Experimental Investigation on Adsorption and Surface Characteristics

Modern Environmental Science and Engineering (ISSN 2333-2581) February 2017, Volume 3, No. 2, pp. 67-81 Doi: 10.15341/mese(2333-2581)/02.03.2017/001 Academic Star Publishing Company, 2017 www.academicstar.us

Experimental Investigation on Adsorption and Surface Characteristics of Salt Hydrates and Hydrophobic Porous Matrix Based Composites Asnakech Lass-Seyoum1, D. Borozdenko1, T. Friedrich1, K. Schönfeld2, J. Adler2, and S. Mack3 1. ZeoSysGmbH, Falkenberger Straße 40, 13088 Berlin, Germany 2. Fraunhofer Institute for Ceramic Technologies and Systems IKTS, Winterbergstrasse 28, 01277 Dresden, Germany 3. Fraunhofer Institute for Interfacial Engineering and Biotechnology IGB, Nobelstraße 12,70569 Stuttgart, Germany Abstract: In the studies reported here more than nine different hygroscopic active salt hydrates and two groups of mesoporous hydrophobic substrates have been used to prepare a wide range of composites for closed sorption thermochemical storage application. The first group of substrate material that has been employed to incorporate the salt hydrates is consisted of both pure and in various form modified attapulgite, whereas the second group involved different kinds of activated carbons. The composites reported here were prepared by direct incorporation of the substrate with a pre-defined concentration of salt hydrate solution. Composites based on porosing agent modified attapulgite and those calcined at a temperature of 400°C present poor thermal stability and unfavorable dynamic performance associated with pressure drop across the bulk material resulting in 15-20% decrease on adsorption capacity. Following a pre-selection of an appropriate material out of a wide range of composites, short and long term cyclic tests were performed. Among all the investigated materials in this work, both pure attapulgite calcined at 550°C (Att550) and Poolkohl (PK) based composites containing CaCl2, MgCl2 and LiCl exhibit an optimum adsorption and surface behavior including high thermal stabilities under hydrothermal conditions over 250 cycles. Key words: salt hydrates, composites, hydrophobic, thermo-chemical, dynamic adsorption

1. Introduction Porous storage materials have received significant interest in recent years for adsorption thermochemical heat storage applications. While several studies have been reported on different approaches taken to identify efficient and economical materials, most of them are focused on silica gel, zeolite and the so-called zeotypemolecular sieves in combination with water as a working fluid [1-14]. These material pairs exhibit good storage characteristics including high storage density, relatively low cost, and in the case of zeolites very high hydrothermal stability. However, the major Corresponding author: Asnakech Lass-Seyoum, Ph.D., research areas/interests: applied surface science research, thermochemical storage. E-mail: [email protected].

limitation of these materials is their high charging temperatures. This significantly limits the storage of thermal heat at a low temperature level that often occurs typically as waste heat or attained from CHP and solar thermal technologies. Currently there is no economically appropriate material that can realize a storage density higher or comparable to silica gel and zeolites, with typical value of around 180 kWh/m³ and 220 kWh/m³ [14] respectively, at a charging temperature below 110°C. This is reasonable, as a considerably high storage density requires a strong intermolecular interaction between the molecules of the working fluid and the solid adsorbent that such a low desorption temperature is unable to overcome this intense interaction [15].

68

Experimental Investigation on Adsorption and Surface Characteristics of Salt Hydrates and Hydrophobic Porous Matrix Based Composites

In the studies reported here a wide range of composite materials of different combinations have been developed and investigated to enable storage of thermal energy at relatively low temperature levels between 90°C and 110°C. The composites investigated here are consisted of hygroscopic active component, typically a mono or multi salt hydrates, embedded on a supporting porous matrix. Decisive factor for the efficiency of such a composite is the interplay of the involved substrate materials, active components and additives. In general the substrate material provides the required surface area and pore system, where the active component can be finely dispersed. In this study two groups of mesoporous materials have been employed as substrate to incorporate the salt hydrates. While the first group is consisted of both pure and in various form modified attapulgite, the second group comprises different kinds of activated carbons. Though a large number of hydrate-forming salts are available, none of this is in a sole form suitable for adsorptive thermochemical storage application. It is because the equilibrium conditions that define the dynamic process, particularly temperature and humidity, usually lie beyond the range of the deliquescence coefficient of the respective salt. Thus the composites adsorbent studied here will have a potential scope in taking advantages of the complementary effect of salt hydrates and the porous matrix materials. Numerous research activities have been performed on developing and investigating composites of different material pairs. These are considered and evaluated mainly from three different views, i.e., application areas, involved materials and investigation methods. So far published studies have shown that composites have an extraordinary potential for application in many areas such as cooling, dehumidification, energy transformation and storage.

For instance a set of representative studies have been reviewed and summarized on two components adsorbents, termed as SWS, where a wide range of salt hydrates and porous matrix have been synthesized and characterized for adsorptive cooling and drying applications [16]. Several other interrelated works done on composites for refrigeration and adsorptive heat transformation applications have been also identified [17-27]. Composite materials are also elucidated as potential candidatesfor adsorptive thermochemical storage application [28-32]. However, most of these investigated composites are preferably based on CaCl2andhydrophilic materials such as silica gels of different pore sizes, zeolites and activated alumina. Due to the hydrophilic nature of the above mentioned substrate materials; it is not possible to certainly determine in advance the ratio of the water uptake by the salt hydrate relative to the one by the carrier material participated in the composite. As a result, it can be challenging to assess the temperature level that is necessary for the dehydration of the storage material. A large number of important investigations have been also accomplished generally based on thermo-analytical methods and structure analysis that enable to assess initial material characteristics [31-36]. However, minor emphasis has been placed on the material cyclic stability from the thermal energy storage aspect. So far only one study has been identified reporting on a cyclic stability of composites for application in thermally driven adsorption heat pump and chiller [39]. Thus, in these studies besides the investigations on the adsorption behavior of the composite, short and long term hydrothermal cyclic tests have been also performed. Moreover, so far there are few studies reported on investigation of composites in an open system, but without focus on the corrosion aspect [28, 30]. Whereas in the studies reported here the composites

Experimental Investigation on Adsorption and Surface Characteristics of Salt Hydrates and Hydrophobic 69 Porous Matrix Based Composites

will be applied in a closed storage system [40] where it is possible to maintain the hydration and dehydration equilibrium conditions using vacuum controlled process parameters. This enables to prevent one of the well-known drawbacks of salt hydrates, i.e., their corrosive behavior [41-43].

2. Experimental part 2.1 Materials Several hygroscopic salt hydrates were selected, as shown in Table 1, as active substances for preparation of a wide range of composites. The bases of the selection criterion were mainly water adsorption Table 1

Thermophysical properties of the used salt hydrates [44-48].

Salt hydrate CaCl2.6H2O MgCl2.6H2O MgSO4.7H2O Na2SO4.10H2O Na2CO3.10H2O Na3PO4.12H2O LiCl.5H2O ZnSO4.7H2O Table 2

capacity, hydration enthalpy and easily availability. All the salt hydrates (Carl Roth GmbH) used here were of lab grade and have been used without further purification. As suitable substrate materials numerous porous hydrophobic materials of various compositions, structure, particle form and size have been employed. All those identified substrate materials exhibit a broad distribution of pores predominantly in the mesoporous range and are hydrophobic by nature. For practical purpose those selected materials were generally classified into two major groups as it has been presented in Table 2. The first group of substrate

MSalt [g/mol] 111 95 120 142 106 164 42 161

MHydrate [g/mol] 219 203 246 322 286 380 132 287

n.MH2O [g/mol] 108 108 126 180 180 216 90 126

H2O content [g/g] 0.973 1.137 1.050 1.268 1.698 1.317 2.143 0.783

Energy density [kWh/kg] 0.458 0.556 0.464 0.486 n.a. n.a. n.a. n.a.

DRH at 20°C [%] 33.3 33.1 91,3 95.6 97.9 99.6 12.4 89.0

Characteristic properties of the used substrate materials. Substrate material (short description) Poolkohl (PK) AFA3-1150W (AFA3) ADA30/60 (ADA) AFA1400 (AFA4) D47/4 (D47) DGK AFA3-1150W (AFA3) ADA30/60 (ADA) Att-400 Att-550 Att-700 20%W 30%M 50%KG

BET surface area Bead size/Pellet ∅ ρ [mm] [g/l] [m²/g] Activated Carbon (AC) 2.5-4.0 480 1250 2.0-3.0 450 1050 0.15-2.0 430 1200 2.0-3.15 450 1450 2.0-4.0 470 1050 1.0-2.0 550 1100 2.0-3.0 450 1050 0.15-2.0 430 1200 Attapulgite calcined at 400°C, 550°C and 700°C temperatures 2.0-3.0 620 121.80 2.0-3.0 600 115.85 2.0-3.0 580 109.47 Attapulgite with porosifying additives 2.0-3.0 765 131.3 2.0-3.0 775 118.2 2.0-3.0 960 75.7

Particle form

pellet pellet fracture pellet pellet granulate pellet fracture fracture fracture fracture fracture fracture fracture

70


material identified in this work is activated carbon (AC), as it displays favourable properties including large internal surface area that lies commonly in the range of 1000-1500 m²/g, high porosity, high surface reactivity and high thermal conductivity. This provides a huge capacity to incorporate salt hydrates. About eight kinds of activated carbons have been identified to develop the intended composites. The second group of porous material used is known as attapulgite. This group of material exhibit free channels of diameter in the range 0.37 to 0.63 nm, enabling incorporation of both water and salt hydrate, thus considered as a potential candidate for making composites. More than ten kinds of attapulgite substrates have been developed mainly through varying the calcination temperature (400°C, 550°C and 700°C) and modification by using three different porosity intensifying additives, i.e., wax (W), maize (M) and diatomite (KG). 2.2 Composites Preparation The composites investigated in this work were made by a direct incorporation of the substrate material with salt hydrate solution of a pre-defined concentration. The following part describes the procedures for preparation of the so far investigated composites. Salt solution (c = 60–100%)

Incorporation in 100% salt hydrate solution (T = 30°C, p = atm, 150 rpm, 72 h)

Composite treatment (filtering, pre-drying (T = 20°C, 24 h)

Cooling (T = 20°C, RH = max. 5%)

Thermal activation (T = 90-120°C, 3-6h)

Substrate material treatment (activation, dust particles removal, macroscopic modification) Fig. 1

Approximately 30-200 g of the selected substrate has been thermally activated to make the pores free of water molecules. The temperature and the activation time vary from 120°C to 150°C and 2-6 hours, respectively depending on the nature and the volume of the substrate used. Thermal relaxation of the activated material has been induced using anhydrous hygroscopic material in order to avoid renewed water uptake. The final weight of the activated substrate and the final water content has been determined. After the cooling process, a homogenous salt hydrate solution of 1-3 ml has been added drop-wise on the substrate material, followed by continuous stirring with 150 rpm for 48-72 hours at a temperature of 25-30°C and atmospheric pressure. The amount of the solution added was set mostly based on the pore volume of the substrate material and the consistency of the mixture. Finally the wet composite was filtered and purified using a small amount of water, in order to remove the rest salt coated on the surface of the substrate material followed by its thermal activation. Summary of composite preparation method is depicted schematically in Fig. 1.

Composite preparation procedure.

2.3 Experimental Characterization Methods So far, wide range of composites has been prepared in a lab scale. In the following section, the different approaches taken on the physico-chemical characterizations and surface analysis of those composites are described. Those parameters that are considered here to identify an optimum composite

were: adsorption capacity, energy density and salt deposition on the surface of the substrate. Moreover, one of the main issues addressed in this paper is the dynamic performance of the composite associated with the mass and heat transfer as well as the pressure drop across the bulk material. 2.3.1 Adsorption and Hydrothermal Cyclic Tests Theoretically many of the prepared composites are

Experim mental Investiigation on Ad dsorption and d Surface Cha aracteristics of Salt Hydra ates and Hydrrophobic 71 Porous Matrrix Based Com mposites

3. Results and Discusssion 3.1 Attapulggite Based Coomposites For practiical purpose, the Att-based compositess are characterisedd in threee phases. Primarily P thhose composite made m based on o pure Attsubbstrates prepared at three diffferent calcinnation tempeeratures (4000°C, 550°C and 700°C) andd three maiinly known salt

hyd drates (CaCll2, MgCl2 aand MgSO4) have beenn chaaracterized. The results attained from f astaticc adsorption at 199.7 mbar partiial pressure are a illustratedd on Fig. 2. From m these resuults, it is cllear that thee stattically determined gravvimetric and d volumetricc adsorption capaccities of thosee three compo osites showedd, with h Δamax = 0.031 g/g, no significant differences d inn term ms of substratte calcinationn temperaturees. However, H com mparing the investigated d compositess with h respect to the involvedd salt hydratee, those withh Mg gSO4 with 0.2218 g/g reveaal by far the least overalll adsorption capaccity, while thoose with CaC Cl2 with 0.4522 g/g showed the highest h valuee. This is in paart due to thee diffferences in thhe relative deeliquescence of the threee saltt hydrates. Thhe variation oof adsorption capacities off thesse compositess could also bbe related to th he mean poree

Adsorption capacity

[g/g] 0.400 0.300 0.200 0.100 0.000 Att5500-CaCl2

Com mposite type

Att700 0-CaCl2

[g/g] [ Adsorption capacity

[g/ml]

0.500

Att400-CaCl2

[g/ml]

0.500 0.400 0.300 0.200 0.100 0.000 0-MgCl2 Att550 0-MgCl2 Att700 0-MgCl2 Att400 Com mposite type

g/g

[g/ml]

0.500 Adsorption capacity

expected to be suitable as a storagge material. It I is therefore addvantageous to t make a pre-selection with w experimentaal methods that t can proovide the ressults within a shoort time rangee. Thus in thesse studies a static and dynamicc adsorption method, m alreaady mentioneed in our previouus works, havve been applied for material screening thhrough deterrmination of their adsorpption properties under u pre-defined equilibbrium condittions [50]. Morreover, thee reversibillity of water w uptake/releaase under hyydrothermal conditions have h been considdered as a pre-requisite p factor for thhose selected com mposites to be regardedd as a potenntial thermochem mical materials for a technical scale s application. Thus several tests were conducted c onn the finally seleccted composiites to determ mine the channges occurred on the adsorbedd/released amoount of waterr and to determinee the establishment of thhe hydrate leevels after multiplle operation cycles. c Evaluuation on a viisual appearance of excess salt on the surface of the composite has h been done as a qualitaative indicatoor of cyclic stability of the investigated matterial. 2.3.2 Surfface and Struccture Analysiis Several teests were coonducted on the pre-seleected composites to t determine the t changes enncountered onn the specific surfface area andd pore volum me resulted from f incorporatingg the porous substrates with w salt hydrates. The measurrement of speecific surfacee area was done d using ASAP P2020 (Microm meritics GmbbH) with nitroogen gas. Samplees were heaated to 300°°C prior to the measuremennt in order to remove r waterr from the parrticle surface and pores. p The poorosity and buulk density vaalues were determ mined by merrcury porosim metry using Auto A Pore IV (Miccromeritics GmbH). G

0.400 0.300 0.200 0.100 0.000 Att400-MgSO4 Att5500-MgSO4 Att700 0-MgSO4 Comp posite type

Fig.. 2 Adsorptioon capacities oof composites based b on puree Att substrate calciined at differen nt temperaturres.

72

Experim mental Investiigation on Ad dsorption and d Surface Cha aracteristics of Salt Hydra ates and Hydrrophobic Porous Matrrix Based Com mposites

radius and specific suurface area of o the involved substrates, as a it is show wn on Fig. 3.. Higher specific surface areaa and lower average poree dimeter off the substrate maaterial favoureed the adsorpption behaviouur of the correspoonding compposite. Furtheermore, posssibly other factorrs, which aree not yet beeen elucidated in detail, may also a play a roole. On the otther hand in order o to identtify the influeence of variatioon of the Att-substrrate calcinaation temperature on the addsorption beehaviour of the respective composite, c d detailed characterizationss in terms of its hydrothermaal material performance p h have been executted. As it hass been already mentioned,, the substrate maaterials explorred are hydroophobic by naature that the therrmodynamic properties of the composites are mainly innfluenced by the involved salt hydrate. The interaction of o water and salt s hydrate inn such a systeem is a mono-variiant type [15,, 21, 23], byy which the water w uptake is acccompanied wiith formation of a fixed num mber of hydratess under defi fined equilibrrium conditiions. However, thhe water uptakke of salt hyddrates can exxtend beyond theirr deliquescennce limits. Thhus, the hydraation process occuurs mainly in i two partiaal processes, i.e., adsorption (D DRH RH) followed by partial absorpption (DRH< RH)). Whereas saturation s wiithout reachinng a solution state is a typicall behaviour obbserved only in a composite buut not in a purre salt hydratee.

Fig.. 3 Influence of specific areaa and average pore diameterr on the t adsorption capacity of Attt550-based composites.

Thus, T generallly the interppretation of those resultss attaained from m hydrotherrmal cycliic stabilityy inveestigations peerformed on a specific com mposite havee beeen achieved on the bases of the range r of thee adsorption capaacity reachedd within thee limit of a defi fined processs time in w which mainly y adsorptionn occurs. In n addition to the above-m mentioned app proach on thee anaalysis of thhe results, bbasic calcullations weree perfformed. Theese include determinatiions of thee aveerage water uptake ∆ ̅ and release ∆ ̅ ass welll as averagge adsorptioon capacity ( . Thee corrresponding experimentall e y attained an nd calculatedd valu ues for the three compoosites are sum mmarized inn Tab ble 3.

Table 3 Aveerage adsorbed d (+∆ ) and deesorbed (-∆ ) water w amount of the investiggated CaCl2 baased compositees. Compositee 1 - 10 cycles 80 cycles 40 cyclees

Att400 (m = 4.144 g) g Att550(m m = 4.221 g) Att700(m m = 3.860 g)

(loss)

[g/g]

∆ ̅ [g]

∆ ̅ [g]

[g/g]

∆ ̅ [g]

∆ ̅ [g]

[g/g]

∆ ̅ [g]

∆ ̅ [gg]

[g/g]]

1.824

1.830

0.441

1.214

1.222

0.294

0.863

00.911

0.214

0.227

1.926

1.983

0.463

1.844

1.904

0.444

1.694

1.776

0.411

0.052

1.598

1.660

0.422

1.593

1.641

0.419

1.484

1.550

0.39 93

0.029

Comparinng the cycllic stabilities of the three t composites, in terms of the adsorbbed and released amount of water, w no significant chaanges have been b recognized within w the firrst 10 cycles. However, inn the following 800 cycles substtantial changees were obserrved. Yet Att400--CaCl2 composite has exhhibited the most m

com mprehensive changes, c in thhat its averag ge adsorptionn capacity ( hass been reduceed by 0.147 g/g g and 0.2277 g/g resulting in approx. 30% % and 50% lo osses after 400 and d 80 cycles, reespectively inn comparison n to the initiall valu ue. Whereas, the specific w water uptake efficiency off Att5 550-CaCl2 annd Att700-CaaCl2 composites obtainedd


after 80 cyclles are 0.411 g/g g and 0.3933 g/g leading to11% t and 7% losses, respectivvely. These values are much m lower thann those dettermined foor Att400-based composites. Thus due to its highher hydrotherrmal stability annd its com mparatively low calcinaation temperature,, Att-550 hass been selecteed as an optim mum substrate forr further use. The dynaamic perform mance of thee Att-550 based composite, in i terms of booth its temperature coursess and the maximuum increase in i a pressure drop duringg the process, is depicted d on Fig. F 4. For practical p purppose both inlet/ouutlet (Tin/Tout) as well as thhose temperaature points withiin the bulk material (T001-T05) are also shown in thee figure. The maximum dynamiic adsorptionn capacity andd the temperature attained from m Att550-CaC Cl2 compositee are 0.401 g/g annd 60.7°C resppectively. Thhis is in consisstent with the resuult attained (00.420 g/g) froom investigattions done under similar s conditions in our previous p workk 65

Att550-CaC Cl2 Δpmax = 12..0 mmH2O amax = 0.401 1 g/g

60

Temperature [°C]

55

T_in T01 T02 T03

50 45 40 35 30 25 20 15 0:00 0:28 0:57 1:26 1:55 2:2 24 2:52 3:21 3:50 4:19 4 4:48 5:16

Adsorrption time [h:m mm]

65

Att550-MgCl2 Δpmax = 2.1 mmH2O amax = 0.3 372 g/g

60 55

Temperature [°C]

50

T_in T01 T02 T03 T04

45 40 35 30 25 20 15

0:14

Fig. 4

0 0:43

1:12

1:40

2:09

2:38

3:077

3:36

Adso orption time [h:m mm]

4:04

Dynaamic performan nce of Att550-b based composiites.

[50]. The therebby determined specific en nergy densityy reacched around 0.285 0 kWh/kkg. So S far there arre no other comparable dyn namic studiess don ne on attaapulgite bassed compossites. From m calo orimetric studdies done onn attapulgite based CaCl2 com mposite by Jäänchen et al.. result of 0..395 g/g andd 0.417 kWh/kg at a sorption ttemperature of o 20°C hass beeen attained [35]. Other sstudies carrieed out on a dyn namic perforrmance of a compositee based onn attaapulgite relateed substrate kknown as verrmiculite andd CaC Cl2 was in the range between 0..06-0.18 g/gg dep pending on the t ratio of the salt hy ydrate to thee substrate [49]. This T result iss significantly y lower thann thosse obtained inn this study. While W CaC Cl2 containiing composite showss com mparable behhaviour in tterms of itss static andd dyn namic perform mances, on MgCl2 baseed compositee slig ght deviationss were observved. In n the second phase p of this work, characcterisations off thosse compositees prepared oon the basis of o a modifiedd Att--550 substratte have been carried out. As has beenn statted in the exxperimental ppart, the mo odification off Att5 550 substratee has been acchieved throu ugh includingg threee types of porosing p addiitives in variious ratios inn order to attain different d porrosity level. Generally G ann imp provement in the porosity llevel has been n achieved byy morre than 6-20% % in compariison to the co orrespondingg unm modified subsstrate (Table 44). To T assess the influence off substrate mo odification onn the adsorption peerformance of the respectiv ve compositee, both h static andd dynamic experiments have beenn perfformed. In Fig. F 5 repressentative resu ults obtainedd from m static addsorption meeasurements have beenn illustrated. Thoose results show that compositess con nsisting modiffied Att550, for the same salt hydrate,, difffer only slighhtly from thosse made from m pure Att5500 substrate. However, H from m the dynam mic performan nce tests (Fig.. 6) at a a dehydrattion temperatture of 110°C C it has beenn observed that coomposites gennerated based d on modifiedd substrates were unstable. u Forr instance, con ncerning the


Table 4

Chaange in porositty of modified and pure Att5550.

Substrrate

P Porosity [%]

BET [m2/g]]

Att5550

53.3

117

20%W W

62.7

131.3

30%M M

66.3

118.15

50%K KG

56.8

75.73

20%W-Ca aCl2 Δpmax = 36 6.1 mmH2O amax = 0.22 20 g/g

6 65 6 60

T_in T01 T02 T03 T04 T05

5 55

T_out

Temperature [°C]

74

5 50 4 45 4 40

[g/g]

Adsorption capacity

0.500

[gg/ml]

3 35 3 30

0.400

2 25 0.300 2 20 0.200

0:00 0:28 0:57 1:26 1:55 2:24 2:52 3:21 3:50 4:199 4:48 5:16

Adsorption n time [h:mm]

0.100

T_in

65

0.000 20%W-C CaCl2

30%M-CaCll2

Compo osite Type

55

[g/g]

0.500

Adsorption capacity

50%KG-CaCl2

[g//ml]

0.400 0.300

50

T02

30%M-CaC Cl2 Δpmax = 25.4 mmH2O 8 g/g amax = 0.278

T03 T04

45 40 35 30

0.200

25

0.100

20 0:00 0:28 0:57 1:26 1:55 2:24 2:52 3:21 3:50 4:19 4:48 5:16

0.000 Att550-MgCl2

20%W--MgCl2

30%M-MgC Cl2

50%KG-MgCl2

Com mposite Type

[g/g]

0.500

Adsorption capacity

Temperature [°C]

Att5550-CaCl2

T01

60

[gg/ml]

0.400 0.300 0.200 0.100 0.000 Att5550-MgSO4

20%W-M MgSO4

30%M-MgS SO4 50%KG-MgSO44

Com mposite Type

Fig. 5 Adsorption behavviour of modiified Att550-b based composites.

dynamic perrformances of o two compoosites compriising CaCl2 and in different raatio modifiedd substrates only o 30%-50% of o the total buulk material was loaded with w water. This unfavourable u e mass transfeer within the bulk b material is mainly m due too the increasse in the presssure drop (Δp), which w is approoximately 3-112 times higher in

Adsorption n time [h:mm]

Fig.. 6 Dynamicc performancee of modified Att550-based d com mposites.

com mparison to thhe pure Att550 based com mposite (Fig.. 4). Moreover, after a 5 dynam mic cycles th he adsorptionn capacities of thee investigatedd composites has declinedd substantially annd a physicaal degradations includingg volu ume expansiion of the bbulk materiaal within thee adsorber has beeen also observved. As A it has been already mentioned, due to thee gen nerally lower hydrothermaal stability an nd inefficientt dyn namic perform mance of the Att550 comp posites madee by using u porositty enhancing additives, fu urther studiess hav ve been purrsued only with those compositess com mprising puree Att550 substtrate. Besides B variattion of the suubstrate materrial, attemptss hav ve been also made m to comppare composiites based onn the kind of activve componennt used. Thuss beyond thee threee already meentioned salt hhydrates, otheers have beenn


employed in i this studiies. In the so far prepared composites the t degree off salt content varied v from 16 1 to 32 wt% depeending on thee type of the salt s hydrate used. u While those CaCl2, MgC Cl2 and LiCl based b composites exhibit a higher h salt content c up too 32wt%, thhose composites containing Na N 2SO4, Na2CO C 3, Na3PO4 and ZnSO4 show w only up to 17 wt%. The effect e of variaation of salt hydrrate on the adsorption behaviour b off the Att550 based compositess, including leevel of hydrattion, has been illuustrated on Fig. F 7. Whilee the equilibrrium adsorption capacities c off composites involving CaCl2 (0.474 g/g) and MgCl2 (00.401 g/g) onnly slightly differ d from each other, o compossite based onn the LiCl (0.657 g/g) exceeeds highly the two aforementiooned composites. The equilibbrium adsorpption capacityy of attapulgite based LiCl composite, with 30% salt content, attaained under paartial pressurre of 15 mbar was reported to be b 0.44 g/g [18]. In the studdies reported here partial presssure of 19.2 mbar has beeen applied. This probably conntributes to the t higher adsorption capaacity in this work compared too the literaturee results. The adssorption-desoorption kineetics and the associated water w uptakee/release ratiio of CaCl2 and LiCl basedd compositess are with 80%-82% at a a dehydration temperature below 100°C comparablle to each other. However inncreasing thee temperature to 110°C resultted a hydrothhermal degraddation of the LiCl L based compposite. Thus, for a possibble application of this particullar compositee, it is intendded to use loower charging tem mperature up to 90°C. Adsorption capacity [g/g]

0.700 0.600

In n addition in all the so farr developed composites, c a unifform distribuution of the salt hydratee across thee substrate granulaate have beenn observed. Representative R e resu ults obtained from f elementtal distribution n maps of Caa in different d regiions of puree and modified Att-basedd com mposites are presented p in Fig. 8. The T final characterisation of Attt550 basedd com mposites focuused on the long-term hydrothermal h l stab bility. For thaat cyclic tests in the range between 2500 and d 400 cycles have h been connducted on th he two finallyy seleected compposites, i.ee., Att550-CaCl2 andd Att5 550-LiCl. Results oobtained from fr thesee inveestigations arre summarizeed on Fig. 9. Those T resultss show that on CaC Cl2 compositee about 0.045 5 g/g decreasee in th he specific addsorption cappacity have been observedd afteer the first 10 1 cycles, thhat is about 9% loses inn com mparison to thhe initial valuue. Through the t followingg 400 0 cycles therre has been some fluctuaations in thee adsorption capaacities but it has stayed at a a properlyy stab ble level arouund 0.40 g/g. T This differs from f the LiCll based compositte result, whhich showed that almostt con nstant hydrothhermal properrties after 250 0 cycles. Thee decrease in speciific adsorptioon capacity was w less than 2% 2 thatt is within thee acceptable vvalue. Moreover, M concerning the dynamic perrformance off Att5 550-CaCl2 determined after 400 cycles thee max ximum tempeerature has shhifted slightly y from 60.7°C C to 59.5ºC 5 (Fig. 4) 4 and its wateer release at a dehydrationn tem mperature off 110°C waas about 87 7%. This iss com mparable to the t result obtained from the dynamicc testts done duringg the first 10 cycles. 3.2 Activated Caarbon Based C Composites

0.500

Analogous A t to those iinvestigationss done onn attaapulgite-basedd composites, detail physccio-chemical

0.400 0.300 0.200 0.100 0.000

Composite typee

Fig. 7

Comp posites of Att5550 and variouss salt hydrates.

Fig.. 8

Pattern off CaCl2 distrib bution across th he granule.

76

Experim mental Investiigation on Ad dsorption and d Surface Cha aracteristics of Salt Hydra ates and Hydrrophobic Porous Matrrix Based Com mposites Att550-C CaCl2

0.500 Adsorption capacity [g/g]

0.450 0.400 0.350 0.300 0.250 0.200 0.150 0.100 0.050 0.000 1

20

40

60 80 100 120 160 1 200 300 40 00 umber of cycles Nu

Att550-LiiCl

0.800

Tab ble 5 Adsorpttion behavior oof AC-based co omposites. Composite Salt content Salt depositt a (visual) [wt%] [g/g] PK-CaCl2 34.5 no 0.510 PK-MgCl2 30.2 no 0.602 28.8 slight 0.225 PK-MgSO4 AFA3-CaCl A 29.5 no 0.480 2 AFA3-MgCl A 28.7 no 0.461 2 ADA-CaCl2 33.5 high 0.236 32.4 high 0.321 ADA-MgCl A 2 AFA4-CaCl A 34.4 slight 0.426 2 AFA4-MgCl A 33.5 no 0.501 2 D47-CaCl2 30.6 no 0.452 D47-MgCl2 30.4 no 0.384 DGK-CaCl2 28.5 high 0.291 30.4 high 0.312 DGK-MgCl D 2 PK-CaCl2

Degree of water release [%]

Adsorption capacity [g/g]

0.700 0.600 0.500 0.400 0.300 0.200 0.100

AFA3-CaCl2

83

AFA4-CaCl2 D47-CaCl2

80 77 74 71 68 0.5

1 10 20 30 40 50 60 70 80 90 100 110 120 140 160 180 200 250

0.000 Nu umber of cycles

hermal stabilitty of Att550-C CaCl2 Fig. 9 Longg-term hydroth and Att550-L LiCl composite..

characterizations have been b executed on composites that have been b preparedd based on several s activvated carbon substtrates. The degreee of salt conntent among those t composites varied only slightly. How wever, the deggree of formaation of visual reest salt depoosits on the external surrface varied signnificantly. Paarticularly thhose composites based on ADA A and DG GK substratess showed a high h degree of sallt deposit. Reesults of thosee activated carrbon based compoosites have been summarizzed in Table 5. Due to thheir insufficieent mechanicaal stability, ADA A and DGK based compposites weree not drawnn in consideratioon for further investigationns. The channge in the degree d of waater release at a temperature range of 900-110°C in terms t of proocess cycles have been b determinned for those stable composites. The results are a illustratedd exemplarilly in Fig. 10.

Fig.. 10

1.5

2.5 3.5 5 4.5 Cycle num mber

5.5 5

Degree of o cyclic water release of PK--composites.

Acccording to thhese results, m minor changees have beenn observed on PK K and AFA4 based compo osites duringg the five process cycles. From F those teest results, itt is apparent that the twoo abo ove mentionned compoosites displlay similarr adsorption behaaviour exceptt for a mino or fluctuationn observed on PK-based P one. Yet th he PK-basedd com mposites posssess higher thermal staability. Thiss favo ourable featuure has beeen also obseerved in ourr prelliminary studdies [50]. Moreover, M P PK-substrate is a recycled activatedd carb bon and available a abbundantly fo or technicall app plication. Thuus due to theese reasons, the t followingg partt of this work w have bbeen focused d on furtherr chaaracterisation of this speciffic compositee. Lik ke the Att-bbased compoosites, the influence i off variiation of salt hydrates on tthe adsorption capacity off the respective PK-based P coomposites haas been alsoo


Adsorption capacity [g/g]

0.700 0.600 0.500 0.400 0.300 0.200 0.100 0.000

Composite type

Fig. 11

Com mposites of PK and various saalt hydrates.

70

PK-CaCl2 Δpmax = 1.8 mm mH2O amax = 0.472 g/g g

65 Temperature [°C]

60

T_in T01 T02 T03 T04 T05

55

T_outt

50 45 40 35 30 25 20 55 2:24 2:52 3:21 3:50 0:00 0:28 0:57 1:26 1:5 Adsorption time [h:mm] 70

PK-MgCll2 Δpmax = 5,4 mm mH2O amax = 0.478 g/g

65 60

T_in n T01 T02 T03

55 Temperature [°C]

investigatedd. The resultss are shown on Fig. 11. The level of the salt s content of o PK-based composites c shhows almost the same s pattern like it has been b observedd on Att-based coomposite. The LiCl (0.605 g/g) containing c coomposite exhhibits the highest adsorption capacity folllowed by CaCl2 (0.516 g/g) and a MgCl2 (00.502 g/g) onees. As it has been b expected thhe adsorptionn capacities of o the PK-based composites, independent of the naturee of the emplooyed salt hydratee, are higherr than that of o the Att-based composites. This can be explained in terms off the differences between thee two-substrrate materials in surface areaa, morphologgy and particcle size as it has been mentiooned in the experimental section of this paper. The dynam mic adsorptioon behaviourr of the PK-based composites has been allso assessed, in order to get information on the masss and heat traansfer withinn the bulk mateerial. Thosse results attained for PK-composiites of CaCl2 and MgCl2 are a shown on Fig. 12. Generallyy regardless of o the used active componeents, PK-based composites exxhibit a favoourable dynaamic performancee in compariison to thosee Att-based ones. o This is botth in terms of the maaximum attaained adsorption capacities c andd heat transfeer within the bulk b material. Inn addition too this favouurable adsorpption behaviour, PK-based coomposites doo not show any physical deeterioration, agglomeratiion or vollume expansion. This T confirmss that those obbservations made m

T04

50 45 40 35 30 25 20 0:00 0:28 0:57 0 1:26 1:55 5 2:24 2:52 3:21 3:50 Ad dsorption time [h h:mm]

Fig.. 12

Dynamicc performance of PK-based composites. c

by other studiess done on caarbonaceous support andd CaC Cl2 based com mposites [38]. Besides B the adsorption behaviour, a relativelyy low w-pressure droop within the bulk materrial has beenn observed on the PK-based coomposites (CaaCl2 = 1.8 & Mg gCl2 = 5.4 mm m H2O) than oof the respectiive Att-basedd com mposites (CaC Cl2 = 12.0 & MgCl2 = 2..1 mm H2O).. Durring the dynaamic process, it was possib ble to attain a max ximum tempperature of 68.1°C. Here the specificc heaat storage dennsity of PK-C CaCl2 compossite has beenn deteermined to bee 0.310 kWh//kg. A comparisonn of the dynnamic perform mance of thee PK--MgCl2composite with the similaarly treatedd substrate attappulgite (Attt-MgCl2) shows s thatt und dernearly thee same presssure drop, the t achievedd adsorption tempperature is hiigher by abou ut 11°C thann the latter compposite. This ccan be attrib buted to thee app prox. 20% redduced adsorpption capacity y (0.372 g/g))

78


mecchanical graanular strenggth tests do one on thee com mposite afterr 25 static and 3 tim mes dynamicc adsorption cyclees. Finally F long-tterm hydrothhermal stability tests havee beeen performed on the two ffinally selecteed, PK-CaCl2 and d PK-LiCl, coomposites. Thhe results of the t long-term m hyd drothermal sttudies carriedd out on PK K-CaCl2 andd PK--LiCl are dispplayed on Figg. 15. While, W PK-CaaCl2 shows a distinct deccline after 100 cycles followed by a roughlyy constant staability over a broad range of cycles, a siggnificant flucctuation havee beeen observed on PK-LiCl.

4. Conclusion C n In n these stuudies a widde range off compositess con nsisting of seeveral salt hyydrates and hydrophobicc substrates were prepared andd characterizeed. From thee variious substratees used to inccorporate the active salt PK-CaCl2

0 0.600 Adsorption capacity [g/g]

and the disttinctly differeent grain size distributionn as well as the difference d in particle geom metry of the two substrates. On the otther hand froom the surface analysis done d after 3 dynaamic cyclic operations o it is clear that as a result of thhe incorporatted salt hydrrate the specific surface areaas of the coomposites haave considerrably decreased coompared withh the pure suubstrate (Fig. 13). Also a signiificant pore size s shift from m macro/messo to micro has beeen observed. However,, PK-based coomposites exhibit a very large l pore diameteer. Thus, no capillary conndensation caan be expected to occur. Instead as it is exxpected onlyy the influence off the surface area a on the effficiency of water w adsorption capacities c of the t compositee was observeed. Moreoverr, improvemeents in granules strength off the composites have been achieved a throough salt hyddrate incorporatioon comparedd to the sttarting substtrate material. Figg. 14 shows thhe results attaained on

0 0.500 0 0.400 0 0.300 0 0.200 0 0.100 0 0.000 1

40

60 80 100 120 160 200 300 400 Numberr of cycles

PK-LiCl

0 0.700

Surfface propertiess vs adsorption n capacity. Adsorption capacity [g/g]

Fig. 13

20

0 0.600 0 0.500 0 0.400 0 0.300 0 0.200 0 0.100

1 10 20 30 40 50 60 70 80 90 100 110 120 140 160 180 200 250

0 0.000 Numberr of cycles

Fig.. 15 Long-teerm hydrotherrmal stability of PK-CaCl2 and d PK-LiCl composites. Fig. 14 Mech hanical strength of compositte vs substrate..

Experimental Investigation on Adsorption and Surface Characteristics of Salt Hydrates and Hydrophobic 79 Porous Matrix Based Composites

hydrate, Att550 and PK have exhibited optimum characteristic features including mechanical and thermal stability. Based on the gravimetric and volumetric adsorption capacities and dynamic energy densities as well as the degree of establishment of a favourable hydration level CaCl2, MgCl2 and LiCl of PK based composites with 0.516 g/g, 0.502 g/g and 0.605 g/g, respectively are found to be the most promising candidates as a storage material for low temperature application. With a minor deviation comparable results with 0.474 g/g, 0.401 g/g and 0.657 g/g have been also attained from Att550 based composites in combination with those three salt hydrates respectively. The experimental results achieved so far have revealed that under pre-defined process conditions using low temperature in the range between 90-110°C a dehydration of the finally selected composites up to 87%, in comparison to the initial water uptake efficiency, was possible. However, more investigations, particularly on identification of optimum process parameters to prevent corrosion have to be made in order to make a firm selection of the composite to use in a technical sorption closed storage system. Thus, further studies will be done on extensive thermochemical heat storage measurements including 1000-2000 cyclic tests under real and controlled process conditions.

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

Acknowledgements The research leading to these results has received funding from the Federal Ministry for Economic Affairs and Energy of Germany under Grant No. 03ESP259 for which it is gratefully acknowledged.

References [1]

[2]

K. C. Ng, H. T. Chua, C. Y. Chung, C. H. Loke, T. Kashiwagi, A. Akisawa and B. B. Saha, Experimental investigation of the silica gel–water adsorption isotherm characteristics, Applied Thermal Engineering 21 (2001) 1631-1642. J. Janchen, D. Ackermann, H. Stach and W. Brosicke, Studies of water adsorption on zeolite and modified

[12]

[13]

[14]

mesoporous materials for seasonal storage of solar heat, Solar Energy 76 (2004) 339-344. Y. I. Aristov, M. Tokarev, A. Freni, I. S. Glaznev and G. Restuccia, Kinetics of water adsorption on silica Fuji Davison RD, Microporous and Mesoporous Materials 96 (2006) 65-71. S. K. Henninger, F. P. Schmidt and H. M. Henning, Water adsorption characteristics of novel materials for heat transformation applications, Applied Thermal Engineering 30 (2010) 1692-1702. D. Dicaire and F. H. Tezel, Regeneration and efficiency characterization of hybrid adsorbent for thermal energy storage of excess and solar heat, Renewable Energy 36 (2011) 986-992. J. Jänchen, K. Schumann, E. Thrun, A. Brandt, B. Unger and U. Hellwig, Preparation, hydrothermal stability and thermal adsorption storage properties of binderless zeolite beads, International Journal of Low-Carbon Technologies 00 (2012) 1-5 L. Bonaccorsi, P. Bruzzaniti, L. Calabrese, A. Freni, E. Proverbio and G. Restuccia, Synthesis of SAPO-34 on graphite foams for adsorber heat exchangers, Applied Thermal Engineering 61 (2013) 848-852. N. Yu, R. Z. Wang and L. W. Wang, Sorption thermal storage for solar energy, Progress in Energy and Combustion Science 39 (2013) 489-514. J. Jänchen and H. Stach, Shaping adsorption properties of nano-porous molecular sieves for solar thermal energy storage and heat pump applications, Solar Energy 104 (2014) 16-18. A. Sapienza, S. Santamaria, A. Frazzica, A. Freni and Y. I. Aristov, Dynamic study of adsorbers by a new gravimetric version of the Large Temperature Jump method, Applied Energy 113 (2014) 1244-1251. J. Jänchen, T. H. Herzog, K. Gleichmann, B. Unger, A. Brandt, G. Fischer and H. Richter, Performance of an open thermal adsorption storage system with Linde type A zeolites: Beads versus honeycombs, Microporous and Mesoporous Materials 207 (2015) 179-184. A. Solé, I. Martorell and L. F. Cabeza, State of the art on gas-solid thermochemical energy storage systems and reactors for building applications, Renewable and Sustainable Energy Reviews 47 (2015) 386-398. J. Jänchen, T. H. Herzog and E. Thrun, Natural zeolites in thermal adsorption storage and building materials for solar energy utilization in houses uses, in: Third Southern African Solar Energy Conference, 11-13 May, 2015. A. Lass-Seyoum, Borozdenko D, Friedrich T, Langhof T, Mack S. Practical Test on a Closed Sorption Thermochemical Storage System with Solar Thermal Energy, Energy Procedia 91 (2016) 182-189.

80


[15] Y. I. Aristov, Challenging offers of material science for adsorption heat transformation: A review, Applied Thermal Engineering 50 (2013) 1610-1618. [16] Y. I. Aristov, G. Restuccia, G. Cacciola, V. N. Parmon, A family of new working materials for solid sorption air conditioning systems, Applied Thermal Engineering 22 (2002) 191-204. [17] Y. I. Aristov, G. Di Marco, M. M. Tokarev and V. N. Parmon, Selective water sorbents for multiple applications, 3. CaCl2 solution confined in micro- and mesoporous silica gels: Pore size effect on the “solidification-melting” diagram, Reaction Kinetics and Catalysis Letters 61 (1997) 147-154. [18] H. J. Chen, Q. Cui, Y. Tang, X. J. Chen and H. G. Yao, Attapulgite based LiCl composite adsorbents for cooling and air conditioning applications, Applied Thermal Engineering 28 (2008) 2187-2193. [19] I. A. Simonova, Simonova, A. Freni, G. Restuccia and Y. I. Aristov, Water sorption on composite “silica modified by calcium nitrate”, Microporous and Mesoporous Materials 122 (2009) 223-228. [20] S. Nakabayashi, K. Nagano, M. Nakamura, J. Togawa, A. Kurokawa, Improvement of water vapor adsorption ability of natural mesoporous material by impregnating with chloride salts for development of a new desiccant filter, Adsorption 17 (2011) 675-686. [21] I. Glaznev, I. Ponomarenko, S. Kirik and Y. I. Aristov, Composites CaCl2/SBA-15 for adsorptive transformation of low temperature heat: Pore size effect, International Journal of Refrigeration 34 (2011) 1244-1250. [22] V. N. Deshmukh and S. V. Joshi, Review: Use of composite adsorbents in adsorption refrigeration, International Journal of Innovative Technology and Creative Engineering 2 (2012) 11-16. [23] Y. I. Aristov and L. G. Gordeeva, Composites “salt inside porous matrix” for adsorption heat transformation: A current state-of-the-art and new trends, Int. J. Low-Carbon Tech. 7 (2012) 288-302. [24] J. V. Veselovskaya, M. M. Tokarev, A. D. Grekova and L. G. Gordeeva, Novel ammonia sorbents “porous matrix modified by active salt” for adsorptive heat transformation: The ways of adsorption dynamics enhancement, Applied Thermal Engineering 37 (2012) 87-94. [25] Bu Xianbiao, Lu Zhenneng and W. Lingbao, Preparation of composite adsorbent with high performance of heat and mass transfer, Chinese Science Bulletin 58 (2013) 3709-3714. [26] L. Gordeeva, A. Grekova, T. Krieger and Y. Aristov, Composites “binary salts in porous matrix” for adsorption heat transformation, Applied Thermal Engineering 50 (2013) 1633-1638.

[27] H. Liu, K. Nagano, A. Morita, J. Togawa and M. Nakamura, Experimental testing of a small sorption air cooler using composite material made from natural siliceous shale and chloride, Applied Thermal Engineering 82 (2015) 68-81. [28] D. Zhu, H. Wu and S. Wang, Experimental study on composite silica gel supported CaCl2 sorbent for low grade heat storage, International Journal of Thermal Sciences 45 (2006) 804-813. [29] H. Wu, S. Wang and D. Zhu, Effects of impregnating variables on dynamic sorption characteristics and storage properties of composite sorbent for solar heat storage, Solar Energy 81 (2007) 864-871. [30] H. Liu, K. Nagano and J. Togawa, A composite material made of mesoporous siliceous shale impregnated with lithium chloride for an open sorption thermal energy storage system, Solar Energy 111(2015) 186-200. [31] A. Jabbari-Hichri, S. Bennici and A. Auroux, Enhancing the heat storage density of silica–alumina by addition of hygroscopic salts (CaCl2, Ba(OH)2, and Li(NO3)), Solar Energy Materials and Solar Cells 140 (2015) 351-360. [32] Y. Yuan, H. Zhang, F. Yang, N. Zhang and X. Cao, Inorganic composite sorbents for water vapor sorption: A research progress, Renewable and Sustainable Energy Reviews 54 (2016) 761-776. [33] M. Tokarev, L. Gordeeva, V. Romannikov, I. Glaznev and Y. I. Aristov, New composite sorbent CaCl2 in mesopores for sorption cooling/heating, International Journal of Thermal Sciences 41 (2002) 470-474. [34] L. Gordeeva, I. S. Glaznev, E. V. Savchenko, V. V. Malakhov, and Y. I. Aristov, Impact of phase composition on water adsorption on inorganic hybrids “salt/silica”, Journal of Colloid and Interface Science 301 (2006) 685-691. [35] J. Jänchen, D. Ackermann, E. Weiler, H. Stach and W. Brösicke, Calorimetric investigation on zeolites, AlPO4’s and CaCl2 impregnated attapulgite for thermochemical storage of heat, Thermochimica Acta. 434 (2005) 37-41. [36] M. Spiridon, O. R. Hauta, M. S. Secula and S. Petrescu, Preparation and characterization of some porous composite materials for water vapor adsorption, Revista de Chimie 63 (2012) 711-714. [37] M. Druske, A. Fopah-Lele, K. Korhammer, H. U. Rammelberg, N. Wegscheider, W. Ruck and T. Schmidt, Developed materials for thermal energy storage: Synthesis and characterization, Energy Procedia 61 (2014) 96-99. [38] K. Korhammer, M. M. Druske, A. Fopah-Lele, H. U. Rammelberg, N. Wegscheider, O. Opel. T. Osterland and W. Ruck, Sorption and thermal characterization of composite materials based on chlorides for thermal energy storage, Applied Energy 162 (2016) 1462-1472.

Experimental Investigation on Adsorption and Surface Characteristics of Salt Hydrates and Hydrophobic 81 Porous Matrix Based Composites [39] S. K. Henninger, G. Munz, K. F. Ratzsch and P. Schossig, Cycle stability of sorption materials and composites for the use in heat pumps and cooling machines, Renewable Energy 36 (2011) 3043-3049. [40] B. Michel, P. Neveu and N. Mazet, Comparison of closed and open thermochemical processes, for long-term thermal energy storage applications, Energy 72 (2014)702-716. [41] A. Solé, L. Miró, C. Barreneche, I. Martorell and L. F. Cabeza, Corrosion of metals and salt hydrates used for thermochemical energy storage, Renewable Energy 75 (2015) 519-523. [42] A. Solé, L. Miró, C. Barreneche, I. Martorell and L. F. Cabeza, Corrosion test of salt hydrates and vessel metals for thermochemical energy storage, Energy Procedia 48 (2014) 431-435. [43] A. J. Farrell, B. Norton and D. M. Kennedy, Corrosive effects of salt hydrate phase change materials used with aluminium and copper, Journal of Materials Processing Technology 175 (2006) 198-205. [44] A. Häberle, F. Trausel, A. J. de Jong and R. Cuypers, A review on the properties of salt hydrates for thermochemical storage, Energy Procedia 48 (2014) 447-452. [45] K. Posern and C. Kaps, Humidity controlled calorimetric investigation of the hydration of MgSO4 hydrates, Journal of Thermal Analysis and Calorimetry 92 (2008) 905-909. [46] L. Greenspan, Humidity fixed points of binary saturated aqueous solutions, Journal of Research of the National Bureau of Standards 81 (1977) 89-96.

[47] J. T. Kelly and A. S. Wexler, Thermodynamics of carbonates and hydrates related to heterogeneous reactions involving mineral aerosol, Journal of Geophysical Research 110 (2005) [48] S. Metzger and J. Lelieveld, Reformulating atmospheric aerosol thermodynamics and hygroscopic growth into fog, haze and clouds, Atmos Chem Phys. 7 (2007) 3163-3193. [49] N. M. Wajid, B. Mempouo, A. Dodo, S. Omer and S. B. Riffa, Experimental study of an adsorption heat storage systems for building applications, Renewable Bioresources 41 (2016) 2052-2059. [50] A. Lass-Seyoum, M. Blicker, D. Borozdenko, T. Friedrich and T. Langhof, Transfer of laboratory results on closed sorption thermo-chemical energy storage to a large-scale technical system, Energy Procedia 30 (2012) 310-320. [51] J. Straszko, M. Olszak-Humienik and J. Możejko, Kinetics of thermal decomposition of ZnSO4•7H2O, Thermochimica Acta 292 (1997) 145-150. [52] K. E. N. Tsoukpoe, H. U. Rammelberg, A. Fopah-Lele, K. Korhammer, A. B. Watts, T. Schmidt and K. L. R. Wolfgang, A review on the use of calcium chloride in applied thermal engineering, Applied Thermal Engineering 75 (2015) 513-531. [53] R. G. Oliveira, R. Z. Wang and T. Xian Li, Adsorption characteristic of methanol in activated carbon impregnated with lithium chloride. Chemical Engineering & Technology 33 (2010) 1679-1686.