Strategies To Control Product Characteristics in Simultaneous

Jan 8, 2019 - interest in zero liquid discharge options has increased.1,2. Several crystallization options have been proposed to cope with a wide rang...
0 downloads 0 Views 1MB Size
Subscriber access provided by Iowa State University | Library

Article

STRATEGIES TO CONTROL PRODUCT CHARACTERISTICS IN SIMULTANEOUS CRYSTALLISATION OF NaCl AND KCl FROM AQUEOUS SOLUTION: SEEDING WITH KCl Frederico M. Penha, Gustavo P. Zago, and Marcelo M. Seckler Cryst. Growth Des., Just Accepted Manuscript • Publication Date (Web): 08 Jan 2019 Downloaded from http://pubs.acs.org on January 8, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

STRATEGIES TO CONTROL PRODUCT CHARACTERISTICS IN SIMULTANEOUS CRYSTALLISATION OF NaCl AND KCl FROM AQUEOUS SOLUTION: SEEDING WITH KCl Frederico M. Penha1, Gustavo P. Zago1, Marcelo M. Seckler1* 1

University of São Paulo, Polytechnic School, Department of Chemical Engineering, Av. Prof. Luciano Gualberto, trav.3,

n.380, 05508-010, São Paulo – SP, Brazil. * Corresponding author: Marcelo Martins Seckler Av. Prof. Luciano Gualberto, trav.3, n.380, 05508-010, São Paulo – SP, Brazil. [email protected] Phone: +55 11 3091-2242

ABSTRACT To meet contemporary industry needs, saline wastewater treatment with zero liquid discharge goals is gaining increasing attention. In this context, evaporative crystallisation has been applied to simultaneously remove all dissolved compounds as a solid residue. By conducting simultaneous crystallisation in a way that the components crystallise as single particles, their downstream separation as products is made simple, so no solid waste is generated. We have recently identified the main phenomena that control particle morphology for batchwise simultaneous crystallisation in the NaCl-KCl-H2O model system seeded with NaCl. In this work, a process strategy is proposed, wherein crystallisation elementary phenomena are tuned by seeding with KCl particles. Formation of multicomponent particles was inhibited with a small seed surface area, because the epitaxial growth of NaCl upon the KCl seeds was hampered. By proper choice of seeding policy, the product may be classified in large particles rich in KCl (purity higher than 90 wt%) and smaller particles rich in NaCl (~75 wt%). Secondary nucleation and epitaxial growth of NaCl on KCl were the main hindrances to the formation of pure populations of each compound.

1. Introduction Industrial wastewater treatment strategies towards water reuse have attracted much attention, since they represent an answer to approach sustainability. In particular, the treatment of saline effluents is important as they are produced in large amounts by several industries. Because of the environmental burden of disposing the dissolved components of the effluent as a brine,

ACS Paragon Plus Environment

Crystal Growth & Design 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 30

interest in zero liquid discharge options has increased 1,2. Several crystallisation options have been proposed to cope with a wide range of compositions 3,4. Whatever the crystallisation option, two streams are generated: water suitable for reuse and a single particulate mixture, which is further disposed of. Consequently, current zero liquid discharge technologies offer a destination to the water in the effluent, but not to the dissolved components. To recover the dissolved components in the effluent as raw materials 5–7, extra effort is required for their purification. A possibility is to conduct crystallisation in such a way that the product is a mixture of single component particles, which are easily separable in downstream operations based on particle density and size, such as those commonly found in the mining industry. The task at hand is thus to gain understanding about simultaneous crystallisation processes to be able to favour the formation of single component particles. It is well known that the presence of a foreign component in the mother liquor can affect crystallisation elementary phenomena and particle characteristics 8,9. Yet, most of the literature about industrial crystallisation refers to the interactions of dissolved components with a single crystallising compound 9–14, with little attention paid to the effect of one solid compound upon the other one. Simultaneous crystallisation is commonly addressed in crystallogenetic studies, i.e. made by mineralogists to understand the morphology and surface topography, among other features, of crystals as found in nature 15, as well as in studies on the growth of thin films, mainly due to their applications to alloys, semiconductors and electronic materials 16,17. It is well established that heterogeneous nucleation and epitaxial growth are important phenomena in simultaneous crystallisation. When heterogeneous nucleation takes place on a foreign substrate, three orientations of the growth deposit might occur with respect to the crystalline lattice of the parent crystal: full lack of orientation; textural but not azimuthal orientation; and both textural and azimuthal orientation (textural orientation refers to the parallelism of the contact planes and azimuthal orientation refers to the parallelism of

ACS Paragon Plus Environment

Page 3 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

the crystallographic directions at the contact plane). Strictly speaking, only the fully oriented type should be called epitaxy as it predicts a strong structural relation between the substrate and the deposit-crystal 18. However, epitaxial growth is often more loosely described as the joint growth of one compound on the surface of a substrate/parent crystal, induced by heterogeneous nucleation and influenced by temperature, supersaturation and by the proximity of crystalline lattice constants 19,20. This view will be adopted here. Royer 21 has established that a small misfit between the crystal lattices of the substrate and the deposit – not higher than 15 % – is necessary for epitaxy to occur. Thereafter many mechanisms have been proposed to determine the physical form of the epitaxial deposits depending on the degree of the lattices mismatch. The smaller the mismatch, the lower is the surface energy at the interface between the crystals, commonly resulting in parallel orientation and formation of layers. For higher values of lattice misfit (e.g. ~11 % for KCl and NaCl 22), 3D nuclei are formed, hence yielding isolated crystals on top of the substrate 23. It has been suggested that, besides the lattice parameters, the actual substrate surface structure – steps, kinks and dislocations – might ease epitaxial growth 16,17,24–27. Simultaneous crystallisation studies have been reported with the NaCl-KCl-H2O system. At temperatures below say 400 o C isomorphic substitution is small, thus only pure compounds of respectively halite and sylvite structure are formed28–30. Some impurity may be incorporated in the single crystals of each compound by liquid inclusions if crystals develop under high supersaturation31,32. The spontaneous nucleation of KCl upon the surface of a dissolving NaCl crystal has been observed even in undersaturated solutions with respect to both salts 33. Takiyama et al. 34 have reported the existence of a continuous solid-solid interfacial region between a KCl deposit and its NaCl substrate, which displayed different physical properties and lower stability compared to the pure compounds. Epitaxial growth of NaCl upon KCl substrates and vice-versa have been reported 35. The KCl surface has been found to become

ACS Paragon Plus Environment

Crystal Growth & Design 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 30

rough with time, whereas the NaCl surface smoothens in cooling simultaneous crystallisation33–35. Yet, these works are limited to static conditions, millimetric-sized crystals and long timescales. In order to address simultaneous crystallisation closer to a zero liquid discharge context we have recently studied the crystallisation of NaCl and KCl in batch operation36. Evaporative crystallisation has been chosen because the objective is to recover pure water (eutectic freeze crystallisation and membrane assisted crystallisation would be other options3). Atmospheric pressure has been chosen for simplicity, so the crystallisation temperature is the normal boiling temperature of the mixture of 110 oC. We have concluded that the proportion of single component particles increased at low supersaturations which were attained with low evaporation rates, small-sized NaCl seeds and high NaCl seed content. However, seeding policies requires more thorough investigation. Seeding is among the most important parameters to tailor crystal product as, under appropriate conditions, it may reduce supersaturation and suppress both primary and secondary nucleation, as well as agglomeration, leading to narrow particle size distributions and high purities. In this contribution, our previous study was extended by considering KCl seeding at a low evaporation rate. KCl seeding has an advantage over NaCl seeding due to the higher growth rate of KCl crystals (in relation to NaCl’s) in the mixed salt solution 35, which favours the preferential formation of KCl particles in the high size end of the PSD. Moreover, the formation of single component particles is expected to be favoured under low evaporation rate because the corresponding low supersaturation is likely to minimise agglomeration and epitaxial growth. Consequently, the aim of the present work was to control the elementary phenomena in simultaneous crystallisation to enhance the formation of single component particles in KCl-seeded batchwise evaporative crystallisation in the NaCl-KCl-H2O model system. ACS Paragon Plus Environment

Page 5 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

2. Experimental 2.1. Materials Saline solutions were prepared with analytical grade reagents (NaCl and KCl) and distilled water. Each experiment was performed with a fresh solution. 2.2. Experimental setup A vertical cylindrical jacketed crystalliser with 0.5 L capacity, baffles and mechanical stirring was used. The evaporation rate was imposed by the jacket circulating fluid, which temperature was controlled by a thermostatic bath (Lauda, Eco RE 620). The crystalliser temperature was monitored by a thermocouple (Pt-100). The top of the crystalliser was connected with a jacketed condenser to collect the evaporated solvent. The water temperature in the condenser jacket was controlled at 1°C by circulation through a thermostatic bath (Lauda, Eco RE 620) and temperature measurement with a thermocouple (K-type) inside the jacket. The condensed vapour was collected in a beaker placed on a scale to monitor the vapour flow rate. 2.3. Experimental conditions The reagents and water were brought into the crystalliser and heated to 107 °C until their complete dissolution. Then seeds were added and the temperature was raised to 110 oC to partly dissolve the seeds, thus removing any fines and fragments originally on their surface. No additional healing procedure was applied to the seeds. At the temperature of 110 oC boiling evaporation started, promoting formation of the solid phases. The amount of condensed vapour was monitored at constant time intervals (5 min) until an evaporation extent of 50 % (mass of evaporated water per initial water mass). Crystallisation was conducted with KCl seeds in two particle sizes and three seed loads, as well as without seeds, as shown in Table 1. All experiments were performed with an evaporation rate of 0.002 gram ACS Paragon Plus Environment

Crystal Growth & Design 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 30

of vapour per gram of initial solution. This evaporation rate is lower than in our previous work (0.003 and 0.005 g/g) to assure low supersaturation values.

Table 1. KCl seeding conditions. Seed size

Seed Content

(μm)

(kgseeds/kginitial_solution)

-

-

E1 / E2 / E3

550

0.05 / 0.10 / 0.25

E4 / E5 / E6

275

0.05 / 0.10 / 0.25

Experiment E0

Slurry samples were taken during the crystallisation process at evaporation extents of 10, 25 and 50%, hereafter called sample 1, 2 and 3 respectively. Sample 1 corresponds to crystallisation of KCl alone, Samples 2 and 3 correspond to simultaneous crystallisation. Sample 3 is the final product. Each sample was immediately vacuum filtered, washed with ethanol (purity > 99.5 %) and then dried at 50 °C for at least 8 h for particle characterisation. 2.4. Seeds preparation KCl seeds were produced from batch cooling crystallisation in two linear cooling conditions: (i) at 11.0 °C.h-1; (ii) at 9.2 °C.h-1. Seeds were filtered, washed in ethanol (Purity > 99.5 %) and dried overnight in an oven at 50 °C. Thereafter crystals were sieved in Tyler series sieves in a magnetic vibrating platform for 1 hour and the fractions of interest (275 μm and 550 μm from procedures i and ii, respectively) were stored. 2.5. Particle characterisation Samples of dry particles were characterised by optical microscopy (Olympus® BX60F-3) and scanning electron microscopy (SEM - JEOL JSM-7401F) coupled with energy dispersive Xray spectroscopy (EDS). The products (Sample 3) particle size distributions were determined by sieving as explained before. Mass based populational densities were calculated. Population ACS Paragon Plus Environment

Page 7 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

coefficients of variation (CV) were also calculated with Equation 1, where L16, L50 and L84 are, respectively, the cumulative characteristic sizes of 16, 50 and 84 % of the sample in mass. 𝐶𝑉 = 100 .

(𝐿84 ― 𝐿16)

(1)

2 .𝐿50

Dominant sizes were considered as the weighted arithmetic mean size between the two adjacent highest populational densities in the particle size distribution. Argentometry with Mohr’s method 37 was applied to determine the mass proportion of KCl in dry crystallisation products in each size range (expressed as grams of KCl per grams of KCl and NaCl times 100). A complete description of the titrations and calculations can be found in our previous work 36. In order to quantify the variation of solids composition with particle size, a segregation index was defined according to Equation 2, where xi is the KCl proportion in size range i, xaverage is the average proportion of KCl in the whole product and Pi is the mass-based population density in size range i in kg.kg-1.m-1.



𝑆𝐼 =

[

|(𝑥𝑖 ― 𝑥𝑎𝑣𝑒𝑟𝑎𝑔𝑒)| 𝑥𝑎𝑣𝑒𝑟𝑎𝑔𝑒

]

× 𝑃𝑖

(2)

∑ 𝑃𝑖

3. Results and discussion 3.1. Seeds Seeds with dominant particle sizes respectively of 550 and 275 μm are shown in Figure 1. They were either single crystals with the cubic habit characteristic of KCl or agglomerates of these primary cubic crystals. Small particles had well-defined edges whereas large particles had slightly rounded edges. Also, a few of the large particles were broken.

ACS Paragon Plus Environment

Crystal Growth & Design 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1. Optical microscopies of the seeds with dominant sizes of 275 μm (left) and 550 μm (right).

Fluid inclusions developed within most crystals, often as six large inclusions symmetrically located with respect to the crystal faces. It is likely that these crystals had hopper faces in early stages of their formation, when the supersaturation was high38,39. As the supersaturation dropped, the faces became flat, generating fluid inclusions in the hopper cavities, which coalesced into the observed single inclusions. This kind of growth was observed here for freshly formed KCl crystals from ternary solutions (see Section 3.4) and has been previously reported by Rohani and Ng 31 for KCl crystals in the presence of several compounds, including NaCl. Coalescence of fluid inclusions has also been observed for batch cooling crystallisation of copper sulphate 32. 3.2. Mass balances Both solid and liquid phase compositions throughout the batches were determined from mass balances for the ternary system KCl-NaCl-H2O and are summarized in Table 2. Solution and solid phase compositions according to the evaporation extent, defined as the proportion of evaporated water with respect to initial water content. The phase diagram at 110 °C,

approximately the boiling point of the initial solution, was obtained with the thermodynamic modelling software OLI Studio 9.2 and is shown in Figure 2. The initial solution composition chosen for the batch experiments was nearly saturated with respect to KCl and undersaturated ACS Paragon Plus Environment

Page 8 of 30

Page 9 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

with respect to NaCl (yellow dot in the figure). KCl seeds were added and the temperature was raised to the boiling temperature. Slight dissolution of the seeds was expected so that the solution composition changed towards the KCl solubility line (blue dot). Upon evaporation, only KCl crystallised and the solution composition followed the KCl solubility line until, at an evaporation extent of 16%, the eutonic condition was reached (grey dot), i.e. the solution in the ternary system is in equilibrium with salts of both solutes40. Thereafter both KCl and NaCl crystallised and the solution composition became invariant. During the batch, the actual solution concentrations were slightly higher than the equilibrium conditions just mentioned, but these differences have been neglected for the present mass balance analysis as the metastable limits for both NaCl and KCl are small.

Table 2. Solution and solid phase compositions according to the evaporation extent, defined as the proportion of evaporated water with respect to initial water content. Sample

0

1

-

2

3

Evaporation extent (%)

0

10

16

25

50

Soluble KCl (g/gH2O)

0.389

0.374

0.365

0.365

0.365

Crystallised KCl (g/gH2O)

-

0.024

0.062

0.098

0.195

Soluble NaCl (g/gH2O)

0.228

0.253

0.274

0.274

0.274

Crystallised NaCl (g/gH2O)

-

-

-

0.025

0.096

ACS Paragon Plus Environment

Crystal Growth & Design 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 30

Figure 2. Solubility curves in the NaCl-KCl-H2O system at 110 °C. Initial condition for batch experiments (yellow dot), initial condition for the beginning of the evaporation (blue dot) and eutonic condition (purple dot) are indicated. 3.3. Supersaturation Phase diagrams do not describe crystallisation mechanisms and kinetics. To this end, it is desirable to take into account supersaturation. Unfortunately, the experimental determination of supersaturation in the NaCl-KCl-H2O system is difficult as the supersaturated and the saturated state are very close to each other. Consequently, a qualitative approach may be applied. Since the supersaturation in a batch crystallisation process is inversely related to the particle specific surface area, the supersaturation in different experimental conditions used in this work may be compared. The supersaturation is considered the highest for the unseeded condition and inversely related to the seed content as well as seed size. 3.4. Batches with 550 μm seeds Seeds and product particle size distributions (PSD) from crystallisation batches with 550 μm sized KCl particles are shown in Figure 3.

ACS Paragon Plus Environment

Page 11 of 30

Table 3 displays the CVs and the dominant sizes. The seeds displayed a unimodal distribution with narrow size dispersion. The products, irrespective of the presence of seeds and of the seed content, resulted in bimodal PSDs with a wide size dispersion. Product dominant sizes for the primary peak for all conditions were within the range of 600-650 μm, whereas the secondary peaks occurred in around 300 μm. Unseeded CV: 26 %

10% CV: 16 %

25% CV: 10 %

KCl 550 μm CV: 7 % 10000

4.00E+03

8000

3.00E+03

6000

2.00E+03

4000

1.00E+03

2000

0.00E+00

Population density (kg.kg-1.m-1)

5.00E+03

5% CV: 12 %

Population density (kg.kg-1.m-1)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

0 50

150

250

350

450

550

Size (μm)

650

750

850

Figure 3. Particle size distribution for the seeds (secondary axis) and for the final samples of experiments performed with 5, 10 and 25 wt% content of 550 μm seeds (primary axis).

Table 3. Coefficients of variation (CV) and dominant sizes (LD). The peaks ‘p1’ and ‘p2’ of the particle size distribution refer to the primary and secondary peaks of the particle size distribution, defined as the peaks at the large and small size ranges, respectively. CV (%) Seeding

Seeds

No seeding

5 wt%

10 wt%

25 wt%

550 µm

7

26

12

16

10

275 µm

3

21

24

25

5 wt%

10 wt%

25 wt%

LD (µm) Seeds (µm) 550 µm

No seeding p1

p2

p1

p2

p1

p2

p1

p2

645

332

616

282

619

285

627

273

ACS Paragon Plus Environment

Crystal Growth & Design 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

275 µm

583

403

Page 12 of 30

564

366

571

372

In the early stages of KCl seeded batches only KCl crystallised. It is likely that crystal growth of the KCl seeds yielded the dominant peak of the PSD, as this peak was larger than the added seed size. At an evaporation extent of 16 % the eutonic condition was reached, sometime thereafter NaCl primary nucleation must have happened. It is likely that the particles originated thereof formed the secondary peak of the PSD. These particles were smaller than those in the primary peak because the NaCl particles residence time was shorter than KCl’s and because the NaCl growth rate in the ternary system is lower than KCl’s 35. The secondary peak was likely also populated with particles generated late in the batch by secondary nucleation of both salts. In the unseeded batch, a similar sequence of phenomena took place, except that the KCl particles were formed from primary nucleation shortly after the start of the evaporation. Consequently, the overall description of the PSD is the same for both seeded and unseeded crystallisations, i.e. primary and secondary peaks of the PSD were mainly formed by KCl and NaCl, respectively. This view will be refined with further particle characterisation shown next. Optical microscopy for products from crystallisation seeded with 550 μm particles (Figure 4) shows populations with two size ranges, which roughly correspond to those found in the PSDs. Particles in each size range are comprised of both single crystals and polycrystalline particles. The latter are constituted by single crystals of multiple sizes.

ACS Paragon Plus Environment

Page 13 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

Figure 4. Optical microscopies for the products of the unseeded batch and for the crystallisation with various amounts of 550 μm seeds.

The large particles are somewhat agglomerated with shapes that resemble the seed agglomerates. They contain fluid inclusions that are spatially distributed in a way that matches the fluid inclusions of the seeds. Both morphological features of these large particles indicate that they were formed by crystal growth upon KCl seeds. Besides, the large particles display abraded (rounded) edges and irregular shapes, due to long exposure to mechanical collisions, as most of them were added as KCl seeds at the start of the batches. Abrasion is also prominent in large particles, in comparison with small particles, due to their relatively large inertia during mechanical collisions. Abrasion is important as it boosts the generation of secondary nuclei. Consequently, the particle shapes suggest that the large KCl particles yielded KCl secondary nuclei which, upon growth, populated the small size range.

ACS Paragon Plus Environment

Crystal Growth & Design 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 30

The small particles do not contain fluid inclusions, so they are either NaCl or KCl crystals formed under low supersaturation. They display well-defined edges because they had little chance to suffer abrasion, as they are small and appeared at a late stage of crystallisation. Figure 5 shows SEM-EDS micrographs for the unseeded condition and Figure 6 shows images for 5 and 25 wt% of 550 μm sized seeds. Only KCl is seen in Sample 1 of each experiment, as expected, since these particles were formed before the eutonic point was reached. Sample 1 of the unseeded experiment was comprised of rather uniformly sized small particles (~100 μm) generated by primary heterogeneous nucleation. The crystals displayed characteristics of Hopper growth. Upon further crystal growth (Figure 5, Sample 2), these KCl crystals lost the hopper habit, with their surfaces becoming flat as the supersaturation decreased with time. These changes were accompanied by the generation of fluid inclusions which ended in the large product particles, as explained before (Section 3.1). As to Sample 1 of the crystallisation with 5 wt% seeds, crystals presented well-defined edges and flat smooth surfaces, consistent with surface-integration-controlled growth upon the added seeds. One concludes that the presence of seeds reduced the supersaturation (in relation to the unseeded condition) because hopper growth was inhibited. Secondary nucleation took place, as evidenced by abraded crystal edges and kinks. For the high seed content of 25 wt%, secondary nucleation was more pronounced, suggested by the presence of completely round crystals. Sample 2 was withdrawn after the eutonic condition had been reached. They comprise single particles of each component as well as multicomponent particles formed by NaCl and KCl primary particles (Figure 5 and Figure 6). Single KCl crystals were formed both by crystal growth upon the seeds and from secondary nucleation. NaCl single crystals were formed mainly by primary heterogeneous nucleation in solution. NaCl particles might also have been

ACS Paragon Plus Environment

Page 15 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

formed by secondary nucleation, yet, this phenomenon is less important here, as the NaCl population is comprised by small crystals. The genesis of the multicomponent particles is more complex. Epitaxial growth is an expected phenomenon since NaCl becomes supersaturated while KCl crystals are suspended in solution and the difference between the lattice parameters of NaCl and KCl is lower than 15 % (~11 %) 22. Although NaCl is known to grow on KCl surfaces by epitaxial growth 35, both oriented and non-oriented NaCl particles can be seen on the surface of KCl. The non-oriented textures may also be associated with agglomeration. In SEM-EDS images for Sample 3 (Figure 5 and Figure 6), taken at 50 % of water removed, multicomponent particles are seen in a much lesser proportion than single component KCl and NaCl particles, compared to Sample 2. This suggests that the NaCl particles that had nucleated on KCl surfaces were engulfed by KCl’s growing surfaces, as KCl growth rate is higher than NaCl’s. Fresh NaCl particles are not frequently observed on KCl parent crystals probably because neither epitaxial growth nor agglomeration are expected at the late stages of the batch due to low supersaturation. Thus, particles at the end of the batch were mainly composed of both large single component particles with irregular shapes (~ 600 μm) and small cubic crystals (~ 250 μm). The large crystals were mainly KCl, with few NaCl epitaxial growth evidences, whereas the small-sized population were identified mainly as NaCl. Some polycrystalline particles are visible in various sizes with different compositions but in small proportion in relation to single crystals.

ACS Paragon Plus Environment

Crystal Growth & Design 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 30

Figure 5. SEM views of particles from unseeded crystallisation. Red and green coloured crystals were identified by EDS as KCl and NaCl, respectively. Grey crystals were not analysed. The magnification is not the same for all images.

Figure 6. SEM views for experiments with 5 and 25 wt% of 550 μm seeds. Red and green coloured crystals were identified by EDS as KCl and NaCl, respectively. Grey crystals were not analysed. The magnification is not the same for all images. 3.5. Batches with 275 μm seeds PSD features for the seeds and products from 275 μm seeded batch experiments are presented in Figure 7 and Table 3. Coefficients of variation (CV) and dominant sizes (LD). The peaks ‘p1’ and ‘p2’ of the particle size distribution refer to the primary and secondary peaks of the particle size distribution, defined as the peaks at the large and small size ranges, respectively.. The overall aspects of the PSD´s are ACS Paragon Plus Environment

Page 17 of 30

similar to those from the 550 μm experiments, i.e. the seeds showed unimodal distribution, whereas the products PSDs were multimodal with size dispersions much wider than the added seeds. However, the primary peaks dominant sizes were smaller and the size dispersions (CVs) were larger. Besides, the population density of the primary peak increased in importance with the seed content, whereas the secondary peak decreased.

5% CV: 21 %

10% CV: 24 %

25% CV: 25 %

KCl 275 μm CV: 3 %

3.00E+03

12000

1.50E+03

6000

0.00E+00

Population dentisy (kg.kg-1.m-1)

18000

4.50E+03

Population dentisy (kg.kg-1.m-1)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

0 50

150

250

350

450

Size (μm)

550

650

750

850

Figure 7. Particle size distribution for the seeds (secondary axis) and for the final samples of experiments performed with 5, 10 and 25 wt% of 275 μm seeds (primary axis). Particles from experiments with 275 μm seeds had some morphological features (Figure 8) similar to those found for experiments with 550 μm seeds. The largest particles were manly agglomerates with fluid inclusions suggesting that they were grown KCl seeds, whereas smaller flat crystals without fluid inclusions were likely either NaCl or KCl crystals formed under low supersaturation. However, differences were seen as well. The relatively small amount of crystal fragments points to the occurrence of less secondary nucleation than in the experiments with 550 μm seeds, likely because of a lower supersaturation associated with a higher surface area. Besides, particles with various sizes were observed, instead of the two ACS Paragon Plus Environment

Crystal Growth & Design 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 30

distinct size ranges seen in 550 μm seeded experiments. This may be attributed to the smaller expected final size for the grown seeds (both because the seeds were smaller and because their increase in size was lower, given the larger surface area for the same amount of deposited salt). The higher surface area available with 275 μm seeds may also have increased NaCl nucleation on KCl seeds surface. This would end up in an increase of mixed composition polycrystals, which also played a role in the observed enlargement of size dispersion with respect to the larger seeds experiment.

Figure 8. Optical microscopies of products from 275 μm seeded batch experiments.

Particles composition using SEM-EDS is shown in Figure 9 for experiments seeded with 5 and 10 wt% of 275 μm seeds. Sample 1, taken before the eutonic was reached, did not contain NaCl crystals, as expected. It was composed of faceted single crystals of cubic habit and agglomerates of KCl (similar to the seeds) with abraded edges. Consequently, in the early stages of the batch, both surface-integration-controlled crystal growth and secondary nucleation took place. For the low seed content, the crystal edges were better defined and the shapes were more regular, when compared to the high seeding experiments, due to less secondary nucleation. Sample 2, taken after the eutonic condition was reached, displayed single KCl crystals as well as polycrystalline multicomponent particles. As discussed previously, such particles may have been formed by agglomeration, by epitaxial growth or by primary heterogeneous nucleation of NaCl on the surface of KCl. Sample 3 presented single

ACS Paragon Plus Environment

Page 19 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

component particles of KCl and NaCl in higher proportion than mixed composition polycrystals. KCl was observed both as large damaged crystals (~500 μm) and as smaller crystals, suggesting that the former ones are grown seeds that originate the latter ones by secondary nucleation.

Figure 9. SEM-EDS analysis for the experiments with 5 and 10 wt% of 275 μm seeds. Red and green coloured crystals were identified by EDS as KCl and NaCl, respectively. Grey crystals were not analysed. (Note that the magnification is not the same for all images) 3.6. Particles chemical composition Figure 10 and Figure 11 show the KCl proportion (solid line) and the population density (dashed line) of the product in each size range. The average KCl proportion in the product is shown (dotted line), which was determined from mass balances displayed in Table 4 and was consistent with measured values (not shown). For the sake of clarity, the PSD is classified in three size ranges, large, intermediate and small, corresponding respectively to the primary peak of the PSD, to the secondary peak and to sizes below the secondary peak of the PSD. When the KCl proportion in the size ranges deviate from the average composition, we refer to the particulate product as segregated. ACS Paragon Plus Environment

Crystal Growth & Design 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 30

Impurities in crystals may derive from isomorphic substitution, insufficient crystal washing and liquid inclusions. For the temperatures of interest, isomorphic substitution of both Na in KCl and K in NaCl are negligible, so the single crystals may be considered pure compounds with either halite or sylvite structure. Impurities from the mother liquor adhered on the crystals surface was under control in laboratory conditions and may be neglected as well. Liquid inclusions have been found to occur. However, most of the inclusions appear to be those already present at the seeds, so we estimate that their influence on crystal purity represents at most a few percent, thus, they were also neglected. Consequently, the only source of impurity within each size fraction arises from mixed composition polycrystals formed by epitaxy or agglomeration.

Table 4. Total solids balance and KCl average proportions. Unseeded

5 wt% seeds 10 wt% seeds 25 wt% seeds

NaCl (g/100gH2O)

9.6

9.6

9.6

9.6

KCl (g/100gH2O)

19.5

27.5

35.5

59.7

Total solids (g/100gH2O)

29.0

37.1

45.1

69.3

KCl proportion (%)

67

74

79

86

For all conditions, the KCl proportion for the large size range was high when compared with the average proportion of the product, often higher than 90 %. The intermediate size showed much lower KCl proportions than the average whereas the small size range presented a KCl proportion higher than or in close proximity to the average. As explained before, the large size range was comprised mainly by KCl particles grown from the added seeds. NaCl crystals were mainly originated from heterogenous primary nucleation in solution, giving rise to the NaCl-rich intermediate size range of the PSD. The small size range was probably populated with grown secondary nuclei of KCl and NaCl crystals. The

ACS Paragon Plus Environment

Page 21 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

low NaCl proportion in the large size range suggests that neither agglomeration nor NaCl epitaxial growth upon KCl parent crystals were dominant phenomena. In general, the most effective segregation was found for high supersaturations, i.e. for the unseeded condition and for low particle surface areas – low seed content and large seeds. This behaviour may be appreciated by inspecting the segregation indexes given in Table 5. This is a surprising result, as in our previous work on simultaneous crystallisation of NaCl and KCl36, segregation was favoured at low supersaturations. However, the supersaturation range was wider than in this contribution – due to changes in the evaporation rate. It is concluded that epitaxial growth and agglomeration, below some limiting supersaturation value, are no longer controlled by supersaturation. Instead, they are favoured at high KCl surface areas, simply because these processes take place at these surfaces.

Table 5. Segregation index for various seed sizes and seed content Segregation index Seed size (µm)

5 wt% 10 wt%

25 wt%

550

0.26

0.23

0.21

275

0.26

0.19

0.14

Unseeded 0.29

Overall, the large and intermediate size fractions are relatively well segregated and combined they represent a significant proportion of the yield. Consequently, downstream size classification of the product is expected to yield a KCl-rich stream of large particles, a NaClrich stream of intermediate sized particles and a small non-segregated stream of small particles. The main phenomena identified as responsible for hampering segregation were epitaxial growth of NaCl upon KCl as well as agglomeration of NaCl and KCl as they lead to the

ACS Paragon Plus Environment

Crystal Growth & Design 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 30

formation of multicomponent particles in the large size range. KCl secondary nucleation is also inconvenient for the separation of the compounds by size. Since secondary nuclei are continuously generated in the batch, KCl particles are present both in the low and in the intermediate NaCl-rich size range. The minimisation of these phenomena has been shown to be possible through proper choice of seed sizes and contents as well as the evaporation rate. For relatively high supersaturations 36, segregation was favoured for large seed surface areas. For the low supersaturations used in this work, on the other hand, segregation was enhanced on lower surface areas, i.e. large seeds in low amounts.

ACS Paragon Plus Environment

Page 23 of 30

KCl

Populational density

KCl

Average proportion

Populational density

4500

3750

75

3375

50

2500

50

2250

25

1250

25

1125

(A)

0

200

KCl

400

Size (µm)

600

(B)

0 0

200

600

75

3.90E+03

75

4125

50

2.60E+03

50

2750

25

1.30E+03

25

1375

0

0.00E+00

0

Populational density (kg.kg-1.m-1) KCl proportion (%)

0

200

400 Size (µm)

600

800

(D)

Average proportion

800

100

(C)

KCl

400 Size (µm)

Populational density 5.20E+03

100

Average proportion

0

0 1000

800

Populational density

5500

Populational density (kg.kg-1.m-1)

0

Populational density (kg.kg-1.m-1)

75

KCl proportion (%)

100

Populational density (kg.kg-1.m-1)

KCl proportion (%)

Average proportion

5000

100

KCl proportion (%)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46

Crystal Growth & Design

0 0

200

400 Size (µm)

600

800

Figure 10. KCl proportion in each size range, mass-based population density and average product KCl proportion for the unseeded condition (A); 5 wt% of 550 μm seeds (B); 10 wt% of 550 μm seeds (C); and 25 wt% of 550 μm seeds (D). ACS Paragon Plus Environment

Crystal Growth & Design

75

3375

50

2250

25

1125

0

(A)

0 0

200

400 Size (µm)

600 KCl

100

KCl proportion (%)

800

KCl

100

Average proportion

2775

50

1850

25

925

0

(B)

0 0

Average proportion

200 Populational density

400 Size (µm)

50

1650

25

825

0

0 200

400 Size (µm)

600

600

800

3300 2475

0

3700

75

75

(C)

Populational density

Populational density (kg.kg-1.m-1)

4500

Populational density (kg.kg-1.m-1)

Populational density

Populational density (kg.kg-1.m-1)

Average proportion

KCl proportion (%)

KCl

100

KCl proportion (%)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46

Page 24 of 30

800

Figure 11. KCl proportion in each size range, mass-based population density and average product KCl proportion for the experiments with (A) 5 wt% of 275 μm seeds; (B) 10 wt% of 275 μm seeds; and (C) 25 wt% of 275 μm seeds.

ACS Paragon Plus Environment

Page 25 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

4. Conclusions Simultaneous crystallisation of KCl and NaCl from its ternary system with water was studied batchwise, using the evaporative method and KCl seeding to minimise the formation of mixed composition particles. In the beginning of the batches, KCl was the only crystallising salt, which underwent surfaceintegration-controlled growth and secondary nucleation. When the eutonic condition was reached, NaCl crystals were formed by primary heterogeneous nucleation, both in solution and on the surface of KCl seeds. The latter subsequently grew epitaxially to form both oriented and non-oriented growth deposits. Because KCl grows faster than NaCl, the large size range was rich in KCl, in proportions higher than 90 %. The intermediate size range was rich in NaCl, but its proportion only reached around 80 % in few cases, due to KCl secondary nucleation. The smallest particles had mixed composition due to secondary nucleation of both salts. Epitaxial growth of NaCl upon KCl as well as agglomeration of NaCl and KCl hamper the downstream separation of the compounds as they lead to the formation of multicomponent particles. It has been found that these processes may be inhibited in the presence of small KCl surface areas (large seeds in low amounts), whereas supersaturation does not play a significant role in the low supersaturation range investigated. Secondary nucleation of both KCl and NaCl, which takes place throughout the batch, is inconvenient for size separation as it promotes the formation of mixed composition populations at the low and the intermediate size ranges. The control of these two elementary phenomena is the key to bypass the main hindrances of this process. The strategies developed in this work evidenced the possibility of favouring the formation of single crystals of both compounds, through low evaporation rates combined with low seed surface areas. However, low evaporation rates entail long operation times. Therefore, secondary nucleation cannot be avoided. ACS Paragon Plus Environment

Crystal Growth & Design 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 30

5. Acknowledgments The financial support of the National Council for Scientific and Technological Development (CNPq), of the Coordination for the Improvement of Higher Education Personnel (CAPES) and of Petrobras are gratefully acknowledged.

6. References (1)

Ghyselbrecht, K.; Huygebaert, M.; Van der Bruggen, B.; Ballet, R.; Meesschaert, B.; Pinoy, L. Desalination of an Industrial Saline Water with Conventional and Bipolar Membrane Electrodialysis. Desalination 2013, 318, 9–18.

(2)

Ranade, V. V.; Bhandari, V. M. Industrial Wastewater Treatment , Recycling , and Reuse Industrial; Butterworth-Heinemann: Oxford, 2014.

(3)

Lu, H.; Wang, J.; Wang, T.; Wang, N.; Bao, Y.; Hao, H. Crystallization Techniques in Wastewater Treatment: An Overview of Applications. Chemosphere 2017, 173, 474–484.

(4)

Chen, Y.; Davis, J. R.; Nguyen, C. H.; Baygents, J. C.; Farrell, J. Electrochemical IonExchange Regeneration and Fluidized Bed Crystallization for Zero-Liquid-Discharge Water Softening. Environ. Sci. Technol. 2016, 50, 5900–5907.

(5)

Kim, D. H. A Review of Desalting Process Techniques and Economic Analysis of the Recovery of Salts from Retentates. Desalination 2011, 270, 1–8.

(6)

Zhou, H.; Bao, Y.; Bai, X.; Ma, R.; Huangfu, L.; Zhang, C. Salt-Forming Regions of Seawater Type Solution in the Evaporation and Fractional Crystallization Process. Fluid Phase Equilib. 2014, 362, 281–287.

(7)

Guan, G.; Yao, C.; Lu, S.; Jiang, Y.; Yu, H.; Yang, X. Sustainable Operation of Membrane Distillation for Hypersaline Applications: Roles of Brine Salinity, Membrane Permeability and Hydrodynamics. Desalination 2018, 445, 123–137.

(8)

Lewis, A.; Seckler, M.; Kramer, H.; Rosmalen, G. van. Industrial Crystallization: Fundamentals and Applications. Cambridge University Press 2015.

(9)

Nyvlt, J.; Ulrich, J. Admixtures in Crystallization; Wiley, 1995.

(10)

Sarig, S. Growth of Potassium Chloride Crystals. J. Cryst. Growth 1977, 39, 255–266. ACS Paragon Plus Environment

Page 27 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

(11)

Akl, M. M.; Nassar, M. M.; Sayed, S. A. Influence of Impurities and Suspension Density on the Crystallization Kinetics of Potassium Chloride. Chem. Ing. Tech. 1991, 63, 935–939.

(12)

König, A.; Emons, H. H.; Nyvlt, J. The Influence of Sodium Chloride on the Driving Force of the Crystallization of Potassium Chloride from Aqueous Solutions at 25 °C. Cryst. Res. Technol. 1987, 22, 13–19.

(13)

Voloshin, A. E.; Kovalev, S. I.; Rudneva, E. B.; Glikin, A. E. Phenomena and Mechanisms of Mixed Crystal Formation in Solutions II. Mechanism of Interface Processes. J. Cryst. Growth 2004, 261, 105–117.

(14)

Sangwal, K. Effects of Impurities on Crystal Growth Processes. Prog. Cryst. Growth Charact. Mater. 1996, 32, 3–43.

(15)

Aquilano, D.; Otálora, F.; Pastero, L.; García-Ruiz, J. M. Three Study Cases of Growth Morphology in Minerals: Halite, Calcite and Gypsum. Prog. Cryst. Growth Charact. Mater. 2016, 62, 227–251.

(16)

Family, F.; Amar, J. G. Kinetics of Epitaxial-Growth and Roughening. Mater. Sci. Eng., B 1995, 30, 149–166.

(17)

Pond, K.; Gossard, A. C.; Lorke, A.; Petroff, P. M. Role of Steps in Epitaxial Growth. Mater. Sci. Eng., B 1995, 30, 121–125.

(18)

Mutaftschiev, B. The Atomistic Nature of Crystal Growth; Springer: Berlin, Heidelberg, 2013.

(19)

Chernov, A. A. Nucleation and Epitaxy. In Modern Crystallography III: Crystal Growth; Springer: Berlin, Heidelberg, 1984.

(20)

Markov, I. V. Crystal Growth for Beginners: Fundamentals of Nucleation, Crystal Growth and Epitaxy; World Scientific, 2003.

(21)

Royer, L. Recherches Experimentalles Sur l’epitaxie on Orientation Mutuelle Des Cristaux Des Espices Differentes. Bull. Soc. Franc. Miner. 1928, 55, 77–159.

(22)

Glikin, A. E.; Plotkina, J. N. V.-2. Disorientation Effects of Epitaxy at Aqueous Media. Mater. Struct. 1999, 6, 155–158.

(23)

Pashley, D. Epitaxy Growth Mechanisms. Mater. Sci. Technol. 1999, 15, 2–8.

(24)

Habelitz, S.; Höche, T.; Hergt, R.; Carl, G.; Rüssel, C. Microstructural Design through ACS Paragon Plus Environment

Crystal Growth & Design 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 30

Epitaxial Growth in Extruded Mica Glass-Ceramics. Acta Mater. 1999, 47, 2831–2840. (25)

Stoneham, A. M.; Harding, J. H. Computer Simulation of Interfaces: What Do We Need to Know? Acta Mater. 1998, 46, 2255–2261.

(26)

Patrone, P. N.; Caflisch, R. E.; Margetis, D. Characterizing Equilibrium in Epitaxial Growth. Epl 2012, 97.

(27)

Sarma, K. R.; Shlichta, P. J.; Wilcox, W. R.; Lefever, R. A. Epitaxy versus Oriented Heterogeneous Nucleation of Organic Crystals on Ionic Substrates. J. Cryst. Growth 1997, 174, 487–494.

(28)

Priya, M.; Mahadevan, C. K. Studies on Multiphased Mixed Crystals of NaCl, KCl and KI. Cryst. Res. Technol. 2009, 44, 92–102.

(29)

Bakker, R. J. Package FLUIDS. Part 4: Thermodynamic Modelling and Purely Empirical Equations for H2O-NaCl-KCl Solutions. Mineral. Petrol. 2012, 105, 1–29.

(30)

Barrett, W. T.; Wallace, W. E. Studies of NaCl-KCl Solid Solutions. I. Heats of Formation, Lattice Spacings, Densities, Schottky Defects and Mutual Solubilities. J. Am. Chem. Soc. 1954, 76, 366–369.

(31)

Rohani, S.; Ng, B. Batch Crystallization of KCl in the Presence of Soluble and Insoluble Impurities. Chem. Eng. Sci. 1992, 47, 367–379.

(32)

Giulietti, M.; Seckler, M. M.; Derenzo, S.; Valarelli, J. V. Changes in Copper Sulfate Crystal Habit during Cooling Crystallization. J. Cryst. Growth 1996, 166, 1089–1093.

(33)

Matsuoka, M.; Yamamoto, K.; Uchida, H.; Takiyama, H. Crystallization Phenomena in Ternary Systems: Nucleation of KCl during Dissolution of NaCl. J. Cryst. Growth 2002, 244, 95–101.

(34)

Takiyama, H.; Yamamoto, K.; Uchida, H.; Matsuoka, M. Observation and Analysis of the Interface Between NaCl Substrate and KCl Deposit Crystal. e-J. Surf. Sci. Nanotechnol. 2007, 5, 113–116.

(35)

Glikin, A. E. Polymineral-Metasomatic Crystallogenesis. Springer Netherlands 2009.

(36)

Penha, F. M.; Zago, G. P.; Nariyoshi, Y. N.; Bernardo, A.; Seckler, M. M. Simultaneous Crystallization of NaCl and KCl from Aqueous Solution: Elementary Phenomena and Product ACS Paragon Plus Environment

Page 29 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

Characterization. Cryst. Growth Des. 2018, 18, 1645-1656. (37)

Harris, D. C. Quantitative Chemical Analysis; W. H. Freeman, 2010.

(38)

Desarnaud, J.; Derluyn, H.; Carmeliet, J.; Bonn, D.; Shahidzadeh, N. Nucleation and Growth of Sodium Chloride in Confined Geometries. SWBSS 2014 - 3rd International Conference on Salt Weathering of Buildings and Stone Sculptures, Brussels. 2014, 17–33.

(39)

Tsukamoto, K. In-Situ Observation of Crystal Growth and the Mechanism. Prog. Cryst. Growth Charact. Mater. 2016, 62, 111–125.

(40)

DeVoe, H. Thermodynamics and Chemistry; Thermodynamics and Chemistry; Prentice Hall, 2001.

ACS Paragon Plus Environment

Crystal Growth & Design 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 30

"For Table of Contents Use Only"

STRATEGIES TO CONTROL PRODUCT CHARACTERISTICS IN SIMULTANEOUS CRYSTALLISATION OF NaCl AND KCl FROM AQUEOUS SOLUTION: SEEDING WITH KCl Frederico M. Penha1, Gustavo P. Zago1, Marcelo M. Seckler1* 1

University of São Paulo, Polytechnic School, Department of Chemical Engineering, Av. Prof.

Luciano Gualberto, trav.3, n.380, 05508-010, São Paulo – SP, Brazil.

Synopsis: In this work, crystallisation phenomena are tuned by seeding with KCl particles. Formation of multicomponent particles was inhibited with a small seed surface area, because the epitaxial growth of NaCl upon the KCl seeds was hampered. By proper choice of seeding policy, the product may be classified in large particles rich in KCl and smaller particles rich in NaCl.

ACS Paragon Plus Environment