materials

Article Transformation Kinetics of Burnt in Freshwater and Sea

Harald Justnes 1,2,* , Carlos Escudero-Oñate 3, Øyvind Aaberg Garmo 3 and Martin Mengede 4 1 SINTEF Community, Stiftelsen for Industriell og Teknisk Forskning (SINTEF), Høgskoleringen 7b, 7034 Trondheim, Norway 2 Faculty of Natural Sciences, Norwegian University of Science and Technology (NTNU), Sem Sælandsvei 12, 7034 Trondheim, Norway 3 Norwegian Institute for Water Research (NIVA), Gaustadalléen 21, 0349 Oslo, Norway; [email protected] (C.E.-O.); [email protected] (Ø.A.G.) 4 Franzefoss Minerals AS, Olav Ingstads vei 5, 1309 Rud, Norway; [email protected] * Correspondence: [email protected]

 Received: 4 September 2020; Accepted: 29 October 2020; Published: 2 November 2020 

Abstract: (CaO), also known as burnt lime, is being considered as a possible treatment to reduce the negative impact of sea urchins on tare forests in northern coastal and blue-green algal blooms in the surrounding of fish-farms. In this respect, the reaction kinetics of burnt lime in contact with sea water has been elucidated and compared to its behaviour in fresh water. In the first minutes of contact between burnt lime and water, it “slaked” as CaO reacted with water to yield calcium (Ca(OH)2). Subsequently, calcium hydroxide reacted with , sulphate and from the sea water to yield (Mg(OH)2), calcium sulphate dihydrate (, CaSO 2H O) and (CaCO ), respectively. In a closed system 4· 2 3 of 1% CaO in natural sea water (where the supply of sulphate, magnesium and carbonate is limited), more than 90% reacted within the first 5 h. It is foreseen that in an open system, like a marine fjord, it will react even faster. The pH 8 of sea water close to the CaO particle surface will immediately increase to a theoretical value of about 12.5 but will, in an open system with large excess of sea water, rapidly fall back to pH 10.5 being equilibrium pH of magnesium hydroxide. This is further reduced to <9 due to the common ion effect of dissolved magnesium in sea water and then be diluted to the sea water background pH, about 8. Field test dosing CaO particles to sea water showed that the pH of water between the particles stayed around 8.

Keywords: ; calcium hydroxide; kinetics; lime; magnesium hydroxide; sea water

1. Introduction

1.1. Applications of Burnt Lime in Sea Water To combat the negative impact of sea urchins on tare forests in northern coastal waters and blue-green algal blooms in the surrounding of fish-farms (monoculture), Brooks et al. [1] and Strand et al. [2] has investigated addition of burnt lime (calcium oxide, CaO) to sea water as a remedial action. The treatment involves a rapid suspension of burnt lime in sea water and immediate discharge into the marine environment. Hence, it becomes of paramount importance to know how CaO particles behaves in contact with sea water and how this substance modifies the water chemistry. The aim of this treatment is to increase tare growth and population . This would provide a habitat for many species and will increase the biodiversity of the marine environment. On a longer term one can imagine harvesting of high-quality sea urchins for the food market as well as a preparedness plan to treat blue-green algal blooms for the same market. In a wider

Materials 2020, 13, 4926; doi:10.3390/ma13214926 www.mdpi.com/journal/materials Materials 2020, 13, 4926 2 of 24

perspective lime can be used to capture CO2 in sea water as an environmental remediation [3,4]. However, designing an efficient treatment strategy to tackle the variable conditions expected in the sea requires increased understanding of the reaction kinetics of CaO in sea water, and this paper is an effort to elucidate this in comparison to fresh water as a novelty. The burnt lime treatment is usually performed by spraying a suspension of CaO particles directly at the surface of the sea and all these particles will interfere with each other in their microenvironment until the gradual dilution is so great that they can be considered as single particles. For instance, the treatment in practice consists of pumping a slurry of 100 kg burnt lime (0.2–0.8 mm diameter) dispersed in 600 litres of sea water onto the surface. This provides an initial CaO concentration in the slurry of 167 g/L, that is taken as a starting point. Assuming that the average particle diameter is 4 3 3 0.5 mm and that they are spherical, the volume of the particle will be 3 πr or 0.0654 mm . Light-burnt 3 lime (burnt at 1060 ◦C) has a typical particle density [5] of ρp = 2 g/cm . The particle is porous since 3 CaO has solid density ρs = 3.35 g/cm ; porosity = (1 ρp/ρs) 100 vol% = 40 vol%, which is also what − × Wuhrer [6] found. This gives for a 167 g/L slurry an outer volume of light-burnt lime of 83.5 cm3 or in other words about 1275 particles per mL. These particles will however immediately react with water as discussed in the next Section about reaction with fresh water. The reaction with fresh water is evaluated first in order to illustrate the difference to sea water in the Section thereafter.

1.2. Chemistry of Burnt Lime (CaO) in Fresh Water Burnt lime reacts violently with water to produce calcium hydroxide (so called slaked lime) according to the reaction below: CaO + H2O = Ca(OH)2 (1) The reaction in Equation (1) releases a heat of Q = 1160 J/g CaO. The heat capacity of water is c = 4.18 kJ/kg K. The increase in temperature of a mass (m) of water due to the reaction in Equation (1) · can be calculated by the formula in Equation (2);

∆T = Q/c m (2) × Using the same concentration of particles as in the preceding Section (167 g/L) and a density of water of 1000 g/L, one find that the water will be heated ∆T = 167 g 1.160 kJ/g/(4.18 kJ/kg K 1 kg) × · × = 46 K = 46 ◦C. This temperature increase is for a closed system (adiabatic, without heat loss) and 100% hydration of the burnt lime particles and will be proportional less for decreasing fractions of reaction. The temperature increase as a function of burnt lime dosage and fraction of hydration is given in Table1 as an illustration.

Table 1. Temperature increase (∆T; ◦C) in water as function of dosage of burnt lime and its hydration fraction as calculated from Equations (1) and (2).

Dosage /Hydration Fraction 200 g/L 150 g/L 100 g/L 50 g/L → ↓ 0.2 11 8 6 3 0.4 22 17 11 6 0.6 33 25 17 8 0.8 44 33 22 11 1.0 56 42 28 14

A light-burnt lime particle is not massive, but porous as shown in Figure1[7]. When the lime particle meets water, this will be sucked into the pores and react immediately to calcium hydroxide while the pore water temperature will increase rapidly to above the boiling point since there is a small amount of water relative to the CaO of the pore wall. If the precipitated calcium hydroxide has blocked the pore (Equation (1) doubles the volume of solid matter) the particle Materials 2020, 13, 4926 3 of 24 willMaterials disintegrate 2020, 13, x FOR due PEER to theREVIEW thermal expansion of the remaining water and/or its vapor pressure.3 of 24 The reaction can be described as chaotic and will not be less chaotic when components from sea water reaction can be described as chaotic and will not be less chaotic when components from sea water interfere as described in next Section. Calcium hydroxide has limited and the solubility interfere as described in next Section. Calcium hydroxide has limited solubility and the solubility actually decreases with increasing temperature as shown in Equation (3) (calculated from tabular actually decreases with increasing temperature as shown in Equation (3) (calculated from tabular data [8]), an anomaly compared to most other compounds. The reaction rate of CaO, on the other hand, data [8]), an anomaly compared to most other compounds. The reaction rate of CaO, on the other increases with increasing temperature and the degree of reaction for a light-burnt lime particle may hand, increases with increasing temperature and the degree of reaction for a light-burnt lime particle reach 80% after 1 h under isothermal conditions, but after <5 min under adiabatic conditions (time to may reach 80% after 1 h under isothermal conditions, but after <5 min under adiabatic conditions reach 60 C according to EN459-2) depending on the fineness of the burnt lime. (time to reach◦ 60 °C according to EN459-2) depending on the fineness of the burnt lime.

Figure 1. AnAn electron electron microscopy image of the pore walls of an unhydrated light-burnt lime particle [[7].7].

Solubility of Ca(OH)2 (g/L) = −0.0117 × T (°C) + 1.924 with r2 = 0.9954 (3) Solubility of Ca(OH) (g/L) = 0.0117 T(◦C) + 1.924 with r = 0.9954 (3) 2 − × In order to calculate the pH of burnt lime slaked in distilled water, one can use the solubility of In order to calculate the pH of burnt lime slaked in distilled water, one can use the solubility of 1.63 g/L 1.63 g/L at 25 °C (from Equation (3)) knowing that calcium hydroxide dissociates as written in at 25 ◦C (from Equation (3)) knowing that calcium hydroxide dissociates as written in Equation (4); Equation (4); 2+ 2+ 2 Ca(OH) (s) = Ca 2++ 2 OH−− with solubility product Ksp = [Ca2+ ] [OH− −2] (4) Ca(OH)2 2 (s) = Ca + 2 OH with solubility product Ksp = [Ca ] ×× [OH ] (4) where the concentrationsconcentrations of in brackets are expressed in molmol/L./L. According to Equation (4) the concentration of hydroxyl ionsions isis twicetwice thatthat ofof thethe dissolveddissolved calciumcalcium hydroxide;hydroxide; 2+ [OH−]− = 2 × [Ca 2+] = 2 × [Ca(OH)2 ](5) (5) [OH ] = 2× [Ca ] = 2 × [Ca(OH) ] since Mw (Ca(OH) ) = 74.08 g/mol, [OH− ] = 2 1.63 g/L/74.08 g/mol = 0.04405 mol/L since Mw (Ca(OH)22) = 74.08 g/mol, [OH ]− = 2·1.63× g/L/74.08 g/mol = 0.04405 mol/L The relation between [OH[OH−−]] and and pH pH at at 25 25 °C◦C is; is; + pH = log [H +] = 14.00 + log [OH−−] (6) pH =− −log [H ] = 14.00 + log [OH ] (6) since since + + 14 H2O H + OH− with dissociation constant Kd = [H ] [OH−] = 1.0 10− at 25 ◦C (7) H↔2O ↔ H+ + OH− with dissociation constant Kd = [H+]× × [OH−] = 1.0·10× −14 at 25 °C (7)

++ + + − + + −14 14 pH = −loglog [H [H] ]== 7 7for for neutral neutral water water since since [H [H] =] [OH= [OH] and−] and [H [H] = √](1.0·10= √(1.0 ). 10− ). − − × [OH−]] = 0.044050.04405 MM forfor dissolved dissolved calcium calcium hydroxide hydroxide gives gives then then pH =pH12.6 = 12.6 for a for solubility a solubility of 1.63 of g /1.63L at 25g/L◦C. at 25 °C. pH will be a bit lower than this as dissolved calcium hydroxide is not completely dissociated and pH will be a bit lower than this as dissolved calcium hydroxide is not completely dissociated species like CaOH+ exists; and species like CaOH+ exists; h +i Ca(OH) 2+ + Ca + OH− Ca(OH) with solubility product K = (8) 2+ − + h 2+i Ca + OH↔ ↔ Ca(OH) with solubility product K = Ca [OH −] (8) [ ] × ×[ ] The geochemical data program GEMS (gems.web.psi.ch) [9,10] with PSI-GEMS thermodynamic data for aqueous species and solids [11] minimizes Gibbs free energy and gives the following composition of a saturated solution of calcium hydroxide at 25 °C:

Materials 2020, 13, 4926 4 of 24

The geochemical data program GEMS (gems.web.psi.ch) [9,10] with PSI-GEMS thermodynamic data base for aqueous species and solids [11] minimizes Gibbs free energy and gives the following Materialscomposition 2020, 13 of, xa FOR saturated PEER REVIEW solution of calcium hydroxide at 25 ◦C: 4 of 24

+2 [Ca[Ca+2] = ]0.016370= 0.016370 M or M 0.65611 or 0.65611 g/L g /L [CaOH+] = 0.004265 M or 0.24345 g/L [CaOH+] = 0.004265 M or 0.24345 g/L [OH−] = 0.037006 M or 0.62910 g/L [OH−] = 0.037006 M or 0.62910 g/L pH = 12.5 pH = 12.5 This gives an amount of dissolved calcium hydroxide of 1.53 g/L instead of the 1.63 g/L obtained This gives an amount of dissolved calcium hydroxide of 1.53 g/L instead of the 1.63 g/L obtained through Equation (3) from [8], which explains the difference in calculated pH together with the presence through Equation (3) from [8], which explains the difference in calculated pH together with the of the species CaOH+. Due to this species the pH as measured by a pH-meter will be lower (12.5) presence of the species CaOH+. Due to this species the pH as measured by a pH-meter will be lower than the corresponding amount of dissolved calcium hydroxide; pH = 12.6 for 1.63 g/L dissolved (12.5) than the corresponding amount of dissolved calcium hydroxide; pH = 12.6 for 1.63 g/L Ca(OH)2, which would have been the result if the solution was titrated rather than measured by a dissolved Ca(OH)2, which would have been the result if the solution was titrated rather than pH-meter. The proceeding evaluations will nevertheless be based on the solubility from Equation (3) measured by a pH-meter. The proceeding evaluations will nevertheless be based on the solubility (1.63 g/L resulting in pH = 12.6) and ignore the presence of species like CaOH+ since it will react with + from Equation (3) (1.63 g/L resulting in pH = 12.6) and ignore2 +the presence of species like CaOH since the different species in sea water as if it was separated as Ca and OH−. it will react with the different species in sea water as if it was separated as Ca2+ and OH−. Another complicating factor for correct pH is that the self-dissociation of water in Equation Another complicating factor for correct pH is that the self-dissociation of water in Equation (7) (7) increases with increasing temperature as temperature is just a measure of molecular movement. increases with increasing temperature as temperature is just a measure of molecular movement. The The pH reference value of 14 will then increase with decreasing temperature, or pH of neutral water pH reference value of 14 will then increase with decreasing temperature, or pH of neutral water will will decrease with increasing temperature as plotted in Figure2. Neutral pH at 5 and 15 ◦C would be decrease with increasing temperature as plotted in Figure 2. Neutral pH at 5 and 15 °C would be 7.4 7.4 and 7.2, respectively, which corresponds to reference values of 14.8 and 14.4 (rather than 14.0 in and 7.2, respectively, which corresponds to reference values of 14.8 and 14.4 (rather than 14.0 in Equation (7)) unless the pH-meter is correcting for temperature. Equation (7)) unless the pH-meter is correcting for temperature.

Figure 2. pH for neutral water as a function of temperature (t) in the range 0–50 ◦C plotted from tabular Figure 2. pH for neutral water as a function of temperature (t) in 5the range2 0–50 °C plotted from data [12]. The regression line follows formula neutral pH = 8 10− t 0.0208 t + 7.4692 with tabular data [12]. The regression line follows formula neutral ×pH = 8 ×× 10−−5 × t2 − 0.0208× × t + 7.4692 regression factor r2 = 0.9999. with regression factor r2 = 0.9999. The pH in a suspension of calcium hydroxide will further increase with decreasing temperature sinceThe the pH solubility in a suspension of calcium of hydroxide calcium hydroxide increases will as shown further by increase Equation with (3). decreasing temperature sinceConsider the solubility distilled of calcium water equilibratinghydroxide increases with natural as shown air containing by Equation 400 (3). ppm or 0.04 vol% carbon Consider distilled water equilibrating with natural air containing 400 ppm or 0.04 vol% . The CO2 can partly be dissolved as the molecule or react with water to hydrogen carbonate (sodioxide. called The carbonic CO2 can ): partly be dissolved as the molecule or react with water to hydrogen carbonate (so called carbonic acid): CO2 + H2O H2CO3 CO2 + H2O ↔↔ H2CO3 (9) [H2CO3] 3 with hydration product Kh = = 1.7 10− at 25 ◦C [CO2 (aq)] × (9) [] with hydration product K = =1.7 × 10 at 25 °C [ ]

The main component of dissolved CO2 will be molecular CO2 (aq), since Kh is so small that the concentration of H2CO3 would be negligible and do not influence pH (i.e., log [H+]). The total solubility of CO2 at 100 kPa CO2 pressure (about 1 atm) and 25 °C is 1.45 g/L. Hydrogen carbonate dissociates according to 2 sequential steps:

Materials 2020, 13, 4926 5 of 24

The main component of dissolved CO2 will be molecular CO2 (aq), since Kh is so small that + the concentration of H2CO3 would be negligible and do not influence pH (i.e., log [H ]). The total solubility of CO2 at 100 kPa CO2 pressure (about 1 atm) and 25 ◦C is 1.45 g/L. Hydrogen carbonate dissociates according to 2 sequential steps: + H CO H + HCO− 2 3 ↔ 3 [ ] + (10) HCO3− [H ] 4 with dissociation constant Ka1 = × = 2.5 10− at 25 ◦C [H2CO3] × + 2 HCO− H + CO − 3 ↔ 3 [CO 2 ] [H+] (11) 3 − × 11 with dissociation constant Ka2 = [ ] = 4.69 10− at 25 ◦C HCO3− × The distribution of the different species for CO2 dissolved in water as a function of pH is shown in Figure3 based on calculations using the equilibrium constants in Equations (9)–(11). The solubility of CO2 is proportional to the partial pressure of carbon dioxide as shown in Equation (12) and increases with decreasing temperature as plotted in Figure4. [CO ] = k P(CO ) (12) 2 × 2 For two different pressures Equation (12) can be reformulated to

[CO2]1 [CO2]2 P2(CO2) k = = and for [CO2]2 = [CO2]1 (13) P1(CO2) P2(CO2) P1(CO2) ×

Figure4 shows that at 15 ◦C and P1(CO2) = 1 atm the concentration of carbon dioxide is [CO2]1 = 2 g/L or 45.45 mM. If k is constant over the whole pressure range, the solubility at P2(CO2) = 0.04 atm will be [CO2]2 = 0.08 g/L or 1.82 mM. When excess calcium hydroxide with pH = 12.6 interact with fresh water with dissolved CO2 the equilibria in Equations (9)–(11) will rapidly be displaced to the right and calcium carbonate will precipitate in an amount corresponding to the amount of dissolved CO2 regardless of species:

Ca(OH)2 + CO2 (aq) = CaCO3 (s) + H2O (14) If the amount of calcium hydroxide is far above saturation the pH will be unchanged at the same time as calcium carbonate is precipitating since the solubility of calcium carbonate is very small ( 0.014 g/L). The calcium concentration for saturated calcium hydroxide is about 22 mM [Ca2+] ≈ and much higher than for saturated calcium carbonate having 0.14 mM [Ca2+], a difference of about 150 times. This means that for thinning of an exactly saturated calcium hydroxide solution, the pH is controlled by dissolved CO2: pH = 14.00 + log {2 ([Ca(OH) ] [CO ])} (15) × 2 − 2 2 In natural fresh water with dissolved carbonate (CO3 −), alkali hydrogen carbonate (HCO3−) and magnesium (Mg2+) dissolved calcium hydroxide will react as follows:

Ca(OH)2 + HCO3− = CaCO3 (s) + H2O + OH− (16)

2 Ca(OH)2 + CO3 − = CaCO3 (s) + 2 OH− (17) 2+ 2+ Ca(OH)2 + Mg = Mg(OH)2 (s) + Ca (18)

The reactions in Equations (16)–(18) leads to precipitates, since the at 25 ◦C for calcite, CaCO3, is 0.014 g/L or 0.14 mM, and for brucite, Mg(OH)2, 0.009 g/L or 0.15 mM. Calcite and brucite are considerably less soluble than (mineral name for calcium hydroxide), Ca(OH)2, with 1.63 g/L or 22 mM (a factor of about 150 in molarity). Materials 2020, 13, 4926 6 of 24 Materials 2020, 13, x FOR PEER REVIEW 6 of 24 Materials 2020, 13, x FOR PEER REVIEW 6 of 24 1 01 -10 -2-1 -3-2 H2CO3/Ct -3 -4 H2CO3/Ct CO2(aq)/Ct -5-4 CO2(aq)/Ct -6-5 HCO3-/Ct HCO3-/Ct -7-6 CO32-/Ct -8-7 CO32-/Ct Log 10 of relative concentration -9-8 Log 10 of relative concentration -9 4 6 8 10 12 4 6pH 8 10 12 pH

Figure 3. Distribution of solute species versus pH for a closed aqueous carbonate system at 25 °C and zeroFigure ionic 3. Distribution strength plotted of solute based species on their versus equilibrium pH for aconstants closed aqueous [13]. carbonate system atat 25 ◦°CC and zero ionic strength plottedplotted basedbased onon theirtheir equilibriumequilibrium constantsconstants [[13].13].

Figure 4. Dissolved CO2 versus temperature (◦C) for carbon dioxide pressure P (CO2) = 1 atm plotted Figure 4. Dissolved CO2 versus temperature (°C) for carbon dioxide pressure P (CO2) = 1 atm plotted from tabular data [14]. fromFigure tabular 4. Dissolved data [14]. CO 2 versus temperature (°C) for carbon dioxide pressure P (CO2) = 1 atm plotted Forfrom fresh tabular water datasaturated [14]. with respect to calcium hydroxide, the reactions in Equations (16) and In natural fresh water with dissolved alkali carbonate (CO32−), alkali hydrogen carbonate (17) will lead to a pH increase since soluble alkali will form2− if the water contains alkali (HCOIn3− ) naturaland magnesium fresh water (Mg 2+with) dissolved dissolved calcium alkali hydroxide carbonate will (CO react3 ), as alkali follows: hydrogen carbonate or alkali hydrogen carbonates. The reaction in Equation (18) will lead to a weak reduction (HCO3−) and magnesium (Mg2+) dissolved calcium hydroxide will react as follows: in pH since it will increase theCa(OH) concentration2 + HCO of3− calcium= CaCO3 ions (s) + that H2O in + turn OH− will suppress the solubility(16) of calcium hydroxide due to theCa(OH) common2 + HCO eff3ect− = accordingCaCO3 (s) to+ H Equation2O + OH− (4). (16) For fresh water undersaturatedCa(OH) with2 + respectCO32− = toCaCO calcium3 (s) + hydroxide, 2 OH− reaction 16 will reduce(17) pH, 2− − reaction 17 not lead to any changeCa(OH) and reaction2 + CO3 18 = reduceCaCO3 pH(s) + if 2 brucite OH is precipitated. (17) Ca(OH)2 + Mg2+ = Mg(OH)2 (s) + Ca2+ (18) On long term, all calcium hydroxide in fresh water will be converted to calcium carbonate as Ca(OH)2 + Mg2+ = Mg(OH)2 (s) + Ca2+ (18) thereThe is an reactions unlimited in Equations supply of CO(16)–(18)2 relative leads to to calcium precipitates, hydroxide since while the solubilities the rate is at dependent 25 °C for calcite, on the CaCOconcentrationThe3, is reactions 0.014 of g/L CO in 2or. Equations Pure0.14 mM, fresh (16)–(18) and water for without brucite, leads to dissolved Mg(OH) precipitates, CO2, 0.0092 sincein equilibriumg/L the or solubilities0.15 mM. with Calcite calcite at 25 °C (CaCOand for brucitecalcite,3) has arepHCaCO considerably 9.93, [ 15is ]0.014 but willg/L less beor soluble somewhat0.14 mM, than and lowerportlandite forif brucite, there (mineral is Mg(OH) dissolved name2, 0.009 CO for2 presentg/Lcalcium or 0.15 as hydroxide), well. mM. BasedCalcite Ca(OH) on and the brucite2 calcite, with 1.63solubilityare considerablyg/L or of 22 0.14 mM mMless (a factor atsoluble 25 ◦ Cof thethanabout pH portlandite 150 should in molarity). be (mineral about 10, name but there for calcium is some uncertaintyhydroxide), in Ca(OH) the solubility.2, with 1.63 g/LFor freshor 22 mMwater (a saturated factor of aboutwith respect 150 in molarity).to calcium hydroxide, the reactions in Equations (16) and (17) willFor freshlead to water a pH saturated increase withsince respect soluble to alkali calcium hydroxides hydroxide, will the form reactions if the waterin Equations contains (16) alkali and carbonates(17) will lead or alkalito a pH hydrogen increase carbonates. since soluble The alkali reaction hydroxides in Equation will (18) form will if thelead water to a weak contains reduction alkali carbonates or alkali hydrogen carbonates. The reaction in Equation (18) will lead to a weak reduction

Materials 2020, 13, 4926 7 of 24

1.3. Chemistry of Burnt Lime (CaO) in Sea Water Atlantic sea water has typical composition listed in Table2 and the sum of components gives a mass of 3.513%, which is a typical salinity. The sum of positive charge is 0.623 M, while the sum of negative charge is 0.622 M, confirming electroneutrality and that all major ions are accounted for. Since sea water consists of a mixture of dissolved salts, one cation does not belong to a specific anion to make a compound.

Table 2. Typical composition of Atlantic sea water.

Species Mw (g/mol) Mass (%) M (mol/L) Na+ 22.990 1.08 0.4825

Cl− 35.453 1.94 0.5620 Mg2+ 24.305 0.13 0.0549 2 SO4 − 96.063 0.27 0.0289 Ca2+ 40.078 0.04 0.0103 K+ 39.098 0.04 0.0105

Br− 79.904 0.0067 0.0009

HCO3− 61.017 0.005 0.0008 2 CO3 − 60.009 0.001 0.0002

+ The greatest difference between sea water and fresh water is the content of Na and Cl−. Sodium chloride is not expected to form compounds with calcium hydroxide, but according to Duschesne and Reardon [16] the solubility of Ca(OH)2 increases with increasing concentration of NaCl due to ion pair formations in their calculations, which were verified experimentally by Johnston and Grove [17] and by Yeatts and Marshall [18]. If a simplified version of sea water is 0.5 M NaCl, the solubility of Ca(OH)2 will increase from 22 to 28 mM. On the other hand, a number of other ions in sea water will interact with Ca(OH)2 and disturb this picture. The second biggest difference is the content of sulphate and much higher concentration of magnesium. The magnesium will react as described in Equation (18), while the sulphate ions will form gypsum: 2 Ca(OH) + SO − +2 H O = CaSO 2H O + 2 OH− (19) 2 4 2 4· 2 The magnesium reaction (Equation (18)) release calcium ions that suppress hydroxide concentration and gypsum formation (Equation (19)) will release hydroxyl ions that potentially reduce the calcium concentration locally relative to solid calcium hydroxide (Equation (4)), so the overall outcome is complex depending on concentration. According to Table2 the molarity of magnesium is 0.0549 M in Atlantic sea water as compared to 0.0289 M for sulphate, so in sea water the magnesium reaction will dominate. The potential compounds formed by interaction of calcium hydroxide and sea water as listed in Table3 are predicted based on possible ion pairs and solubility data [19]. The magnesium compounds with higher molar solubility than brucite, Mg(OH)2, are not expected to form on its expense. An experimental program was then initiated to measure which compounds that are actually formed when burnt lime is added to sea water. Materials 2020, 13, x FOR PEER REVIEW 8 of 24

The magnesium compounds with higher molar solubility than brucite, Mg(OH)2, are not expected to form on its expense. An experimental program was then initiated to measure which compounds that are actually formed when burnt lime is added to sea water.

Table 3. Potential compounds that can form by interaction of CaO with sea water.

Compounds Mw (g/mol) Solubility (g/L) Concentration (mM) Ca(OH)2 74.09 1.77 23.89 CaCO3 a 100.09 0.014 0.14 Mg(OH)2 58.32 0.009 0.15 MgCO3 b 84.31 0.106 1.26 CaSO4·2H2O 172.17 2.5 14.52 Materials 2020, 13, 4926 8 of 24 MgCO3·Mg(OH)2·3H2O c 196.68 - - 3MgCO3·Mg(OH)2·3H2O b 365.31 0.4 1.09 b TableMgCO 3. Potential3·5H2O compounds that174.39 can form by interaction1.8 of CaO with 10.32 sea water. MgCO3·3H2O b 138.36 1.6 11.56 a Can beCompounds found as the polymorphs calcite, Mw (g aragonit/mol)e and vaterite Solubility which (g can/L) be differentiated Concentration by XRD. (mM) b Not expected to form as the molar solubility is greater than for Mg(OH)2. c Solubility not known, but Ca(OH)2 74.09 1.77 23.89 the molar solubility is expected to be greater than for Mg(OH)2 and it is not likely to form. a CaCO3 100.09 0.014 0.14

2. Materials Mg(OH)and Methods2 58.32 0.009 0.15 MgCO b 84.31 0.106 1.26 2.1. Materials 3 CaSO 2H O 172.17 2.5 14.52 The burnt lime4· (CaO)2 used in all tests is of industrial grade from Miljøkalk AS, Verdal, Norway. MgCO Mg(OH) 3H O c 196.68 - - There are two3· types denoted2· 2 “Fine, burnt lime” and “Coarse, burnt lime” and the physical difference b can 3MgCObe observed3 Mg(OH) directly2 3H in2 FigureO 5, while365.31 the sieving curves are 0.4 shown in Figure 6. The 1.09 sieving was performed in· accordance· with EN 933-1. Other properties of the burnt lime samples are given in Table MgCO 5H O b 174.39 1.8 10.32 4. 3· 2 MgCO 3H O b 138.36 1.6 11.56 3· 2 a Can be found as the polymorphsTable calcite, 4. Properties aragonite and of fine vaterite and whichcoarse canburnt be dilime.fferentiated by XRD. b Not expected c to form as the molar solubility is greater than for Mg(OH)2. Solubility not known, but the molar solubility is Property Method Fine CaO Coarse CaO expected to be greater than for Mg(OH)2 and it is not likely to form. Specific surface (m2/g) BET 1.53 1.12 2. Materials and MethodsActive CaO (%) ASTM C25-11 E1 94 91 CO2 residue (%) ASTM C25-11 1.6 2.7 2.1. Materials Reactivity in fresh water - - Time to reach 60 °C (s) EN 459-2 100 250 The burnt limeSemi-adiabatic (CaO) used inΔT all (°C) tests is of industrial grade 75 from Miljøkalk 75 AS, Verdal, Norway. There are two typesReactivity denoted in “Fine, sea water burnt lime” and “Coarse, burnt- lime” and- the physical difference can be observed directlyTime to in reach Figure 605 °C, while (s) the sievingEN 459-2 curves are 300 shown in Figure 500 6. The sieving was performed in accordanceSemi-adiabatic with EN Δ 933-1.T (°C) Other properties of the burnt 70 lime samples 70 are given in Table4.

FigureFigure 5. Physical 5. Physical appearance appearance of of finefine CaOCaO to the the left left and and coarse coarse CaO CaO to the to theright. right.

Table 4. Properties of fine and coarse burnt lime.

Property Method Fine CaO Coarse CaO Specific surface (m2/g) BET 1.53 1.12 Active CaO (%) ASTM C25-11 E1 94 91

CO2 residue (%) ASTM C25-11 1.6 2.7 Reactivity in fresh water -- Time to reach 60 ◦C (s)EN 459-2 100 250 Semi-adiabatic ∆T(◦C) 75 75 Reactivity in sea water -- Time to reach 60 ◦C (s)EN 459-2 300 500 Semi-adiabatic ∆T(◦C) 70 70 Materials 2020, 13, 4926 9 of 24 Materials 2020, 13, x FOR PEER REVIEW 9 of 24

Figure 6. SievingSieving curves curves for for fine fine and and coarse CaO (burnt lime).

2.2. Equipment Equipment

2.2.1. Characterization Tools for the Solid Phase 2.2.1. Characterization Tools for the Solid Phase Isothermal calorimetry of pastes was performed with a TAM (TA Instruments, New Castle, Isothermal calorimetry of cement pastes was performed with a TAM (TA Instruments, New DE, USA) Air instrument set to 15 C in closed glass vials. The glass vials were coupled to an external Castle, DE, USA) Air instrument set◦ to 15 °C in closed glass vials. The glass vials were coupled to an mixing device with 4 syringes with 1 mL capacity each for injection of either fresh or sea water. 70 mg of external mixing device with 4 syringes with 1 mL capacity each for injection of either fresh or sea coarse or fine fraction of CaO was weighed into the glass vial, the syringes filled with 4 mL fresh or sea water. 70 mg of coarse or fine fraction of CaO was weighed into the glass vial, the syringes filled with water and the whole arrangement closed in the isothermal calorimeter for thermal equilibration verified 4 mL fresh or sea water and the whole arrangement closed in the isothermal calorimeter for thermal by zero heat flow. The stirring motor was the only part outside the calorimeter. Each experiment was equilibration verified by zero heat flow. The stirring motor was the only part outside the calorimeter. repeated 3 times and the heat flow curves used was an average of 3. A dosage of 70 mg CaO/4 mL Each experiment was repeated 3 times and the heat flow curves used was an average of 3. A dosage corresponds to 17.5 g/L. of 70 mg CaO/4 mL corresponds to 17.5 g/L. Thermogravimetric analysis (TGA) was performed with a Mettler Toledo TGA/SDTA 851 Thermogravimetric analysis (TGA) was performed with a Mettler Toledo TGA/SDTA 851 (Greifensee, Switzerland). Samples were analysed with a heating rate of 10 C/min between 40–900 C. (Greifensee, Switzerland). Samples were analysed with a heating rate of 10◦ °C/min between 40–900◦ All measurements were performed in nitrogen atmosphere with a flow rate of 50 mL/min. The calcium °C. All measurements were performed in nitrogen atmosphere with a flow rate of 50 mL/min. The hydroxide content in a hydrated CaO sample was calculated from the mass loss between 400–550 C calcium hydroxide content in a hydrated CaO sample was calculated from the mass loss≈ between◦ ≈ caused by its dehydration. The corresponding temperature range for thermal dehydration of 400–550 °C caused by its dehydration. The corresponding temperature range for thermal dehydration magnesium hydroxide is 300–400 C and the decarbonation of precipitated calcium carbonate for of magnesium hydroxide is 300–400◦ °C and the decarbonation of precipitated calcium carbonate for temperatures >600 C. The exact boundaries were chosen from the peak in the DTG curve. temperatures >600 °C.◦ The exact boundaries were chosen from the peak in the DTG curve. The mineralogy was detected using a Bruker D8-Advance with a generator of x-ray The mineralogy was detected using a Bruker D8-Advance with a generator of x-ray KRISTALLOFLEX K 760-80F (power: 3000 W, voltage: 20–60 KV and current: 5–80 mA) (Bruker KRISTALLOFLEX K 760-80F (power: 3000 W, voltage: 20–60 KV and current: 5–80 mA) (Bruker D8- D8-Advance, Billerica, MA, USA) with a tube of RX with copper anode was employed to record the Advance, Billerica, MA, USA) with a tube of RX with copper anode was employed to record the diffractograms of the CaO in the kinetic experiments. All the measurements were taken at 25 C from diffractograms of the CaO in the kinetic experiments. All the measurements were taken at 25 °C◦ from 10 to 70 2θ with a step size of 0.05 2θ and 3 s time per step. 10 to 70 2θ with a step size of 0.05°◦ 2θ and 3 s time per step. The morphology and the local chemical composition of the materials was explored in the The morphology and the local chemical composition of the materials was explored in the time- time-course experiments using by Scanning Electron Microscopy (SEM) coupled to Energy Dispersive course experiments using by Scanning Electron Microscopy (SEM) coupled to Energy Dispersive X- X-Ray Analysis (EDX). For these analyses a Hitachi S-3000 N Electron Microscope (Tokyo, Japan) Ray Analysis (EDX). For these analyses a Hitachi S-3000 N Electron Microscope (Tokyo, Japan) coupled to an EDX Bruker Esprit 1.8 unit (Billerica, MA, USA) was employed. 2.2.2. Analysis in the Liquid Phase 2.2.2. Analysis in the Liquid Phase Fine CaO was added to natural sea water sampled from the Trondheim Fjord, city, Norway, Fine CaO was added to natural sea water sampled from the Trondheim Fjord, city, Norway, at at temperatures 5 or 15 C in a surplus dosage of 10 g CaO/L during stirring. The experiments were temperatures 5 or 15 °C◦ in a surplus dosage of 10 g CaO/L during stirring. The experiments were performed in a container open to the atmosphere and the temperature of the surrounding environment performed in a container open to the atmosphere and the temperature of the surrounding was kept constant either by immersion in an ice-water bath (5 C) or kept in a thermostatic room set environment was kept constant either by immersion in an ◦ice-water bath (5 °C) or kept in a thermostatic room set at 15 °C. Before starting the experiments, the water was pre-adjusted to the required temperature by a thermostat.

Materials 2020, 13, 4926 10 of 24

at 15 ◦C. Before starting the experiments, the water was pre-adjusted to the required temperature by a thermostat. The ion composition of the applied sea water is listed in Table5. Samples of 400 µL was ≈ drawn from this suspension by a syringe at the times 0 (sea water before addition of CaO), 1, 3, 10, 15 and 30 min. Thereafter solution was drawn every hour until 8 h and a final sampling after 24 h. The sampled suspension was filtered through 0.45 µm syringe-filter and 100 µL of the filtrate was diluted to 10 mL using MilliQ water and analysed by Ion Chromatography (Dionex®, Sunnyvale, CA, USA) for F , Cl , Br , SO 2 , CO 2 (given as µS min by the instrument), Na+, Mg2+,K+ and Ca2+. − − − 4 − 3 − · The analytical system was based on a double channel so both cations and anions were analysed under the same run. A Dionex IonPac CS16 column was used for detection of cations while a column Dionex AS18 coupled with a column CG5A was used for measurement of anions. The pH was measured by a pH-meter in a GLP/GMP environment calibrated just before the experiments towards pH = 7.0 and a 10.012 standard, as well as corrected with respect to temperature.

Table 5. Major ion composition and pH of the sea water used.

Species Mw (g/mol) ppm M (mol/L) Na+ 22.990 10,981.1 0.4776

Cl− 35.453 18,553.6 0.5233 Mg2+ 24.305 1301.6 0.0536 2 SO4 − 96.063 2835.3 0.0295 Ca2+ 40.078 399.0 0.0100 K+ 39.098 394.0 0.0101

Br− 79.904 0.05 0.0000

F− 18.998 1.25 0.0001 pH - 7.88 -

2.2.3. Exposure Assays of CaO

Fine CaO was dispersed into sea water in a dosage of 5 g CaO per 500 mL sea water at 5 ◦C under slow stirring, and the suspension was kept under constant stirring between each sampling. Solids from the suspensions was rapid filtered off using filter paper on Büchner funnel attached to a vacuum pump followed by washing with 99% to halt the reaction. The solids were then added to a plastic tube with screw cap together with 15 mL ethanol and shaken for 30 s. This was followed by centrifugation at 1500 rpm and the liquid phase removed. Finally, the samples were freeze dried under vacuum for 24 h before sealed storage prior to further solid-phase analysis. The sampling was performed after 1, 5, 15 and 30 min, as well as after 1, 2, 5, 8, 24 h (1 day) and 2, 3, 4, 5, 6 and 7 days. The solids produced in the time-course transformation experiments were investigated by SEM/EDX for topology and semi-quantitative local chemical composition. X-ray diffraction (XRD) was applied to follow qualitatively how the crystalline phases developed as a function of time, while TGA was used to determine the content of some phases quantitatively.

3. Results

3.1. Influence of Burnt Lime on Sea Water Composition as Function of Time The salinity of the applied sea water in Table5 is 3.448%, which is marginally lower than average Atlantic sea water given in Table2 (3.513%). The sum of positive charge is 0.6149 mol /L while the sum of negative charge is 0.5824 mol/L. The lack of negative charge (∆ = 0.0325 mol/L) is attributed to carbonate and hydrogen carbonate not measured, as well as uncertainty in measurements. Materials 2020, 13, 4926 11 of 24

The dosage of 10 g/L CaO in sea water corresponds to 13.2 g/L Ca(OH)2, which is about eight times higher than its solubility. The result is a white suspension often referred to as «milk of lime». 2+ 2+ 2 2 The evolutions of magnesium (Mg ), calcium (Ca ), sulphate (SO4 −), carbonate (CO3 −) and pH in sea water as a function of time until 8 h after addition of burnt lime are plotted in Figures7–11, respectively. The ion concentrations after 24 h are presented in Table6. Materials 20202020,, 1313,, xx FORFOR PEERPEER REVIEWREVIEW 1111 ofof 2424

Figure 7. Concentration of magnesium in sea water at T = 5 °C (dashed line) and 15 °C (solid line) as FigureFigure 7. Concentration 7. Concentration of of magnesium magnesium in in sea sea waterwater at T == 55 °C◦C (dashed (dashed line) line) and and 15 °C 15 (solid◦C (solid line) line)as as function of time after adding 1% (w/v) fine CaO. functionfunction of time of time after after adding adding 1% 1% (w (w/v/)v fine) fine CaO. CaO.

Figure 8. Concentration of calcium in sea water at T = 5 °C (dashed line) and 15 °C (solid line) as FigureFigure 8. Concentration 8. Concentration of of calcium calcium in in seasea waterwater at T == 55 °C◦C (dashed (dashed line) line) and and 15 15°C ◦(solidC (solid line) line) as as functionfunctionfunction of time ofof timetime after afterafter adding addingadding 1% 1%1% (w ((w/v//)vv) fine) finefine CaO. CaO.CaO.

FigureFigure 9. Concentration 9. Concentration of of sulphate sulphate in in seasea waterwater at T T == 555 °C°C◦C (dashed(dashed (dashed line)line) line) andand and 1515 °C 15°C (solid◦(solidC (solid line)line)line) asas as functionfunctionfunction of time ofof timetime after afterafter adding addingadding 1% 1%1% (w ((w/v//)vv) fine) finefine CaO. CaO.CaO.

Materials 2020, 13, 4926 12 of 24 Materials 2020, 13, x FOR PEER REVIEW 12 of 24

Figure 10. Relative concentration of carbonate in sea water at T = 5 °C (dashed line) and 15 °C (solid FigureFigure 10. Relative10. Relative concentration concentration of of carbonate carbonate inin seasea water at at T T == 5 5°C◦ C(dashed (dashed line) line) and and 15 °C 15 (solid◦C (solid line)line) as function as function oftime of time after after adding adding 1% 1% ( w(w/v/v))fine fine CaO.CaO.

FigureFigure 11. pH11. pH in seain sea water water at at T T= =5 5◦ C°C (dashed (dashed line) and and 15 15 °C◦C (solid (solid line) line) as asfunction function of time of time afterafter addingadding 1% (1%w/v ()w fine/v) fine CaO. CaO. Table 6. Compositional change of sea water at 5 and 15 C the first 24 h after adding 1% w/v fine CaO. Table 6. Compositional change of sea water at 5 and 15 °C◦ the first 24 h after adding 1% w/v fine CaO.

TemperatureTemperature 5 5◦ °CC Temperature Temperature 15 15 °C◦ C SpeciesSpecies TimeTime 0 0 h h Time Time 24 h Time Time 0 0h h Time Time 24 24h h Magnesium, Mg2+ 1302 mg/L 236 mg/L 1265 mg/L 2.1 mg/L Magnesium,Magnesium, Mg Mg2+ 1302 1302 mg mg/L/L 236 mg/L mg/L 1265 1265 mg/L mg/L 2.1 2.1mg/L mg /L Calcium, Ca2+ 399 mg/L 2056 mg/L 386 mg/L 2361 mg/L Calcium, 2Ca+ 399 mg/L 2056 mg/L 386 mg/L 2361 mg/L Calcium, Ca 2− 399 mg/L 2056 mg/L 386 mg/L 2361 mg/L Sulphate, SO42− 2835 mg/L 2662 mg/L 2764 mg/L 2648 mg/L 2 2− Sulphate,Carbonate, SO4 CO− 32− 0.0462835 mgμS·min/L 0.035 2662 μ mgS·min/L 0.045 2764 μS·min mg/L 0.061 2648 μS·min mg/ L 2 Carbonate, CO3 − 0.046 µS min 0.035 µS min 0.045 µS min 0.061 µS min 3.2. Change of Solids Composition as Function· of Time for CaO· in Contact with· Sea Water · 3.2. ChangeThe of result Solids from Composition thermogravimetry, as Function TG of (mass Time loss for CaOas function in Contact of temperature) with Sea Water is plotted in Figure 12 for fine, burnt lime (CaO) as received and in Figure 13 for the solids obtained after 1% CaO has beenThe resultdispersed from in thermogravimetry, sea water for two days. TG (mass The si lossgnal aswith function maximum of temperature) for the derived is plottedcurve (DTG) in Figure at 12 for fine,≈ 400 burnt °C in limeFigure (CaO) 12 is asdue received to the decomposition and in Figure of 13 Ca(OH) for the2 to solids CaO and obtained H2O, while after 1%the CaOsignal has at ≈ been dispersed740 °C in is sea caused water by for decomposition two days. The of signal CaCO with3 to maximumCaO and CO for2.the The derived signal just curve looks (DTG) large at in400 this C in ≈ ◦ Figureaccurate 12 is due measurement to the decomposition due to the sc ofale, Ca(OH) but theto bulk CaO calculation and H O, from while the the curves signal only at calculates740 C is causedto 2 2 ≈ ◦ 2.09% Ca(OH)2 and 3.85% CaCO3 with the remainder being 94.06% CaO. The XRD of the same sample by decomposition2.09% Ca(OH)2 ofand CaCO 3.85%3 toCaCO CaO3 with and the CO remainder2. The signal being just 94.06% looks CaO. large The in XRD thisaccurate of the same measurement sample revealed largely signals from CaO with traces of Ca(OH)2 and CaCO3. It is difficult to avoid that a due to the scale, but the bulk calculation from the curves only calculates to 2.09% Ca(OH)2 and 3.85% CaCO3 with the remainder being 94.06% CaO. The XRD of the same sample revealed largely signals from CaO with traces of Ca(OH)2 and CaCO3. It is difficult to avoid that a material as reactive as fine CaO will pick up some moisture and maybe carbon dioxide from air during handling. Materials 2020, 13, x FOR PEER REVIEW 13 of 24 Materials 2020, 13, x FOR PEER REVIEW 13 of 24 material as reactive as fine CaO will pick up some moisture and maybe carbon dioxide from air material as reactive as fine CaO will pick up some moisture and maybe carbon dioxide from air duringMaterials handling.2020, 13, 4926 13 of 24 during handling.

Figure 12. The thermogravimetry curves (TG) in black and its derivative (DTG) in red for fine CaO as FigureFigure 12.12. The thermogravimetry curves curves (TG) (TG) in in black black and and its its derivative derivative (DTG) (DTG) in inred red for for fine fine CaO CaO as received. asreceived. received.

Figure 13. The thermogravimetry curves (TG) in black and its derivative (DTG) in red for the solids of Figure 13. The thermogravimetry curves (TG) in black and its derivative (DTG) in red for the solids 1%Figure CaO 13. dispersed The thermogravimetry in sea water for curves 2 days. (TG) in black and its derivative (DTG) in red for the solids of 1% CaO dispersed in sea water for 2 days. of 1% CaO dispersed in sea water for 2 days. Similar calculations as for received CaO can be done when it has been reacting with sea water for Similar calculations as for received CaO can be done when it has been reacting with sea water someSimilar time, for calculations instance from as for the received curves after CaO two can days be done reaction when in Figureit has been13. The reacting signals with with sea maxima water for some time, for instance from the curves after two days reaction in Figure 13. The signals with atfor some140, 370, time, 510 for and instance 730 C arefrom due the to curves thermal after decomposition two days reaction of gypsum in Figure (loses two 13. waterThe signals molecules), with maxima≈ at ≈ 140, 370, 510◦ and 730 °C are due to thermal decomposition of gypsum (loses two water magnesiummaxima at ≈ hydroxide 140, 370, 510 (loses and one 730 water °C are molecule), due to therma calciuml decomposition hydroxide (loses of gypsum one water (loses molecule) two water and molecules), magnesium hydroxide (loses one water molecule), calcium hydroxide (loses one water calciummolecules), carbonate magnesium (loses onehydroxide molecule (loses carbon one dioxide). water molecule), From the calcium TG/DTG hydroxide data, the bulk (loses composition one water molecule) and calcium carbonate (loses one molecule carbon dioxide). From the TG/DTG data, the ofmolecule) solids from and reaction calcium of carbonate CaO with (loses sea water one atmolecule any point carbon in time dioxide). can be calculatedFrom the forTG/DTG the five data, phases: the bulk composition of solids from reaction of CaO with sea water at any point in time can be calculated CaSObulk composition4 H2O (gypsum), of solids Mg(OH) from2 reaction(brucite), of Ca(OH) CaO with2 (portlandite), sea water at CaCO any point3 (sum in of time crystal can modificationsbe calculated for the ·five phases: CaSO4⸳H2O (gypsum), Mg(OH)2 (brucite), Ca(OH)2 (portlandite), CaCO3 (sum of calcite,for the aragonitefive phases: and CaSO vaterite),4⸳H2O as (gypsum), well as remaining Mg(OH) CaO2 (brucite), (burnt Ca(OH) lime) from2 (portlandite), the material CaCO mass3 weighed (sum of in, after subtracting the other five phases. The chemical reasoning in Section 1.3 postulate that only Materials 2020,, 13,, 4926x FOR PEER REVIEW 14 of 24 crystal modifications calcite, aragonite and vaterite), as well as remaining CaO (burnt lime) from the these phases will be present, and this is also confirmed by XRD after reacting 1% CaO with sea water material mass weighed in, after subtracting the other five phases. The chemical reasoning in Section for two days. The bulk composition of solids as a function of reaction time for 1% fine CaO in sea 1.3 postulate that only these phases will be present, and this is also confirmed by XRD after reacting water is shown in Table7 for the first day and in Table8 for a week, while the content of remaining 1% CaO with sea water for two days. The bulk composition of solids as a function of reaction time CaO is plotted in Figure 14. The deviation from 100% in Tables7 and8 is due to adsorbed moisture. for 1% fine CaO in sea water is shown in Table 7 for the first day and in Table 8 for a week, while the The results reveal that more than 97% CaO has reacted after one day and as much as 96% after just 5 h. content of remaining CaO is plotted in Figure 14. The deviation from 100% in Tables 7 and 8 is due to adsorbed moisture. The results reveal that more than 97% CaO has reacted after one day and as Table 7. Bulk composition (%) of solids from reaction of 1% fine CaO with sea water versus time at much as 96% after just 5 h. 5 ◦C for the first 24 h.

Table 7.Time Bulk composition 1 min (%) 5 of min solids 15from min reaction 30 min of 1% fine 1 h CaO 2with h sea 5 hwater 8 versus h 24time h at 5 °C forTime the (min)first 24 h. 1 5 15 30 60 120 300 480 1440 GypsumTime 1 0.0 min 5 0.0 min 15 0.0 min 30 0.0 min 1 0.7 h 2 0.0 h 5 8.5h 8 4.3h 24 0.4h

Mg(OH)Time (min)2 9.01 17.75 7.115 14.830 16.160 120 5.0 300 17.0 480 15.2 1440 12.7 Gypsum 0.0 0.0 0.0 0.0 0.7 0.0 8.5 4.3 0.4 Ca(OH)2 50.4 49.0 55.4 66.2 61.3 70.3 62.7 69.9 76.4 Mg(OH)2 9.0 17.7 7.1 14.8 16.1 5.0 17.0 15.2 12.7 CaCO3 7.7 8.2 5.7 8.7 7.0 4.9 6.4 6.0 6.3 Ca(OH)2 50.4 49.0 55.4 66.2 61.3 70.3 62.7 69.9 76.4 CaO 30.1 20.9 30.4 7.6 12.3 18.8 4.2 3.2 2.9 CaCO3 7.7 8.2 5.7 8.7 7.0 4.9 6.4 6.0 6.3 SumCaO 97.330.1 95.820.9 98.630.4 97.37.6 12.3 97.5 18.8 99.0 4.2 98.8 3.2 98.6 2.9 98.7 Sum 97.3 95.8 98.6 97.3 97.5 99.0 98.8 98.6 98.7 Table 8. Bulk composition (%) of solids from reaction of 1% fine CaO with sea water versus time at 5Table◦C for 8. 1–7Bulk days. composition (%) of solids from reaction of 1% fine CaO with sea water versus time at 5 °C for 1–7 days. Time 1 d 2 d 3 d 4 d 5 d 6 d 7 d GypsumTime 0.4 1 3.1d 2 d 8.4 3 d 4 5.3 d 5 d 5.8 6 d 9.57 d 8.2

Mg(OH)2 Gypsum12.7 0.4 16.4 3.1 17.68.4 12.35.3 5.8 21.39.5 16.08.2 14.7 Mg(OH)2 12.7 16.4 17.6 12.3 21.3 16.0 14.7 Ca(OH)2 76.4 72.6 64.6 73.4 62.2 68.8 70.0 Ca(OH)2 76.4 72.6 64.6 73.4 62.2 68.8 70.0 CaCO3 6.3 5.8 8.6 8.5 9.7 5.3 6.4 CaCO3 6.3 5.8 8.6 8.5 9.7 5.3 6.4 CaO 2.9 1.0 0.0 0.0 0.0 0.0 0.0 CaO 2.9 1.0 0.0 0.0 0.0 0.0 0.0 SumSum 98.7 98.7 98.8 98.8 99.299.2 99.4 99.4 99.0 99.0 99.6 99.699.3 99.3

Figure 14. Remaining content of CaO as a function of time after 1% finefine CaO have been mixed in sea water at 5 ◦°C.C.

The reactivity of CaO have also been followed by isothermal calorimetry at 15 ◦°C.C. The average heat flowflow curves (mW(mW/g/g CaO) for three parallel samples at dosage 17.5 g CaOCaO/L/L are compared for finefine and coarse burnt lime at the upper part of FigureFigure 1515,, whilewhile thethe correspondingcorresponding cumulativecumulative heat curves are plotted at the lower part of FigureFigure 1515.. The heat flowflow curves have a bimodal form with a firstfirst immediate rapid reaction rate and a secondary increase in the range 30 min (fine(fine CaO) to 70 min (coarse CaO). The secondary reaction was more pronounced for fine CaO. Note that a point on the

Materials 2020, 13, 4926 15 of 24

Materials 2020, 13, x FOR PEER REVIEW 15 of 24 (coarse CaO). The secondary reaction was more pronounced for fine CaO. Note that a point on the heat flow curve in Figure 15 (upper part) represent the rate of an exothermal reaction while the slope of an heat flow curve in Figure 15 (upper part) represent the rate of an exothermal reaction while the slope increasing curve represents acceleration. For the cumulative heat in Figure 15 (lower part), a value of of an increasing curve represents acceleration. For the cumulative heat in Figure 15 (lower part), a about 928 J/g CaO corresponds to a reaction degree of 80% since the reaction heat for burnt lime in value of about 928 J/g CaO corresponds to a reaction degree≈ of ≈ 80% since the reaction heat for burnt pure water is 1160 J/g CaO, unless there are other exothermal reactions involved like precipitation of lime in pure water is 1160 J/g CaO, unless there are other exothermal reactions involved like brucite. According to the plots in Figure 15, coarse CaO release 928 J/g (80% reaction) after 390 min, precipitation of brucite. According to the plots in Figure 15, coarse CaO release 928 J/g (80% reaction) while fine CaO develop same amount of energy in 226 min. Note that the reaction rate under adiabatic after 390 min, while fine CaO develop same amount of energy in 226 min. Note that the reaction rate conditions (developed heat is conserved and temperature rises as in Tables1 and4) will be much faster under adiabatic conditions (developed heat is conserved and temperature rises as in Tables 1 and 4) than under isothermal condition (i.e., constant temperature) as in Figure 15. will be much faster than under isothermal condition (i.e., constant temperature) as in Figure 15.

Figure 15. The heat flow flow ( top) and cumulative heat (bottom) under isothermalisothermal conditions at 15 ◦°CC when 17.5 g CaO is mixed with 1 litre seasea water.water.

Analogous isothermal experiments as as for sea wa waterter were performed for 17.5 g CaO per litre distilled water at 15 °C.◦C. The heat flow flow curves for fine fine and coarse CaO are plotted at the upper part of Figure 16,16, while the corresponding cumulative heat curves curves are are shown shown at at the the lower lower part part of of Figure Figure 16.16. Evidently the reaction of burnt lime in fresh water is much more intense than in sea water. The time for coarse CaO to reach 928 J/g J/g was 154 min (less than half of that in sea water), while fine fine CaO used 58 min to reach the same heat evolution compared toto 226226 minmin inin seasea water.water.

Materials 2020, 13, 4926 16 of 24 Materials 2020, 13, x FOR PEER REVIEW 16 of 24

Figure 16. TheThe heat heat flow flow ( toptop)) and cumulative heat ( bottom) under isothermal conditions at 15 ◦°CC when 17.5 g CaO is mixed with 1 litre distilled water.

4. Discussion Discussion 4.1. Influence of Burnt Lime on Sea Water Composition as Function of Time 4.1. Influence of Burnt Lime on Sea Water Composition as Function of Time When comparing the evolution of magnesium in Figure7 with that of calcium in Figure8 one notice When comparing the evolution of magnesium in Figure 7 with that of calcium in Figure 8 one that the concentration of magnesium drops while concentration of calcium increases correspondingly. notice that the concentration of magnesium drops while concentration of calcium increases This is in accordance with Equation (18) postulated from a chemical consideration. From Table7 it correspondingly. This is in accordance with Equation (18) postulated from a chemical consideration. is evident that the concentration of magnesium in sea water at 15 C has dropped 1263 mg/L after From Table 7 it is evident that the concentration of magnesium in ◦sea water at 15 °C has dropped 24 h while the concentration of calcium has increased by 1975 mg/L. Considering the atomic mass of 1263 mg/L after 24 h while the concentration of calcium has increased by 1975 mg/L. Considering the Mg being 24.305 g/mol and that of Ca being 40.078 g/mol, this corresponds to a drop of magnesium atomic mass of Mg being 24.305 g/mol and that of Ca being 40.078 g/mol, this corresponds to a drop of 52 mM and an increase in calcium of 49 mM which is equal in light of measurement uncertainty. of magnesium of 52 mM and an increase in calcium of 49 mM which is equal in light of measurement The corresponding numbers at 5 C are a 44 mM drop for magnesium and a 41 mM increase for calcium, uncertainty. The corresponding◦ numbers at 5 °C are a 44 mM drop for magnesium and a 41 mM again about equal. increase for calcium, again about equal. The reason why the ion exchange between calcium and magnesium is faster at 15 C than at 5 C The reason why the ion exchange between calcium and magnesium is faster at 15◦ °C than at◦ 5 the first 8 h (see Figures7 and8) is that it is likely di ffusion controlled and diffusivity is lower at lower °C the first 8 h (see Figures 7 and 8) is that it is likely diffusion controlled and diffusivity is lower at temperatures. The temperature dependence of diffusion coefficients [20] can be expressed by lower temperatures. The temperature dependence of diffusion coefficients [20] can be expressed by ( !) ∆∆EEa 1 1 1 1 D(T) = D(Tref) exp a − (20) D(T) = D(Tref) × exp R Tref − T (20) R Tref T where ΔEa = activation energy for the reaction (assumed 40 kJ/Mol), R = universal gas constant = 8.3144598 J/K·mol, Tref, e.g., 288.15 K (15 °C) versus T = 278.15 K (5 °C) in this case.

Materials 2020, 13, 4926 17 of 24

where ∆Ea = activation energy for the reaction (assumed 40 kJ/Mol), R = universal gas constant = 8.3144598 J/K mol, T , e.g., 288.15 K (15 C) versus T = 278.15 K (5 C) in this case. · ref ◦ ◦ The ratio between the diffusion coefficients at 15 and 5 ◦C is 1.8 using the numbers for Equation (20). The drop in magnesium concentration after 3 h is 948 and 566 mg/L at 15 and 5 ◦C, respectively, giving a ratio of 1.67. Using the numbers for calcium gives a corresponding ratio of 1.86. This is a strong indication that the Ca2+ Mg2+ replacement in the material is diffusion controlled. ↔ The exchange of calcium with magnesium agrees with Nishimuta and Seki [21] studying effects of liming for the improvement of mariculture grounds in a model system. They found that lime transmuted gradually into magnesium hydroxide as a result of the exchange of Ca2+ for Mg2+, and that lime (CaO) increased pH of the sea water and consequently sulfide and ammonium ion were eluted from the bottom mud. The change in sulphate concentration is somewhat chaotic to start with as seen from Figure9. It is possible that precipitated magnesium hydroxide blocks some of the surface of the CaO particles in the start, but after a while the concentration decreases successively until it flattens out until 24 h being the latest measuring point. The solubility of gypsum is 1.82 g/L and 1.96 g/L at 5 and 2 15 ◦C, respectively, which corresponds to 1020 and 1090 mg SO4 −/L. The measured concentrations are far higher than that (2662 and 2648 mg/L) after 24 h and shows that the reaction between sulphate anions and calcium hydroxide to gypsum (Equation (19)) is a relatively slow process. The decay in in the sulphate concentration in the first 24 h is not more than 4–6%. The carbonate concentration drops and levels out on a plateau when 1% fine CaO is added sea water (Figure 10). The value of the plateau cannot be compared to the solubility of calcium carbonate since it is not an absolute value, but a relative value measured as the area under a curve. The change in pH as function of time when 1% fine CaO is added sea water in Figure 11 shows that pH increases rapidly (within 1 min) from the pH of sea water of 7.9 at 5 ◦C and 7.7 at 15 ◦C to 10.6 and 10.1, respectively. The low measured pH relative to calcium hydroxide is because the CaO dosage was so small that Mg2+ was not depleted as seen from Figure7. Hence, the pH is regulated by precipitated magnesium hydroxide. This will also be the case at open sea where magnesium is in excess. The dissolution of magnesium hydroxide can be written as;

2+ 2+ 2 12 Mg(OH) (s) = Mg + 2 OH− with Ksp = [Mg ] [OH−] = 5.61 10− (21) 2 × × 2+ Dispersing magnesium hydroxide in fresh water means that [OH−] = 2 [Mg ] and [OH−] = p3 4 2Ksp = 2.2 10− corresponding to pH = 10.4. From Figure7 it can be seen that after 8 h at 15 ◦C × q 2+ 2+ 6 [Mg ] = 149 g/L = 6.13 mM and [OH ] = Ksp/[Mg ] = 29.8 10 corresponding to pH = 9.5 − × − while it is measured 9.8. This is just serving as an illustration and it is without taking into temperature effect or complex ions etc. At 5 ◦C it seems like situation is far from equilibrium. At open sea where magnesium is in surplus and concentration is 53.6 mM (Table4), the pH at equilibrium would have been reduced to 9.0 at 25 ◦C and lower than that at lower temperatures since solubility of magnesium hydroxide decreases with decreasing temperature (0.009 g/L at 18 ◦C and 0.04 g/L at 100 ◦C[19]) unlike calcium hydroxide (Equation (3)).

4.2. Change of Solids Composition as Function of Time for CaO in Contact with Sea Water The results in Table7 reveal that calcium hydroxide is formed immediately (first minute) and in a relatively large amount when 1% fine CaO is mixed with sea water at 5 ◦C, since only water is needed, and it is in large excess. Magnesium hydroxide and calcium carbonate are also formed relatively quickly even though the amounts of reactants are limited in this “closed system” and is controlled by amount of CaO relative to sea water and the content of magnesium and carbonate in the sea water. In practical applications the system is “open”, and the supply of reactants are unlimited. The availability of reactants then relies on a combination of current/convection and diffusion of species. It is noticeable from Table7 that the formation of gypsum occurs slower than the other compounds and appears first after 5 h. The presence at this time is confirmed by XRD and can be seen as prismatic Materials 2020, 13, x FOR PEER REVIEW 18 of 24

in Figures 15 and 16. Later calcium hydroxides crystallizes to hexagonal plates in line with its that penetrates through the precipitated mass of magnesium hydroxide. Examples of this are shown in Figure 17. Images (a) and (b) in Figure 17 depicts seemingly amorphous precipitations (likely magnesium Materialshydroxide)2020, 13 on, 4926 the lime particles after 5 and 15 min, respectively, while image (c) after two days shows18 of 24 that calcium hydroxide has crystallized to small hexagonal plates penetrating the precipitated grey mass on the particle surface. Magnesium hydroxide will appear darker grey than calcium hydroxide crystals by SEM in Figure 17. Another feature disclosed by SEM is that a grey mass is precipitated since it is composed of lighter elements, but the physical density may also affect the grey levels. earlyHexagonal on the particle. crystals of This calcium is likely hydroxide to be magnesium also appear hydroxide on the particle and surface may be after the reason 4 d in image why burnt(d) and lime hydratesimage (e) slower shows in various sea water crystal than inshapes fresh on water the partic as seenle fromsurface the after calorimetry seven days. curves Image in Figures (f) focuses 15 and on 16. Laterwell-crystallized calcium hydroxides prisms crystallizesof gypsum at to an hexagonal age of three plates days. in line Note with that its the crystal magnification structure indicated that penetrates by throughthe dotted the precipitatedline at the bottom mass of magnesiumthe images is hydroxide. 50 μm for image Examples (f) and of this200 μ arem shownfor the other in Figure images. 17.

Figure 17. A selection of images from the SEM investigation as a function of time for lime particle; (a) 5 min, (b) 15 min,(c) 2 days, (d) 4 days and (e) 7 days, as well as (f) after 3 days showing gypsum crystals.

Images (a) and (b) in Figure 17 depicts seemingly amorphous precipitations (likely magnesium hydroxide) on the lime particles after 5 and 15 min, respectively, while image (c) after two days shows Materials 2020, 13, 4926 19 of 24 that calcium hydroxide has crystallized to small hexagonal plates penetrating the precipitated grey mass on the particle surface. Magnesium hydroxide will appear darker grey than calcium hydroxide Materials 2020, 13, x FOR PEER REVIEW 19 of 24 since it is composed of lighter elements, but the physical density may also affect the grey levels. Hexagonal crystals of calcium hydroxide also appear on the particle surface after 4 d in image (d) and image (e) shows various crystal shapes on the particle surface after seven days. Image (f) focuses on well-crystallizedFigure 17. A prisms selection of of gypsum images from at an the age SEM of investigation three days. as Note a function that theof time magnification for lime particle; indicated (a) by the dotted5 min, line (b) at 15 the min, bottom (c) 2 days, of the (d) images4 days and is 50 (e)µ 7m days, for imageas well (f)as ( andf) after 200 3 µdaysm for showing the other gypsum images. crystals. 4.3. Burnt Lime Added to Sea Water in an “Open System” 4.3. Burnt Lime Added to Sea Water in an “Open System” Very few trials have been performed to measure pH when burnt lime has been dosed directly to openVery sea. few In suchtrials anhave “open been system”performed will to measure the calcium pH when hydroxide burnt lime formed has been during dosed hydration directly haveto accessopen to sea. magnesium In such an and“open other system” components will the calcium for neutralization, hydroxide formed and pH during will be hydration a question have of timeaccess and distanceto magnesium from individual and other burnt components lime particles. for neutralization, and pH will be a question of time and distanceThe following from individual test was burnt performed lime particles. in the river Murray in Canada as described in the (Aquaculture and RuralThe Development) following test DFARD was performed Tech Report in #253 the [22river]: The Murray sea water in justCanada outside as thedescribed river mouth in the had (Aquaculture and Rural Development) DFARD Tech Report #253 [22]: The sea water just outside the pH 8.1, while the pH in a trough of slaked lime (i.e., calcium hydroxide) was measured to 12.6. When the river mouth had pH 8.1, while the pH in a trough of slaked lime (i.e., calcium hydroxide) was slaked lime was thrown into the sea, the pH within the cloud of particles was up to 9. The sea water about measured to 12.6. When the slaked lime was thrown into the sea, the pH within the cloud of particles 10 m from the trough, but directly above the treatment zone, had pH in the range 8.2–8.3. The pH values was up to 9. The sea water about 10 m from the trough, but directly above the treatment zone, had dropped rather rapidly back to the natural background level of 8.1 close to the treatment zone. pH in the range 8.2–8.3. The pH values dropped rather rapidly back to the natural background level of In8.1 2017close [to23 ]the and treatment experiment zone. was performed in the Slettnes fjord near Hammerfest in northern Norway,In where2017 [23] a suspension and experiment of burnt was lime performed was spread in the over Slettnes the sea fjord where near pH-meters Hammerfest were in mountednorthern at 4.98Norway, and 6.28 where m depths. a suspension A photo of ofburnt the boatlime sprayingwas spread lime over suspension the sea where is shown pH-meters in Figure were 18 mounted, while the pHat logs 4.98 forand the 6.28 whole m depths. period A (2photo h and of 24the min)boat arespraying plotted lime in suspension Figure 19. is shown in Figure 18, while theThe pH measurementlogs for the whole at depth period 4.98 (2 h m and starts 24 atmin) 7.8 are and plotted climbs in to Figure pH 8.0 19. within 20 min after spraying starts andThe the measurement highest pH at measured depth 4.98 within m starts the at period 7.8 and of climbs 2 h and to 24pH min 8.0 within was 8. 20 1. min The after measurement spraying at depthstarts 6.28 and m the started highest with pH pHmeasured 7.2 and within climbed the peri to 7.9od withinof 2 h and 20 min24 min and was stabilized 8. 1. The at measurement 8.0 within the measurementat depth 6.28 period m started and with with pH 8.0 7.2 as theand highest climbed value to 7.9 read. within It is20 worth min and noting stabilized that all at values 8.0 within measured the aremeasurement within the range period to beand expected with 8.0 in as natural the highest sea water. value read. It is worth noting that all values measuredThe measured are within pH the at openrange seato be is expected in line with in natural the pH sea estimation water. <9 using Equation (21) and sea 2+ water concentrationThe measured of pH Mg at openat temperatures sea is in line with<20 ◦ thC.e pH estimation <9 using Equation (21) and sea waterBased concentration on the experimental of Mg2+ at temperatures data retrieved, <20 °C. observations like those depicted in Figure 17 and theoreticalBased evaluation, on the experimental the reaction data progress retrieved, when obse burntrvations lime is like mixed those into depicted the sea canin Figure be described 17 and as sketchedtheoretical in Figure evaluation, 20a–f. the reaction progress when burnt lime is mixed into the sea can be described as sketched in Figure 20a–f.

FigureFigure 18. 18.Photo Photo of of the the boatboat sprayingspraying lime suspension of of coarse coarse CaO CaO over over the the sea. sea.

Materials 2020, 13, 4926 20 of 24

MaterialsMaterials 2020, 2020 13, x, 13 FOR, x FOR PEER PEER REVIEW REVIEW 20 of 2420 of 24

Figure 19. pH measured in open sea in the area where a suspension of coarse CaO was sprayed. Figure 19. pH measured in open sea in the area where a suspension of coarse CaO was sprayed. Figure 19. pH measured in open sea in the area where a suspension of coarse CaO was sprayed.

(a)

(a)

(b)

Figure 20. Cont. Materials 2020, 13, x FOR PEER REVIEW 21 of 24 Materials 2020, 13, 4926 21 of 24 (b)

(c)

(d)

Figure 20. Cont.

Materials 2020, 13, 4926 22 of 24 Materials 2020, 13, x FOR PEER REVIEW 22 of 24

(e)

(f)

FigureFigure 20. 20.(a) (a Sketch) Sketch of of the the «open«open system» system» sea sea water water with with temperature temperature profile profile and different and di ffions;erent (b ions;) (b) SketchSketch of an agglomerate agglomerate of ofburnt burnt lime lime particles; particles; (c) Phase (c) Phase 1: Immediate 1: Immediate reaction reaction between between burnt lime burnt limeand and sea sea water—a water—a surface surface layer layer of of calcium calcium hydroxide hydroxide is formed; is formed; (d) (dPhase) Phase 2: The 2: The surface surface reaction reaction formingforming calcium calcium hydroxide hydroxide continues continues and magnesiumand magnesium hydroxide hydroxide is precipitated; is precipitated; (e) Phase (e) 3Phase: Structural 3: compressionStructural and compression formation and of crystals;formation ( fof) Phase crystals; 4: ( Allf) Phase burnt 4: lime All burnt particles lime are particles completely are completely reacted and transformedreacted and to othertransformed solids. to other solids.

5. Conclusions5. Conclusions TheThe reaction reaction of burntof burnt lime lime in in sea sea water water has has beenbeen discussed and and compared compared to tofresh fresh water. water.

A particle of burnt lime (CaO) added to sea water will immediately react to form calcium hydroxide (called slaking) and more than 90% of the lime would have reacted within 5 h in a “closed Materials 2020, 13, 4926 23 of 24 system”. This process is expected to be even faster in an “open system” such as the open sea or a fjord. The slaking process is extremely exothermic and will be able to increase the temperature of the surroundings of the particles. However, a remarkable or persistent increase of local sea water temperature would be extremely unlikely due to the low doses applied and the large water reservoir. The reaction of burnt lime in sea water is slower than in fresh water. This is probably due to the large concentration of magnesium in sea water that provokes a rapid precipitation of magnesium hydroxide onto the surface of the particle and hinders the progress of the hydration reaction of CaO. The slaking process will increase pH locally from the natural level of about 8 to a theoretical maximum of 12.5 near the surface of the particle on a micro level. In an “open system” the pH will be within the variation in natural sea water (7.8–8.5) a few cm away from the particle, as shown in practical field experiments. In parallel to the slaking process, dissolved calcium hydroxide will react with magnesium ions in sea water and precipitate less soluble magnesium hydroxide. The equilibrium pH of magnesium hydroxide in fresh water is 10.5, but in equilibrium with the magnesium concentration in sea water it is calculated to < 9 as observed. All carbonate, hydrogen carbonate and dissolved CO2 in the vicinity of formed calcium hydroxide will be consumed and yield calcium carbonate. Same happens with sulphate, that will react and form gypsum, but somewhat slower than the magnesium and carbonate reactions. It is likely however that the magnesium reaction will dominate in an “open system”. The product formed at equilibrium after dosing burnt lime in the sea will be a mixture of magnesium hydroxide (that possibly will be slowly converted to magnesium carbonate over time), calcium carbonate and gypsum (i.e., calcium sulphate dihydrate). All calcium oxide and hydroxide will be consumed and no traces of them will remain at equilibrium. How long time that will take depends on the thinning rate probably being a function of the particle size, the temperature and the water renewal due to natural currents. Correspondingly will all dissolved calcium be diluted and eventually reach the natural background level of sea water. According to the data retrieved in this study, the reaction will be complete within a few days or maximum a week. It is worth noting that all of the final products magnesium carbonate, calcium carbonate and gypsum occur naturally and can be regarded harmless for the environment.

Author Contributions: Conceptualization, M.M.; Methodology, C.E.-O. and H.J.; Validation, Ø.A.G.; Investigation, C.E.-O. and H.J.; Writing—Original Draft Preparation, H.J.; Writing—Review & Editing, Ø.A.G., C.E.-O., H.J. and M.M. Visualization, M.M., C.E.-O. and H.J.; Supervision, M.M.; Funding Acquisition, M.M. All authors have read and agreed to the published version of the manuscript. Funding: This work was supported by grants 217663 and 269272 from the Research Council of Norway. Acknowledgments: Flagship Fjord and Coast at the Fram Centre in Tromsø helped to facilitate the field work. Thanks to Alf Børre Tangeraas who assisted during the lab and field trials, to Franzefoss AS for financial and practical support, and to Cermaq ASA for help with logistics and all the other support we have asked for. Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

References

1. Brooks, S.J.; Georgantzopoulou, A.; Johansen, J.T.; Mengede, M. Determining the risk of calcium oxide (CaO) particle exposure to marine organisms. Mar. Environ. Res. 2020, 156, 104917. [CrossRef][PubMed] 2. Strand, H.K.; Christie, H.; With Fagerli, C.; Mengede, M.; Moy, F. Optimizing the use of quicklime (CaO) for sea urchin management—A lab and field study. Ecol. Eng. X 2020, 100018. [CrossRef] 3. Khseshgi, H.S. Sequestering atmospheric carbon dioxide by increasing ocean alkalinity. Energy 1995, 20, 915–922. [CrossRef] 4. Renforth, P.; Jenkins, B.G.; Krüger, T. Engineering challenges of ocean liming. Energy 2013, 60, 442–452. [CrossRef] Materials 2020, 13, 4926 24 of 24

5. Wuhrer, J. Physikalisch-chemische Untersuchungen über den Zustand des Branntkalkes und über die Vorgänge und Einflüsse beim Brennen. Zement-Kalk-Gips 1953, 10, 354–368. 6. Wuhrer, J. Wissenschaftliche und verfahrenstechnische Probleme des Kalkbrennens. Chemie Ingenieur Technik 1958, 1, 19–30. [CrossRef] 7. Mengede, M. In house R&D by Franzefoss Minerals. Quicklime Sea Water 2017. 8. Dean, J.A. (Ed.) Lange’s Handbook of Chemistry, 13th ed.; McGraw-Hill: New York, NY, USA, 1985; pp. 10–19. ISBN 0-07-016192-5. 9. Wagner, T.; Kulik, D.A.; Hingerl, F.F.; Dmytrievava, S.V. Gem-selector geochemical modelling package: TSolMod library and data interface for multicomponent phase models. Can. Miner. 2012, 50, 1173–1195. [CrossRef] 10. Kulik, D.A.; Wagner, T.; Dmytrieva, S.V.; Kosakowski, G.; Hingerl, F.F.; Chudnenko, K.V. GEM-Selector geochemical modelling package: Revised algorithm and GEMS3K numerical kernel for coupled simulation codes. Comput. Geosci. 2013, 17, 1–24. 11. Hummel, W.; Berner, U.; Curti, E.; Pearson, F.J.; Thoenen, T. Nagra/PSI chemical thermodynamic data base 01/01. Radiochim. Acta 2002, 90, 805–813. [CrossRef] 12. Temperature Dependence of the pH of Pure Water. Available online: https://chem.libretexts.org/Core/ Physical_and_Theoretical_Chemistry/Acids_and_Bases/Acids_and_Bases_in_Aqueous_Solutions/The_ pH_Scale/Temperature_Dependence_of_the_pH_of_pure_Water (accessed on 24 September 2020). 13. Stumm, W.; Morgan, J.J. (Eds.) Aquatic Chemistry-Chemical Equilibria and Rates in Natural Waters, 3rd ed.; John Wiley & Sons, Inc.: New York, NY, USA, 1996; pp. 157–163. 14. Dean, J.A. (Ed.) Lange’s Handbook of Chemistry, 15th ed.; McGraw-Hill: New York, NY, USA, 1999; pp. 5.4–5.5. 15. pH of Common and Bases. Available online: https://www.aqion.de/site/191 (accessed on 24 September 2020). 16. Duchesne, J.; Reardon, E.J. Measurement and prediction of portlandite solubility in alkali solutions. Cem. Concr. Res. 1995, 25, 1043–1053. [CrossRef] 17. Johnston, J.; Grove, C. The solubility of calcium hydroxide in aqueous salt solutions. J. Amer. Soc. 1931, 53, 3976–3991. [CrossRef] 18. Yeatts, L.B.; Marshall, W.L. Aqueous systems at high temperature. XVIII. Activity coefficient behavior of calcium hydroxide in aqueous sodium nitrate to the critical temperature of water. J. Phys. Chem. 1967, 71, 2641–2650. [CrossRef] 19. Lide, D.R. (Ed.) Handbook of Chemistry and Physics, 71st ed.; CRC Press: New York, NY, USA, 1990; p. 4–77. 20. Justnes, H.; Kim, M.-O.; Ng, S.; Qian, X. Methodology of Calculating Required Chloride Diffusion Coefficient for Intended Service Life as Function of Cover in Reinforced Marine Structures. Cem. Concr. Comp. 2016, 73, 316–323. [CrossRef] 21. Nishimuta, A.; Seki, M. Effects of Lime for the Improvement of Mariculture Grounds. Nippon Suisan Gakkaishi 1983, 49, 353–358. [CrossRef] 22. Ramsay, A.; Gill, K.; Morrison, A.; Mac Nair, N. Hydrated Lime Application by the PEI Aquaculture Industry; Technical Report #253; Prince Edvard Island (PEI) Department of Fisheries, Aquaculture and Rural Development (DFARD): Montague, PE, Canada, 2014; 28p. 23. Garmo, Ø.; With Fagerli, C.; Segtnan, O.A. Surveillance of pH during Dosing of Burnt Lime in the Slettnes Fjord; NIVA memo 1587/17; Norwegian Institute for Water Research: Oslo, Norway, 2017. (In Norwegian)

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).