intrinsic kinetic properties of glasses. According to the surface reaction controlling mechanism, the initial dissolution rate may be measured in conditions far from solution saturation. According to the standard leaching test condition (Strachan et al., 1981), the property of normalized mass loss, Li, reflects the initial dissolution behavior (forward dissolution rate) of the glass phase in pure water. This approach can provide a simple way to take the intrinsic kinetics of the glass phase as a function of the chemical composition of glass phase under certain conditions.

(14)

is the free energy of the where AGglass,hydration the free energy of component i glass, AGi,hydration in the glass, and Xi the mole fraction of component i in the glass. Plodinec et al. (1984a, 1984b) applied this approach to predict the chemical durability of various types of glasses covering a broad range compositions. The relative durabilities of over 300 natural and man-made glasses (including natural obsidians, tektites, basalt, pure SiO,, modern window glass, simulated medieval window glasses and nuclear waste glasses) have been compared based on their relative thermodynamic stabilities, expressed as the free energies of hydration (Jantzen, 1988a; 1988b). The linear relationships between the free energy of hydration of a given

3.2. Prediction of the dissolution behaviors of the glass phases of coal fly ashes in an aqueous solution There is a proportion of glass phase in coal fly ashes. Compared with other glass phases of solid wastes, the glass phases contained in coal fly ash have been studied in their compositions, structures, density distribution and individual particles (Hulett et al., 1980; Ramsden and Shibaoka, 1982; Roy et al., 1985; Hemmings and Berry, 1986; Hemmings et al., 1987; McCarthy et al., 1987; Hemmings and Berry, 1988; Odler and Zysk, 1989). We take the glass phase of coal fly ash as an example to study the dissolution or alteration characteristics of waste glass phases in aqueous solutions. It is known that the composition of glass phases in coal fly ashes may vary from ash to ash and

101

J. Yan, I. Nerehieks / The Science of the Total Environment 172 (1995) 95-118 Table 1 Chemical composition and coal sources of coal fly ash samples Sample

1 2 3 4 5 6

Coal sources

Major chemical composition (%) SiO,

Al,O,

Fe@3

59.80 43.20 47.88 57.76 60.53 66.41

22.10 21.00 24.63 27.38 25.01 24.49

3.30 24.20 1.31 0.43 0.34 0.07

Fe0

CaO

12.00 3.95 3.08 0.68

10.10 1.60 2.12 1.71 1.07 0.29

MgO

Na,O

K,O

TiO

others

1.10 1.00 1.65 1.37 0.95 0.50

2.10 0.50 0.23 0.74 0.40 0.08

0.40 2.20 0.64 1.66 1.68 0.28

0.62 0.90 1.07 1.28 1.13 1.06

1.64 6.60 6.18 1.66 3.78 3.08

Sub-bituminous coal High-Fe bituminous coal Bituminous coal Bituminous coal Bituminous coal Bituminouscoal

Data sources: the composition of sample 1 is obtained from Hemmings and Berry, 1986; sample 2 is obtained from Hemmings et al., 1987; and samples 3, 4, 5, and 6 are obtained from Ramsden and Shibaoka, 1982.

from particle to particle. In consideration of the heterogeneity of the glass phase in composition, the six samples of glass phases used in this studyinclude that the original coals of ashes are bituminous and sub-bituminous, the types of ash are the high-Ca and low-Fe fly ash, low-Ca and high-Fe fly ash, and various silicon contents in ashes. Because density fraction may reveal a marked speciation in glass composition of coal fly ash (Hemmings and Berry, 1986; Hemmings et al., 19871, the glass phases in various density fractions of an ash have been studied (samples 1 and 2). According to two types of glass particle (solid glass and vesicular glass), the glass phases

of individual particles in coal fly ash have also been taken into account (samples 3 to 6). The chemical composition of ashes and the coal sources of the samples used in this study are listed in Table 1. The mineralogical contents for some samples are shown in Table 2. On the basis of the glass chemical compositions of various ashes, density fractions and individual particles of coal fly ashes, the forward dissolution rates of different waste glass phases in aqueous solution were calculated by means of the hydration free energies of the glass phases according to the method discussed in the last section. The thermodynamic data used to calculate the hydra-

Table 2 Major mineralogical contents of the ashes (%) for sample 1 and 2 in density fraction Sample 1

Density fractions (g/cm31

Phases

< 0.79

0.79-1.50

1.50-2.00

2.00-2.50

2.50-2.85

> 2.85

Glass a-quartz Mullite

84 1 15

77 1 22

76 2 22

77 10 13

89 11 0

97 3 0

Sample 2

Density fractions (g/cm3)

Phases

< 0.79

0.79-1.50

1.50-2.00

2.00-2.50

2.50-2.75

2.75-2.85

2.85-3.00

> 3.00

Glass a-quartz Mullite Hematite Spine1

81 0 19 0 0

61 3 36 0 0

58 7 35 0 0

862 11 27 0 0

55 24 18 1 2

63 10 24 1 2

68 5 22 2 3

34 2 0 29 35

Data sources: the sample 1 is obtained from Hemmings and Berry, 1986; sample 2 is obtained from Hemmings et al.. 1987.

102

J. Yan, I. Neretnieks

/The

Science

of the Total Environment

tion free energies of glass phases are obtained from Plodinec et al. (1984a and 1984b). The chemical compositions of various glass phases and calculation results of the dissolution rates are given in Table 3 - Table 6. For the samples of glass phase based on the density distribution, the average value of dissolution rate and glass composition is the weighted average of all density fractions of a sample (samples 1 and 2). For the samples of individual glass particles, the average dissolution rate of glass phase in a coal fly ash is the average value of all investigated glass particles in the sample, and the glass phase that the dissolution rate is closest to the average dissolution rate is considered as average glass in a coal fly ash (samples 3, 4, 5, 6). Table 6 shows the calculation results of various individual particles for sample 5. Other samples are treated in a similar way. Temperature effects have been taken into account in the calculation of dissolution rates of waste glass phases under various temperatures. For many silicate glass phases, including waste glasses and some natural glasses and silicate minerals, the dissolution rates usually follow the Arrhenius temperature dependence in dilute aqueous solutions: k = Aexp( -E,/RT)

95-118

Table 7 shows some apparent activation energies of waste and natural glass phases. In this study, we use 65 kJ/mole as apparent activation energy to calculate the temperature effects on forward dissolution rates for all glass phases of the coal fly ashes. Another important kinetic parameter of waste glass dissolution is the final dissolution rate. This parameter reflects the alteration property of the glass phase when the aqueous phase reaches saturation with respect to the major component (usually Si) of the glass phase. According to the similarity of multi-component silicate glass phases in dissolution kinetics, we compare the final dissolution rates with corresponding forward rates for some multi-component silicate glasses (Table 8). The two dissolution rates differ by two to three orders of magnitude. Therefore, we assume that the final dissolution rate is lower by 2.5 orders of magnitude than the forward dissolution rate for the glass phase of coal fly ash. The final dissolution rates of the six samples are given in Table 9. 4. Simulation glass phase

of the dissolution

behaviors of waste

In order to assess the dissolution behaviors of waste glass phases in possible disposal environments, particularly for long-term dissolution or alteration of the glass phase, the geochemical modeling of dissolution of waste glass phases is performed based on the dissolution kinetics of

(16)

where k is the reaction rate constant, A is a constant, E, the activation energy constant, R the gas constant, and T the absolute temperature. Table

172 (1995)

3

Dissolution rates calculated for the glasses of high-Ca and low-Fe coal fly ash according to the density fraction (Sample 1) Density Cg/cm3)

Fraction (%I

Raw ash < 0.79 0.79-1.50 1.50-2.00 2.00-2.50 2.50-2.85 > 2.85 Average

1.4 7.6 23.2 28.8 29.2 9.9

Chemical composition (%) SiO,

f&O,

Fe203

CaO

M@

Na,O

K,O

Totals

~&rnoO

;mol/m’s)

59.8 64.7 67.6 69.7 63.2 51.2 43.1 59.6

22.1 23.5 18.1 15.6 23.3 20.9 19.2 20.0

3.3 3.3 3.2 2.3 2.6 4.7 6.4 3.6

10.1 2.8 5.9 5.2 6.8 17.5 23.9 11.1

1.1 1.3 1.3

2.1 3.6 2.6 2.2 2.1 2.3 1.7 2.2

0.4 0.1 0.3 0.4 0.3 0.2 0.3 0.3

99.3 99.0 96.5 99.2 98.3 96.3 98.0

9.205 7.347 9.079 6.958 - 8.507 - 17.19 -3.64

2.52 x 3.30 x 2.75 x 3.24x 1.76 x 4.34 x 1.13 x

1.1

0.9 1.5 1.7 1.2

10-l’ 10-11 lo-” lo-” lo-” lo-l* 10-10

The data source of glass composition and density fraction: Hemmings and Berry, 1986; AG is the hydration free energy of glass phase;r is the forward dissolution rate for Si in pure water and is calculated from the AG at 25°C; the average composition of glass phase is weighted average of the density fractions.

J. Yan, I. Neretnieks

/ The Science

of the Total Environment

I72 (1995)

103

95-118

Table 4 Dissolution rates calculated for the glasses of low-Ca and high-Fe coal fly ash according to the density distribution (Sample 2) Density (g/cm3)

Fraction (%o)

Raw ash < 0.79 0.79-1.50 1.50-2.00 2.00-2.50 2.50-2.75 2.75-2.85 2.85-3.00 > 3.00 Average

0.3 1.8 3.8 8.0 9.0 52.8 8.7 15.6

Chemical composition (%) SiO,

Also,

Fe203

CaO

WO

Na,O

K,O

Totals

kY/rnof)

;mol/m’

s)

43.2 59.4 66.7 67.0 66.9 60.8 54.7 51.1 44.0 54.9

21.0 22.7 9.8 4.5 8.4 9.8 12.1 11.4 29.9 14.0

24.2 8.2 12.0 16.1 14.5 18.6 20.3 25.2 13.7 18.7

1.6 0.6 0.8 1.1 1.2 2.3 2.8 2.9 4.4 2.7

1.0 1.6 1.8 1.7 1.3 1.5 1.8 1.6 1.9 1.7

0.5 0.7 0.8 0.9 0.8 0.8 1.0 0.8 0.5 0.9

2.2 5.5 6.2 6.3 4.8 3.9 4.7 3.9 2.1 4.3

98.7 98.1 97.6 97.9 97.7 97.4 96.9 96.5 97.2

4.862 2.038 9.079 5.372 3.720 - 0.644 - 0.782 1.895 0.651

3.95 x 6.30 x 2.75 x 4.17 x 4.65 x 7.16 x 6.81 x 4.21 x 6.13 x

lo-” 10-l’ lo-l1 10-11 10-l’ 10-l’ lo-” lo-l1 10-l’

The data source of glass composition and density fraction: Hemmings et al., 1987; AG is the hydration free energy of glass phase; Y is the forward dissolution rate for Si in pure water and is calculated from the AG at 25°C; the average composition of glass phase is weighted average of the density fractions.

waste glass. Because the initial (forward) and final dissolution rates represent two limits of kinetic properties of waste glass during dissolution processes, the modeling is focused on these limit situations. The surface layer of the reacted glass phase is considered to have little or no effect as a diffusion barrier. Trace heavy metals, Zn, Cu, Pb and Cr, have been incorporated with the glass phase to examine the release of the trace elements during the dissolution of the glass phase. The content of every trace element in the glass phases is about 0.1% (weight percentage), close to the higher content of the elements in coal fly ash (Theis and Gardner, 1990; Murarka et al., 1991). A landfill is considered to be the disposal environment of solid waste. About 10% of the average precipitation (0.5-0.8 m3/m2 year in Sweden) is assumed to flow through the waste deposit site in the simulation. 4.1. Simulation system and geochemical modeling

A volume unit in a combustion residue monofill is considered as a fixed bed flow through reactor. Glass phase dissolution takes place together with release of other components of solid waste in the reactor. The dissolution process of waste glass is evaluated by aqueous phase composition of the system outflow when the system reaches a steady state. This system is shown in Fig. 1. A simple model is proposed for simulation of dissolution of the waste glass phase. The major assumptions of

the model are: l

The state of the system is mainly controlled by kinetic parameters of glass dissolution and reaction time.

Water Inflow

u ,

I

‘..:..:..:..:..:.

::...:.-.‘.

.:..:. *:. .:. -:. .:. ..‘...‘. *.a. ..‘. ..>: . . . . . . . . . .. . . . . . . .

Equilibrium reactions in aqueous phase

I

I~:::‘::::i 1~::~::~ ......* ..,..... u outflow

Fig. 1. A fixed bed flow-through reactor system used to simulate dissolution of waste glass.

0.07

0.34

0.43

1.31

Fe203

12.00 24.84 3.95 3.91 3.08 4.25 0.68 0.21

FeOa 0.64 0.52 1.66 3.35 1.68 5.23 0.28 2.56

K,O 0.23 0.71 0.61 0.40 0.33 0.08 0.09

Na,O 2.12 2.42 1.74 0.26 1.07 0.25 0.29 0.25

CaO 1.07 1.41 1.28 1.17 1.68 0.24 1.06 0.53

TiO,

96.62

99.67

100.9

103.8

Totals

15.70

6.874

8.786

- 8.447

4.777

11.13

10.85

11.13

AG(kJ/mol) Mean S.D.

1.643

0.6630

11.00

29.74

Mean

r(mol/m*

1.711

0.9979

20.53

26.75

S.D.

s> X lo-”

aTotal iron is expressed as Fe0 for the glass phases; AG is the hydration free energy of glass; r is the forward dissolution rate for Si in pure water and is calculated from the AG at 25°C; S.D. means standard deviation; The data source of chemical compositions of glass particles: Ramsden and Shibaoka, 1982 (sample 3 is the Liddell fly ash; sample 4, the Munmorah fly ash; sample 5, the Vales Point fly ash; and sample 6, the Wallerawang fly ash).

1.65 3.55 1.37 1.52 .\ 0.95 0.56 0.50 0.31

47.88 47.33 57.76 58.07 60.53 63.61 66.41 61.09

3 Raw ash Average glass 4 Raw ash Average glass 5 Raw ash Average glass 6 Raw ash Average glass

24.63 23.75 27.38 32.05 25.01 25.20 24.49 31.58

Chemical composition(%) SiO, ~2% NO

Samples

Table 5 Dissolution rates calculated for the glass phases according to the individual particles in coal fly ash (Samples 3, 4, 5 and 6)

k p.

3 &

s 3 ts, 21 3

5

J. Yan, I. Neretnieks Table 6 Dissolution Particle

rates calculated Chemical

No

Solid glass 1 2 3 4 5 6 7 8 9 10 11 12 13 Vesicular

glass 14 15 16 17 18 19 20 21 22 23 24 2.5

/The

for the glass phases composition

Science of the Total Environment

according

to the individual

particles

192 (1995)

in coal fly ash (Sample

(%)

5)

AG(KJ/mol)

r(mol/m3s>

SiO,

Also,

MS

FeOa

K,O

Na,O

CaO

32.17 40.54 41.91 44.47 46.06 51.01 52.12 54.52 54.68 57.49 65.87 66.70 71.45

12.97 15.79 18.74 19.75 33.33 32.14 33.86 23.73 19.49 23.06 18.56 13.05 20.93

2.08 0.43 12.56 4.43 8.10 6.17 0.51 0.69 0.57 5.52 0.04 0.54 0.47

44.43 1.31 17.00 26.85 0.62 5.62 10.53 4.68 1.42 2.22 0.10 16.37 0.90

0.40 3.75 1.83 0.89 0.57 1.76 0.92 2.72 4.72 0.53 12.05 0.81 7.67

0.26 0.72 0.12 0.21 0.10 0.50 0.20 0.41 0.48 0.29 2.30 0.10 0.57

1.16 0.51 0.05 0.78 11.13 0.05 0.43 1.04 0.35 8.91 0.22 0.55 0.24

1.84 38.10 5.20 1.12 0.22 2.65 0.39 9.59 17.64 1.55 0.34 0.36 0.20

- 18.24 26.40 - 15.95 - 13.45 9.761 2.000 7.477 12.70 16.87 - 2.397 - 2.088 4.506 6.807

3.51 1.81 3.44 2.68 1.76 4.55 2.39 1.31 7.88 8.88 9.83 4.41 3.56

x x x x x x x x x x x x x

10” lo-l3 lOWI lo-l1 lo-l1 10-12 10-” lOW12 lo-l3 10-12 10-l’ 10-12 1O-‘2

47.46 53.17 54.72 56.96 58.52 65.25 65.36 65.53 65.63 67.26 67.68 70.45

41.62 44.25 39.88 31.99 26.86 21.53 15.65 21.25 24.46 20.82 24.91 18.59

0.66 0.58 0.43 1.82 0.48 0.72 0.30 0.44 0.59 0.70 0.73 0.91

0.74 1.08 1.36 5.65 1.43 1.05 1.51 1.00 1.94 1.32 1.61 1.20

0.71 0.89 1.72 0.71 1.50 9.49 1.22 1.46 2.19 7.11 2.80 4.70

0.38 0.26 0.24 0.24 0.53 0.30 0.00 0.33 0.17 0.22 0.48 0.04

0.20 0.21 0.87 0.37 0.50 0.11 0.87 0.74 0.58 0.07 0.31 0.08

0.12 0.00 0.40 1.44 1.03 0.87 14.33 0.60 2.81 0.12 0.84 0.81

14.98 15.23 13.51 10.68 14.64 3.376 22.27 15.60 14.71 7.109 13.24 11.69

8.60 9.36 1.19 1.76 1.11 4.96 4.83 1.10 1.23 3.22 1.52 1.19

x x x x x x x x x x x x

10-13 lo-r3 10-12 lo-l2 lo@12 lo-” lo-l3 10-12 lo-l2 10ml’ lo-= lOW’2

aTotal iron expressed as FeO; AG is the hydration is calculated from the AG at 25°C; The data source

TiO,

105

95-118

free energy of glass; r is the forward dissolution rate for Si in pure water and of chemical compositions of glass particles: Ramsden and Shibaoka, 1982.

o Dissolution of the glass phase takes place in a pH buffering environment (because of the release of buffering components during dissolution of solid waste). o Dissolution of the waste glass is congruent for all glass components including trace metals. o Except for the dissolution reaction of the glass phase, other reactions quickly reach equilibria in the system. A geochemical computer code-STEADYQL is used to simulate the dissolution processes of the waste glass phases. This computer program is based on a quasi-steady-state model that considers chemical processes in three time scales: fast reversible processes in the aqueous phase, described in terms of chemical equilibrium; slow

solid dissolution processes described by kinetic equation and the steady-state solution is found; and very slow processes, which are considered to be invariant in time for the solution of the system. The numerical algorithm incorporates these principles and is formulated by: mass action equations, rates of slow processes and fluxes, and flux balance and mole balance. A mathematical description is given by Furrer et al. (1989, 1990) in more detail. 4.2. Simulation parameters and conditions

The samples used for simulation are the same as ones in section 3.2. The simulation parameters and conditions may be divided into system parameters, kinetic parameters and dissolution reactions. Most of the parameters are selected on

106

J. Yan, I. Neretnieks

/ The Science of the Total Environment

I72 (1995)

95-118

Table 7 Apparent activation energies of dissolution for silicate glass phases Glass phases

Activation energy &J/m00

Conditions

Ref.

General waste glasses High-silica glass Tholeiitic glass Basaltic glass Tektite glass Quartz cu-Cristobalite P-Cristobalite Amorphous silica

60-90 73.2 + 6.3 65 60 79.6 67.4-76.6 68.7 65.0 60.9-64.9

Various compositions Distilled water, 6-70°C Seawater, 3-60°C Aqueous solution, 50-200°C o-100°C Aqueous solution, O-300’% Aqueous solution, 0-300°C Aqueous solution, 0-300°C Aqueous solution, 0-300°C

1 2 3 4 5 6 6 6 6

References: (1) White, 1986; (2) Barkatt et al., 1981; (3) Crovisier et al., 1985; (4) Guy and Schott, 1989; (5) White, 1988; (6) Rimstidt and Barnes. 1980.

the basis of the environments of waste disposal and the physical and chemical properties of coal fly ashes. The systemparameters:

Reactor depth H = l(m) 0 Inflow rate R, = 2 x lo-’ (m3m-%-r) 0 Inflow H,COt CH,COg y = lo-’ (mol dmp3) (Pco, = 10-3.50 atm) . Buffering system pH = 9 . Porosity of the ash in reactor 19= 0.5 0 Bulk density of the ash p = 1.25 (g cme3> . Specific surface area S, = 0.5 (rn’g-‘>

l

The kinetic parameters:

Dissolution rates of the waste glasses (mol rn-‘s-l) Forward rates Final rates Samples 13 x lo-lo 3.57 x lo-l3 1 Table 8 The final dissolution rates compared with corresponding forward rates for multi-component silicate glasses Glasses

SAN60 SM58 JSS-A SRL-131 PNL76-68 Basalta

1.7 1.5 3.0 r&/m2 d) 1.5 rs,,(g/m2 d) 0.005 0.003 0.0025 0.033 570 600 90 300 rfor/rfin

1.8 0.0075 240

3-20a O.la 30-200

ameans natural basalt glasses and the dissolution rates are expressed as wm/lOOO years;rf,,.. the forward dissolution rate; rti: the final dissolution rate (at Si-saturation). Data sources: Grambow et al., 1985b; Grambow et al., 1986; Grambow and Strachan, 1988; Iseghem and Grambow, 1988.

2 3 4 5 6

6.13 x lo-r1 2.97 x 10-l’ 1.10 x lo-l* 6.63 x lo-l2 1.64 x lo-r1

1.94 9.40 3.48 2.10 5.19

x x x x x

lo-l3 lo-l3 lo-l3 lo-l4 lo-l4

Dissolution reactions: Samples Reactions

Glass + 1.879Hf --f SiO, (aq) + 0.396AIf3 + 0.046Fe3+ + 0.200Ca2+ + O.O78Na+ + 0.031Mg2+ + 0.0019Cr3+ + 0.0016Cu2+ + 0.0005Pb2+ + 0.0015Zn2+ + 0.9395H 0 Glass + 1.951H+ + SiO, (aq) + 0.3iOf+~l+~ + 0.236Fe3+ + 0.054Ca2+ + O.l28Na+ + 0.047Mg2+ + 0.0019Cr3+ + 0.0016Cu2+ + 0.0005Pb2+ + 0.0015Zn2+ + 0.9755H20 Glass + 3.442H+ -+ SiO, (aq) + 0.592Al+3 + 0.435Fe3+ + 0.055Ca2+ + O.O14Na+ + 0.112Mg2+ + 0.0019Cr3+ + 0.0016Cu2+ + 0.0005Pb2+ + 0 OO15Zn2+ + 1.72lH,G Glass + 2.313H’ + SiO, (aq) + 0.650AI+3 + 0,056Fe3’ + 0.005Ca2’ + O.O94Na+ + 0.039Mg2+ + 0.0019Cr3+ + 0.0016CuZf + 0.0005Pb2+ + 0.0015Zn2’ + 1156H,O Glass + 1.727H’ + SiO, (aq, + 0.466Al+3 + 0.056Fe3’ + 0.004Ca2’ + 0.114Naf + 0.013Mg2+ + 0.0019Cr3+ + 0.0016Cu2+ + 0.0005Pb2+ + 0.0015Zn2+ + 08635H,O Glass + 1.933Hf -+ SiO, (aq) + 0.610AIf3 + 0.003Fe3’ + 0.004Ca2’ + 0.056Na’ + 0.008Mg2 + + 0.0019Cr3+ + 0.0016Cu2’ + 0.0005Pb2+ + 0.0015Zn2+ + 0.9665H20

J. Yan, I. Neretnieks

/ The Science

of the Total Environment

107

172 (1995) 95-118

Table 9 Estimation of the final dissolution rates for the glass phases of coal fly ashes Samples rsn (mol/m%)

1 3.57 x lo- l3

2 1.94 x lo- l3

3 9.40 x 10-13

4.3. Simulation results In order to investigate the dissolution or alteration behaviors of waste glass phases during a long-term period, the results of simulation of waste glass dissolution are given in two forms. One shows the potential contribution of waste glass dissolution for the concentrations of released species in the aqueous phase of system outflow (Table 10). It can be used to evaluate the influence of glass dissolution or alteration on release concentration of a species. The potential concentration stands for the available concentration of a species dissolved due to glass dissolution if other reactions do not affect the concentration of the species in the system. In this way, the glass dissolution or alteration can be separated from other additional reactions, for example precipitation or adsorption, in the complex processes. Only the results for the forward dissolution rates are shown. For the final dissolution rate, the concentrations are 316 times (2.5 orders of magnitude) lower. Another form of the simulation results is given in the specific flux of dissolution or alteration of the glass phase with respect to the cross section area of the reactor (Table 11). The specific fluxes of various species of the glass phase express the relationships between the amount of dissolution or alteration of waste glass and reaction time in the system. These values are obtained under the dissolution reaction having reached steady state in the system and at two limited dissolution kinetic conditions. The values are shown in Table 11, only for the forward dissolution rates. One of the most important behaviors of toxic elements (heavy metals) during glass dissolution is that these species may be retarded into the aqueous solution by various kinds of trapping or retention effects. If the dissolution of glass phase is not a limiting factor, their release rates should be lower than that of glass matrix dissolution.

4 3.48 x lo- l3

5 2.10 x lo- l4

6 5.19 x lo- l4

Solubility limit, co-precipitation (Grambow, 1982; Petit et al., 1989,199O) and adsorption (Barkatt et al., 1981; Lee and Clark, 1986; Petit et al., 1990) are considered as main mechanisms that account for the retardation mechanisms. In this study, we only assess the solubility limit for glass constituents after glass network degradation. An equilibrium program PHREEQE (Parkhurst et al., 1980) is used to evaluate the aqueous chemistry and potential precipitation of secondary solid phases for the system outflow. The selection of complexation reactions in aqueous phase and dissolution/precipitation reactions of the potential minerals is based on Rai et al., 1987 (Rai and Zachara, 1984; Eary et al., 1990; Mattigod et al., 1990) and our early work (Yan and Neretnieks, 1994a). The speciation of species in aqueous phase and the saturation states (saturation index SI = log (LAP/K), where the IAP is the ion activity product and the K is the solubility constant of the mineral) can be obtained from these geochemical calculations. The calculations are performed under two situations or the alkaline condition (pH = 9, it is the same as the system outflow) and the weakly acidic condition (pH = 4, it simulates that the buffering capacity of the system is consumed). The saturation states of the outflow solutions from simulation system are shown in Table 12 15. 5. Discussion We would like to compare our simulation results with actual measured data. However, because there are some limitations in test methods, sample separation and of difficulties to take into account the time factor for investigation of the glass dissolution in solid wastes and because real waste always contains mineral phases with different kinetics and compositions, this has not been possible. Furthermore the simulations are

x 10 x 101 X lo1 x 101 x 10’ X 10’

7.98 x 3.32 x 2.49 x 1.21 x 5.85 X 1.86 x

100 loo lo1 101 10-l 10’

9.27 x 2.61 x 1.83 x 1.04 x 7.03 x 9.17 x

10-l 10’ 10’ 100 lo- 2 1o-3

4.03 x 5.97 x 2.32 x 9.33 x 5.02 x 1.22 x

100 10-l lo- ’ 10-Z 1O-3 10-Z

625 x 10-l 5.19 x lo- 1 4.72 X 10’ 7.27 x 10’ 1.63 x lo-’ 2.44 x lo- 2

2.02 1.10 4.21 1.87 1.26 3.05

1.13 x 6.13 x 2.97 x 1.10 x 6.63 x 1.64 x

1 2 3 4 5 6

lo- lo lo-” 10-l’ 10-10 lo-l2 lo-”

Potential concentrations of the species in an aqueous phase (mol/l) Al Fe Ca SiO, Mg

r

(mol/m2 s)

Samples No. 3.02 x 1.66 x 6.32 x 2.80 x 1.88 x 4.58 x

Zn lo-’ 10-2 lo-* 10-Z 1O-3 10-3

3.22 x 1.77 x 6.74 x 2.98 x 2.01 x 4.89 x

cu lo- 2 lo- 2 lo-’ lo- 2 lo- 3 10-3

1.01 x 5.53 x 2.11 x 9.33 x 6.28 x 1.53 x

Pb 10-2 10-3 10-2 10-3 lo- 4 10-d

3.83 x 2.10 x 8.00 x 3.54 x 2.39 x 5.80 x

Cr lo- 2 10-2 lo-’ 10-Z 1O-3 10-3

Table 10 Simulation results for glass phase dissolution in a fixed bed flow-through reactor system for forward dissolution rate I. The potential contribution of glass dissolution for the concentrations of an aqueous phase. (for a final dissolution rate all concentrations are 316 times lower)

Samples No.

1.13 x 6.13 x 2.97 x 1.10 x 6.63 x 1.64 x

lo-‘0 10-r’ 10-l’ 10-10 10-l’ lo-”

;mol/m2s) 5.04 2.09 1.57 7.65 3.69 1.18

1.27 x 6.97 x 2.66 x 1.18 x 7.92 x 1.93 x

lo3 lo2 lo3 10s 10’ 102

Al

SiO, x x x x x x

102 102 103 lo2 10’ 10’

5.85 x 1.65 x 1.16 x 6.59 x 4.44 x 5.79 x

Fe 10’ 10’ lo3 10’ 10s 10-l

2.54 x 3.77 x 1.46 x 5.89 x 3.17 x 7.71 x

Ca lo2 10’ 10’ IO0 10-r 10-l

Reactor area specific fluxes of species (mol/m2 year) 3.94 3.28 2.98 4.59 1.03 1.54

Mg x x x x x x

101 IO’ 10’ 10’ 100 10s

1.19 x 1.05 x 3.99 x 1.77 x 1.19 x 2.89 x

Zn 100 100 100 10” 10-l 10-l

2.04 x 1.12 x 4.25 x 1.88 x 1.27 x 3.09 x

CU 10’ 100 10’ 10s 10-r 10-l

6.36 x 3.49 x 1.33 x 5.89 x 3.96 x 9.64 x

Pb 10’ 10-r 10s 10-r lo-’ lo-’

2.42 x 1.33 x 5.05 x 2.24 x 1.51 x 3.66.x

Cr 10’ 100 100 10” 10-l lo- ’

Table 11 Simulation results for glass phase dissolution in a fixed bed flow-through reactor system for forward dissolution rate II The amount of dissolution or alteration of the glass phase in the system per m2 of landfill area (for a final dissolution rate all data of glass dissolution or alteration are 316 times lower)

110 Table 12 The saturation Mineral

J. Yan, I. Neretnieks

states of the system Formula

outflow

/The

with

Science

respect

CaCOa CaMg(COJ, SiO,(am) SiO A&& NOH&d Fe(OH),bm) Al,Si,Os(OH), Ai,Si,O,(OH), Ca02AI,Si,OlO(OH), Mg$i,O,(OH), Mg,Si,O,.s(OH)&I,O MgC0,3.H,0 NaAiSiaOs NaAlSisOs NaAlSi,O,H,O Cr(OH)s(am) Cr(OH),(x = 0.6) Cr(OH),(x = 0.1)

cue cUSiO,(OH), PbC03 PbO(yellow) ZnCOs Zn(OH),(am) ZnO

to potential

minerals

(for

172 (1995)

forward

95-118

dissolution

rate and pH = 9)

Sl=log(lAP/K) Samples

CALCITE DOLOMITE Si02a CHALCEDY GIBBSITE AKOH), a Fe(OH), a KAOLINIT HALLOYSI MONTMORI CHRYSOTL SEPIOLIT NESQUEHO L ALBITE H ALBITE ANALCIME CriOH); CdOH); Cr(OH); * cuco, Cu(OH), TENORITE DIOPTASE CERRUSIT MASSICOT PbSiO, SMITHSON Zn(OH),a zINcITE ZnSiO,

of the Total Environment

1

0.6506 0.6215 5.7680 6.6076 7.8469 5.1569 8.0811 30.9972 29.4622 44.0306 28.2679 28.7671 - 4.6468 32.3593 31.0493 25.3872 5.7174 7.5474 9.3774 - 3.8370 4.4858 6.2116 10.0820 1.4905 2.7652 12.9114 - 2.0015 2.2413 3.5572 15.2234

2

3

4

5

6

0.3337 1.0405 4.7072 5.5468 6.8649 4.1749 8.0607 26.6863 25.1513 37.0282 25.1632 25.4173 - 3.2350 27.2527 25.9427 21.5667 4.8609 6.6909 8.5209 - 3.4404 3.8938 5.3944 8.4293 2.0830 2.1438 11.2292 - 0.8476 2.4066 3.4971 14.1026

0.5186 1.4758 6.2660 7.1056 8.4458 5.7558 9.3575 33.2800 31.7450 47.4192 31.4521 31.5270 - 3.9275 34.2538 32.9438 26.6947 6.1437 7.9737 9.8037 - 3.2294 4.8068 6.6217 10.9011 2.0132 3.0903 13.1346 - 1.5932 2.3630 3.7679 15.9321

-0.1594 1.0720 5.3547 6.1943 7.8434 5.1534 8.0653 30.0500 28.5150 41.8581 27.4453 27.7759 - 3.0455 30.5112 29.2017 24.0660 5.5030 7.3330 9.1630 - 2.8009 4.3006 5.9129 9.4835 2.6798 2.6195 12.3525 - 0.5937 2.4277 3.6300 14.8829

- 1.0874 - 1.4778 2.8084 3.6480 5.2047 2.5147 5.8984 19.4034 17.8684 25.3426 14.9366 15.7923 - 3.8374 18.3953 17.0853 14.7730 3.0098 4.8398 6.6698 - 3.9753 2.3681 3.7037 5.0047 1.4926 0.3976 7.5843 - 0.6777 1.5856 2.5113 11.2180

-0.7732 - 1.0160 3.3240 4.1636 5.8022 3.1122 5.1015 21.6531 20.1187 28.7491 16.5136 17.6506 - 3.7624 20.6602 19.3502 16.4981 3.4899 5.3199 7.1499 - 3.5321 2.8377 4.1975 5.9899 1.9230 0.8786 8.5808 - 0.3248 1.9650 2.9148 12.1371

Cr(OH)t are considered as the solid solutions Cr,Fe,,(OH)a .~ __land Cr(OH)?* and Zachara, 1988). For &OH):, x = 0.6 and Cr(OH):*, x = 0.1.

focused on two limiting situations of glass phase dissolution. Therefore, it is difficult to compare our modeling results directly with the current experimental and field observation data of leaching from the solid wastes. So far, only some nuclear waste glasses and natural glasses have been systematically studied under the well-controlled laboratory conditions. Compared with nuclear waste glass, the glass phases in combustion residues are mixed with other minerals or components instead of being a relative homogeneous phase such as nuclear glass. This characteristic makes the dissolution behaviors of the glass phase more difficult to investigate experimentally. For

that are formed

by Cr(OH),

and Fe(OH)s

(Rai

example, the final dissolution rate of glass phase can indirectly be measured from the solution concentration changes of the soluble species dissolved from the glass matrix after the solution is saturated with respect to silica (Freude et al., 1985; Grambow; 1987). For the glass phases in solid wastes, however, it is not easy to find a soluble species that only exists in the glass phase and can be accurately determined by current analytical methods. Similar problems also appear in the measurements of other dissolution behaviors of the glass phases. Another critical reason for the absence of glass dissolution data is that the glass phases in most investigations are considered

111

J. Yan, I. Neretnieks / The Science of the Total Environment 172 (1995) 95-118 Table 13 The saturation Mineral

states of the system

outflow

Formula

with

respect

CaCO, CaMdCO,), SiOJam) SiO, ANOH), MOH),(am> Fe(OH)&m) AI,Si,O,(OH), Al,Si,O,(OH), Ca0.2Al,Si,01,(OH),

Mg&O,(OH), Mg,Si30,.,(0H),H,0 MgC0,3H,O NaAISl,Os NaAlSi,O, NaAlSi,06H,0 Cr(OH)Jam) Cr(OH),(x = 0.6) Cr(OH),(x = 0.1)

cue CuSiO,(OH)Z PbCO, PbO(yellow) ZnCO, Zn(OHkJam) ZnO

minerals

(for final

dissolution

rate and pH = 9)

sz = log (LAP/K) Samples

CALCITE DOLOMITE SiO,a CHALCEDY GIBBSITE AltOH),a Fe(OH),a KAOLINIT HALLOYSI MONTMORI CHRYSOTL SEPIOLIT NESQUEHO L ALBITE H ALBITE ANALCIME CdOH); Cr(OH)il; Cr(OH); * cuco, Cu(OH), TENORITE DIOPTASE CERRUSIT MASSICOT PbSiO, SMITHSON Zn(OH), a ZINCITE ZnSiO,

to potentail

1

- 0.3560 - 1.3814 1.4501 2.2897 3.8716 1.1816 4.4886 14.0057 12.4707 17.3310 10.2046 10.4059 - 4.4281 11.5567 10.2467 9.3075 1.7485 3.5785 5.4085 - 5.2868 1.0248 2.3456 2.3031 0.2871 - 0.8544 4.9739 - 1.9047 0.3269 1.2378 8.5861

as an inert material. Little or no attempt is made to relate their contributions to leaching of the solid wastes. In this study, we do not primarily aim at providing an accurate description for the dissolution behaviors of the glass phases in the solid wastes. We try to address an important issue of whether the glass phases can be regarded as inert materials or if they under landfill conditions significantly could contribute to the release of toxic metals. The results can be used to direct our future work even if this answer is quasi quantitative or needs to be modified. The validity must be tested by more accurate theoretic and experimental investigations. On the other hand, it is impossible to reproduce the long-term behaviors of the waste glass in

2

3

- 0.9580 - 1.8413 1.1898 2.0293 3.5133 0.8233 4.9428 12.7681 11.2331 15.4137 9.6938 9.6322 - 4.2849 10.5696 9.2596 8.5812 1.5098 3.3398 5.1698 -5.4135 0.7594 2.0799 1.7774 0.1846 - 1.0960 4.4720 - 2.0155 0.0775 0.9880 8.0760

- 0.7222 - 0.9856 1.7741 2.6137 4.3396 1.6496 5.7846 15.5912 14.0562 19.5547 13.1888 12.9324 - 3.6701 12.5655 11.2555 9.9910 2.0406 3.8706 5.7006 - 4.9780 1.3499 2.6720 2.9522 0.5586 - 0.5653 5.5870 - 1.6135 0.6344 1.5465 9.2189

4 - 1.7941 - 2.5591 1.4171 2.2567 4.0659 1.3759 4.5436 14.3283 12.7933 17.3925 10.4681 10.5266 -4.1675 11.7342 10.4242 9.5180 1.7264 3.5564 5.3864 -5.1730 0.9880 2.3088 2.2333 0.4131 - 0.8790 4.9163 - 1.7808 0.3002 1.2110 8.5263

5 - 2.7712 - 4.9009 0.2485 1.0880 2.8024 0.1124 3.3801 9.4632 7.9282 9.9764 3.7734 4.1174 - 5.5300 5.9285 4.6185 4.8818 0.6074 2.4374 4.2674 - 6.2607 - 0.1872 1.1329 -0.1105 - 0.6211 - 2.0016 2.6251 - 2.8438 - 0.8503 0.0597 6.2064

6 - 2.4383 - 4.4450 0.6332 1.4727 3.2946 0.6046 2.4928 11.2171 9.6821 12.5941 4.9487 5.5419 - 5.4073 7.6416 6.3316 6.2101 0.9826 2.8126 4.6426 - 5.8871 0.1988 1.5189 0.6602 - 0.2572 - 1.6252 3.3862 - 2.4745 - 0.4687 0.4415 6.9728

a given disposal conditions by laboratory experiments. There are many reasons for this. Among the most important are that kinetics may change during the experiment in not observable ways, different phases including mineral and surface phases are present and these may react differently than the glass phases and obscure the release from the latter. The time to reach steady state dissolution is also often much larger than what can be allowed in experiments. Nevertheless the type of simulations we have made give important insights into what potential release might be expected under long term landfill conditions and what factors may influence the release from the amorphous glass phases.

112

.l. Yan, I. Neretniekx

Table 14 The saturation Mineral

states of the system

outflow

Formula

/ The Science of the Total Environment

with

respect

AI(

CaCO, CaMg(CO,), SiO,(am) SiO, AI(

a

Fe(OH),a KAOLINIT HALLOYSI MONTMORI CHRYSOTL SEPIOLIT NESQUEHO L ALBITE H ALBITE ANALCIME CdOH): CdOH); Cr(OH);* cuco, Cu(OH), TENORITE DIOPOTASE CERRUSIT MASSICOT PbSiO, SMITHSON Zn(OH), a ZINCITE ZnSiO,

Al(OH),(am) Fe(OH)&m) Al,Si,OS(OH), Al,Si,O,(OH), CaO~2AI,Si,010(OH)2

Mg,Si,O,(OH), Mg2Si307,5(OH)3H20 MgCO,.SH,O NaAlSi,Os NaAISi,Os NaAlSi,O,H,O

Cr(OH),(am) Cr(OH),(x Cr(OH),(x

= 0.6) = 0.1)

cue CuSiO,(OH), PbCO, PbO(yellow) ZnCO,

Zn(OH),(am) ZnO

minerals

(for

forward

95-118

dissolution

rate and pH = 4)

sz = log(i?lAP/K) Samples

CALCITE DOLOMITE SiO,a CHALCEDY GIBBSITE

to potentail

172 (1995)

1

- 2.2505 -5.1802 5.8156 6.6552 2.1104 - 0.5796 5.0935 19.6132 18.0782 29.1249 - 1.6055 8.9439 - 7.5292 21.8981 20.5881 14.8845 0.8534 2.6834 4.5134 -5.1628 - 3.9268 - 2.2071 1.7170 - 0.4885 - 6.3068 3.8871 - 4.8909 - 7.7349 - 6.4252 5.2887

From predictions of the kinetic properties of waste glass phase, it can be found that the difference in dissolution rate is about one order of magnitude for various density fractions of the glass phase in a coal fly ash. High density glass phase appears to have a higher dissolution rate. This difference reflects the variation of chemical composition of the glass phase in various density fractions. The silicon content exerts one of the most important influences on the chemical durability of the glass phase. High density glass phase has lower silicon content and higher content of calcium or iron oxide. These factors together make high density glass phase have a higher dissolution rate because reducing the silicon content of glass phase or raising CaO or Fe0 content, will

2

3

- 2.3291 - 4.5876 5.1064 5.9459 2.4311 - 0.2589 5.8536 18.6119 17.0769 26.6080 - 2.1435 7.8888 -6.1855 19.8209 18.5109 13.7407 1.1004 2.9304 4.7604 - 4.6442 - 3.7113 - 2.2157 1.2233 0.0022 - 6.3433 3.1413 -4.3721 - 7.5192 - 6.4336 4.5710

- 2.3201 - 4.2010 6.3258 7.1653 2.2875 - 0.4025 6.2478 21.0829 19.5479 31.5896 1.5731 11.7072 - 6.7656 23.2748 21.9648 15.6559 0.9502 2.7802 4.6102 - 4.7029 - 3.8281 -2.0132 2.3259 0.0027 - 6.0816 4.6224 - 4.4309 - 7.6361 - 6.2312 5.9927

4 - 2.8236 - 4.6236 5.7492 6.5888 2.3343 - 0.3557 5.1690 19.9117 18.3767 29.0864 - 1.4579 8.9671 - 6.3503 21.9536 20.6436 15.0228 0.8580 2.6880 4.5180 - 4.1394 - 3.9342 - 2.2309 1.6432 0.5417 - 6.3237 3.8037 - 3.8674 - 7.7422 - 6.4488 5.1986

5 - 5.6706 - 10.5345 3.1315 3.9710 2.3124 - 0.3776 3.6595 14.2648 12.7298 18.1154 - 13.6230 - 2.7091 -8.3111 11.5715 10.2615 7.6261 0.0303 1.8603 3.6903 - 5.2788 - 4.1970 - 2.8613 - 1.2373 - 1.2827 - 7.6392 - 0.1295 - 5.0067 - 8.0049 - 7.0792 1.9505

6 - 3.9377 - 7.0637 4.1364 4.9759 2.6471 - 0.0429 3.5245 16.9681 15.4331 22.8454 - 8.0807 2.6083 - 6.6457 15.4102 14.1002 10.4358 0.8365 2.6665 4.4965 - 4.4573 - 3.9440 - 2.5842 0.0206 0.1348 - 6.7661 1.7485 -4.1854 - 7.7521 - 6.8023 3.2323

decrease the hydration free energy of glass phase. This influence of chemical composition on the durability of the glass phase is also found in the glass phases of individual particles. Larger differences of dissolution rates can be found between individual particles. This may exhibit the heterogeneity of waste glass phase in chemical compositions. However, in comparison with the chemical compositions of the raw ashes, the composition of the average glass phase in an ash sample is very close to the composition of its raw ash, particularly for the major components. This means that the glass phase, which has very high or low dissolution rate, may only be a small fraction in the glass phases of an ash sample. Similar characteristics may also be found in the glass phases of

J. Yan, I. Neretnieks

/The

Science of the Total Environment

113

172 (1995) 95-118

Table 15 The saturation states of the system outflow with respect to potentail minerals (for final dissolution rate and pH = 4) Mineral

Formula

SI = log (LAP/K) Samples

CALCITE DOLOMITE SiO,a CHALCEDY GIBBSITE

ANOH), a Fe(OH), a KAOLINIT HALLOYSI MONTMORI CHRYSOTL SEPIOLIT NESQUEHO L ALBITE H ALBITE ANALCIME

Cr(OH),a CdOH); Cr(OH);* cllco, CdOH), TENORITE DIOPTASE CERRUSIT MASSICOT PbSiO, SMITHSON Zn(OH),a ZINCITE ZnSiO,

CaCO, CaMg(COJ2

SiO,(am) SiO, AI(

AI( Fe(OH)&m) A,Si,O,(OH), Al,Si,O,(OH), CaO.2Al,Si,O,,(OH),

Mg,Si,O,(OH), Mg,SisO,,s(OH),H,O MgC0,3H,O NaAlSiaOs NaAlSisO, NaAlSi,O,H,O

Cr(OH),(am) Cr(OH),(x = 0.6) CrtOH),(x = 0.1)

cue CuSiO,(OH)a PbCO, PbO(yellow) ZnCO,

Zn(OH),(am) zno

1

-5.5437 - 11.7434 1.5311 2.3707 1.4403 - 1.2497 1.8859 9.3050 7.7700 10.2407 - 19.9879 - 9.5875 - 9.6024 4.3322 3.0222 2.0020 - 2.4038 - 0.5738 1.2562 - 6.4803 -5.1125 - 3.7917 - 3.7532 - 3.8013 - 9.8867 - 3.9774 - 6.2083 - 8.9205 - 8.0097 - 0.5803

3

2 - 6.2912 - 12.4982 1.2635 2.1031 1.1965 - 1.4935 2.3940 8.2818 6.7468 8.5639 -20.5224 - 10.3891 - 9.6086 3.4392 2.1292 1.3770 - 2.5309 - 0.7009 1.1291 - 6.6591 - 5.2837 - 3.9632 - 4.1920 - 3.9801 - 10.0582 - 4.4165 - 6.3872 - 9.0917 - 8.1813 - 1.0196

various density fractions of an ash sample. Therefore, it is reasonable that the average glass phase should represent the glass phases of a coal fly ash on their chemical durabilities.

In glass 10-i’ been

- 5.8785 - 11.2654 1.8911 2.7307 1.7110 - 0.9790 3.1395 10.5679 9.0329 12.1333 - 16.9391 - 6.9579 - 8.7936 5.2513 3.9413 2.5597 - 2.1820 - 0.3520 1.4780 - 6.2834 - 4.9527 - 3.6305 - 3.2333 - 3.5069 - 9.6280 - 3.3586 - 6.0114 - 8.7607 - 7.8485 - 0.0591

4 - 7.1999 - 13.3505 1.5027 2.3422 1.5835 - 1.1065 1.9254 9.5346 7.9996 10.1879 - 19.9082 - 9.5818 - 9.5532 4.4546 3.1446 2.1529 - 2.4715 -0.6415 1.1885 - 6.5357 -5.1715 - 3.8506 - 3.8406 - 3.8526 - 9.9416 - 4.0607 - 6.2637 - 8.9795 - 8.0687 - 0.6678

5 - 8.1669 - 15.6891 0.3116 1.1511 0.7877 - 1.9023 0.9318 5.5601 4.0251 3.7991 - 26.3527 - 15.8616 - 10.9225 - 0.9194 - 2.2294 - 2.0292 -3.1737 - 1.3437 0.4863 - 7.4118 - 6.0299 - 4.7098 - 5.8901 -4.7111 - 10.7832 - 6.0934 - 7.1398 - 9.8379 - 8.9278 - 2.7181

6 - 7.8716 - 15.3055 0.6992 1.5388 1.1296 - 1.5604 - 0.0031 7.0191 5.4841 6.0682 - 25.3177 - 14.5257 - 10.8345 0.6383 - 0.6717 - 0.8593 - 2.9187 - 1.0887 0.7413 -7.1132 - 5.7330 - 4.4128 - 5.2055 - 4.4223 - 10.4959 - 5.4185 - 6.8413 - 9.5411 - 8.6310 - 2.0336

this study, the forward dissolution rates of phase of coal fly ashes are from 6.63 x to 2.97 X 10-l’ mol rn-‘s-l, which have shown in Table 3 - 5. In comparison with

Table 16 Comparison with other glass phases on the dissolution rates Glass phase Tholeiite basalt glass Synthetic basalt glass ABS118 waste glass SRL-131 waste glass Glass of coal fly ash Glass of coal fly ash

r&

(mol rnm2sK1)

1.31 x 3.92 x 4.01 x 1.28 x 2.97 x 6.63 x

10-s 10-m lo-‘0 1O-g 10-10 lo-l2

Ref. Guy and Schott, 1989 Byers et al., 1985 Christensen et al., 1986 Lokken and Strachan, 1984 This study (sample 3) This study (sample 5)

‘rfor is the forward dissolution rates in Si (25°C) and calculated from the testing values which are obtained from the references.

114

J. Yan, I. Neretnieks

/The

Science of the Total Environment

other glass phases that are similar to the glass phase of coal fly ash in chemical composition, the dissolution rates of glass phase of coal fly ashes are close to or lower than basalt glasses and some nuclear waste glasses (Table 16). Our modeling results indicate that if one were to neglect adsorption or precipitation of secondary phases, some of the metals may be released in millimolar to tens of millimolar concentrations from the glass phases. In some circumstances this may be higher than acceptable. The primary release from the glass phase thus cannot always be neglected. In our simulations we have so far not accounted for the formation of secondary phases or adsorption of the toxic metals. The solubilities of some of the metals are exceeded and secondary precipitates might form. This means that precipitation/dissolution of the secondary solid phases may control the compositions of the leaching solutions and solubility limit will become a retention mechanism for the release of species from waste glass alteration. Similarly, other trapping mechanisms such as ion-exchange and adsorption may also play important roles in retarding some species into aqueous solution. However there are some limitations for these retention mechanisms, many factors (pH, the property of the secondary solid phase, the redox status of the system, complexing ligands) affect the intensity and capacity of these processes. For example, under acidic conditions, the solubility limit may not be reached for many elements, in particular for transition metals. Comparing the saturation indices in Table 14 and 15 with these in Table 12 and 13, it is seen that under the acidic conditions the saturation indices with respect to many minerals will decrease and some minerals will be dissolved into aqueous phase again when it becomes acid. Although other retention processes may reduce the release rates of toxic elements when the glass dissolution is not a limiting factor, these mechanisms usually are closely related to the properties of the toxic elements themselves and environmental conditions, and can not provide a permanent barrier against release of toxic elements because the products of glass alteration do not have the

172 (1995) 95-118

durability of the glass phase and the secondary phases are much more quickly leached than glass phases in certain conditions. These results also point out that the pH buffering capacity of the combustion residues may be very important. In addition to kinetic properties of waste glass, the characteristics of solid waste and disposal environments also affect the dissolution or alteration of waste glass phases and contribute to the leaching of solid waste. When coal fly ash is disposed in a landfill, a low-flow or near closed system provides a relatively long-term reaction time, and the small particles of coal fly ash make a large surface area available for glass dissolution. In this case, system parameters may be as important as the kinetic properties of waste glass. The dissolution kinetics of waste glass may shift with changes of the system conditions. For example, the dissolution rate of waste glass may rise to near the forward dissolution rate in a fast-flow system, but the contribution of glass dissolution to the concentration of aqueous solution is still decided by the flux and mass balances of the system. Therefore, the contribution of the waste glass phase in the leaching of solid waste should be a function of the glass durability, time scale and system properties. It is not enough to only take into account the low alteration rate of waste glass phase. 6. Conclusions On the basis of the prediction of kinetic properties, the chemical durabilities of the glass phases in the coal fly ashes are similar to the basalt glass and some nuclear glasses. According to geochemical modeling, the glass phase dissolution is not always a limiting factor in the leaching processes of combustion residues (coal fly ashes). Under certain disposal conditions, the glass phase in combustion residues (coal fly ash) can not provide a permanent barrier for release of toxic elements. Additional reactions associated to toxic elements influence their release behaviors. Some retention mechanisms such as precipitation of secondary phase may become controlling factors but only under some limited conditions. In the leaching processes of combustion residues, the

J. Yan, I. Neretnieks

/ The Science

of the Total Environment

contribution of glass phases is decided not only by their kinetic behaviors but also by their other properties related to dissolution reactions and disposal environment. Special experimental methods need to be developed for investigation of waste glass dissolution in solid waste in order to provide more information to modify the research approaches. Acknowledgements The financial support of the Swedish Waste Research Council (AFR) is gratefully acknowledged. References Aagaard, P. and H.C. Helgeson, 1982. Thermodynamic and kinetic constraints on reaction rates among minerals and aqueous solutions. I. theoretical considerations. Am. J. Sci., 282: 237-285. Abrajano, Jr. T.A., J.K. Bates and J.J. Mazer, 1989. Aqueous corrosion of natural and unclear waste glasses II. Mechanisms of vapor hydration of nuclear waste glasses. J. NonCryst. Solids, 108: 269-288. Abrajano, T.A., J.K. Bates, A.B. Woodland, J.P. Bradley and W.L. Bourcier, 1990. Secondary phases formation during nuclear waste-glass dissolution. Clays. Clay Miner., 38: 537-548. Adams, P.B., 1988. Glass corrosion theory a tool for understanding the past, designing for the present and predicting the future. Mater. Res. Sot. Symp. Proc., 125: 115-127. Advocat, T., J.L. Crovisier, B. Fritz and E. Vernaz, 1990. Thermokinetic model of borosilicate glass dissolution: contextual affinity. Mater. Res. Sot. Symp. Proc., 176: 241-248. Advocat, T., J.L. Crovisier, E. Vernaz, G. Ehret and H. Charpentier, 1991. Hydrolysis of R7T7 nuclear glass in dilute media: Mechanisms and rate as a function of pH. Mater. Res. Sot. Symp. Proc., 212: 57-64. Andrade, A, Y.M.A. Coenegracht, G.G. Hollman, M. Janssen-Jurkovicova, H.S. Pietersen, S.P. Vriend and R.D. Schuiling, 1987. Leaching characteristics of fly ash after four years of natural weathering. Mater. Res. Sot. Symp. Proc., 86: 81-98. Barkatt, A., J.H. Simmons and P.B. Macedo, 1981. Corrosion mechanisms and chemical durability of glass media proposed for the fixation of radioactive wastes. Nucl. Chem. Waste Manage., 2: 3-23. Barkatt, A., B.C. Gibson, P.B. Macedo, C.J. Montrose, W. Sousanpour, A. Barkatt, M-A. Boroomand, V. Rogers and M. Penafiel, 1986. Mechanisms of defense waste glass dissolution. Nucl. Tech., 73: 140-164. Bourcier, W.L., D.W. Peiffer, K.G. Knauss, K.D. Mckeegan and D.K. Smith, 1990. A kinetic model for borosilicate glass

172 (1995)

95-118

115

dissolution based on the dissolution affinity of a surface alteration layer. Mater. Res. Sot. Symp. Proc., 176: 209-216. Bourcier, W.L., 1991. Overview of chemical modeling of nuclear waste glass dissolution. Mater. Res. Sot. Symp. Proc., 212: 3-18. Bunker, B.C., D.R. Tallant, T.J. Headley, G.L. Turner and R.J. Kirkpatrick, 1988. The structure of leached sodium borosilicate glass. Phys. Chem Glass., 29: 106-120. Byers, C.D., M.J. Jercinovic, R.C. Ewing and K. Keil, 1985. Basalt glass: an analogue for the evaluation of the long-term stability of nuclear waste form borosilicate glasses. Mater. Res. Sot. Symp. Proc., 44: 583-590. Chick, L.A. and L.R. Pederson, 1984. The relationship between reaction layer thickness and leach rate for nuclear waste glasses. Mater. Res. Sot. Symp. Proc., 26: 635-642. Christensen, H., H-P. Hermansson and I.-K. Bjiimer, 1986. Leaching of simulated nuclear waste glass under dynamic conditions. Adv. Ceram., 20: 475-485. Conradt, R., H. Roggendorf and H. Scholze, 1985. Investigation on the role of surface layers in HLW glass leaching. Mater. Res. Sot. Symp. Proc., 50: 203-210. Crovisier, J.L., B. Fritz, B. Grambow and J.P. Eberhart, 1985. Dissolution of basaltic glass in seawater: experiments and the thermodynamic modeling. Mater. Res. Sot. Symp. Proc., 50: 273-280. Crovisier, J.L., J. Homrorez and J.P. Eberhart, 1987. Dissolution of basaltic glass in seawater: Mechanism and rate. Geochim. Cosmochim. Acta., 51: 2977-2990. Cunnane, J.C., J.K. Bates, W.L. Ebert, X. Feng, J.J. Mazer, D.J. Wronkiewicz, J. Sproull, W.L. Bourcier and B.P. McGrail, 1993. High-level nuclear-waste borosilicate glass: a compendium of characteristics. Mater. Res. Sot. Symp. Proc., 294: 225-232. Doremus, R.H., 1975. Interdiffnsion of hydrogen and alkali ions in a glass surface. J. Non-Cryst. Solids, 19: 137-144. Drad, J.-C., G.D. Mea, A. Paccagnella, J.-C. Petit and L. Trotignon, 1988. The aqueous dissolution of alkali silicate glasses: reappraisal of mechanisms by H and Na depth profiling with high energy ion beams. Phys. Chem. Glass., 29: 249-255. Eary, L.E., D. Rai, S.V. Mattigod and C.C. Ainsworth, 1990. Geochemical factors controlling the mobilization of inorganic constituents from fossil fuel combustion residues: II. Review of the minor elements. J. Environ. Qual., 19: 202-214. Eberendu, A.R.N. and K.E. Daugherty, 1984. The quantitative determination of glass in slag by infrared spectroscopy. Cem. Con. Res., 14: 873-883. Eighmy, T.T., J.D. Eusden, Jr., K. Marsella, J. Hogan, D. Domingo, J.E. Krzanowski and D. Stlmplli, 1994. Particle petrogenesis and speciation of elements in MSW incineration bottom ashes. Environmental Aspects of Construction with Waste Materials. Stud. Environ. Sci., 60, Elsevier, Amsterdam, pp. 111-136.

116

J. Yan, I. Neretnieks

/The

Science of the Total Environment

EPRI, 1981. Trace-element and phase relations in fly ash. EPRI EA-1822, Electric Power Research Institute, Palo Alto, California. EPRI, 1983. Physical-chemical characteristics of utility solid wastes. EPRI EA-3236, Electric Power Research Institute, Palo Alto, California. Ewing, R.C. and M.J. Jercinovic, 1987. Natural analogues: Their application to the prediction of the long-term behavior of nuclear waste forms. Mater. Res. Sot. Symp. Proc., 84: 67-83. Freude, E., B. Grambow, W. Lutze, H. Rabe and R.C. Ewing, 1985. Long-term release from high waste glass-Part IV: The effect of leaching mechanism. Mater. Res. Sot. Symp. Proc., 44: 99-106. Furrer, G., J. Westall and P. Sollins, 1989. The study of soil chemistry through quasi-steady-state models: I. Mathematical definition of model. Geochim. Cosmochim. Act., 53: 595-601. Furrer, G., P. Sollins and J. Westall, 1990. The study of soil chemistry through quasi-steady-state models: II. Acidity of soil solution. Geochim. Cosmochim. Act., 54: 2363-2374. Grambow, B., 1982. The role of metal ion solubility in leaching of nuclear waste glasses. Mater. Res. Sot. Symp. Proc., 11: 93-102. Grambow, B., 1985. A general rate equation for nuclear waste glass corrosion. Mater. Res. Sot. Symp. Proc., 44: 15-27. Grambow, B., H.P. Hermansson, I.K. Bjiimer and L. Werme, 1985a. Glass/water reaction with and without bentonite present - experiment and model. Mater. Res. Sot. Symp. Proc., 50: 187-194. Grambow, B., M.J. Jercinovic, R.C. Ewing and C.D. Byers, 1985b. Weathered basalt glass: A natural analogue for the effects of reaction progress on nuclear waste glass alteration. Mater. Res. Sot. Symp. Proc., 50: 263-272. Grambow, B., H.P. Hermansson, I.K. Bjiimer, H. Christensen and L. Werme, 1986. Reaction of nuclear waste glass with slowly flowing solutions. Advan. Ceram., 20: 465-474. Grambow, B., 1987. Nuclear waste glass dissolution: mechanism, model and application. JSS-TR-87-02, Swedish Nuclear Fuel and Waste Management Co., Stockholm. October. Grambow, B., H.U. Zwicky, G. Bart, I.K. Bjiirner and L.O. Werme, 1987. Modeling of the effect of iron corrosion products on nuclear waste glass performance. Mater. Res. Sot. Symp. Proc., 84: 471-481. Grambow, B. and D.M. Strachan, 1988. A comparison of the performance of nuclear waste glasses by modeling. Mater. Res. Sot. Symp. Proc., 112: 713-724. Grambow, B., W. Lutze and R. Miiller, 1992. Empirical dissolution rate law for the glass R7T7 contacting halite- and silica-saturated brines. Mater. Res. Sot. Symp.. Proc., 257: 143-150. Gutierrez, B., C. Pazos and J. Coca, 1993. Characterization and Leaching of coal fly ash. Waste Manage. Res., 11: 279-286.

172 (1995) 95-118

C. and J. Schott, 1989. Multisite surface reaction vs. transport control during the hydrolysis of a complex oxide. Chem. Geol., 78: 181-204. Helgeson, H., W.M. Murphy and P. Aagaard, 1984. Thermodynamic and kinetic constraints on reaction rates among minerals and aqueous solutions. II. Rate constants, effective surface area and the hydrolysis of feldspar. Geochim. Cosmochim. Act., 48: 2405-2432. Hemmings, R.T. and E.E. Berry, 1986. Speciation in size and density fractionated fly ash. Mater. Res. Sot. Symp. Proc., Guy,

65: 91-104.

Hemmings, R.T., E.E. Berry, B.J. Cornelius and B.E. Scheetz, 1987. Speciation in size and density fractionated fly ash II. characterization of a low-Calcium, high fron fly ash. Mater. Res. Sot. Symp. Proc., 86: 81-98. Hemmings, R.T. and E.E. Berry, 1988. On the glass in coal fly ashes: recent advances. Mater. Res. Sot. Symp. Proc., 113: 3-38.

Hench, L.L., 1988. Corrosion of silicate glasses: An overview. Mater. Res. Sot. Symp. Proc., 125: 189-200. Hjelmar, O., 1990. Leachate from land disposal of coal fly ash. Waste Manage. Res., 8: 429-449. Hulett, L.D., A.J. Weinberger, K-J. Northcutt and M. Ferguson, 1980. Chemical species in fly ash from coal-burning power plant. Science, 210: 1356-1358. Iseghem, V.P. and B. Grambow, 1988. The long-term corrosion and modeling of two simulated belgian reference high-level waste glasses. Mater. Res. Sot. Symp. Proc., 112: 631-639.

Jantzen, C.M., 19SSa. Prediction of glass durability as a function of environmental conditions. Mater. Res. Sot. Symp. Proc., 125: 143-159. Jantzen, C.M., 19SSb. Pourbaix diagram for the prediction of waste glass durability in geologic environments. Mater. Res. Sot. Symp. Proc., 112: 519-530. Kirby, C.S. and J.D. Rimstidt, 1993. Mineralogy and surface properties of municipal solid waste ash. Environ. Sci. Tech., 27: 652-660.

Knauss, K.G., W.L. Bourcier, K.D. McKeegan, C.I. Merzbacher, S.N. Nguyen, F.J. Ryerson, D.K. Smith, H.C. Weed and L. Newton, 1990. Dissolution kinetics of a simple analogue nuclear waste glass as a function of pH, time and temperature. Mater. Res. Sot. Symp. Proc., 176: 371-381. Kosson, D.S., H. van der Sloot, H. Holmes and C. Wiles, 1991. Leaching properties of untreated and treated residues tested in the USEPA program for evaluation of treatment and utilization technologies for municipal waste combustor residues. Proceedings of WASCON ‘91 Conference (Environmental Implications of Construction with Waste Materials), Maasitricht, The Netherlands, November 10-14, Elsevier Science Publishers B.V. Lasaga, A.C., 1981. Rate laws of chemical reactions. Kinetics of geochemical processes, Reviews in Mineralogy, vol. 8. In: A. Lasaga and R.J. Kirkpatrick (Eds), Mineralogical Society of America.

J. Yan, I. Neretniek

/ The Science of the Total Environment

Lasaga, A., 1984. Chemical kinetics of water-rock interactions. J. Phys. Res., 89: 4009-4025. Lee, C.T. and D.E. Clark, 1986. Effects of solution cations on waste glass leaching. Adv. Ceram., 20: 541-550. I&ken, R.O. and D.M. Strachan, 1984. Long-term leaching of two simulated waste glasses. Adv. Ceram., 8: 39-48. Lutze, W., 1988. Silicate glasses. Radioactive Waste Forms for the Future, Chapter 1, Elsevier Science Publishers B.V., New York. Mattigod S.V., D. Rai, L.E. Eary and C.C. Ainsworth, 1990. Geochemical factors controlling the mobilization of inorganic constituents from fossil fuel combustion residues: I. Review of the major elements. J. Environ. Qual., 19: 188-201. McCarthy, G.J., O.E. Manz, D.M. Johansen, S.J. Steinward and R.J. Stevenson, 1987. Correlations of chemistry and mineralogy of Western U.S. fly ash. Mater. Res. Sot. Symp. Proc., 86: 109-112. McGrail, B.P., A. Kumar and D.E. Day, 1984. Sodium diffusion and leaching of simulated nuclear waste glass. J. Am. Ceram. Sot., 67: 463-467. Michaux, L., E. Mouche and J.-C. Petit, 1992. Geochemical modeling of the long term dissolution behavior of the French nuclear glass R7T7. Appl. Geochem., Supplementary Issue, No. 1: 41-54. Mouche, E. and E. Vernaz, 1988. Aqueous corrosion of the French LWR solution reference glass: first generation model. Mater. Res. Sot. Symp. Proc., 112: 703-712. Murarka, I.P., D. Rai and CC. Ainsworth, 1991. Geochemical basis for predicting leaching of inorganic constituents from coal-combustion residues. In: D. Friedman (Ed), ASTM STP 107.5, Am. Sot. Test. Mater., pp. 279-288. Murphy, WM. and H.C. Helgeson, 1987. Thermodynamic and kinetic constraints on reaction rates among minerals and aqueous solutions. III. activated complexes and the pH-dependence of the rates of feldspar, pyroxene, wolastonite and olivine hydrolysis. Geochim. Cosmochim. Acta., 51: 3117-3153. Nogues, J.L., E.Y. Vernuz and N. Jacquet-Francillon, 1985. Nuclear glass corrosion mechanism applied to the French LWR reference glass. Mater. Res. Sot. Symp. Proc., 44: 89-98. Norton, G.A., R. Markuszewski and H.R. Shanks, 1986. Morphological and chemical characterization of iron-rich fly ash fractions. Environ. Sci. Tech., 20: 409-413. Odler, I. and K.-H. Zysk, 1989. Investigation on the composition of individual fly ash and CFB-ash particles. Mater. Res. Sot. Symp. Proc., 136: 131-137. Parkhurst, D.L., D.C. Thorstenson and L.N. Plummer, 1980. PHREEQE-A computer program for geochemical calculations. Report USGS/WRI 80-96, NTIS Tech. Rep. PB81-167801. Paul, A., 1977. Chemical durability of glasses: a thermodynamic approach. J. Mater. Sci., 12: 2246-2268. Paul, A., 1990. Chemistry of glasses (Second Edition), Chapter 6, Chapman and Hall, New York. Petit, J.-C., J.-C. Dran, L. Trotignon, J.-M. Casabonne, A.

172 (1995) 95-118

111

Paccagnella and G.D. Mea, 1989. Mechanism of heavy element retention in hydrated layers formed on leached silicate glasses. Mater. Res. Sot. Symp. Proc., 127: 33-40. Petit, J.-C., G. Della, J.-C. Dran, M.-C. Magonthier, P.A. Mando and A. Paccagnella, 1990. Hydrated-layer formation during dissolution of complex silicate glasses and minerals. Geochim. Cosmochim Acta., 54: 1941-1955. Plodinec, M.J., C.M. Jantzen and G.G. Wicks, 1984a. Thermodynamic approach to prediction of the stability of proposed radiowaste glasses. Adv. Ceram., 8: 491-495. Plodinec, M.J., CM. Jantzen and G.G. Wicks, 1984b. Stability of radioactive waste glasses assessed from hydration thermodynamics. Mater. Res. Sot. Symp. Proc., 26: 7.55-762. Rai, D. and J.M. Zachara, 1984. Chemical attenuation rates, coefficients, and constants in leachate migration. Vol. 1 A critical review. EPRI EA-3356 V.l, Electric Power Research Institute, Palo Alto, California, Rai, D., C.C. Ainsworth, L.E. Eary, S.V. Mattigod and D.R. Jackson, 1987. Inorganic and organic constituents in fossil fuel combustion residues. Volume 1: A critical review, EPRI EA-51 76, Electric Power Research Institute, Palo Alto, California. Rai, D. and J.M. Zachara, 1988. Chromium reactions in geologic materials. EPRI EA-5741, Electric Power Research Institute, Palo Alto, California. Ramsden, A.R. and M. Shibaoka, 1982. Characterization and analysis of individual fly ash particles from coal-fired power stations by a combination of optical microscopy, electron microscopy and quantitative electron microprobe analysis. Atmos. Environ., 16: 2191-2206. Rimstidt, J.D. and H.L. Barnes, 1980. The kinetics of silicawater reactions. Geochim. Cosmochim. Act., 44: 1683-1699. Roy, D.M., K. Luke and S. Diamond, 1985. Characterization of fly ash and its reactions in concrete. Mater. Res. Sot. Symp. Proc., 43: 3-20. Strachan, D.M., B.O. Barnes and R.P. Turcotte, 1981. Standard leach tests for nuclear waste materials. Sci. Basis Nucl. Waste Manage., 3: 347-354. Theis, T.L. and K.H. Gardner, 1990. Environmental assessment of ash disposal. CRC Critical ReviewsTM. Environ. Control, 20: 21-42. van der Sloot, H.A., E.G. Weyers, D. Hoede and J. Wijkstra, 1985. Physical and chemical characterization of pulverized-coal ash with respect to cement-based applications. EC-178, Netherlands Energy Research Foundation, November. White, W.B., 1986. Dissolution mechanisms of nuclear waste glasses: A critical review. Adv. Ceram., 20: 431-442. White, W.B., 1988. Glass hydration mechanism with application to obsidian hydration dating. Mater. Res. Sot. Symp. proc., 112: 355-366. Yan, J. and I. Neretnieks, 1993a. Glass phases of solid waste and leaching (dissolution) mechanisms. Department of Chemical Engineering, KAT 93/11, Royal Institute of Technology, Sweden.

118

J. Yan, I. Neretnieks

/The

Science of the Total Environment

Yan, J. and I. Neretnieks, 1993b. Morphology, mineralogy and chemical speciation of solid wastes. Department of Chemical Engineering, RAT 93/14, Royal Institute of Technology, Sweden. Yan, J. and I. Neretnieks, 1994a. Aspects of reaction mechanisms and kinetics of waste glass phases with an aqueous solution. Department of Chemical Engineering and Technology RAT 94/9, Royal Institute of Technology, Sweden.

172 (199.5) 95-118

Yan, J. and I. Neretnieks, 1994b. A general conceptual model for dissolution of the solid waste-combustion residues. Department of Chemical Engineering and Technology RAT 94/19, Royal Institute of Technology, Sweden. Zwicky, H.U., B. Grambow, C. Magrabi, E.T. Aeme, R. Bradley, B. Barnes, Th. Graber, M. Mohos and L.O. Werme, 1989. Corrosion behavior of British Magnox waste glass in pure water. Mater. Res. Sot. Symp. Proc., 127: 129-136.