Comparison of the Effect of Coaddition of Li Compounds and Addition of a Single Li Compound on Reactivity and Structure of Magnesium Hydroxide.

Mg(OH)2 is a chemical heat storage material that is studied for the utilization of 300-350 °C waste heat. In this study, LiCl and LiOH were coadded to Mg(OH)2, and the reactivity and structural evolution were investigated. In the hydration of samples at 200 °C subsequent to dehydration at 270 °C, Mg(OH)2 with coadded LiCl and LiOH showed excellent hydration reactivity, with a heat output density of 1053 kJ kg-1. The coaddition of LiCl and LiOH enhanced both the dehydration and the hydration reactivity of Mg(OH)2. X-ray diffraction analysis indicated that the addition of LiOH to Mg(OH)2 promoted the decomposition of Mg(OH)2 and the diffusion of water on the surface of Mg(OH)2, whereas the addition of LiCl to Mg(OH)2 promoted these processes in the bulk phase of Mg(OH)2.

Introduction

In recent years, global environmental problems, such as global warming and depletion of energy resources, have become critical issues. Many techniques have been developed toward alleviating these problems. For instance, power generation by using renewable energy sources, such as solar, wind, and geothermal energy, has been researched. However, these techniques are limited by restrictions related to process implementation, low efficiency, and the gap between energy supply and demand. Conditions for solar and geothermal power generation are especially limited by weather and location. This energy gap can be compensated by utilizing waste heat (industrial waste heat, geothermal heat, and solar heat) with the help of chemical heat storage materials. Chemical heat storage materials can store thermal energy semi-permanently, implying that the stored thermal energy can be exploited for power generation as desired.

Heat storage techniques can be classified as sensible, latent, and chemical heat storage techniques. Sensible and latent heat storage techniques are very simple, and therefore, can be applied for practical use, but the period of energy storage is short. In contrast, the chemical heat storage technique is associated with much larger energy density (0.5–1 kW h/kg), and the period of energy storage is theoretically unlimited.[1−4] Thus, chemical heat storage allows for long-distance transport of the stored heat, which can then be reused whenever required.

Chemical heat storage materials can utilize waste heat via chemical reactions, especially gas–solid reactions.[1−4] Several reaction systems such as Ca(OH)2/CaO,[5−7] MgCl2·6H2O/MgCl2, and MgSO4·7H2O/MgSO4[8,9] have been investigated. For storage of heat in the 450–550 °C range, the dehydration of Ca(OH)2 is suitable, but it is unsuitable for storage of low-medium temperature heat (200–300 °C). Despite the low temperature of ∼100 °C required for dehydration of salt hydrates such as MgSO4·7H2O and MgCl2·6H2O, the reversibility of this process for these materials is quite low, where hydration (heat output) requires a long time. The Mg(OH)2/MgO system has been studied for the storage of low-medium (300–350 °C) temperature heat.[10−15] The dehydration temperature for Mg(OH)2 is quite low compared with that for Ca(OH)2, and the hydration reactivity of MgO is much higher than that of MgSO4. The Mg(OH)2/MgO reaction is represented by eq where ΔHr0 is the enthalpy change for the dehydration reaction, and ΔHcnd0 is the enthalpy change for the condensation of water.

In previous studies, several additives, such as cetyl trimethyl ammonium bromide, lithium chloride, and lithium hydroxide-modified Mg(OH)2, have been shown to enhance the reactivity of Mg(OH)2.[11−15] In particular, LiCl-added Mg(OH)2 and LiOH-added Mg(OH)2 were much more efficiently dehydrated at 270–300 °C than pure Mg(OH)2.[12−14] Therefore, we believe that the Mg(OH)2/MgO system has a significant potential for thermal energy storage at 200–300 °C, which constitutes a large part of the industrial waste heat in Japan. However, the hydration rate of MgO with these added salts and that of pure MgO at 180–200 °C is sluggish. Thus, applicable reaction conditions for the Mg(OH)2/MgO system are still limited.

Knoll et al. studied the relationship between the specific surface area and the reactivity of the Mg(OH)2/MgO system.[16] However, there are few studies on the effect of additives on the reactivity of materials; one such study was carried out by Shkatulov et al.[15] The effect of LiCl or LiOH addition on the reactivity and structure of the Mg(OH)2/MgO system has not yet been reported. If the mechanism can be elucidated, it can be extended to other chemical heat storage materials with added salts.

In this study, we propose the use of LiCl and LiOH coadded Mg(OH)2 to enhance the reactivity of Mg(OH)2 and investigate the effect of adding the Li compounds on the reactivity and structure of Mg(OH)2. In several similar studies, the effect of two different Li compounds (LiCl and LiOH) on the reactivity of the Mg(OH)2/MgO system has not been reported. Moreover, the role of LiCl or LiOH on the dehydration reactivity of Mg(OH)2, which has not been revealed earlier, was clarified in this study.

Results and Discussion

Comparison of the Effect of Addition of a Single Li Compound Versus Coaddition of Li Compounds on the Dehydration Reactivity of Mg(OH)2

Figure shows the dehydration behavior of all samples when heated to 600 °C. The y-axis represents the mole fraction of Mg(OH)2 based on eq , and the x-axis represents the temperature. The purple line shows the dehydration behavior of Mg(OH)2–W. The dehydration of Mg(OH)2 with added LiCl and/or LiOH progressed at lower temperatures. Interestingly, the coaddition of LiCl and LiOH to Mg(OH)2 (especially L10/LO10) resulted in dehydration at lower temperatures than in the case of single addition (L10, LO20). For example, at 270 °C, the slope of the thermogravimetric (TG) curve of L10/LO10 was significantly larger than that observed for any other sample. These results show that the coaddition of LiCl and LiOH to Mg(OH)2 enhanced the dehydration reactivity of Mg(OH)2 to a greater extent than the addition of LiCl or LiOH alone.

Figure 1

Dehydration behavior of all samples heated to 600 °C.

Although Mg(OH)2–W did not undergo complete dehydration, the X-ray diffraction (XRD) patterns of all of the samples heated to 600 °C showed that no peaks were derived from Mg(OH)2 (Figure S1). A previous study showed that pretreatment at temperatures lower than 600 °C did not lead to the complete removal of structurally bound water.[17] An IR analysis revealed that the peak at around 3700 cm–1 corresponding to isolated hydroxyl groups on the Mg(OH)2 surface did not disappear (Figure S2).[18,19] This peak was shifted to 3730 cm–1 after Mg(OH)2–W was heated at higher than 400 °C. This behavior clearly indicates that the hydroxyl groups of Mg(OH)2 surface were converted to the ones of MgO surface at higher than 400 °C. However, when we carefully looked at the absorption band at around 3730 cm–1, a small component due to the hydroxyl groups of Mg(OH)2 still existed at 3700 cm–1.[18,19] Therefore, this result might cause the incomplete dehydration of Mg(OH)2. Thus, the mole fraction did not reach 0%, as shown in Figure . The dehydration conversions of all of the samples, however, were more than 90%. Hence, we considered that the dehydration was complete. Further research is required to elucidate this effect in detail and to investigate other samples.

Table shows the peak temperature for dehydration of all of the samples derived from the differential thermogravimetric (DTG) curve (Figure a–e), based on Figure . The peak temperature is the temperature at which the dehydration rate is the fastest. Here, the dehydration rate was calculated from the measured weight change and temperature and is the sample weight differentiated by the temperature. A lower peak temperature means that the sample can be dehydrated at a lower temperature, which indicates that the sample is suitable for lower temperature heat storage. L10/LO10 showed the lowest peak temperature of all of the samples. Therefore, L10/LO10 is expected to store low-grade waste heat more efficiently than L10 and LO20, which have been studied previously.[12−14] These results indicated that the temperature range at which Mg(OH)2 can be efficiently dehydrated was increased by the coaddition of LiCl and LiOH. Moreover, a detailed investigation of the dehydration kinetics analysis at 200–300 °C is required in future work.

Table 1

Peak Temperature of all Samples

sample	peak temperature [°C]
Mg(OH)2–W	371
L10	317
LO20	342
L5/LO5	314
L10/LO10	305

Figure 2

DTG curve of (a) Mg(OH)2–W, (b) L10, (c) LO20, (d) L5/LO5, and (e) L10/LO10 heated to 600 °C.

Effect of Coaddition of the Li Compounds on Hydration Reactivity of MgO

Figure shows the dehydration and hydration behavior of all of the samples. The sample was first heated at 270 °C for 30 min and subsequently hydrated at 110 °C for 80 min under a mixture of Ar gas and saturated water vapor at 57.8 kPa. Under these conditions, the Δxd value for Mg(OH)2–W and that for LO20 was only 9.50 and 55.2%, respectively. The Δxd value for L10, L5/LO5, and L10/LO10 was 89.7, 89.0, and 95.0% respectively, indicating that these three samples are suitable for use under the stated conditions. However, L10/LO10 was dehydrated the fastest of all of the samples, similar to the case in Figure . All samples could be reversibly (de)hydrated. Thus, Mg(OH)2 with coadded LiCl and LiOH should be useful as a chemical heat storage material.

Figure 3

Dehydration and hydration behavior of all of the samples; Td = 270 °C, Th = 110 °C, and PH = 57.8 kPa.

Table shows the reaction conversion for all samples. The Δx2 value for L10 was much higher than that of Mg(OH)2–W and LO20. This is probably because of the high hygroscopicity of LiCl. The mole fraction change Δx2 indicates the formation of the n hydrate of LiCl or LiOH, as expressed in eqs and 4. This result demonstrates that LiCl is easily converted into LiCl·nH2O than LiOH, so that the hydration of MgO is promoted.[13]nH2O in LiCl·nH2O probably interacted with the interface between MgO and LiCl·nH2O.[13]

Table 2

Reaction Conversion for All Samples; Td = 270 °C, Th = 110 °C, and PH = 57.8 kPa

sample	Δxd [%]	Δx1 [%]	Δx2 [%]
Mg(OH)2–W	9.50	7.80	2.30
L10	89.7	98.0	52.5
LO20	55.2	54.5	4.40
L5/LO5	89.0	81.1	26.4
L10/LO10	95.0	90.1	51.1

Figure shows the hydration behavior of L10, L5/LO5, and L10/LO10 at 110 and 200 °C for 80 min after dehydration at 270 °C under Ar flow for 30 min. Although L10 and L5/LO5 were well-hydrated at 110 °C, only 30–40% hydration was achieved at 200 °C. Generally, hydration does not progress effectively at higher temperatures because it is an exothermic reaction. Thus, it is natural that the hydration reactivity of the samples declined with the increasing temperature.[13] However, L10/LO10 was well-hydrated at 110 and 200 °C, although the apparent hydration reaction rate decreased at 200 °C. Therefore, the hydration reactivity of L10/LO10 was higher at high hydration temperatures as compared with that of the other samples.

Figure 4

Hydration behavior of L5/LO5, L10/LO10, and L10 at 110 and 200 °C; Td = 270 °C, Th = 110 and 200 °C, and PH = 57.8 kPa. The blue line shows the hydration behavior at 110 °C, and the red line shows that at 200 °C. “◆”, “●”, and “■” indicate the hydration behavior of L5/LO5, L10/LO10, and L10, respectively.

Figure shows the heat output density for the hydration reaction of all samples at three different hydration temperatures (110, 170, and 200 °C); the orange dotted line shows the heat output density of 1000 kJ kg–1, which is the recommended target value considering the economy and repayment of the initial cost in Japan.[20] The heat output density at 110 °C exceeded 1000 kJ kg–1, except in the case of LO20 (only 55.2% dehydration of LO20 was achieved at 270 °C; Figure ). However, except for LO20 and L10/LO10, the heat output density decreased dramatically with increasing temperature for the reason mentioned above. The only sample for which the heat output density exceeded 1000 kJ kg–1 at all temperatures was L10/LO10 (1185, 1069, and 1053 kJ kg–1 at 110, 170, and 200 °C, respectively). The amount of LiCl and LiOH might be insufficient for improving the hydration reactivity of MgO for L5/LO5 under the stated conditions. Thus, L10/LO10 had a high hydration reactivity even at a high temperature (200 °C), which can reduce heat loss during the dehydration and hydration cycles. Thus, this sample can release high-temperature and high-qualified heat in the heat output operation. This feature is very valuable for the heat output operation and is much better than that of L10, which was previously studied.[12,13] These results show that the dehydration and hydration reactivity of Mg(OH)2/MgO can be more effectively enhanced by the coaddition of LiCl and LiOH than by the individual addition of LiCl or LiOH. In the former case, the dehydration temperature shifts more dramatically toward a lower level, and the hydration reactivity of MgO is enhanced to a greater extent, as shown in Figures and 6. MgCO3/MgO with an added LiNO3–KNO3 eutectic showed much better reactivity than the pure MgCO3/MgO system in a previous study.[21] Hence, the addition of binary salts can improve the reactivity of other chemical heat storage materials.

Figure 5

Heat output density at various hydration temperatures for all of the samples; Td = 270 °C; Th = 110, 170, and 200 °C; and PH = 57.8 kPa.

Figure 6

Hydration behavior of all samples at 200 °C; Td = 350 °C, Th = 200 °C, and PH = 57.8 kPa.

The heat output density of LO20 was almost constant, independent of the temperature. Therefore, LiOH possibly has good effects on the hydration reactivity of MgO, such as maintaining high hydration reactivity at high hydration temperatures. Thus, LiCl and LiOH might exert independent effects on the hydration of MgO. Although MgO can plausibly absorb water because of the hygroscopicity of the added LiCl,[13] the hydration of MgO at high temperature may be promoted by the LiOH addition (Figure ).

After dehydration at 350 °C, hydration of the samples was evaluated to clarify the effect of the addition of LiCl and/or LiOH on the hydration reactivity of MgO. Figure shows a comparison of the reactivity of all samples when hydrated at 200 °C for 80 min under a mixture of water vapor (at 57.8 kPa) and Ar gas after dehydration at 350 °C for 30 min. The Δx1 value for Mg(OH)2–W and L10 was only 2.89 and 16.8%, respectively. On the other hand, the value of Δx1 for LO20 was 49.7%. The Δx1 value for L5/LO5 and L10/LO10 was 89.4 and 94.3%, respectively, and the hydration of L5/LO5 and L10/LO10 was almost complete at 200 °C, even after dehydration at 350 °C. The hydration reactivity of MgO was especially improved by LiCl and LiOH coaddition. From this result, we conclude that the coaddition of LiCl and LiOH enhanced not only the dehydration reactivity, but also the hydration reactivity at 200 °C, of Mg(OH)2/MgO. In particular, LiOH addition promoted this enhancement at 200 °C to a greater extent than LiCl addition. Surprisingly, Δx1 for L5/LO5 after dehydration at 350 °C was much higher than that after dehydration at 270 °C. After dehydration at 350 °C, the specific surface area might have increased, because of which the reactive area for the hydration of MgO was larger than that in the case of dehydration at 270 °C.[22] As the hydration of L10/LO10 after dehydration at 270 °C was almost complete, the reactivity of L10/LO10 was higher than that of L5/LO5.

This result shows that LiOH addition promoted the hydration of MgO at 200 °C. Interestingly, the coaddition of LiCl and LiOH promoted the dehydration and hydration of Mg(OH)2/MgO to a much greater extent than the single addition of LiCl or LiOH. However, further studies are required to understand why LiCl and LiOH coaddition greatly promoted the hydration of MgO at 200 °C.

Figure shows the hydration behavior of L10/LO10 at 110 °C with variation in the water vapor pressure (7.4–57.8 kPa); the relationship between the water vapor pressure and saturated temperature is shown in Table . As shown in Figure , the extent of hydration of MgO decreased when the water vapor pressure was lowered. This behavior agrees with the fact that the number of water molecules decreased at lower water vapor pressure and with a previous study of L10.[13] For PH = 19.9, 12.3, and 7.4 kPa, the Δx1 value for L10/LO10 was only 59.1, 27.1, and 11.5%, respectively. Below PH = 19.9 kPa, the hydration reactivity of L10/LO10 declined dramatically. Thus, the coaddition of LiCl and LiOH could not enhance the hydration reactivity at low water vapor pressure.

Figure 7

Hydration behavior of L10/LO10 at various water vapor pressures; Td = 270 °C, Th = 110 °C, and PH = 7.4–57.8 kPa.

Table 6

Relationship between PH and Saturated Vapor Temperature (Ts)

Ts [°C]	PH2O [kPa]
40	7.4
50	12.3
60	19.9
70	31.2
80	47.4
85	57.8

Heat output at higher temperature can be achieved by increasing the water vapor pressure.[23] However, the hydration reactivity at low water vapor pressure must be enhanced for the utilization of low-grade waste heat. For instance, the temperature of waste heat is supposed to be as low as 30 °C. Therefore, if the low water vapor pressure can be used efficiently in the heat output operation, the application scope of the Mg(OH)2/MgO system can be widened, for example, a cooling system.[24]

Sample Characterization by XRD Analysis for Evaluating the Effect of Addition of Li Compounds on the Structure of Mg(OH)2

Figure shows the XRD patterns of all of the samples before reaction. This figure shows the peaks of the brucite structure of Mg(OH)2 for all samples. The data indicate the dominance of the brucite structure of Mg(OH)2 in all of the samples, although LiCl and/or LiOH were added to Mg(OH)2. No peaks of LiCl were detected in the LiCl-added samples (L10, L10/LO10, and L5/LO5) because the LiCl species were probably dissolved in the Mg(OH)2 phase, well-dispersed throughout the Mg(OH)2 particles, or converted into solution, which is not detectable during the experiment. In the future work, we must investigate the presence of LiCl by techniques such as XAFS. LiOH might be partially converted to Li2CO3 during preparation, which agrees with the fact that LiOH is generally contaminated with Li2CO3.[25] This accounts for the detection of the Li2CO3 peak.

Figure 8

XRD patterns of all samples.

Because of the difficulty in detecting any changes in the XRD pattern in Figure , two of the strongest peaks derived from the Mg(OH)2 (001) and (002) planes of each sample are expanded in Figure a,b. The black dotted line indicates the position of the Mg(OH)2 (001) or (002) peaks of Mg(OH)2–W. For Mg(OH)2 with added LiCl and/or LiOH, the peaks of the (001) and (002) planes shifted toward lower and higher angle, respectively.

Figure 9

Expanded XRD patterns of all samples; the black dotted line shows the peak position of Mg(OH)2 (001) or (002) plane of Mg(OH)2–W: (a) (001) plane and (b) (002) plane.

We evaluated this change from only two planes, but the evaluation of all measured planes is required. The lattice parameters and volumes of all samples were calculated based on all of the measured XRD peak positions to clarify the factors causing the peak shift. Table shows the calculated lattice parameters and volumes for all samples. These values were calculated by the least-squares method using all measured XRD peak positions. The measurement was carried out in triplicate to reduce the error.

Table 3

Calculated Lattice Parameters and Lattice Volume of All Samples Based on Measured Peak Position

sample	lattice parameter: a [Å]	lattice parameter: c [Å]	lattice volume: V [Å3]
Mg(OH)2–W	3.145 ± 0.003	4.763 ± 0.007	47.12 ± 0.15
L10	3.147 ± 0.001	4.787 ± 0.031	47.42 ± 0.30
LO20	3.146 ± 0.004	4.783 ± 0.007	47.33 ± 0.08
L5/LO5	3.146 ± 0.002	4.775 ± 0.005	47.27 ± 0.12
L10/LO10	3.145 ± 0.001	4.766 ± 0.013	47.15 ± 0.16

The lattice parameters and volumes of Mg(OH)2 with added LiCl and/or LiOH were larger than those of Mg(OH)2–W. Thus, the values seemed to increase by addition of LiCl and/or LiOH, where Li+ ions may have been substituted into the Mg2+ sites.[26] The effective ionic radius of Li+ (0.76 Å) is larger than that of Mg2+ (0.72 Å),[27] indicating that the lattice parameters increased because of Li+ ion substitution. The Li+ ion substitution could disrupt the charge balance at the substituted sites because of the replacement of a divalent cation (Mg2+) by a monovalent cation (Li+). Therefore, lattice defects such as oxygen vacancies might be generated to compensate the disruption in the charge balance caused by Li+ ion substitution. At such sites, the decomposition of Mg(OH)2 may occur readily.[14]

The increase in lattice parameters was small. However, even this small difference might influence the reactivity of the Mg(OH)2/MgO system. Further, the effective ionic radii of Li+ and Mg2+ are very close to each other, and even a slight change in the lattice parameters is reasonable.

The increase in the lattice parameter c, as shown in Table , is indicative of expansion of the interlayer distance in the brucite structure based on the first-principles calculation, as cited from a previous article.[28] This plausibly promoted release/insertion of water molecules from/into the structure of Mg(OH)2/MgO, which could lead to the enhancement of the rate of dehydration and hydration of Mg(OH)2/MgO.

In summary, the (1) lattice defects and (2) expansion of the interlayer distance in the brucite structure by Li+ ion substitution plausibly influence the dehydration of Mg(OH)2. On the other hand, point (2) has a more profound effect on the hydration of MgO.

The results of this study qualitatively indicate that the lattice parameters increase upon LiCl and/or LiOH addition when compared to those of pure Mg(OH)2. Nonetheless, more analysis is required to clearly reveal the effects mentioned in this study, by drastically increasing the amount of Li compounds added to Mg(OH)2 or by simulation using first-principles calculations.

Table shows that L10 had a lower peak temperature for the dehydration reaction, and that dehydration progressed at lower temperature compared with LO20, meaning that LiCl and LiOH have different effects on the dehydration of Mg(OH)2. To elucidate the effect of the addition of LiCl and LiOH on the dehydration of Mg(OH)2, each sample was heated to various temperatures, and XRD analysis was carried out for the samples after dehydration.

Figure a shows the XRD patterns of Mg(OH)2–W after heating to various temperatures. The black dotted line shows the peak position of Mg(OH)2–W after heating to 310 °C, and the red dotted line shows that of MgO after heating to 380 °C. This figure shows the appearance of the highest intensity peak at 42° corresponding to the MgO (200) plane in the XRD pattern of the sample treated at 310–320 °C, whereas the highest intensity peak corresponding to the Mg(OH)2 (002) plane at 38° disappeared for the sample treated at 370–380 °C. These observations are indicative of the decomposition of Mg(OH)2 and that the diffusion of water molecules on the surface of Mg(OH)2 started at 310–320 °C. Likewise, the decomposition of Mg(OH)2 and the diffusion of water molecules in the bulk phase were completed at 370–380 °C.

Figure 10

(a). XRD patterns of Mg(OH)2–W after heating to various temperatures (310–380 °C); the black dotted line shows the peak position of Mg(OH)2 after heating to 310 °C, and the red dotted line shows that of MgO after heating to 380 °C. (b) XRD patterns of L10 after heating to various temperatures (250–320 °C); the black dotted line shows the peak position of Mg(OH)2 after heating to 250 °C, and the red dotted line shows that of MgO after heating to 310 °C. (c) XRD patterns of LO20 after heating to various temperatures (230–300 °C); the black dotted line shows the peak position of Mg(OH)2 after heating to 300 °C, and the red dotted line shows that of MgO after heating to 300 °C. (c)′ XRD patterns of LO20 after heating to various temperatures (300–370 °C); the black dotted line shows the peak position of Mg(OH)2 after heating to 300 °C, and the red dotted line shows that of MgO after heating to 370 °C. (d) XRD patterns of L5/LO5 after heating to various temperatures (240–320 °C); the black dotted line shows the peak position of Mg(OH)2 after heating to 240 °C, and the red dotted line shows that of MgO after heating to 310 °C. (e) XRD patterns of L10/LO10 after heating to various temperatures (220–310 °C); the black dotted line shows the peak position of Mg(OH)2 after heating to 220 °C, and the red dotted line shows that of MgO after heating to 300 °C.

The same measurement was carried out for other samples, and the XRD patterns are presented in Figure b–e. Table shows the temperatures at which the peak corresponding to the MgO (200) plane at around 42° appeared and that of the Mg(OH)2 (002) plane at around 38° disappeared for all samples. For LO20, the MgO (200) peak at ∼42° appeared at a much lower temperature compared with that of Mg(OH)2–W, and the Mg(OH)2 (002) peak at ∼38° disappeared at much lower temperature for L10 compared with that of Mg(OH)2–W. These results indicate that LiOH promoted the decomposition of Mg(OH)2 and the diffusion of water molecules on the surface of Mg(OH)2, whereas LiCl promoted the decomposition of Mg(OH)2 and the diffusion of water molecules in the bulk phase. Therefore, it is possible that LiCl and LiOH act cooperatively in the dehydration of Mg(OH)2.

Table 4

Temperature of Disappearance of Mg(OH)2 (002) Peak at 38° and Appearance of MgO (200) Peak at 42° for All Samples

sample	temp. of disappearance [°C]	temp. of appearance [°C]
Mg(OH)2–W	370–380	310–320
L10	310–320	250–260
LO20	360–370	230–240
L5/LO5	310–320	240–250
L10/LO10	300–310	220–230

From the discussion presented above, it is deduced that the decomposition of Mg(OH)2 and the diffusion of water molecules both on the surface and in the bulk phase of Mg(OH)2 were promoted in the case of L10/LO10. Thus, the dehydration of L10/LO10 progressed at the lowest temperature (Figure ).

Figure shows a plot of the heat-treatment temperature versus the logarithm of the intensity ratio of the MgO (200) and Mg(OH)2 (002) peaks from the XRD patterns (Figure ). The orange dotted line indicates the point where the y-axis value is equal to 0, meaning that the intensity of MgO (200) and Mg(OH)2 (002) peaks are equal.

Figure 11

Heating temperature vs log plot; y axis shows the logarithm of intensity ratio for Mg(OH)2 and MgO, and x axis shows the heating temperature [°C].

In the case of Mg(OH)2–W, the plot was on the highest temperature side, indicating that the dehydration of Mg(OH)2–W progressed at higher temperature when compared with the case of other samples. The figure clearly shows that the dehydration of L10/LO10 progressed at the lowest temperature because the plot fell on the lowest temperature side. This trend is consistent with the TG analysis presented in Figure and Table .

Conclusions

Mg(OH)2 was singly doped with LiCl or LiOH and codoped with LiCl and LiOH to enhance its reactivity for application as a chemical heat storage material. The reactivity of LiCl/LiOH/Mg(OH)2 was also investigated, and the effect of the addition of LiCl and/or LiOH on the structure of Mg(OH)2 was analyzed.

LiCl/LiOH/Mg(OH)2 (L10/LO10) showed much better dehydration and hydration reactivity than LiCl/Mg(OH)2 (L10) and LiOH/Mg(OH)2 (LO20). In particular, almost complete hydration of L10/LO10 was achieved at 200 °C after dehydration at 350 °C, whereas pure Mg(OH)2, L10, and LO20 were hardly hydrated. Thus, L10/LO10 can be used to supply high temperature and qualified heat in heat output operations. Therefore, the coaddition of LiCl and/or LiOH enhances both the dehydration and hydration reactivity to a greater extent than the addition of LiCl or LiOH individually. On the other hand, the coaddition of LiCl and LiOH could not enhance the hydration reactivity of MgO at low water vapor pressure. This is one of the main issues to be resolved for heat utilization over a wide temperature range. XRD analysis indicates that the Li+ ions substituted into the Mg2+ site cause interlayer expansion in the brucite structure, where the lattice parameters increased upon addition of LiCl and/or LiOH. The Li ion substitution plausibly induced the formation of lattice defects such as oxygen vacancies because of the disruption of the charge balance at the substituted sites. These effects promoted the dehydration and hydration of Mg(OH)2 and MgO. However, further studies are required for the clear elucidation of these effects by the dramatic addition of Li compounds (LiCl and/or LiOH) or by simulation, such as first-principles calculations. It is plausible that LiOH promoted the decomposition of Mg(OH)2 and the diffusion of water molecules on the surface of Mg(OH)2, whereas LiCl promoted the decomposition of Mg(OH)2 and the diffusion of water molecules in the bulk phase. Further studies are required to elucidate the effect of coaddition of LiCl and LiOH on the hydration of MgO. We must also confirm the cycle repeatability of dehydration and hydration in the future work.

Experimental Section

Sample Preparation

To prepare LiCl and LiOH coadded Mg(OH)2 (referred to as LiCl/LiOH/Mg(OH)2) and LiCl- or LiOH-added Mg(OH)2 (referred to as LiCl/Mg(OH)2 and LiOH/Mg(OH)2, respectively), LiCl·H2O (99.9%, Wako Pure Chemical Industries, Ltd.), LiOH·H2O (Wako Pure Chemical Industries, Ltd.), and Mg(OH)2 (99.9%, 0.07 μm, Wako Pure Chemical Industries, Ltd) were used as precursors.

LiCl/Mg(OH)2, LiOH/Mg(OH)2, and LiCl/LiOH/Mg(OH)2 were prepared by the impregnation method. First, aqueous LiCl and/or LiOH was prepared from LiCl·H2O and/or LiOH·H2O and ultrapure water. Thereafter, authentic powder Mg(OH)2 was impregnated with the solutions and stirred for 30 min. Subsequently, the water was evaporated using a rotary evaporator at 40 °C. Finally, the samples were dried at 120 °C overnight. All samples were obtained as white powders. LiCl-added Mg(OH)2 with a Mg(OH)2/LiCl mole ratio of 100:10 (referred to as L10), LiOH-added Mg(OH)2 with Mg(OH)2/LiOH = 100:20 (referred to as LO20), LiCl and LiOH coadded Mg(OH)2 with Mg(OH)2/LiCl/LiOH = 100:5:5 and 100:10:10 (referred to as L5/LO5 and L10/LO10, respectively) were prepared by this method. For comparison, Mg(OH)2 without the Li compounds, referred to as Mg(OH)2–W, was prepared by the same method. All of the prepared samples are shown in Table . Our previous studies showed that L10 and LO20 were the best mixing ratios with respect to the activation energy for dehydration and for enhancing the dehydration and hydration reactivity of LiCl/Mg(OH)2 and LiOH/Mg(OH)2, respectively.[12−14] Hence, these samples were used for the comparison of the reactivity in this study.

Table 5

Mixing Ratio of the Prepared Samples

sample	mixing ratio [mole ratio]
Mg(OH)2–W	Mg(OH)2 without Li compounds
L10	Mg(OH)2/LiCl = 100:10
LO20	Mg(OH)2/LiOH = 100:20
L5/LO5	Mg(OH)2/LiCl/LiOH = 100:5:5
L10/LO10	Mg(OH)2/LiCl/LiOH = 100:10:10

Evaluation of Reactivity Using Thermobalance

The reactivity of all samples was measured by using a thermobalance (TGD-9600 series, ADVANCE RIKO, Inc.). The samples (∼20 mg) were charged into a Pt cell and then heated under a constant Ar flow. The mole fraction of Mg(OH)2 was calculated as shown below, based on the TG data measured by using the thermobalance.[12−14]Here, wini is the initial weight (the weight of Mg(OH)2 at 200 °C) [mg], wfin is the weight of Mg(OH)2 that reacted theoretically (the weight of MgO) [mg], w is the weight of Mg(OH)2 in the sample during the reaction [mg], MMg(OH) is the molecular weight of Mg(OH)2 [g mol–1], MMgO is the molecular weight of MgO [g mol–1], and x is the mole fraction of Mg(OH)2 [—]. In this study, we assumed that LiCl and LiOH included in the sample never react; therefore, the weights of LiCl and LiOH in the sample were subtracted from the total sample weight.[12−14,29]

Figure shows the TG curve for the sample dehydrated at 270 °C and then hydrated at 110 °C. The first weight loss is attributed to the removal of physically adsorbed water. The sample weight did not change at 200 °C; therefore, the sample weight at 200 °C is used as the initial sample weight in this study.

Figure 12

TG curve for the sample dehydrated at 270 °C under 100 mL min–1 Ar flow and subsequently hydrated at 110 °C with Ar gas and water vapor mixture at 57.8 kPa.

Dehydration Reaction

The conditions for the dehydration reaction are shown below; the sample was heated from room temperature (20–25 °C) to various temperatures (220–800 °C) at a rate of 10 °C min–1 under Ar flowing at 100 mL min–1.

Hydration Reaction

For the hydration reaction, to remove physically adsorbed water, the sample was heated at 120 °C for 30 min at a rate of 20 °C min–1 under Ar flowing at 100 mL min–1. The sample was then dehydrated at 270 or 350 °C for 30 min at a heating rate of 20 °C min–1 under Ar flowing at 100 mL min–1. The sample was then hydrated at constant hydration temperatures (110, 170, and 200 °C) for 80 min at a heating rate of −20 °C min–1 by introducing a gas mixture of water vapor at 57.8 kPa and Ar gas. Subsequently, the introduction of water vapor was stopped, and the sample was dried at the hydration reaction temperature for 30 min.

Figure shows the dehydration and hydration profiles of L10/LO10. As shown in Figure , x0 is the mole fraction of Mg(OH)2 before the hydration reaction [—], xh is that of Mg(OH)2 at the end of hydration [—], xc is that of Mg(OH)2 at 10 min after the end of hydration [—], Δxd is the dehydration reaction conversion [%], Δx1 is the hydration reaction conversion [%], and Δx2 is the reaction conversion by physical water adsorption [%]; Δxd, Δx1, and Δx2 are calculated as percentages, as shown below

Figure 13

Data for hydration reaction test; dehydration temperature (Td) = 270 °C under 100 mL min–1 Ar flow and hydration temperature (Th) = 110 °C with Ar gas and water vapor mixture at 57.8 kPa.

To evaluate the effect of the hydration reaction temperature (Th) on the hydration of MgO, Th was varied as 110, 170, and 200 °C, and the water vapor pressure was fixed at 57.8 kPa. To evaluate the effect of the water vapor pressure on the hydration of MgO, the water vapor pressure varied in the range of 7.4–57.8 kPa by controlling the water flow using a microfeeder (NP-KX-101, NIHON SEIMITSU KAGAKU Co. Ltd.), and the Ar balance and Th was fixed at 110 °C. Table shows the relationship between the water vapor pressure (PH) and saturated temperature (Ts).

Evaluation of Heat Output Density

The heat output densities of all samples were calculated from the enthalpy change of the hydration reaction and that during the sorption of water vapor (eqs and 11). Herein, the heat output density is expressed as the heat output density per unit weight (kg) of the sample. It was assumed that the enthalpy change for the sorption of water vapor corresponds to that of the condensation of water (eq ). The heat output densities were calculated as followswhere Qr is the heat output density of the hydration reaction [kJ kg–1], Qs is the heat output density for the condensation of water [kJ kg–1], and Mhyd is the weight of magnesium hydroxide per kg of sample [kg]. ΔHhyd and ΔHad are the enthalpy change of the reaction and condensation per kg magnesium hydroxide [kJ kg–1], respectively.[13,14]

Sample Characterization by XRD

To investigate the crystal structure of the samples, XRD analysis was carried out in triplicate to reduce the error by using an Ultima IV (Rigaku Corp.) X-ray diffractometer. The 2θ range was 10–150°, the scan rate was 10.0° min–1, and the scan width was 0.01°. Cu Kα radiation was used with a generator voltage of 40 kV and 40 mA current. Before measurement, the sample was dried at 120 °C overnight to remove physically adsorbed water.

To analyze the structure of pure Mg(OH)2 and the Li compounds-added Mg(OH)2, the lattice parameters and volumes of all of the samples were calculated by the least-squares method. These values were adjusted to match the measured peak positions by this method.[30] The 90% confidence interval for the calculations was estimated by Student’s t-distribution, based on eq .where N is the number of measurements (in this study, N = 3), and u is the standard deviation.