Crystallization

Evaporator Fouling 101 - Sodium Salt Crystallization and Soluble-scale Fouling Christopher L. Verrill, and Wm. James Frederick, Jr. Institute of Paper Science and Technology Georgia Institute of Technology, Atlanta, GA USA ABSTRACT This paper reviews the fundamentals of fouling mechanisms in black liquor evaporators, specifically crystallization fundamentals applied to soluble evaporator scales. Much has been learned recently about crystallization of Na2CO3, Na2SO4 and double salts of these species, and its relationship to fouling in black liquor evaporators and high solids concentrators. This recently acquired information, together with solubilities and mass balances, has been used to evaluate the conditions that lead to fouling in black liquor evaporators, and to guide the development of alternative process configurations and modes of operation to control the rate of deposition of soluble scales. INTRODUCTION The capacity of many black liquor evaporators and high-solids concentrators suffers from fouling of their heat-transfer surfaces. The fouling that occurs in about one third of these evaporators and concentrators results from deposition of Na2CO3 and Na2SO4 salts. The deposits formed are referred to as soluble scale because these sodium salts are moderately soluble in water and can be removed by washing with hot water. In the 1970s, Grace showed that soluble scaling in rising-film (i.e., LTV) evaporators occurred when black liquor was concentrated above the solubility limit of burkeite, a solid solution of approximate composition 2Na2SO4⋅Na2CO3 [1]. By limiting the product solids below the solubility limit of burkeite, soluble-scale fouling was largely eliminated in older, rising-film evaporators. With the introduction of falling-film evaporators and high-solids concentrators, soluble-scale problems appeared again. The fouling characteristics in these units were different from the earlier burkeite scale problems: the onset of scale deposition was erratic, not predictable, and seemed to be associated with high ratios of Na2CO3 to Na2SO4 in the black liquor. Fouling was rapid, and evaporator bodies often had to be shut down and boiled out every few days [2]. Black liquor evaporators and high-solids concentrators become crystallizing evaporators as soon as the total solids concentration in the black liquor exceeds the solubility limit of the Na2CO3 and Na2SO4 that it contains. Soluble-scale deposition in these evaporators and concentrators is a crystallization problem, controlled by crystallization phenomena and the influence of fluid flow and heat transfer on those phenomena. The results of on-going, extensive bench- and pilot-scale studies of the crystallization behavior of Na2CO3 and Na2SO4 during black liquor evaporation have led to a much clearer understanding of fundamental processes involved in the fouling of heat-transfer surfaces by soluble scale, and how this type of fouling can be controlled (subject of this paper). We have been working with several pulp manufacturers to apply the results obtained in these studies to control scale deposition in their industrial black liquor evaporators. An example application is described in a companion paper, “Evaporator Fouling Mitigation - Case Studies.” Advances are also being made in understanding two other forms of evaporator fouling that are described in the companion paper, sodium oxalate deposition (a form of soluble scale) and calcium carbonate scaling. Frederick and Adams have provided a concise review other types of evaporator fouling including aluminosilicate scale, fiber, soap, and lignin deposits [3]. PRINCIPLES OF CRYSTALLIZATION The formation and growth of crystals, whether in suspension or as scale on surfaces, involves the processes illustrated in Figure 1. The principles of crystallization and how they relate to crystallization of Na2CO3 and Na2SO4 from aqueous solutions and black liquor have been presented elsewhere [4-8]. Here we summarize the key points. When water is evaporated from black liquor, the concentrations of the dissolved Na2CO3 and Na2SO4 eventually reach and exceed their solubility, a thermodynamic limit to the amount of solute that can be dissolved at equilibrium.

The amount by which their concentrations exceed their solubility is referred to as the level of supersaturation achieved. Supersaturation is necessary for nucleation and growth of crystals. A supersaturated solution is not thermodynamically stable, but supersaturation can be maintained up to a certain limit, referred to as the metastable limit. When a solution is concentrated to the metastable limit, primary nucleation of crystals occurs spontaneously. The metastable limit is not a thermodynamic value, and can depend, for example, upon the rate of solvent removal, or the presence of dissolved species or suspended particles other than the species that crystallize.

Supersat’nCrystal

SizeDistribution

Growth

Nucleation

SurfaceArea

Feed ProductCrystals

Supersat’nCrystal

SizeDistribution

Growth

Nucleation

SurfaceArea

Feed ProductCrystals

Figure 1. Crystal formation and growth processes [9]. Figure 2 illustrates the solubility and metastable limits for solutions of Na2CO3 and Na2SO4 in water. The region between the solubility and metastable limits is known as the metastable zone. Solutions whose concentrations fall within the metastable zone will support growth of crystals or crystalline deposits that already exist, but will not produce new crystals by primary nucleation within the time frame of the observation. For many solutes, the metastable zone is narrow, of the order of 1% of their solubility. The data in Figure 2 show that the metastable zone can exceed 10% of the solubility for Na2CO3 and Na2SO4 in water.

0.20

0.32

0.28

0.24

0.36

0.20

0.32

0.28

0.24

0.36

50 80 110 140 170 20050 80 110 140 170 200

Temperature, oC

Wei

ght f

ract

ion

Na2C

O3+

Na2S

O4

in s

olut

ion

Solubilitylimit

STABLEREGION

METASTABLEREGION

UNSTABLEREGION

Metastablelimit

Figure 2. The solubility and metastable limits versus temperature for the system Na2CO3-Na2SO4-water. The mole ratio of Na2CO3 to Na2SO4 is 1:2 for this data. Adapted from Shi and Rousseau [4]. New crystals are formed by either primary or secondary nucleation. Primary nucleation occurs when the metastable limit is exceeded. When it occurs, large numbers of small crystals form very rapidly and the dissolved solute concentration typically drops below the metastable limit. Secondary nucleation results from crystal collisions; it produces crystals more slowly and steadily. Once crystals or crystalline deposits are formed, they will continue to grow as long as the solution remains supersaturated. They will not grow in saturated solutions, and will redissolve, albeit slowly, in solutions whose concentrations are below their solubility limit.

CRYSTAL PHASES Crystallization from Aqueous Solutions We now know that at least five unique crystal species can be produced when black liquor is concentrated above the solubility of Na2SO4, and Na2CO3 at 90 to 150oC. These crystal species are burkeite, dicarbonate, Na2SO4, Na2CO3, and Na2CO3·H2O. Which species is crystallized and its exact composition is directly related to the concentrations of carbonate and sulfate in the solution from which the crystals are produced (Figure 3).

0.2

0.4

0.6

0.8

1

0.2

0.4

0.6

0.8

1

0.2 0.4 0.6 0.8 10.2 0.4 0.6 0.8 1

Na2SO4Regions: Na2SO4Regions:N

a 2C

O3/(

Na 2

SO4

+ N

a 2C

O3)

mol

e ra

tios

in s

olid

Na2CO3/(Na2SO4 + Na2CO3) mole ratios in solution

Burkeite Na2CO3

Dicar-bonate

0.2

0.4

0.6

0.8

1

0.2

0.4

0.6

0.8

1

0.2 0.4 0.6 0.8 10.2 0.4 0.6 0.8 1

Na2SO4Regions: Na2SO4Regions:N

a 2C

O3/(

Na 2

SO4

+ N

a 2C

O3)

mol

e ra

tios

in s

olid


Burkeite Na2CO3

Dicar-bonate

Figure 3. Compositions of the crystals formed (excluding residual or structural water) in the sulfate, burkeite, dicarbonate, and carbonate regions as a function of solution composition. Solid circles are experimental data at 115 °C. Adapted from Shi et al. [7]. To generate the data in Figure 3, the composition of the first crystals produced was measured during evaporative crystallization of aqueous solutions of sodium carbonate and sodium sulfate initially having molar ratios ranging from 1:5 to 12:1. According to the first solid species crystallized, the solutions were classified into regions. In addition to the expected phases, a new, previously unidentified crystalline phase was predominant for solutions in the composition range 0.833 < xNa2CO3

< 0.889, where xNa2CO3 is the mole ratio of Na2CO3 to Na2CO3 + Na2SO4 in

solution. This new species, termed dicarbonate, is distinguished from burkeite by substantially higher carbonate content and a different x-ray diffraction (XRD) pattern. During black liquor evaporation, only burkeite and dicarbonate are important with respect to fouling [8]. Neither of these crystal species has a fixed chemical composition. The composition of burkeite can vary from 1:3.5 to nearly 1:1 moles of Na2CO3 per mole Na2SO4. The composition of dicarbonate can vary from 1.5:1 to nearly 3:1 moles of Na2CO3 per mole Na2SO4. Immediately after they are formed, the crystals of the two species behave differently. Burkeite crystals are initially very small, but they agglomerate rapidly and grow to larger crystals (Figure 4a). Dicarbonate crystals are also initially very small, but higher levels of supersaturation achieved with dicarbonate results in far more crystals upon primary nucleation. These crystals (Figure 4b) do not grow as rapidly, and agglomerate more slowly. The much larger number of crystals and their slower growth and agglomeration make dicarbonate agglomeration on heat-transfer surfaces more likely.

Figure 4. a) Burkeite crystals produced during evaporative crystallization of aqueous solutions containing sodium carbonate and sodium sulfate in a molar ratio of 1:2; b) Crystals of dicarbonate from solutions with a carbonate-sulfate molar ratio of 5:1. From Shi [5]. Crystallization from Black Liquor Similar crystallization experiments were carried out with black liquor at 130°C. The results for black liquor are shown in Figure 5, along with the aqueous solution crystallization data from Figure 3. There is an additional composition region, 0.68 < xNa2CO3

< 0.82, in which both burkeite and dicarbonate were identified in the crystal phase by XRD analysis. The other composition regions are the same as in Figure 3.

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 1

Na2SO4Regions: NewBurkeite+ New

Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id

2CO3/(Na2SO4 + Na2CO3) mole ratios in solution

Burkeite Na2CO3

0

0.2

0.4

0.6

0.8

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 10.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite Na2CO3

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite Na2CO3

0

0.2

0.4

0.6

0.8

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 10.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite Na2CO3

Burkeite +Dicar-bonate

Dicar-bon-ate

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite Na2CO3

0

0.2

0.4

0.6

0.8

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 10.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite Na2CO3

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite Na2CO3

0

0.2

0.4

0.6

0.8

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 10.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite Na2CO3


Dicar-bon-ate

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite Na2CO3Na2CO3

0

0.2

0.4

0.6

0.8

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 10.2 0.4 0.6 0.8 1

Na2SO4Regions: New

0

0.2

0.4

0.6

0.8

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 10.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite Na2CO3Burkeite

+ New

Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite Na2CO3

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 1

Na2SO4Regions:

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3


Burkeite Na2CO3NewBurkeite+ New

Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3


Burkeite Na2CO3

0

0.2

0.4

0.6

0.8

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 10.2 0.4 0.6 0.8 10

0.2

0.4

0.6

0.8

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 10.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3


Burkeite Na2CO3


Dicar-bon-ateNa2SO4Regions: NewBurkeite

+ New

Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3


Burkeite Na2CO3


Dicar-bon-ate

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 1

Na2SO4Regions:

Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3


Burkeite

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 1

Na2SO4Regions:

Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3



0

Crystallization @ 130 C (Liquor A)Crystallization @ 130 C (Liquor B)Crystallization @ 115 C (Aqueous Solution)

Na2SO4Regions:

Na

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite Na2CO3

0

0.2

0.4

0.6

0.8

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 10.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite Na2CO3

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite Na2CO3

0

0.2

0.4

0.6

0.8

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 10.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite Na2CO3


Dicar-bon-ate

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite Na2CO3

0

0.2

0.4

0.6

0.8

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 10.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite Na2CO3

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite Na2CO3

0

0.2

0.4

0.6

0.8

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 10.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite Na2CO3


Dicar-bon-ate

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id



0

0.2

0.4

0.6

0.8

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 10.2 0.4 0.6 0.8 1

Na2SO4Regions: New

0

0.2

0.4

0.6

0.8

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 10.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite Na2CO3Burkeite

+ New

Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite Na2CO3

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 1

Na2SO4Regions:

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3


Burkeite Na2CO3NewBurkeite+ New

Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3


Burkeite Na2CO3

0

0.2

0.4

0.6

0.8

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 10.2 0.4 0.6 0.8 10

0.2

0.4

0.6

0.8

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 10.2 0.4 0.6 0.8 1


Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3


Burkeite Na2CO3


Dicar-bon-ateNa2SO4Regions: NewBurkeite

+ New

Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3


Burkeite Na2CO3


Dicar-bon-ate

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 1

Na2SO4Regions:

Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3


Burkeite

0

0.2

0.4

0.6

0.8

0.2 0.4 0.6 0.8 1

Na2SO4Regions:

Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3



0

Crystallization @ 130 C (Liquor A)Crystallization @ 130 C (Liquor B)Crystallization @ 115 C (Aqueous Solution)

Na2SO4Regions:

Na Figure 5. Comparison of the crystals compositions obtained from black liquor crystallization at 130°C and aqueous solution crystallization at 115°C. Adapted from Shi [5]. The regions where dicarbonate or both burkeite and dicarbonate are found in the crystalline phase is of importance for industrial black liquor evaporators. Data collected during a 1998 survey of black liquor evaporator fouling [2] indicate that for about 60% of these evaporators, the weak black liquor composition falls within the range 0.68 <

xNa2CO3 < 0.89. As discussed below, the composition of the dissolved sodium salts in the remaining liquors from that survey would move into the dicarbonate composition range when concentrated to about 70% dry solids content. NUCLEATION AND CRYSTAL GROWTH During evaporation of water from sodium carbonate and sodium sulfate solutions, more than one solid phase could be crystallized as the solution composition changes during crystallization. Solutions whose compositions are initially in the burkeite crystallization region shift toward, and ultimately into, the dicarbonate crystallization region as crystallization proceeds. This occurs because the mole ratio of Na2SO4/Na2CO3 in burkeite crystals always exceeds 1, so that more Na2SO4 than Na2CO3 is depleted from solution when burkeite crystallizes. The data in Figure 6 are for an aqueous solution of Na2CO3 and Na2SO4 whose composition was initially in the burkeite crystallization region (Na2CO3/Na2SO4 mole ratio of 3:1). During evaporation of water, two primary nucleation events were observed; nucleation of burkeite beginning at a total solids content just above 35%, followed by nucleation of dicarbonate at a total solids content of 42%. In both cases, a large number of fine crystals were created abruptly during nucleation, but the number of fine crystals was about five times greater when dicarbonate crystallized. Solutions whose compositions are initially in the dicarbonate crystallization region produce crystals that are enriched in sulfate compared with the solution. As dicarbonate crystals are produced, the composition of the solution shifts toward the Na2CO3 crystallization region. Once the solution composition has reached the boundary of the Na2CO3 crystallization region, however, both dicarbonate and Na2CO3 crystallize to maintain a non-changing solution composition [5].

2Na2CO3•Na2SO42Na2CO3•Na2SO4

Num

ber o

f par

ticle

s co

unte

d (s

-1)

Num

ber o

f par

ticle

s co

unte

d (s

-1)

Nucleation ofburkeite

Nucleation ofdicarbonate

Total Solids Content, %

2Na2CO3•Na2SO42Na2CO3•Na2SO42Na2CO3•Na2SO42Na2CO3•Na2SO4

Num

ber o

f par

ticle

s co

unte

d (s

-1)

Num

ber o

f par

ticle

s co

unte

d (s

-1)

Nucleation ofburkeite

Nucleation ofdicarbonate

Total Solids Content, % Figure 6. Number of 1-10 μm crystals in suspension during evaporation of an aqueous solution containing sodium carbonate and sodium sulfate in a molar ratio of 3:1. Two nucleation events are seen, namely, the nucleation of burkeite and of dicarbonate. Adapted from Shi et al. [7]. The amount of fine crystals produced during a primary nucleation event depended upon the Na2CO3 to Na2SO4 ratio in the system (Figure 7). The data in Figure 7 suggests that, for solutions initially in the burkeite crystallization region, the number of small crystals produced when burkeite nucleates increases as the Na2CO3 to Na2SO4 ratio increases. In Figure 7, the initial total salt concentrations were the same in each experiment, and the evaporation rates were approximately the same. The longer time to achieve primary nucleation with increasing Na2CO3 to Na2SO4 ratio therefore indicates that the degree of supersaturation required for nucleation increased with increasing Na2CO3 to Na2SO4 ratio. A second major nucleation event occurs at about 330 minutes into the evaporation run in the solutions with a Na2CO3 to Na2SO4 ratio of 4:1 and 5:1. This second event corresponds to nucleation of dicarbonate. The amount of fine crystals produced during these nucleation events is about the same as was produced during primary nucleation of burkeite. This work clearly showed that dicarbonate crystals will not grow on existing burkeite crystals, but will nucleate and grow as a separate solid phase.

0

5000

10000

15000

0 100 200 300 400

Time, min

coun

ted,

s-1

1:2 3:1

4:15:1

Na2CO3:Na2SO4

Size Range 1-10 mmNum

ber o

f par

ticle

s -

0

5000

10000

15000

0 100 200 300 400

Time, min

coun

ted,

s-1

1:2 3:1

4:15:1

Na2CO3:Na2SO4

Size Range 1-10 mmNum

ber o

f par

ticle

s -

Figure 7. Number of 1-10 μm crystals in suspension during evaporation of solutions of sodium carbonate and sodium sulfate at different molar ratios [8]. EFFECTS OF CALCIUM ON THE SOLID SPECIES Shi et al. [6] discovered in this work that a small amount of calcium ion acts as a nucleation inhibitor for both burkeite and dicarbonate crystals. When nucleation was inhibited, higher levels of supersaturation developed before primary nucleation occurred, and many more crystals were produced upon nucleation. Conversely, when the calcium ions were removed by sequestration with EDTA, primary nucleation occurred at lower total concentrations of Na2CO3 and Na2SO4 in the burkeite, dicarbonate, and Na2CO3 crystallization regions, and fewer crystals were produced during nucleation. The inhibition of burkeite and dicarbonate nucleation and crystal growth by calcium ions was verified in a set of pilot evaporator experiments [10, 11]. Figure 8 illustrates the impact of calcium ion on nucleation of burkeite and dicarbonate during evaporation of water from black liquor. When no calcium was added, primary nucleation of burkeite occurred at about 55% total solids content, but without producing many small crystals. A second nucleation event, during which dicarbonate began to crystallize occurred at about 68% total solids content, and produced a large number of small crystals. When 100 ppm CaCO3 was added to the liquor, relatively few small crystals were produced when the black liquor was concentrated to the same total solids content.

2000

1000

050 55 60 65 7570

Total solids content, wt-%

Num

ber o

f par

ticle

s co

unte

d, s

-1

100 ppm CaCO3

no CaCO3

2000

1000

0

2000

1000

050 55 60 65 757050 55 60 65 7570

Total solids content, wt-%

Num

ber o

f par

ticle

s co

unte

d, s

-1

100 ppm CaCO3

no CaCO3

Figure 8. Primary nucleation indicated by rise in number of 1-10 μm crystals in suspension during evaporation of black liquor with (a) no CaCO3 added and (b) 100 ppm CaCO3 added, based on dry black liquor solids. Adapted from Shi et al. [12].

SCALING CHARACTERISTICS The previous work of our research group has identified two mechanisms by which soluble-scale deposits form during evaporation of black liquor. They are (a) agglomeration of fine crystals on heat-transfer surfaces following a primary nucleation event, and (b) crystal growth from a supersaturated solution below the metastable limit. These two scale-deposition processes have different characteristics. However, both can contribute to scaling within the same evaporator effect. One possible scenario is agglomeration of fine particles on a heat-transfer surface serving as a scale initiation mechanism and scale growth continuing via supersaturation-driven growth of crystals on the scale surface. Primary nucleation of dicarbonate crystals from black liquor causes rapid fouling of heat-transfer surfaces. The metastable limit for sodium salts in aqueous solutions and in black liquor can be more than 10% higher than the solubility limit. This high level of developed supersaturation means that a large amount of potential scale is produced during a nucleation event – equivalent to about 10 t/day for a mill that produces 1000 adt/day of pulp. The fine crystals produced during nucleation agglomerate rapidly. In semi-batch pilot evaporator trials, where black liquor was concentrated to above the point where primary nucleation of dicarbonate crystals occurred, the heat-transfer surface always began to foul rapidly just before the onset of primary nucleation. Primary nucleation of burkeite crystals from black liquor did not cause rapid fouling of heat-transfer surfaces in pilot evaporator experiments [11]. For a range of liquor compositions tested in the pilot evaporator, the results in Figure 9 show that burkeite nucleation occurs at 45-55% total solids and dicarbonate nucleation is at 60-70% total solids. The onset of surface scale, determined from analysis of heat transfer data, corresponds closely with the nucleation of dicarbonate.

Figure 9. Pilot falling-film evaporator results showing soluble scale fouling begins when dicarbonate starts to crystallize during evaporation of black liquor. Adapted from Euhus et al. [11]. An important consideration for diagnosing soluble scale fouling problems is the inverse solubility of these sodium salts. At typical operating temperatures, solubility decreases as temperature is increased. Thus, in a falling film evaporator the level of supersaturation of dicarbonate will be higher in the film of liquor contacting the heated surface than in the sump [10]. This explains why the onset of fouling occurs at slightly lower total solids than dicarbonate nucleation for all of the cases shown in Figure 9. Once crystals or crystal nuclei are attached to a heat-transfer surface, scale will grow as long as there is some level of supersaturation in the liquor. The rate of scale growth depends on the degree of supersaturation, the total surface area of suspended crystals, and the crystallization kinetics of the particular crystal being grown. An important

Black liquor and pilot evaporator run number

Dry

sol

ids

cont

ent,

wt-%

40

45

50

55

60

65

70

75

A-2 A-3 A-4 A-5 A-6 A-7B A-8 A-9 A-9B A-10 A-11 B-1 B-2 B-4 B-5A-2 A-3 A-4 A-5 A-6 A-7B A-8 A-9 A-9B A-10 A-11 B-1 B-2 B-4 B-5

Burkeite nucleationDicarbonate nucleationOnset of fouling


unknown is the mechanism(s) of initiation of scale on a clean heat-transfer surface. Agglomeration of the fine crystals produced during a nucleation event is likely one mechanism. The importance of surface nucleation per se has not yet been established. However, initial experiments indicate that burkeite crystals can be induced to nucleate on a heat-transfer, while dicarbonate crystals form in the bulk and subsequently adhere to the surface. These early observations need further analysis [10]. APPLICATION TO SCALE CONTROL From a practical perspective, three questions must be answered when designing or optimizing black liquor concentrators to minimize Na2CO3-Na2SO4 scales:

• When will fouling occur? • Where will fouling occur? • How can fouling be avoided or at least minimized?

To answer these questions with respect to dicarbonate fouling, we need to know (a) in which evaporator effect and body dicarbonate will begin to crystallize, and (b) whether it will deposit on heat-transfer surfaces in that effect and body as it crystallizes. Falling film evaporator effects operate best at a steady product solids level, either well above or just below a point of salt crystallization. Fluctuating solids levels that cause an effect to “swing” through crystallization points can dramatically increase fouling and plugging rates by creating periodic massive nucleation events. Whether dicarbonate fouls heat transfer surfaces when it crystallizes depends upon whether dicarbonate grows on suspended crystals, or whether supersaturation and primary nucleation occur. A situation expected to suffer from dicarbonate scaling is illustrated in Figure 10. The initially horizontal but later decreasing curves show the percentages or amounts of dissolved Na2CO3 and Na2SO4 in the system at any point during the evaporation process. The vertical, dashed lines mark the product solids concentration of black liquor exiting the various bodies of a hypothetical falling-film evaporator and forced circulation concentrator. In this illustration, the transition from burkeite to dicarbonate crystallization occurs in the 1B falling-film body. The crystal population in body 1B contains mainly burkeite and very little dicarbonate. Dicarbonate will not crystallize on burkeite crystals, and there are insufficient dicarbonate crystals to relieve supersaturation. Body 1B will foul when supersaturation reaches the metastable limit (crystallization point) and a massive nucleation event occurs. Our understanding of crystallization during black liquor evaporation suggests that operating an effect well above a crystallization point can help reduce episodes of surface fouling and tube plugging by maintaining a stable population of suspended salt crystals [8]. In this case, supersaturation is kept low by slow growth of the suspended crystals and the chance of nucleation of a large number of new fine particles is minimized. It is the fine particles that can more easily “stick” to the tube surface and foul the evaporator. This improved mode of operation is illustrated in Figure 11. Since the product solids content from the 1B body is well above the transition point from Burkeite to dicarbonate crystallization, there will be a substantial population of dicarbonate crystals in the liquor. Dicarbonate will be able to crystallize on the suspended crystals. The high degree of supersaturation needed for primary nucleation and subsequent fouling will never be reached. To make the elevated solids content strategy work, it is essential to maintain suspended crystals in the falling-film effects at all times. A means to start up on recirculated product liquor after on-line washes and off-line boil outs will be needed. This provides a population of “seed” crystals to prevent high supersaturation from developing when the evaporator is recovering from a wash or boil out. It may also help to stabilize crystallization of Burkeite and dicarbonate by 1) maintaining the highest practical liquor volume in the falling-film effect sump and 2) increasing the recirculation rate [3]. Another possible response to excessive soluble scale fouling rates in falling-film evaporators is to decrease the product solids below the solubility limit for the sodium salts, e.g., allowing final product solids to drop so that body 1A is below 50% and 1B is below 60% in Figure 10. This action may prevent formation of soluble scale, but can create problems downstream. When forced-circulation concentrators are “pushed” to make up for lost evaporation capacity in falling film effects, excessive vapor velocities can entrain liquor causing pluggage of the mist eliminators and contamination of the condensate in upstream effects. Allowing the final product solids to drop may also have adverse impacts on recovery boiler emissions and ash chemistry.

3Dry solids content of black liquor, wt-%

100%

50% 60% 70% 80%40%

% of Na2CO3and Na2SO4that remain in solution

BurkeiteNo Crystals Burkeite Dicarbonate

Na2CO3

Na2SO4

45% 55% 65% 75%

1A FF 1B FF Conc.2nd FF

Figure 10. Illustration of hypothetical evaporator where dicarbonate fouling is likely to occur in body 1B.

4Dry solids content of black liquor, wt-%

100%

50% 60% 70% 80%40%


BurkeiteNo Crystals Burkeite Dicarbonate

Na2CO3

Na2SO4

45% 55% 65% 75%


Figure 11. Illustration of hypothetical evaporator system where dicarbonate fouling should not occur; falling-film body 1A is operating with stable populations of Burkeite and body 1B has a stable population of dicarbonate.

There is no proven model for predicting the point of dicarbonate crystallization in black liquors. To provide a useful tool, an empirical model was developed from existing data to predict the crystallization of burkeite and dicarbonate as a function of the relative concentrations of Na2CO3 and Na2SO4 in solution [13]. The apparent equilibrium relationships obtained were solved with the constraints of Shi’s experimentally-determined relationship between solution and crystal composition [5] and the conservation of mass and charge. The resulting model predicts the changing concentrations of the sodium salts in solution and the changing composition of the crystals formed as evaporation proceeds. CONTINUING FUNDAMENTAL STUDIES On-going laboratory studies of soluble scaling is focused on providing a sound fundamental basis for testing compounds that could advantageously modify nucleation and growth of scale. Species present in black liquor (Ca, soap) are of particular interest because these may explain why certain liquors and/or mill conditions can cause sudden, dramatic capacity loss in high solids evaporator equipment. We are currently making a series of solubility measurements over the temperature range of 100-145ºC using a batch crystallization apparatus. This data is needed for developing thermodynamic property correlations for the variable-composition double salts, burkeite and dicarbonate. These correlations will be implemented into first-principals solubility models. Studies of the transient effects in industrial equipment are also underway. Pursuit of these more challenging aspects of evaporator fouling has required development of new experimental methods to monitor changing black liquor composition during evaporative crystallization and to determine the composition of the potentially unstable solids that form as suspended crystals or heat-transfer surface scale. These tools include headspace-GC techniques for rapid determination of black liquor solubility, means to quickly extract and isolate crystal samples from evaporating black liquor, and quantitative solid phase characterization by XRD techniques. SUMMARY The causes of dicarbonate fouling in black liquor evaporators have become understood during the past seven years. As a result, we now understand the relationship between black liquor composition, crystallization of burkeite and dicarbonate, evaporator configuration, and fouling. This information is beginning to be used to solve dicarbonate fouling problems in black liquor evaporators around the world. ACKNOWLEDGMENTS The contents of this paper are a result of the financial support and guidance provided by the U.S. Department of Energy, the National Science Foundation, State of Georgia Traditional Industries Program, Andritz Corporation, Mead Corporation, Potlatch Corporation, Weyerhaeuser Company, International Paper Company, Lasentec Corporation, and the member companies of the Institute of Paper Science and Technology. REFERENCES 1. Grace, T. M., “Survey of Evaporator Scaling in the Alkaline Pulp Industry,” Project 3234, Report 1, The Institute of Paper Chemistry, Appleton, WI (September 22, 1975). 2. Schmidl, W., and Frederick, W. J., “Current Trends in Evaporator Fouling,” Proc. 1998 International Chemical Recovery Conference, Tappi Press, Atlanta, pp. 367-377 (1998). 3. Frederick, W. J. Jr., and Adams, T. N., “Black Liquor Evaporator Fouling,” 2004 Kraft Recovery Short Course Notes, Tappi Press, Atlanta, section 3.3, pp. 1-9 (2004). 4. Shi, B., and Rousseau, R. W., “Crystal Properties and Nucleation Kinetics from Aqueous Solutions of Na2CO3 and Na2SO4,” Industrial and Engineering Chemistry Research 40(6):1541-1547 (2001).

5. Shi, B., “Crystallization of Solutes That Lead to Scale Formation in Black Liquor Evaporation,” Ph.D. Thesis, Georgia Institute of Technology (April, 2002). 6. Shi, B., Frederick, W. J. Jr., and Rousseau, R. W., “Effects of Calcium and Other Ionic Impurities on the Primary Nucleation of Burkeite,” Industrial and Engineering Chemistry Research 42(12):2861-2869 (2003). 7. Shi, B., Frederick, W. J. Jr., and Rousseau, R. W., “Nucleation, Growth, and Composition of Crystals Obtained from Solutions of Na2CO3 and Na2SO4,” Industrial and Engineering Chemistry Research 42(25):6343-6347 (2003). 8. Frederick, W. J. Jr., Shi, B., Euhus, D. D., and Rousseau, R. W., “Crystallization and Control of Sodium Salt Scales in Black Liquor Concentrators,” Tappi J. 3(6):7-13 (2004). 9. Mullin, J. W., Crystallization, 4th ed., Butterworth-Heinemann, Oxford (2001). 10. Euhus, D. D., “Nucleation in Bulk Solutions and Crystal Growth on Heat-Transfer Surfaces during Evaporative Crystallization of Salts Composed of Na2CO3 and Na2SO4,” Ph.D. Thesis, Georgia Institute of Technology (September 2003). 11. Euhus, D. D., Rousseau, R. W., Frederick, W. J. Jr., Schmidl, W., Lien, S. J., Thorn, P. A., and Smith, P. K., “Eliminating Sodium Salt Fouling in Black Liquor Concentrators: Crystallization Behavior and Fouling in Pilot Evaporation Trials,” Paper 52-3 at the TAPPI Fall Technical Conference, Chicago, IL (October 2003). 12. Shi, B., Rousseau, R. W., and Frederick, W. J. Jr., “Nucleation of Burkeite from Aqueous Solutions and Black Liquor,” Proc. 2001 International Chemical Recovery Conference, PAPTAC, Montreal, pp. 177-180 (2001). 13. Soemardji, A. P., Verrill, C. L., Frederick, W. J. Jr., and Theliander, H., “Prediction of Crystal Species Transition in Aqueous Solutions of Na2CO3 and Na2SO4 and Kraft Black Liquor,” Tappi J. 3(11):27-32 (2004).

1

Georgia Institute of Technology • IPST at Georgia Tech500 10th Street NW • Atlanta, GA 30332-0620

404-894-1838 • www. ipst.gatech.edu

Christopher L. VerrillW. J. Frederick, Jr.

Evaporator Fouling 101Sodium Salt Crystallization and Soluble-Scale Fouling

Soluble Scale

• Deposition of soluble scale is a serious problem in black liquor evaporators– Weak BL contains Na2CO3 and

Na2SO4

– Na-CO3-SO4 salts crystallize from BL at about 50% dry solids content

– Often these salts deposit as scale

Example of Severe Fouling

#1 effect of falling-film unit (56-62% solids)

90% tubes plugged after 6 months operation

2

Objectives of Tutorial• Review the basic concepts of

crystallization and soluble-scale deposition

• Summarize the results of crystallization studies and fouling by Na2CO3 and Na2SO4 salts from black liquor

• Examine implications to control soluble-scale fouling

• Case studies (paper 34-2)

Background

• In 1970’s problems with burkeite fouling in LTVs operating above solubility limit– Improvements by liquor recirculation – Solved by limiting maximum product solids

• Soluble scale problems reappeared in late 80’s in falling-film units– Episodes of sudden, dramatic capacity loss– Aided by avoiding high CO3/SO4 ratios

Key Factors in Sodium Salt Scaling

• Solubility of Na2SO4 and Na2CO3

• Crystallization of Na2SO4 and Na2CO3

• Attachment of scale to surface

3

Formation of Sodium Salt Crystals and ScaleDuring Evaporation

45% dry solidscontent



Scale

Black liquor at

Crystal Formation and Growth

Growth

Nucleation

SurfaceArea

BlackBlackLiquorLiquor Supersat’nSuper-

saturation

Growth

Nucleation

SurfaceArea

ProductProductCrystalsCrystalsProductProductCrystalsCrystals

CrystalSize

Distribution

CrystalSize

Distribution

What happens here • depends upon evaporation conditions• Impacts scale formation

Solubility and Metastable Limits vs. Temperature for Na2CO3-Na2SO4-Water

Na2CO3 to Na2SO4mole ratio = 1:2

0.200.20

0.320.32

0.280.28

0.240.24

0.360.36

0.200.20

0.320.32

0.280.28

0.240.24

0.360.36

5050 8080 110110 140140 170170 2002005050 8080 110110 140140 170170 200200

Temperature, Temperature, ooCC

Weight fractionWeight fractionNaNa22COCO33 + Na+ Na22SOSO44in solutionin solution

Solubilitylimit

STABLESTABLEREGIONREGION

METASTABLEMETASTABLEREGIONREGION

UNSTABLEUNSTABLEREGIONREGION

Metastablelimit

4

Solubility and Metastable Limits vs. Temperature for Na2CO3-Na2SO4-Water

Na2CO3 to Na2SO4mole ratio = 1:2

0.200.20

0.320.32

0.280.28

0.240.24

0.360.36

0.200.20

0.320.32

0.280.28

0.240.24

0.360.36

5050 8080 110110 140140 170170 2002005050 8080 110110 140140 170170 200200

Temperature, Temperature, ooCC

Weight fractionWeight fractionNaNa22COCO33 + Na+ Na22SOSO44in solutionin solution

Solubilitylimit

Metastablelimit

Crystal Formation during Evaporationof Water from Aqueous Na2SO4

Number ofNumber ofparticlesparticlescounted (scounted (s--11))

00 3030 6060 909000 3030 6060 9090

PrimaryPrimary

SecondarySecondary

00 3030 6060 9090

PrimaryPrimary

SecondarySecondary00

200200

400400

600600

800800

10001000

00 3030 6060 9090

Time, minTime, min

PrimaryPrimary

SecondarySecondary

11--10 um10 um

1111--20 um20 um

2121--50 um50 um

5151--500 um500 um

1.1. Nucleation producesNucleation producesvery small crystalsvery small crystals

2.2. Small crystalsSmall crystalsagglomerateagglomerateto larger onesto larger ones

Green & Frattali, 100oC

Schroeder, 150oC

This study, 115oC

Crystals vs. Solution Composition for Na2CO3-Na2SO4-H2O

0.2

0.4

0.6

0.8

1

0.2

0.4

0.6

0.8

1

0.2 0.4 0.6 0.8 10.2 0.4 0.6 0.8 1

Na2SO4Regions: Na2SO4Regions:

Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite Na2CO3

Dicar-bonate

0.2

0.4

0.6

0.8

1

0.2

0.4

0.6

0.8

1

0.2 0.4 0.6 0.8 10.2 0.4 0.6 0.8 1

Na2SO4Regions: Na2SO4Regions:

Na 2

CO

3/(N

a 2SO

4+

Na 2

CO

3) m

ole

ratio

s in

sol

id


Burkeite Na2CO3

Dicar-bonate

Burkeite

Na2SO4

Dicar-bonate

Na2CO3

Frederick, Shi, Euhus, Rousseau, Tappi J. 3(6):7-13 (2004)

5

Dicarbonate is a Distinct Phase

Na2CO3⋅H2O

Dicarbonate

115°C

125°C

135°C

105°C

Dicarbonate Nucleation Produces More Small Particles than Burkeite

1-10 μm particles

00

50005000

1000010000

1500015000

00 1010 200200 300300 400400

Time, minTime, min

1:23:14:15:1

Size Range 1-10 mm

Na2CO3:Na2SO4Na2CO3:Na2SO4

Size Range 1-10 mm

Number ofNumber ofparticles particles counted, scounted, s--11

in solutionin solutionNaNa22COCO33

NaNa22COCO33 + Na+ Na22SOSO44

in crystalsin crystals

NaNa22COCO33

NaNa22COCO33 + Na+ Na22SOSO44

Na2SO4

Burkeite

00 0.20.2 0.40.4 0.60.6 0.80.8 1.01.000

0.20.2

0.40.4

0.60.6

0.80.8

1.01.0

Burkeite +Dicarbonate

Dicarbonate

Crystals vs. Solution Composition for Na2CO3 and Na2SO4 in Black Liquor

Na2CO3

6

Primary nucleation of Na salts from black liquor(a) no CaCO3 added

(b) 100 ppm CaCO3added

a.a.

b.

Num

ber o

f par

ticle

s co

unte

d (s

-1)

Solids content (wt-%)

a.a.

b.

Num

ber o

f par

ticle

s co

unte

d (s

-1)


Ca2+ Impacts Nucleation

Nucleation Events During Evaporation3:1 molar Na2CO3-Na2SO4 in water


Num

ber o

f par

ticle

s co

unte

d (s

-1)


Num

ber o

f par

ticle

s co

unte

d (s

-1)

1-10 um

11-20 um

21-50 um

51-500 um


NucleationNucleationof Burkeiteof Burkeite

Nucleation of Nucleation of dicarbonatedicarbonate

0

1

2

3

4

5

6

7

35 45 55 65 75

Dry solids content, wt-%

Wt-%

cry

stal

s in

BLS

Total Na2SO4 + Na2CO3= 8.2 wt-% of BLS

Burkeite

Dicarbonate

Formation of Sodium Salt Crystals and Scale During Evaporation

7

Crystallization in Steady-Stateversus Non-Steady-State Conditions• At steady-state conditions:

– Constant crystal concentration, composition, and particle size

– Crystal growth relieves supersaturation– Secondary nucleation produces new crystals slowly

• Under non-steady-state conditions:– Crystal concentration, composition, and particle

size change with time– High supersaturation may develop– Primary nucleation may produce new nuclei rapidly

over a short time interval

Sodium Salt Fouling Begins Before Bulk Dicarbonate Crystallization

Black liquor and pilot evaporator run number

Dry

sol

ids

cont

ent,

wt-%

40

45

50

55

60

65

70

75

A-2 A-3 A-4 A-5 A-6 A-7B A-8 A-9 A-9B A-10 A-11 B-1 B-2 B-4 B-5


Dicarbonate Crystals

500 μm

Burkeite Crystals

100 μm

Why? Dicarbonate Crystals are Smallerand Form Agglomerates

8

Key Findings fromPilot Evaporator Studies

• Fouling during BL evaporation depends on– Carbonate-to-sulfate ratio in liquid– Evaporation rate– Degree of supersaturation that develops

• Calcium ion increases supersaturation and makes fouling more likely

• Agglomeration of fine dicarbonate crystals on heat transfer surfaces is the likely fouling mechanism

Implications to Scale Control

• Dicarbonate nucleation is a cause of BL concentrator fouling at >50% solids– Maintain liquid-phase molar CO3/SO4 < 2.1– Adjust liquor composition in the concentrator

effects

• Controlling supersaturation is critical for managing fouling and plugging– Maintain steady levels of Ca and other NPEs

Three questions to answer when designing or optimizing black liquor concentrators to minimize Na2CO3-Na2SO4 scales:

1. When can fouling occur?

2. Where can fouling occur?

3. How can fouling be avoided or at least minimized?

9

Dry solids content of black liquor, wt-%

100%

50% 60% 70% 80%40%


BurkeiteNo Crystals Dicarbonate

Na2CO3

Na2SO4

45% 55% 65% 75%


Location of burkeite-to-dicarbonate transition where dicarbonate fouling will

likely occur in FF body 1B

Illustration, not real data

Raising product solids shifts the burkeite-to- dicarbonate transition to a point where

dicarbonate fouling should not occur

Illustration, not real data Dry solids content of black liquor, wt-%

100%

50% 60% 70% 80%40%


BurkeiteNo Crystals Dicarbonate

Na2CO3

Na2SO4

45% 55% 65% 75%


Implications to Scale Control

• A stable population of suspended crystals provides means to relieve supersaturation and avoid scaling– Operate effects well above total solids content

where dicarbonate nucleation occurs– Start up problem effects on product liquor to

provide dicarbonate seed crystals

10

Predicting Dicarbonate Formation

• Dicarbonate tendency can be evaluated from liquor composition entering effect experiencing crystallization

Crystallization region Dissolved CO3/(CO3+SO4)

mole fraction

Dissolved Na2CO3/Na2SO4

wt. fraction Burkeite < 0.68 < 1.6 Burkeite + Dicarbonate 0.68 - 0.83 1.6 - 3.7 Dicarbonate 0.83 - 0.89 3.7 - 6.0 Carbonate > 0.90 > 6.7

Predicting Dicarbonate Formation

• Modeling tools are needed to predict:– Total solids content when crystallization begins– Dissolved [species] when crystallization begins– First type of crystals formed– Transitions to other crystal types

• Problems with this approach:– Little Na-CO3-SO4-H2O thermodynamic data at

>100°C– No thermodynamic properties of dicarbonate

Empirical Model

Solve the apparent equilibrium constant:

012

2 βββ ++= xxa

where a is fit from Shi’s data for each region:

a)-(3-23

a-24

6 ][CO ][SO ][Na1)( +=calcKeq

The Model

⎟⎟⎠

⎞⎜⎜⎝

⎛+

= −−

−

24

23

23

SOCOCO

x

11

Empirical Crystallization Model Results

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

20% 30% 40% 50% 60%

Total solids content, wt. %

Wei

ght P

erce

nt

Diss. CO3, wt. % d.s.Diss. SO4, wt. % d.s.cryst. mass, wt. %

Model for Salt-water systemInitial CO3 / SO4 mole ratio = 2.5 / 1

High Performance EvaporatorsWhere Next?

• This work has provided knowledge and tools to evaluate fouling problems

• Need to apply these to improving mill operations and document results

• Some fundamental questions remain about crystallization behavior

Research Needs / Future Work• Thermodynamic data for improved models• Investigation of dicarbonate stability and

time-varying effects• Quantification of Ca2+ inhibition effects• Field demos of scale control strategies• Fundamentals of scale initiation

mechanisms

12

Acknowledgements

• Major financial support– U. S. DOE contract DE-FC36-99-GO10387– Evaporator Fouling Consortium

• Ahlstrom (now Andritz) • Weyerhaeuser• Mead (now MeadWestvaco)• Potlatch• International Paper• Lasentec

– State of Georgia TIP3 program

Acknowledgements

• Major contributions to project results– Ronald W. Rousseau– Bing Shi (GaTech PhD 4/02)

– Dan Euhus (Ga Tech PhD 9/03)

– Wolfgang Schmidl– Steven Lien, Alan Ball– Nikolai DeMartini– Alfi Soemardji & Marta Bialik (Chalmers)

Selected Recent Literature

– Frederick, Shi, Euhus, Rousseau, Tappi J. 3(6):7 (Jun 2004)– Soemardji, Verrill, Frederick, Theliander, Tappi J. 3(11):27

(Nov 2004).– Verrill, Giehl, Ratnieks, “Manipulation of Crystallization to

Resolve Severe Concentrator Scaling,” paper 8-5 IntChemical Recovery Conf, Charleston, SC (Jun 2004) & O papel (Oct 2004)

– Schmidl, Verrill, Frederick, Ball, DiMartini, “Experimental Determination and Modeling of Sodium Salt Solubility in High-Solids Kraft Black Liquor,” paper 52-1 Tappi Fall Tech Conf, Chicago (Oct 2003)

Crystallization

Documents

total solids

dry solids

black liquor

black liquor

na2so4 na2co3

solids concentrators

black liquor

sodium carbonate