Preservation of carbon dioxide clathrate hydrate in the presence of trehalose under freezer conditions

To investigate the preservation of CO2 clathrate hydrate in the presence of sugar for the novel frozen dessert, mass fractions of CO2 clathrate hydrate in CO2 clathrate hydrate samples coexisting with trehalose were intermittently measured. The samples were prepared from trehalose aqueous solution with trehalose mass fractions of 0.05 and 0.10 at 3.0 MPa and 276.2 K. The samples having particle sizes of 1.0 mm and 5.6–8.0 mm were stored at 243.2 K and 253.2 K for three weeks under atmospheric pressure. The mass fractions of CO2 clathrate hydrate in the samples were 0.87–0.97 before the preservation, and CO2 clathrate hydrate still remained 0.56–0.76 in the mass fractions for 5.6–8.0 mm samples and 0.37–0.55 for 1.0 mm samples after the preservation. The preservation in the trehalose system was better than in the sucrose system and comparable to that in the pure CO2 clathrate hydrate system. This comparison indicates that trehalose is a more suitable sugar for the novel frozen carbonated dessert using CO2 clathrate hydrate than sucrose in terms of CO2 concentration in the dessert. It is inferred that existence of aqueous solution in the samples is a significant factor of the preservation of CO2 clathrate hydrate in the presence of sugar.

Carbonated drinks, e.g. soda, beer, champagne etc., are popular all over the world because they have a feeling of refreshment. Solid carbonated foods, like a candy or jelly, are also available in markets. However, the number of solid carbonated foods products is less than that of carbonated drinks and the solid foods are not as popular as carbonated drinks because CO 2 solubility in ice or other solid is extremely less than the solubility in liquid water. One of the solutions of this problem is clathrate hydrate.
Clathrate hydrate, also called gas hydrate, is a crystalline solid composed of water (host molecule) and other molecules (guest molecule), e.g. methane, ethane, carbon dioxide etc. Guest molecules are trapped in the molecular-level cages that are composed of hydrogen-bonded water molecules. Clathrate hydrate can store large amount of gas molecules. For example, CO 2 concentration in clathrate hydrate is 298 kg/m 3 1 , which is 20-50 times higher than that in carbonated water, 6-15 kg/m 3 2 . Then, clathrate hydrate has a great potential for the solid carbonated food.
In general, high pressure and low temperature condition is necessary to thermodynamically stabilize clathrate hydrate. For example, phase equilibrium pressure of CO 2 clathrate hydrate is 1.318 MPa at 273.8 K 3 . Preservation of CO 2 clathrate hydrate under thermodynamically stable condition is often practically difficult because such high pressure is needed. This difficulty can be overcome using a unique phenomenon observed for clathrate hydrate that is called "self-preservation" or "anomalous preservation" [4][5][6][7] . Some kinds of clathrate hydrate anomalously slowly decompose below the water freezing temperature despite under their thermodynamically unstable conditions. It is suggested that the anomalous preservation phenomenon is appeared by an ice layer covering clathrate hydrate particles which is formed because of clathrate hydrate dissociation 5,6 . Ice plays an important role for the anomalous preservation of clathrate hydrate. Numerous observations of decomposition of CO 2 clathrate hydrate are reported 7,8 . Takeya and Ripmeester 7 reported that decomposition rate of CO 2 clathrate hydrate decreased at the temperatures between from 220 K to 260 K under atmospheric pressure. CO 2 clathrate hydrate showed the anomalous preservation. Sun et al. 8 reported the preservation of CO 2 clathrate hydrate samples at 253 K and 258 K under atmospheric pressure for three weeks. After the three-week preservation, CO 2 concentration in the samples exceeded that in carbonated water. CO 2 clathrate hydrate can be preserved at thermodynamically unstable temperature under atmospheric pressure below the water freezing temperature.
Using the anomalous preservation, CO 2 clathrate hydrate is focused as a novel carbonated food [8][9][10][11][12][13] . Peters et al. reported a rapid production process of the novel carbonated dessert using CO 2 clathrate hydrate 9 , a safety test for a model of the package of the dessert 10 and a sensory test for the dessert 11 . Makiya et al. 12 reported the formation pressure and temperature of clathrate hydrate which is formed from CO 2 + ethanol at temperature from 254 K to 268 K. Sato et al. 13 reported the preservation of CO 2 clathrate hydrate samples coexisting with sucrose at 253 K and 258 K under atmospheric pressure for three weeks. After the three-week preservation, CO 2 concentration in the samples coexisting with sucrose exceeded that in carbonated water. However, the concentration is lower than that in pure CO 2 clathrate hydrate samples because of presence of concentrated sucrose aqueous solution in the samples 13 . This report indicates that higher CO 2 concentration in novel carbonated desserts using CO 2 clathrate hydrate can be achieved.
One of the candidates for the sugar is trehalose. Trehalose is a disaccharide of glucose and eutectic temperature in water-trehalose system is 270.7 K 14 . Then, ice and trehalose dihydrate phases are stable in water-trehalose system at 253 K and 258 K under atmospheric pressure, i.e. domestic freezer conditions. Therefore, CO 2 concentration in CO 2 clathrate hydrate samples coexisting with trehalose after three week preservation may exceed the concentration in the samples coexisting with sucrose and that in carbonated water. Trehalose has been used in a variety of research applications and commercial products 15 . Especially in food industry, trehalose has unique characteristics; keeping vegetables and fruits fresh, protecting vitamins against heating, protecting food products against low-temperature and freezing stress, etc. 15 . Trehalose may have strong advantages as a sugar for the dessert using CO 2 clathrate hydrate.
In the present study, to investigate the preservation of CO 2 clathrate hydrate in the presence of trehalose, mass fractions of CO 2 clathrate hydrate in CO 2 clathrate hydrate samples coexisting with trehalose were intermittently measured by mass measurement of CO 2 in the samples for three weeks. The samples were prepared from trehalose aqueous solution with trehalose mass fractions of 0.05 and 0.10 at 3.0 MPa and 276.2 K. Particle sizes of the samples were 1.0 mm and 5.6-8.0 mm. These values correspond to the sizes of the ice particles in frozen dessert such as sorbet, smoothie, and frozen drink. The prepared samples were stored at 243.2 K and 253.2 K for three weeks under atmospheric pressure. We also performed powder X-ray diffraction (PXRD) measurements to identify the structures of crystalline compounds in the samples after the preservation and to calculate the mass fractions of CO 2 clathrate hydrate and ice in the samples after the preservation. The mass fractions of CO 2 clathrate hydrate and ice obtained from PXRD measurements were independently determined from the mass measurements.

Results and Discussion
The CO 2 clathrate hydrate samples prepared from trehalose aqueous solution with 0.05 and 0.10 trehalose mass fractions at 276.2 K and 3.0 MPa were stored for three weeks at 243.2 K and 253.2 K under atmospheric pressure. Particle sizes of the samples were 1.0 mm and 5.6-8.0 mm. The mass fractions of CO 2 clathrate hydrate in the samples were 0.87-0.94 and 0.93-0.97 for the samples with 0.10 and 0.05 trehalose mass fractions before the preservation experiments, respectively. Therefore, the amount of CO 2 clathrate hydrate and trehalose in the samples was greater than 0.97 in mass fraction. This indicates that most of the water was converted to CO 2 clathrate hydrate.
Results of the three-week preservation experiments for all samples at 243.2 K and 253.2 K are shown in Figs 1 and 2, respectively. In the first two days of the preservation, the mass fractions decreased from 0.87-0.97 to 0.40-0.61 for 1.0 mm diameter samples and to 0.79-0.95 for 5.6-8.0 mm diameter samples at both 243.2 K and 253.2 K. Afterwards, the mass fractions still remained around 0.4-0.6 for 1.0 mm diameter samples and around 0.6-0.9 for 5.6-8.0 mm diameter samples during the preservation. At the end of the preservation, the mass fractions still remained 0.37-0.55 for 1.0 mm diameter samples and 0.56-0.76 for 5.6-8.0 mm diameter samples. The mass fractions of CO 2 clathrate hydrate in the samples after the three-week preservation are specified in Table 1. All of the mass fractions after the preservation significantly exceeded 0.02 that is equivalent to carbonated water 2 . This indicates that CO 2 clathrate hydrate in the samples was anomalously preserved after the preservation although the samples were stored under thermodynamically unstable conditions of CO 2 clathrate hydrate. The mass fractions in 5.6-8.0 mm diameter samples were higher than those in 1.0 mm diameter samples after the preservation at both 243.2 K and 253.2 K. This trend is consistent with the results of the pure CO 2 clathrate hydrate preservation 8 . In contrast, the mass fractions after the preservation have not changed depending on the trehalose mass fraction and the preservation temperature in the present preservation experiments.
Duplicate preservation experiments were performed to confirm the reproducibility of the preservation experiments for the samples with 0.10 trehalose mass fraction at 243.2 K. The results of the two independent runs were shown in Fig. 3 as circle and square plots for 1.0 mm and 5.6-8.0 mm diameter samples, respectively. The results of both runs were consistent for each particle size of the samples within the combined uncertainty of the mass fraction of CO 2 clathrate hydrate. This consistency supports the reproducibility of the preservation of CO 2 clathrate hydrate in the presence of trehalose. Both difference of the mass fractions between two runs and deviation of each mass fraction for 1.0 mm diameter samples tended to be smaller than those for 5.6-8.0 mm diameter samples. This trend indicates that distribution of CO 2 clathrate hydrate among the sample particles for 1.0 mm diameter samples were more homogeneous than that for 5.6-8.0 mm diameter samples.
We performed PXRD measurements to identify the crystallographic structures of crystalline compounds in the samples after the three-week preservation. Figures 4 and 5 show PXRD peak patterns obtained with the 1.0 mm diameter samples prepared from 0.10 mass fraction trehalose solution and both stored and measured at 243.2 K and 253.2 K, respectively. In both of the samples, hexagonal ice, structure I clathrate hydrate and trehalose dihydrate were observed. The existence of hexagonal ice and trehalose dihydrate in the samples is reasonable because the eutectic point temperature and composition of water-trehalose system were reported to be 270.7 K and 0.298 mass fraction of trehalose, respectively 14  coexisting with trehalose was confirmed by PXRD measurements. We also performed PXRD measurements to calculate the mass fractions of CO 2 clathrate hydrate and ice in the samples with 0.10 trehalose mass fraction after the preservation at 100 K by the Rietveld method using RIETAN-FP program 16 . The mass fractions of CO 2 clathrate hydrate and ice were independently determined from the mass measurements. The mass fractions of CO 2 clathrate hydrate and ice in the samples after the preservation were calculated to be 0.28 and 0.62 for the 1.0 mm diameter samples stored at 243.2 K, and 0.37 and 0.53 for the 1.0 mm diameter samples stored at 253.2 K, respectively. The mass fractions of CO 2 clathrate hydrate in the samples obtained from PXRD measurements were approximately 0.1 lower than those obtained from the mass measurements. This difference may be due to the decomposition of CO 2 clathrate hydrate during powdering process of PXRD measurements. The mass fractions of hexagonal ice in the samples after the preservation were approximately equal to decreases in the mass fractions of CO 2 clathrate hydrate in the samples obtained from the mass measurements. It indicates that most of the hexagonal ice was produced by decomposition of CO 2 clathrate hydrate during the preservation.  We visually observed the surfaces of the samples before and after the preservation experiments. Figure 6 shows the surfaces of 5.6-8.0 mm diameter samples prepared from 0.10 trehalose mass fraction aqueous solution. Both the observations and the preservation were performed at 253.2 K. Before the preservation, the surfaces were smooth as shown in Fig. 6a and some bubbles were observed at the surfaces as shown in Fig. 6b. Because the bubbles were observed at the surfaces, liquid and gas existed at the surfaces. It is reported that pure CO 2 clathrate hydrate decomposed to metastable super-cooled water below the water freezing temperature 17 . Although CO 2 solubility in trehalose aqueous solution was not found in the literature, it is reported that CO 2 solubility both in sucrose aqueous solution 18 and in glucose aqueous solution 19 were lower than that in pure water and decreased with an increase in concentration of sucrose and glucose in solution. CO 2 solubility in the sucrose solution with sucrose mass fraction of 0.12 18 and in the glucose solution with glucose mass fraction of 0.12 19 are approximate 10% and 20% lower than that in pure water, respectively. Thus, the effect of CO 2 in trehalose aqueous solution on the depression of water freezing temperature may be approximately the same as or lower than that in pure water. The depression of water freezing temperature by CO 2 in trehalose aqueous solution is estimated to be up to 0.2 K under 0.1 MPa of CO 2 2 . Thus, trehalose aqueous solution is thermodynamically unstable at 253.2 K 14 . The liquid observed at the surfaces is regarded as super-cooled water or super-cooled trehalose aqueous solution. After the preservation experiments, the surfaces changed to powder-like and the edges of the samples became round  samples and preservation temperature). a Data are expressed as the mean ± the combined uncertainty of the mass fraction due to the uncertainty of the measurements and standard deviation of measured mass fractions (n = 3). as shown in Fig. 6c. These observed changes at the surfaces indicate that CO 2 clathrate hydrate decomposed at the surfaces. According to the visual observation before and after the preservation, CO 2 clathrate hydrate in the samples decomposed to super-cooled water or solution at the surfaces. Results of the PXRD measurements and the visual observation indicate that the super-cooled water or solution was transformed to hexagonal ice at the surfaces.    Figures 7 and 8 show the comparison of the preservation in the trehalose system with the preservation in the pure CO 2 clathrate hydrate system 8 and in the sucrose system 13 at 253.2 K for 5.6-8.0 mm samples and for 1.0 mm (with sucrose 13 and trehalose) or 2.8 mm (without sugars 8 ) samples, respectively. In Fig. 8, the results for two difference diameters are shown because the smallest diameter of the samples without sugars 8 was 2.8 mm. For both 5.6-8.0 mm and 1.0 mm samples, the preservation in the trehalose system was comparable to that in the pure system 8 , whereas better than that in the sucrose system 13 . Note that the preservation for 1.0 mm samples with trehalose was comparable to that for 2.8 mm samples without sugars although the diameter of samples with trehalose was smaller than that without sugars. This comparison indicates that trehalose is a more suitable sugar for the novel carbonated dessert using CO 2 clathrate hydrate than sucrose in terms of CO 2 concentration in the dessert. The anomalous preservation may appear by an ice layer that is formed by dissociation of clathrate hydrate 5,6 In the samples coexisting with sucrose 13 , sucrose aqueous solution may exist because the preservation temperatures were higher than the eutectic temperature of water-sucrose system. Sato et al. 13 discussed that the  sucrose aqueous solution in the samples influenced the ice formation and thus the preservation in the sucrose system was inferior to that in the pure system. In contrast, CO 2 clathrate hydrate in the samples coexisting with trehalose decomposed to ice at the surfaces of the samples because the preservation temperature was lower than the eutectic temperature of trehalose-water system that is reported to be 270.7 K 14 . Better preservation with trehalose than that with sucrose 13 may be ascribed to the absence of aqueous solution in the samples. It is inferred from the comparison that existence of aqueous solution in the samples is a significant factor of the preservation of CO 2 clathrate hydrate in the presence of sugar.

Methods
Sample preparation. The CO 2 clathrate hydrate samples coexisting with trehalose were prepared in the pressure vessel as shown in Fig. 9. The inner dimensions of the vessel were diameter 80 mm, height 40 mm and volume about 200 cm 3 . The vessel was placed in a temperature-controlled bath filled with ethylene glycol aqueous solution at 276.2 K. A platinum resistance thermometer (with uncertainly of ± 0.1 K) and a pressure transducer (GC31-174-L7N18, Nagano Keiki Co., Ltd., with uncertainly of ± 0.05 MPa) were used to measure the temperature and pressure inside the vessel. In the vessel, an impeller was driven at 400 rpm to agitate trehalose aqueous solution, gas and CO 2 clathrate hydrate particles. To prepare the samples, we supplied approximate 50 g of trehalose aqueous solution with trehalose mass fraction of 0.05 or 0.10 to the vessel. Trehalose dihydrate (Hayashibara Co., Ltd., 0.98 mass fraction certified purity) and laboratory-made deionized and distilled water (electrical conductivity was less than 0.5 × 10 −4 S/m) were used. To discharge the air inside the vessel, CO 2 gas (Toyoko Kagaku Co., Ltd., 0.999 volume fraction certified purity) was supplied to the vessel and discharged using a vacuum pump till the partial pressure of the air decreased to 0.01 kPa or lower. Then, CO 2 gas was supplied to the vessel at 3.0 MPa and 276.2 K to form CO 2 clathrate hydrate. This pressure is higher than the phase equilibrium pressure for CO 2 clathrate hydrate forming system at 276.2 K, 1.717 MPa 20 . During preparing the samples, the pressure in the vessel decreased because of CO 2 clathrate hydrate formation. Then, CO 2 gas was recharged in the vessel at 3.0 MPa until no further pressure reduction was observed. After that, the vessel was transferred to liquid nitrogen bath and cooled to 219.2 K. During the cooling procedure, the pressure inside the vessel was reduced in a stepwise manner, approximately along the pressure-temperature line of the ice + vapor + CO 2 clathrate hydrate equilibrium. The samples were removed from the vessel at 219 K under atmospheric pressure. The samples were immediately crushed using a chilled mortar and pestle, and sieved into particles with diameters of 1.0 mm and 5.6-8.0 mm. Preservation experiments. During the preservation experiments, the samples were stored in plastic bottles at 243.2 K or 253.2 K under atmospheric pressure as shown in Fig. 10. The plastic bottles were set in stainless-steel cans placed in a temperature-controlled bath and mechanically stabilized by stainless-steel blocks used as sinkers. The temperature in the bottles was controlled by ethylene glycol aqueous solution in temperature-controlled bath. The temperature inside the cans was intermittently measured using a platinum resistance thermometer to confirm that the sample temperature was kept at the prescribed value, 243.2 K or 253.2 K. During the three-week preservation, approximate 1.5 g samples were intermittently removed from the bottles. The removed samples were used to the mass measurements to determine the mass fractions of CO 2 clathrate hydrate in the samples.
To determine the mass fractions of CO 2 clathrate hydrate in the samples, we measured CO 2 mass in the samples. The removed samples were divided into three portions. We placed the portions in three closed containers and measured mass of the samples in the container using an electronic balance (Sefi IUW-200D, with uncertainly of ± 0.1 mg). Then, the containers were heated to 293.2 K to decompose all of CO 2 clathrate hydrate in the samples. After the decomposition, the containers were cooled to 253.2 K, and the containers were opened once to release CO 2 gas evolved from CO 2 clathrate hydrate into atmosphere. After that, the containers were reheated to 293.2 K to dissociate ice in the containers and mass of the samples was measured using the electronic balance. The mass measurements and cooling were repeated until no further mass reduction was observed, that is, mass reduction was smaller than 1 mg. The difference between the first and last mass of the samples is the CO 2 mass in the