Composition for long-term latent heat storage

Abstract

The invention relates to a composition for use in a long-term latent heat storage device, the composition comprising a non-eutectic mixture of a crystallizing phase-change material and a vitrifying phase-change material.

Description

[0001] Composition for long-term latent heat storage

[0002] Technical field of the invention

[0003] The present invention relates to materials for long-term latent heat storage utilizing the supercooling effect. In particular, the present invention relates to the use of such materials for long-term latent heat storage. Furthermore, the present invention relates to long-term latent heat storage devices that utilize the supercooling effect.

[0004] Background of the invention

[0005] In recent decades, energy and its use have increasingly come into focus. Over the years, awareness of the finite nature of fossil fuels and the problems associated with their use has grown. Climate change, for example, presents society with ever-new challenges.

[0006] To prevent further global warming, climate-damaging emissions must be reduced. Data on German final energy consumption clearly show that a shift in heat generation is necessary. In the future, process efficiency must be increased, and the recycling of waste heat into energy systems must also be improved. However, waste heat is often generated at a different time than it can be used. Thermal energy storage is an important tool for bridging this time gap.

[0007] In the field of thermal energy storage, there are various technologies that can be divided into three groups: sensible, latent, and thermochemical energy storage.

[0008] Sensible storage is divided into solid and liquid, latent storage into solid-liquid, liquid-gas and solid-solid, while thermochemical storage is divided into sorption and reaction processes (see Figure 1).

[0009] Sensitive storage systems, which operate by sensing the heating of the storage material, are well-known and widely used today. Hot water storage systems are particularly popular in this area. However, they are limited by the specific heat capacity of the materials used and therefore do not represent the optimum in terms of stored energy per kilogram or cubic meter. Thermochemical storage systems offer great potential, but are technically complex and still in their early stages of development. They use chemically reversible reactions or sorption to store energy.

[0010] Latent heat storage (LHS) combines the advantages of both of the aforementioned technologies: they can store more heat than defined solely by their specific heat capacity and are technically not too complicated. This technology uses phase change to store energy.

[0011] As can be seen in Figure 2, for example, at the melting point of a material, significantly more energy is stored during a phase transition than in a comparable purely sensible heat storage process. However, it is also evident that sensible heat storage processes also occur before and after the phase transition.

[0012] Latent heat storage systems used for technical purposes utilize the process of spontaneous crystallization. During heat storage, a crystalline substance is gradually melted until it is completely molten. During this process, the temperature of the entire system does not increase. Only then does the system transition to sensible heat storage. If energy is withdrawn from the storage system, the liquid crystallizes again and releases the energy back into the environment. However, this type of latent heat storage system is therefore unsuitable for long-term storage.

[0013] If a substance cools below its melting point without crystallizing, this process is called supercooling. No crystal structure forms; instead, the amorphous structure of the liquid is retained, and a supercooled melt forms. The various possible phases, depending on the temperature and their specific volume, are shown in Figure 3.

[0014] All matter fundamentally strives for the solid state of a crystal, as this is the state of matter with the lowest energy. However, some substances are unable to achieve this state, or only with difficulty. Such substances are highly viscous, and their building blocks cannot arrange themselves into a crystal lattice quickly enough upon cooling. In addition to viscosity, a rapid cooling rate also promotes supercooling, as nucleation and crystal growth are time-dependent. First, a supercooled melt forms. With further cooling, the glass transition point is reached, above which the liquid is frozen as amorphous glass. Figure 4 shows the dependencies between melt, crystal, and glass. The specific volume of a melt and supercooled melt decreases continuously with decreasing temperature until a kink occurs at the glass transition temperature Tg, and the volume changes parallel to that of the crystal.This kink can be explained by a sharp increase in viscosity and the freezing of the supercooled melt into glass. The glass transition point is defined by a viscosity of 10. 12 Pa s.

[0015] Figure 4 illustrates the relationship between stored heat and temperature for latent heat storage. In conventional LWS, supercooling often poses a problem because the materials are technically more difficult to control. Figure 4 a) shows a case of supercooling followed by crystallization. In some cases, however, nucleation and thus crystallization are completely absent, as shown in Figure 4 b). If crystallization is completely stalled, only the sensible heat is stored. Therefore, such materials are fundamentally unsuitable for LWS.

[0016] For this reason, the state of the art has primarily focused on methods for suppressing supercooling rather than utilizing it. For example, nucleating substances or seed crystals are often deliberately added to the material compositions to promote crystallization. Rough surfaces, for example, are also used to promote heterogeneous crystallization. Less well-known is the addition of plasticizing substances to increase the free volume of the materials and thus facilitate nucleation.

[0017] However, the long-term use of latent heat storage systems would also be interesting. This would combine the advantages of high energy density and simple technology, making it possible, for example, to utilize waste heat from certain processes at different times.

[0018] Summary of the invention

[0019] A major problem with long-term storage using latent heat storage systems is the spontaneous or slow crystallization of the materials used. This often compromises the long-term stability of the systems.

[0020] It is therefore an object of the present invention to find a composition for latent heat storage devices that can be used to exploit the advantage of supercooling. To do this, the composition must be able to be supercooled for a long time and yet still be triggered to crystallize in this state. Compositions often cannot be supercooled stably or not at all because crystallization is triggered with a time delay or spontaneously. On the other hand, it is often the case that compositions can be subcooled stably but can then no longer be triggered to crystallize or can only be triggered with very high energy expenditure. It has now surprisingly been found that the object described above is achieved by a composition for use in a long-term latent heat storage device, wherein the composition comprises a non-eutectic mixture of a crystallizing phase change material and a vitrifying phase change material.

[0021] It has further been surprisingly found that the above-described object is achieved by using the composition according to the invention for long-term latent heat storage.

[0022] Furthermore, it was surprisingly found that the above-described object is achieved by using the composition according to the invention in a long-term latent heat storage device.

[0023] Finally, it was surprisingly found that the above-mentioned object is achieved by a long-term latent heat storage device, wherein the long-term latent heat storage device comprises the composition according to the invention.

[0024] An advantage of the present invention is that stable latent heat storage with a subcooling effect can be achieved, so that even latently stored heat can be provided in a long-term stable manner and can be retrieved when needed.

[0025] Short description of the figures

[0026] Figure 1 shows possible thermal storage options.

[0027] Figure 2 shows the difference between sensible and latent heat storage.

[0028] Figure 3 shows relevant phase transitions in a solid-liquid system.

[0029] Figure 4 shows possible processes during subcooling.

[0030] Figure 5 shows a signal of a melting peak of a DSC measurement with the characteristic quantities.

[0031] Figure 6 shows the results (enthalpy of fusion, melting temperature) of the DSC investigations for selected sugar alcohols.

[0032] Figure 7 shows the results (specific heat capacity) of the DSC investigations for xylitol, D-sorbitol and erythritol, as well as literature data (DelBarrio: del Barrio, EP, et al., Characterization of different sugar alcohols as phase change materials for thermal energy storage applications, Solar Energy Materials and Solar Cells, 2017, 159, p. 560-569; Tong: Tong, B., et al., Thermodynamic investigation of several natural polyols, Journal of thermal analysis and calorimetry, 2009, 95(2), p. 469-475).

[0033] Figure 8 shows the results (specific heat capacity) of the DSC investigations for isomalt, D-mannitol, maltitol and dulcitol, as well as literature data (DelBarrio: del Barrio, EP, et al., Characterization of different sugar alcohols as phase change materials for thermal energy storage application, Solar Energy Materials and Solar Cells, 2017, 159, p. 560-569; Cheng Ze: Cheng, Z., et al., Low-temperature heat capacity and standard thermodynamic functions of D-galactose and galactitol, Chemical Research in Chinese Universities, 2015, 31 (6), p. 987-991 ; Daneshyari: Knyazev, A., et al., Thermodynamic properties of myoinositol, The Journal of Chemical Thermodynamics, 2018, 116, p. 76-84).

[0034] Figure 9 shows the DSC signal of the first measurement cycle of a) crystallizing D-mannitol and b) supercooling maltitol.

[0035] Figure 10 shows the degree of supercooling u of the measured pure substances for two different sample sizes (0.01 g (DSC; a)) and 100 g (3SK, b)).

[0036] Figure 11 shows images of all samples from the long-term tests of the pure substances over 7 days at ambient temperature in the hot state (day 0) and as a supercooled melt after 1 day and 7 days.

[0037] Figure 12 shows the results (enthalpy of fusion, melting temperature) of the DSC investigations for eutectic mixtures of selected sugar alcohols.

[0038] Figure 13 shows the results (specific heat capacity) of the DSC investigations for eutectic mixtures of selected sugar alcohols.

[0039] Figure 14 shows the degree of supercooling u of the measured eutectic mixtures for two different sample sizes (0.01 g (DSC; a)) and 100 g (3SK, b)).

[0040] Figure 15 shows optical investigation of subcooling during the long-term test of eutectic mixtures over 7 days.

[0041] Figure 16 shows the optical investigation of supercooling during the long-term test of further eutectic mixtures over 7 days. Figure 17 shows the degree of supercooling u as a function of the weight fraction of a) D-mannitol and b) dulcitol in non-eutectic mixtures measured by DSC.

[0042] Figure 18 shows images of non-eutectic mixtures of isomalt with different proportions of isomalt, recorded at 70 °C as supercooled melts.

[0043] Figure 20 shows the structure of the 10 l tank: a) Section through the 10 l tank and b) Representation of the location of the thermocouples. The red circle represents the outside, the green the center, and the blue the inside. The three gray dots of the thermocouples are used for the air triggering method. Subsequently, all seven thermocouples are used for the stirrer triggering method.

[0044] Figure 21: P&L flow diagram of the test bench including the 70-liter storage tank (latent heat storage d). The thermocouples of the storage tank are labeled with a letter depending on their penetration depth (S = short (45 mm), M = middle (100 mm), L = long (155 mm)).

[0045] Figure 22 shows the temperature curve in the PCM during cooling without crystallization in the 10 l storage tank; for the assignment of the thermocouples, see Figure 20.

[0046] Figure 23 shows the temperature distribution in the PCM during crystallization initiation in the 10 l storage tank at a trigger temperature of 70 °C with a) the anchor stirrer, b) the ViscoJet stirrer and c) with air.

[0047] Definitions

[0048] The term "latent heat storage" or "latent heat storage" as used herein refers to a process for storing heat, wherein at least a portion of the supplied thermal energy is stored in the form of enthalpy of transformation, preferably enthalpy of fusion ("latent heat"). The stored energy is hidden because, as long as the phase transition is not fully completed, the temperature of a substance does not rise further despite the addition of heat. Thus, during latent heat storage, the temperature of the substance rises, at least for a time, while further energy input does not.

[0049] The term "long-term latent heat storage" as used herein refers to latent heat storage as defined above, which provides heat over a long period of time, preferably more than one day, more preferably more than one week, and most preferably more than one month. Providing is preferably understood to mean producing a supercooled melt that does not crystallize spontaneously during the specified period, but whose crystallization remains selectively induced.

[0050] The term “inoculation” as used herein refers to a process for deliberately causing a supercooled melt to crystallize. A distinction is made between homogeneous inoculation, heterogeneous inoculation, mechanical inoculation, and electrical inoculation. In homogeneous inoculation, seed crystals of the melt material are added to the melt. Homogeneous inoculation can also be achieved by locally limited, strong cooling of the supercooled melt. In heterogeneous inoculation, seed crystals that are not made of the melt material are added to the melt. Heterogeneous inoculation can preferably also be achieved by rough surfaces, such as storage surfaces. Mechanical inoculation can mean any type of introduction of mechanical energy into the supercooled melt. Mechanical inoculation is preferably selected from the list consisting of stirring, shaking, momentum transfer, gas injection, and ultrasound.

[0051] The term "phase change material" or "phase-change material (PCM)" as used herein refers to the material used in latent heat storage. This material undergoes a phase change, preferably from solid to liquid. In general, any material that exists in the solid and liquid phases within the desired temperature range and whose melting point lies within this temperature range is suitable for storing latent heat. However, not every material is suitable for storing latent heat for a long time. A phase-change material that enables long-term latent heat storage is capable of forming a supercooled melt that does not crystallize spontaneously or only does so after a long time, such as days, weeks, or months. However, this melt can be further crystallized by initiating it. Another parameter for a suitable phase-change material is the amount of energy that can be stored in the phase change.The higher this energy, the more suitable the material is. Furthermore, the potential energy density per volume and / or mass is another parameter for suitability as a phase-change material.

[0052] The term "supercooling" or "supercooled melt" as used herein refers to a process step or its result in which a melt is cooled below the melting point of the molten material without crystallization of the material having occurred. In a supercooled melt, the material is therefore in a liquid, or at least viscous, amorphous form. Thus, the material has no long-range order. The term "vitrification" or "vitrified melt" as used herein refers to a process in which a supercooled melt is cooled below the glass transition temperature T g The viscosity of the melt reaches a value of 10 12Pa s or higher. Such a melt no longer behaves like a liquid to the observer and is therefore no longer susceptible to crystallization.

[0053] Detailed description of the invention

[0054] As described above, the present invention relates to a composition, a use, and a long-term latent heat storage device. These embodiments are described in detail below. Further details can be found in the experimental section.

[0055] Composition of the invention

[0056] As described above, the present invention relates to a composition for use in a long-term latent heat storage device, wherein the composition comprises a non-eutectic mixture of a crystallizing phase change material and a vitrifying phase change material.

[0057] As explained above, long-term heat storage requires that the composition is capable of forming a supercooled melt, which can nevertheless still be deliberately induced to crystallize by triggering.

[0058] Known phase-change materials often exhibit good crystallization behavior. This means that these phase-change materials crystallize readily upon cooling and can release their heat in the process. Such phase-change materials are referred to herein as crystallizing phase-change materials. Such crystallizing phase-change materials are suitable for building latent heat storage devices, which are capable of releasing heat directly, i.e., without any time delay. The sole advantage of these heat storage devices is their higher energy density. Often, attempts are made to avoid supercooling and thus the formation of a supercooled melt.

[0059] In contrast to crystallizing phase change materials, there are also vitrifying phase change materials. Such phase change materials form a supercooled melt that vitrifies upon appropriate subcooling. Vitrifying phase change materials have the disadvantage that they can no longer crystallize once vitrified. Therefore, in the state of the art, care is taken in latent heat storage systems to ensure that the phase change materials used are not vitrifying phase change materials. It has now been surprisingly discovered that by mixing a crystallizing phase change material with a vitrifying phase change material, the long-term stability of the supercooled melt of a phase change material can be improved. The supercooled melt can still be deliberately induced to crystallize by triggering.This makes it possible to produce a composition for a phase change material that is suitable for long-term latent heat storage.

[0060] The compositions according to the invention are preferably suitable for long-term latent heat storage in the field of industrial waste heat or process heat, where the industrial waste heat or process heat lies in the temperature range of 90°C to 250°C. The temperature ranges for latent heat storage can be roughly divided into the following areas: refrigeration technology (-18°C to 18°C), building services (25°C to 90°C), process heat (100°C to 250°C), and high-temperature applications (over 300°C). Phase-change materials that generally function in the process heat temperature range are salt hydrates, paraffins, sugar alcohols, and nitrates. Therefore, the crystallizing phase-change material and / or the vitrifying phase-change material is preferably selected from the group consisting of salt hydrates, nitrates, paraffins, and sugar alcohols.More preferably, the crystallizing phase change material and the vitrifying phase change material are preferably selected from the group consisting of salt hydrates, nitrates, paraffins and sugar alcohols.

[0061] However, some of these materials also have disadvantages. For example, salt hydrates are often less stable, as they tend to phase segregate when used above 100°C. The water of crystallization escapes during repeated cycling. This water is not reabsorbed into the solid structure and forms a second, aqueous phase. This reduces the usable volume fraction of the PGM and thus the storage capacity. Like nitrates, they tend to corrosive the container material. Problematic with paraffins are their flammability and low thermal conductivity. Furthermore, none of these substance groups cover the entire temperature range from 100°C to 250°C.

[0062] Therefore, the crystallizing phase change material and / or the vitrifying phase change material more preferably comprise a sugar alcohol. Most preferably, the crystallizing phase change material and / or the vitrifying phase change material more preferably comprises a sugar alcohol. Sugar alcohols are organic carbon compounds and are the hydrogenated form of carbohydrates. The basic structure of all sugar alcohols is as shown in formula (I).

[0063] The position of the OH group determines the type of sugar alcohol. This group of materials possesses several of the desired PCM properties. They often possess unusually high enthalpy at moderate densities and have a melting temperature within the desired range. They are also non-flammable, non-toxic, and non-corrosive and often of natural origin. Today, sugar alcohols are used as sugar substitutes in the food industry, as well as in the pharmaceutical and cosmetic sectors. For this reason, they are considered safe to use and readily available at a reasonable price. A major advantage for on-demand heat storage is their pronounced supercooling behavior. Since this behavior has often been considered disadvantageous, these substances have been little studied as PGMs.

[0064] Preferably, the sugar alcohols of the phase change materials of the composition of the invention are selected from the group consisting of D- Iditol, L-Iditol, D-Threitol, L-Talitol, D-Talitol, L-Threitol, L-Sylitol, X-Sorbitol D-Sorbitol, D-Lactitol Monohydrate, Arabinitol, Trehalosedihydrate, Adonitol, L-Arabitol, D-Arabitol, Erythritol, Glycero-gulo-Heptitol, D-gylcero-D-ido-heptitol, Isomalt, Palatinitol, Lactitol, Maltitol, Volatile, L-Mallitol, L-Mallitol D-Mannitol, D-Erythro-D-galacto-octitol, Maltotriitol, Perseitol, Dulcitol, L-Trehalose, D-Trehalose, Myo-Inositol, and Pentaerythritol.

[0065] More preferably, the sugar alcohols of the phase-change materials of the composition according to the invention are selected from the group consisting of xylitol, isomalt, D-sorbitol, erythritol, maltitol, mannitol, dulcitol, and myo-inositol, most preferably from the group consisting of xylitol, isomalt, D-sorbitol, erythritol, maltitol, D-mannitol, dulcitol, and myo-inositol. These materials have been found to best cover the process heat temperature range (see Figure 6). Even more preferably, the sugar alcohols of the crystallizing phase-change interval are selected from the group consisting of erythritol, D-mannitol, dulcitol, and myo-inositol. Furthermore, even more preferably, the sugar alcohols of the vitrifying phase-change material are selected from the group consisting of xylitol, D-sorbitol, maltitol, and isomalt.

[0066] Preferably, the crystallizing phase-change material and the vitrifying phase-change material are present in a weight ratio ranging from 60:40 to 40:60. Particularly preferably, the crystallizing phase-change material and the vitrifying phase-change material are present in a weight ratio ranging from 50:50 to 40:60. It has been found that the interplay of supercoolability and resolvability is best balanced in this weight ratio.

[0067] In order to function well in the temperature range of the process heat, the crystallizing phase change interval and / or the vitrifying phase change interval have a melting temperature T m in the range of 90 to 250 °C. Even more preferably, the crystallizing phase change material has a melting temperature T m in the range of 105 to 250 °C. Even more preferably, the vitrifying phase change material has a melting temperature T min the range of 90 to 150 °C. Most preferably, the crystallizing phase change material has a melting temperature T m in the range of 105 to 250 °C and the vitrifying phase change material has a melting temperature T m in the range of 90 to 150 °C. Such a combination has been shown to result in an improved balance between supercoolability and triggerability.

[0068] Preferably, the crystallizing phase change material and / or the vitrifying phase change material has an enthalpy of fusion Ah m in the range of 100 to 350 J / g. More preferably, the crystallizing phase change material has an enthalpy of fusion Ah m in the range of 245 to 350 J / g. Also more preferably, the vitrifying phase change material has an enthalpy of fusion Ah m in the range of 100 to 245 J / g.

[0069] In a preferred embodiment of the present invention, the crystallizing phase-change material and / or the vitrifying phase-change material comprises a plasticizer, preferably water. This increases the free volume and reduces the viscosity, making crystallization easier to initiate. This can improve the crystallization readiness of the composition. Use of the invention

[0070] The present invention also relates to the use of the composition according to the invention for long-term latent heat storage. As shown above and as verified in the experimental section, the composition according to the invention can be supercooled with long-term stability while still remaining selectively crystallizable.

[0071] Therefore, the present invention also relates to the use of the composition according to the invention in a long-term latent heat storage device.

[0072] Long-term latent heat storage of the invention

[0073] Ultimately, the present invention is directed to a long-term latent heat storage device for long-term latent heat storage, wherein the long-term latent heat storage device comprises the composition according to the invention as a phase change material. Preferably, the entire phase change material of the long-term latent heat storage device consists of the composition according to the invention.

[0074] Experimental part

[0075] Measurement methods a) Differential scanning calorimetry (DSC)

[0076] DSC was used to determine the phase change temperatures and enthalpies, as well as the specific heat capacity. A dynamic differential scanning calorimeter (DSC) from Netzsch Gerätebau GmbH, type 204 F1 Phoenix, was used to investigate these thermal properties. A T-sensor was installed in the measuring cell. An intracooler was connected for cooling, limiting the measuring range to -85 °C to 600 °C. Nitrogen was used as the protective and purge gas for all measurements. After each sensor replacement, the gas flow was increased from 20 ml / min to 30 ml / min. The differential scanning calorimeter was recalibrated on average every 200 measurements, as recommended by the manufacturer, using the calibration materials indium, tin, zinc, and bismuth. Cold-welded aluminum crucibles with a volume of 40 pL were used for the measurements. To prevent a pressure increase in the crucibles, the lids were perforated.A sample mass of 10±1 mg was weighed in each sample. A Shimadzu UniBloc AP125WD balance was used for this purpose. The balance has a weighing range of up to 52 g and an accuracy of 0.02 mg. For precise characterization of the eutectic mixtures, the samples were pulverized for 50 s using a Fritsch Pulverisette 23 before weighing. -1 homogenized for 2.5 min.

[0077] Basically, two different measurement methods were used, one to determine the phase change temperatures and enthalpies and a second to determine the specific heat capacity.

[0078] For the measurement method for phase change temperatures and enthalpies, the maximum measurement temperature was T = T m + 30 K and the minimum T = T cr- 30 K. For materials exhibiting high subcooling, 20 °C was set as the lower temperature limit. In some cases, the cycle was extended to 0 °C. The measurement range was always extended when phase transitions occurred near 20 °C.

[0079] Each material was measured three times, with each measurement consisting of three cycles. Each cycle consisted of a heating phase, a 10-minute isothermal phase, and a cooling phase. A ten-minute isothermal phase was also used between the individual cycles, as well as before and after the measurement. The average value of the three measurements was calculated. For the detailed investigation of the pure substances, as well as the eutectic and non-eutectic mixtures, a heating and cooling rate of 1 K / min was used.

[0080] The specific heat capacity was measured according to DIN 51007. The measurement began with a fifteen-minute isotherm at 20 °C. This was followed by a heating phase at 10 K / min to at least 30 K above T m , which was followed by another 15-minute isotherm. After a cooling phase, the measurement was considered complete. First, both the empty sample and the empty reference crucible were measured to determine the baseline. Once a reproducible baseline could be determined, a sapphire crystal was placed in the sample crucible and measured. The actual samples were then measured. It was important to use crucibles with as similar a weight as possible. Here, too, the individual materials were measured three times to then calculate the average.

[0081] The Proteus software from Netzsch Gerätebau GmbH was used to evaluate the measurement results. The measurement signal was displayed as heat flow versus time or temperature. A schematic representation of a melting peak is shown in Figure 5. As is common in the literature, the onset temperature was defined as the melting point (T m ) or crystallization temperature T cr used. The onset temperature T m is the intersection of the tangent through the inflection point of the peak and the extension of the left baseline. The peak temperature Tp eak is the temperature at the peak's apex. The area below the peak corresponds to the phase change enthalpy Ah. The area is calculated by connecting the baselines before and after the peak, where the connection can be linear, horizontal, tangential, sigmoidal, or Bezier. If the glass transition point is also within the measurement range, this is additionally evaluated. The temperature of the inflection point is considered the glass transition temperature T. g .

[0082] To investigate the temperature resistance of the materials, measurements were conducted using the simultaneous thermal analysis (STA) 409 PC / PG Luxx from Netzsch Gerätebau GmbH. This operates similarly to DSC, but in addition to heat flow, it also detects mass loss. The aim is to determine the temperature at which material loss and thus destruction processes occur in the phase-change material. A maximum temperature of 250 °C was set for this purpose, as this is the upper limit temperature for the applications considered here. A heating rate of 20 K / min was used, and the material was purged with nitrogen. In this case, aluminum oxide crucibles were used. The analysis is also performed using the Proteus software from Netzsch Gerätebau GmbH. b) Three-layer calorimetry (3SK)

[0083] Three-layer calorimetry was performed to determine the phase change temperature and enthalpy in larger volumes compared to DSC. A three-layer calorimeter (3SK) from w&a was used for this purpose. The supercooling behavior, in particular, could be investigated in detail, as the probability of crystallization also increases with increasing volume. The advantages of 3SK are the enclosed and defined sample volumes, which do not exchange with the surroundings. This provided detailed measurement results in a larger volume than DSC, but with lower accuracy.

[0084] The sample size was 100 g. In the calorimeter, the temperature of the shell and the sample was monitored using type K thermocouples and recorded every minute with a resolution of 0.2 K using a Comark data logger. The calorimeter was operated in a Memmert GmbH & Co. KG UF55Plus furnace, which also controlled the heating program. Heating was carried out for 24 hours at a fan speed of 100% and a damper opening of 0%, followed by cooling for 24 hours at the same fan speed but with a damper opening of 100%. The target temperature was set to T m +15 K for heating and 20 °C for cooling. Since the furnace does not have active cooling, this was done against ambient temperature.

[0085] The samples were examined in transparent polyethylene terephthalate bags. The samples were weighed on a PNJ 12000-1 M precision balance from KERN & SOHN GmbH. The balance can weigh a maximum of 12 kg and a minimum of 5 g with a calibration value of 1 g. The powdered material was filled into the 15 cm long bags and then sealed with aluminum tape.

[0086] The eutectic mixtures were weighed to a total of 100.0 g using an ENTRIS224I-1 S analytical balance from Sartorius Lab Instruments GmbH & Co. KG. This balance has a weighing range of up to 220 g and a standard deviation of 0.1 mg. The two components were then mixed in a mortar and pestle until a homogeneous crystalline powder was obtained. This powder was then poured into the prepared bags. c) X-ray diffraction

[0087] To verify the solid structure of the eutectic mixtures, X-ray diffraction was used randomly. This allows us to determine whether mixed crystals have formed or whether the components are still present in pure form.

[0088] The pure substances were added directly to the sample plate and measured. To determine vitrification, the PCM was added to the sample plate and then melted in the furnace. The molten PCM was then cooled to ambient temperature.

[0089] To test for mixed crystals, the material was first weighed into a beaker (ENTRIS224I-1 S, Sartorius Lab Instruments GmbH & Co. KG) and melted in a furnace (UF55Plus, Memmert GmbH & Co. KG). A sample was then taken and, after crystallization, crushed in a mortar. This sample was then placed on the sample plate and measured. Examples a) Materials used

[0090] The materials used below can be found in Table 1.

[0091] Table 1 : Materials as used in the examples. b) Tests carried out

[0092] 1. Hypothermia - all materials

[0093] To determine the undercooling of all materials in Table 1, the onset temperature of crystallization in DSC was used in the first step. These experiments were designed to determine the undercooling and its long-term stability. The degree of undercooling was determined from the percentage difference from complete undercooling to room temperature (100%), see Equation (I).

[0094] 2. Supercooling and dissolution - pure substances and eutectics

[0095] Since subcooling is volume-dependent, it was subsequently investigated in 10 g to 20 g samples. To investigate subcooling on this scale, the pure substances and eutectics were weighed into a beaker in 10 g samples using a balance (PNJ 12000-1 M, KERN & SOHN GmbH). The substances were melted in an oil bath using a stirrer plate with integrated heating (IKA® C-MAG HS 7). The samples were then cooled to ambient temperature in the fume hood. The temperature was monitored during heating (Type K thermocouple inside the sample) and cooling (Type K surface thermocouple outside the sample).

[0096] Using a spatula, the surface is checked for its condition and examined to see whether crystallization can be triggered.

[0097] 3. Long-term stability undercooling / reproducibility - pure substances and eutectics

[0098] The samples were then left to rest at ambient conditions for seven days. Changes during the cooling and long-term test were photographed and documented.

[0099] To verify reproducibility, the process was repeated for the pure substances and eutectics after the seven days. The surface was then examined for its condition using a spatula and investigated for potential crystallization initiation.

[0100] 4. Hypothermia and triggering - non-eutectics

[0101] To investigate supercooling, 20 g samples of the mixtures were weighed into beakers using a Sartorius ENTRIS224I-1S balance. The samples were melted in an oven at temperatures approximately 20 °C above their respective melting points. After removal from the oven, they were weighed again. They were then cooled in a fume hood under ambient conditions, while their temperature was also monitored using a Type K surface thermocouple at the bottom of the beaker. Changes, such as nucleation, were documented with photographs. When the sample reached 70 °C, it was stirred by hand with a spatula to initiate crystallization. If this was successful and the sample showed no crystallization up to 70 °C, it was heated again. During the subsequent cooling process, in addition to the surface thermocouple, the temperature in the sugar alcohol itself was measured using a Type K thermocouple.This time, the material is completely cooled to at least 30 °C and any abnormalities are documented again.

[0102] The subsequent release using a spatula was carried out at 30 °C or lower temperatures. For each material combination, a composition was determined that was sufficiently subcooled and releaseable.

[0103] 5. Long-term stability undercooling / reproducibility - non-eutectics

[0104] To investigate the long-term stability of supercooling, 10 g of the mixture was weighed into a beaker (ENTRIS224I-1S, Sartorius Lab Instruments GmbH & Co. KG). The samples were then melted in a furnace and subsequently cooled to ambient conditions. Following this, they were monitored for seven days, as were the pure substances and eutectics, and changes were recorded photographically. c) Results Pure substances (comparative examples) Characteristics

[0105] The characteristics of the pure substances listed in Table 1 are shown in Figure 6. Myo-inositol has by far the highest melting temperature, while xylitol has the lowest. Erythritol and dulcitol, in particular, stand out due to their high enthalpy. Isomalt, on the other hand, has the lowest enthalpy of fusion.

[0106] Without wishing to be bound by any theory, it is assumed that the different configuration of the chiral carbon atoms, and thus their conformation, is crucial for the thermal properties. This is evident in the significant differences between D-sorbitol, D-mannitol, and dulcitol, which have the same molar mass and are also diastereoisomers.

[0107] The measured specific heat capacities of the pure substances are shown in Figures 7 and 8. Xylitol and erythritol show a significantly flatter slope than the other substances. At 40 °C, D-sorbitol has the lowest specific heat capacity at 1.29 J / (g K) and dulcitol the highest at 1.64 J / (g K). Substances with a higher melting range tend to also have a higher specific heat capacity. Since the specific heat capacity directly influences the overall capacity or performance of a storage device via the sensible heat, high values ​​close to the melting temperature are desirable. For all sugar alcohols, the c p Value in the liquid phase at least 1 J / (g K) above that in the solid phase.

[0108] Hypothermia

[0109] The results of the subcooling tests (DSC, see b1) of the pure substances according to Table 1 are shown in Table 2.

[0110] Table 2: Crystallization enthalpy Ahcr and temperature T cr , as well as the temperature difference AT between melting temperature T m and crystallization temperature T cr If no crystallization occurred, 20 °C was used as Tcr to determine the difference AT.

[0111] The substances can therefore be divided into two groups. One half shows crystallization in DSC (crystallizing phase-change material), and the other half shows no crystallization (vitrifying phase-change material), marked with in Table 2. As an example of a crystallizing phase-change material, the DSC signal of D-mannitol is shown in Figure 9 a). A peak in the measurement signal can be seen upon both heating and cooling. This is where melting or crystallization occurs.

[0112] The difference between T m and T CTis referred to as AT in Table 2 and provides a value for assessing the subcooling performance. For non-crystallizing materials, 20 °C was used as the reference value for T cr All crystallizing materials exhibit supercooling, with myo-inositol showing the lowest at 39.1 °C and dulcitol the highest at 73.0 °C.

[0113] As shown in Figure 9 b) using maltitol as an example, no peak and thus no crystallization can be detected for the group of non-crystallizing sugar alcohols upon cooling in the DSC. This proves that xylitol, D-sorbitol, isomalt, and maltitol, in a sample size of 0.01 g, supercool to room temperature. Even in the subsequent two DSC cycles, no post-crystallization, remelting, or crystallization could be measured for any of these materials. Consequently, stable supercooling behavior can be assumed.

[0114] Since the probability of nucleation and thus crystallization increases with increasing volume, in addition to DSC, tests were also conducted in the 3SK (see b). In the 3SK, the sample was 10,000 times larger than in the DSC. It is assumed that materials that crystallize in small volumes do so even earlier in larger volumes, thus further decreasing the degree of supercooling. Consequently, only substances that do not crystallize in DSC were investigated on a larger scale. Thus, xylitol, D-sorbitol, and maltitol were investigated in the 3SK.

[0115] Figure 10 shows the degree of supercooling u for all DSC (0.01 g) and 3SK samples (100 g). Substances that subcool to room temperature (20 °C) are assigned a degree of supercooling of 100%. Substances that do not completely subcool are assigned a percentage depending on their crystallization temperature.

[0116] Because myo-inositol exhibits the highest melting temperature and the smallest temperature difference between melting and crystallization, it also exhibits the lowest degree of supercooling at 19.5%. It is followed by D-mannitol (35.5%), dulcitol (44.1%), and erythritol (57.7%). Thus, among the group of crystallizing sugar alcohols, erythritol delivers the best results. Xylitol, D-sorbitol, and maltitol exhibit stable supercooling even in the larger volume.

[0117] Long-term stability

[0118] In addition to the general tendency toward supercooling, the stability of this supercooling is also crucial. Long-term tests lasting seven days were conducted to determine any changes in the supercooled melt's exposure to ambient temperature and conditions. The changes during this time were recorded visually and are shown in Figure 11.

[0119] As can be seen in the first column, the sugar alcohols form a transparent liquid when hot and molten. The substances were kept at this temperature in the oven until they were completely melted and no streaks or bubbles were visible.

[0120] These streaks indicate incomplete melting. Bubbles are formed by gas inclusions. For example, the moisture contained in the solids can cause gaseous water to escape. This is favored by the fact that some sugar alcohols have hygroscopic properties. Heterogeneous nucleation can occur on the surface of these gas bubbles, which requires significantly less energy than homogeneous nucleation. In addition, pressure surges occur when the bubbles burst, which can further promote crystallization.

[0121] At the beginning of the experiments, no difference was observed between the materials. The greatest change occurred during the first 24 hours. Xylitol exhibited a fibrous, white surface, while the other materials remained transparent. D-sorbitol exhibited a wavy surface, and maltitol had several cracks running through the material. After seven days, these changes had further increased, but no new effects appeared. All materials had a hard surface at the end of the test series, but D-sorbitol was still deformable.

[0122] The whitish color of xylitol already indicates crystallization, and the crystal nuclei are clearly visible. These form the center of the outward-growing, fibrous crystals. Accordingly, xylitol does not provide long-term stability under supercooling, but rather crystallizes slowly over a period of at least 24 hours at ambient temperature.

[0123] Overall, D-sorbitol and maltitol are shown to be stable supercooling melts, which are still transparent at the end of the test series.

[0124] Triggering

[0125] Substances that tend to crystallize can be easily crystallized by stirring. These substances include the first group of substances that crystallize in DSC: erythritol, D-mannitol, dulcitol, and myo-inositol. Stirring the substances at temperatures close to their melting point can reduce supercooling and trigger crystallization. Xylitol can also be crystallized because, as shown in long-term experiments, this sugar alcohol also does not undergo stable supercooling over time.

[0126] D-sorbitol and maltitol, like isomalt, exhibit strong supercooling. Crystallization at room temperature is not possible. d) Results of eutectic mixtures (comparative examples)

[0127] Preparation of eutectic mixtures

[0128] The composition of the mixtures is calculated according to equations (ll-IV) for components A and B.

[0129] XA,E + XB,E = 1 CD

[0130] Only mixtures in which a component constituted at least 7 percent by weight of the mixture were investigated. For smaller admixtures, the influence of the component was assumed to be too small to have a significant impact on the material properties. A table with the exact composition details can be found in Table 3. To validate the mixture composition of the eutectics determined using equation (II), the DSG signal was monitored. When the eutectic point was reached, a single, narrow peak without tailing was observed.

[0131] Table 3: Composition of the eutectic mixtures used in molar and weight fractions, as well as the melting temperature determined according to equation (II). The enthalpy was determined once according to equation (III) and compared with the more accurate equation (IV).

[0132] Characteristics

[0133] The measured enthalpies of fusion and melting temperatures are shown in Figure 12. The lower melting temperatures of the mixtures compared to the pure substances are clearly visible. Furthermore, the mixtures are often close to the values ​​of one of their components.

[0134] All mixtures have enthalpies above 150 J / g. Since, with the exception of Xyl-Sor E, Ery-Mal E, and Dul-Ino E, one component accounts for more than 70 percent of the mixture, this explains the similarity of the properties to the corresponding pure substance.

[0135] Due to their high enthalpies and variable melting temperatures, these mixtures can significantly expand the range of sugar alcohols that can be used as PGMs.

[0136] The specific heat capacities are shown in Figure 13. In the solid state, Ery-Mal E provided by far the steepest slope of the measurement curve. In addition, the c p Values ​​significantly higher than those of the others. The other mixtures showed a more similar curve. Xyl-Mal E and Sor-Mal E have a somewhat steeper curve than the other three, thus demonstrating a greater temperature dependence. With the exception of Ery-Mal E, the specific heat capacity of all substances is significantly higher in the liquid state than in the solid state.

[0137] The mixtures of maltitol with D-mannitol, dulcitol, and myo-inositol each contained at least 70% maltitol by weight. Since maltitol is the main component of the mixture, this substance also determines the c p Value. The mixture containing erythritol deviated significantly from the others. Erythritol is believed to be responsible for these deviations. However, erythritol has a specific heat capacity of 1.4 J / (gK) at 40 °C, which is significantly lower than the value of the mixture.

[0138] Hypothermia

[0139] Figure 14 shows the influence of sample volume on the degree of subcooling of the DSC and 3SK samples. The exact crystallization temperatures of the 0.01 g samples in the DSC are listed in Table 5.

[0140] Table 4: Representation of the crystallization temperature, enthalpy, and glass transition temperature of eutectic mixtures determined by DSC. If no crystallization occurred, 0 °C was used as Tcr used to determine the difference AT.

[0141] Of the 15 eutectics investigated, 12 undercooled 100% in DSC with a sample volume of 0.01 g. Only Man-Dul E, Man-Ino E, and Dul-Ino E did not undercool as much, but only by less than 50%. These substances are combinations of sugar alcohols, which do not undercool completely even as pure substances.

[0142] In addition to the substances that were already not completely supercooled in the DSC, Ery-Man E, Ery-Dul E and Ery-Ino E also no longer show complete supercooling in the larger volume.

[0143] Eutectics in which the crystallizing substances erythritol, D-mannitol, and dulcitol constitute the majority of the mixture did not exhibit stable supercooling behavior, even when combined with other sugar alcohols. The fact that Dul-Ino E exhibited better supercooling behavior on a larger scale than on a smaller scale indicates the instability of the supercooling. Furthermore, unlike the DSC measurements, the 3SK did not perform a triplicate determination, which is why a larger measurement error must be assumed.

[0144] Despite the increased probability of nucleation due to the larger volume, the six PCMs Xyl-Mal E, Sor-Mal E, Ery-Mal E, Mal-Man E, Mal-Dul E, and Mal-Ino E showed stable supercooling. Thus, it can be assumed that maltitol has a positive effect on supercooling.

[0145] Long-term stability

[0146] The substances tested in the 3SK were then subjected to long-term testing under ambient conditions. Figure 15 shows the results for the mixtures containing maltitol. The remaining substances are shown in Figure 16.

[0147] As a liquid melt directly after heating, all substances presented a uniform appearance. The first differences appeared within the first 24 hours. Xyl-Mal E remained a transparent melt. Sor-Mal E was also transparent, but with an uneven surface. Mal-Man E showed slightly cloudy changes on the surface, while Mal-Dul E and Mal-Ino E were completely transparent, but with stress cracks. After seven days, Xyl-Mal E was fully crystallized. Sor-Mal E appeared unchanged from day 1. Ery-Mal E, Mal-Man E, and Mal-Dul E showed increased surface turbidity. Mal-Ino E showed comparatively few nuclei on the surface.

[0148] The crystallization rate in the highly viscous melt was very slow, which is why nucleation progressed slowly, with the exception of Xyl-Mal E. Mal-Dul E and Mal-Ino E showed surface crystallization after 7 days, although they showed signs of vitrification after day 1. Even after 7 days, the vitrified stress cracks in Mal-Ino E were still visible. Since the samples had direct contact with the environment and maltitol is highly hygroscopic, it is possible that these samples attracted atmospheric moisture.

[0149] In the volume-dependent supercooling test, all mixtures containing maltitol showed complete supercooling. The closed measurement system therefore had a positive effect on supercooling. In the open system of the long-term test series, Mal-Man E showed surface changes after just 24 hours. After seven days, the mixtures containing xylitol and erythritol had crystallized. Accordingly, Xyl-Mal E and Ery-Mal E did not show good supercooling behavior. Mal-Man E, Mal-Dul E, and Mal-Ino E showed a tendency toward crystallization exclusively on the surface and were therefore more stable in their supercooling, with Mal-Ino E showing the most stable supercooling. As the maltitol content in the mixture increased, the supercooling also seemed to become more stable. The only exception to this was Sor-Mal E, which consisted primarily of D-sorbitol, but showed the smallest differences over the 7 days.Therefore, Sor-Mal is considered the most stable eutectic mixture.

[0150] Triggering

[0151] The eutectic mixtures, which were investigated for their long-term stability during supercooling, were also investigated for the deliberate induction of their crystallization. In this series of experiments, this was determined by stirring the substance by hand.

[0152] As already seen in Figure 15, Mal-Dul E and Mal-Ino E exhibited stress cracking at room temperature, indicating vitrification. This was confirmed by DSC. The vitrification temperature was determined to be 32.3 °C for Mal-Man E, 39.6 °C for Mal-Dul E, and 47.1 °C for Mal-Ino E. Since all three temperatures are above room temperature, dissolution of these substances was therefore not possible. Since Ery-Mal E (Tm = 110.3 °C) and Xyl-Mal E (Tm = 90.7 °C) began to crystallize spontaneously, dissolution at room temperature was no longer possible. At temperatures above room temperature, however, crystallization could be initiated and accelerated by stirring. Sor-Mal E was the only eutectic mixture that allowed penetration with a spatula at room temperature; however, its viscosity is too high, and stirring or dissolution was not possible.In summary, none of the eutectic mixtures is suitable for the targeted induction of crystallization. The high vitrification temperatures, which are clearly related to the maltitol concentration in the mixtures, are a major obstacle here. The use of other initiation methods that introduce more energy into the PGM may be helpful, especially with Sor-Mal E, but initiation must not consume more energy than it gains. e) Results of non-eutectic mixtures (inventive examples).

[0153] As previously discussed, pure sugar alcohols can be divided into two groups: those that exhibit crystallization (erythritol, D-mannitol, dulcitol, and myo-inositol) and those that vitrify (D-sorbitol, maltitol, and isomalt).

[0154] By combining vitrifying and crystallizing substances from these two groups, a composition with the desired properties can be formed. Consequently, D-sorbitol, maltitol, or isomalt are combined with erythritol, D-mannitol, and dulcitol to achieve a non-eutectic composition that simultaneously provides stable supercooling and allows crystallization to be triggered.

[0155] Hypothermia

[0156] First, the influence of the vitrifying ZA on the supercooling of the substances erythritol, D-mannitol, and dulcitol was illustrated at the smallest scale. Subsequently, the volume influence was investigated at a 20 g scale, followed by the long-term stability of the supercooling. Figure 17 shows the degree of supercooling at the smallest scale (0.01 g) for mixtures containing dulcitol and D-mannitol.

[0157] Erythritol and its mixtures showed no crystallization in DSC and therefore all had a degree of supercooling of 100%. Their representation is therefore omitted. Mixtures with D-mannitol showed complete supercooling up to approximately 55 wt% mannitol. The exception was the mixture with isomalt, which only stopped showing complete supercooling at 70 wt%. Thus, isomalt had the best supercooling properties with D-mannitol, followed by D-sorbitol and maltitol. Maltitol provided the least supercooling at 60 wt% D-mannitol. The mixtures with dulcitol all stopped showing complete supercooling only at 74 wt% dulcitol. Here, maltitol and isomalt produced almost identical results, while D-sorbitol led to lesser supercooling. In summary, on this scale, erythritol mixtures subcool best, followed by dulcitol and D-mannitol.Erythritol, as a pure substance, tended to experience greater supercooling than the other substances; accordingly, even small amounts of crystallization inhibitors were sufficient to completely prevent crystallization in DSC. The tendency toward pure substance supercooling was also reflected for D-mannitol and dulcitol. In addition, isomalt appeared to be able to suppress crystallization effectively for all substances. Maltitol also yielded good results with erythritol and dulcitol, but this effect was not present with D-mannitol. D-sorbitol showed the poorest results, especially with high amounts of crystallizing material.

[0158] In the next step, subcooling was investigated on a larger scale at ambient conditions to determine the influence of volume. In 20 g samples in a beaker, subcooling to 70 °C was first investigated. If this was successful, the next step was to investigate subcooling to ambient temperature.

[0159] As an example, Figure 18 shows images of mixtures with isomalt of different compositions. The melts were subcooled to 70 °C. At the highest isomalt content, no crystallization was observed. Only with Iso-Dul were small air bubbles still visible in the mixture. The same applies to 60 and 50 wt.% isomalt. The first crystallization nuclei can only be seen at 40 wt.% isomalt.

[0160] A similar picture was observed for mixtures with D-sorbitol and maltitol. At 40 wt% D-sorbitol and above, erythritol crystallization was suppressed. For D-mannitol and dulcitol, at least 60 wt% D-sorbitol was required. In combination with maltitol, D-mannitol and dulcitol required at least 50 wt% maltitol, but with erythritol, only 40 wt%.

[0161] This confirmed that the supercooling of erythritol, D-mannitol, and dulictol can be influenced by the addition of vitrifying materials, even on a larger scale. In each case, 40 to 60 wt.% of the vitrifying material was required to effectively suppress crystallization. To precisely define the combination of materials, further supercooling experiments were conducted for mixtures with this composition. This time, however, supercooling down to ambient temperature was investigated.

[0162] Table 5 lists the maximum amounts of the crystallizing substance in the mixture. Larger amounts of this component lead to nuclei forming in the melt. As previously shown, the enthalpy also increases with increasing amounts of the crystallizing substance, so the maximum value listed here is always used for further investigations.

[0163] Table 5: Composition of the most promising non-eutectic mixtures

[0164] Both D-sorbitol and maltitol required over 50 wt.% to successfully suppress crystallization on this scale. For isomalt, exactly 50 wt.% was required with erythritol and D-mannitol, while 60 wt.% was required in combination with dulcitol. This means that for all substances, at least 50 wt.% of vitrifying material was required to suppress crystallization at room temperature.

[0165] Long-term stability

[0166] The long-term stability of these substances was then examined. This is shown in Figure 18. As before, all substances had the same appearance of a transparent liquid in the liquid state. Initial changes occurred over the next 24 hours. All erythritol mixtures showed surface nuclei. However, Mal-Ery showed the fewest nuclei. The nuclei and crystallization edges of Sor-Ery and Iso-Ery were clearly visible. However, these substances were still largely transparent. Sor-Man also showed surface crystallization, but this was noticeably whiter in color. All other substances showed no change on the first day. After seven days, both Sor-Ery and Sor-Man had a more intense white color. Mal-Ery and Iso-Ery also appeared less transparent. Mal-Man and Iso-Man were the only substances that showed no change. Mal-Dul and Iso-Dul both showed small, white nuclei. Iso-Dul also had stress cracks.Although several substances showed a color change after seven days, complete crystallization did not appear to have occurred. Had this been the case, a solid white substance would have resulted. Erythritol showed the most stable supercooling with maltitol. For D-mannitol, no differences were found between maltitol and isomalt. The same applies to dulcitol, although stress cracks indicated vitrification and thus a tendency towards more stable supercooling. In general, the long-term stability of supercooling with D-sorbitol is less stable than with maltitol or isomalt. Furthermore, since these mixtures also used the largest amounts of crystallization-inhibiting substances, it can be assumed that D-sorbitol is less suitable for this purpose than the other two substances.

[0167] Triggering

[0168] All material compositions that had previously been tested for supercooling were first tested for their initiability at approximately 70 °C. All materials containing less than 80 wt.% of the vitrifying component were also initiable. This was demonstrated through laboratory tests. In each case, crystallization was initiated by mechanical stirring with a spatula. Crystallization was clearly recognizable by a color change from transparent to white. The more crystallizing material the mixture contained, the easier it was to initiate crystallization. The materials that delivered the best results in terms of supercooling and thermal properties were then tested for their initiability at ambient temperature. The composition corresponds to Table 5. At temperatures below 30 °C, Sor-Ery could be stirred, but no crystallization was detectable.The actual initiation of crystallization presumably requires significantly more energy. Iso-Ery was also still malleable but could no longer be fully initiated. Mal-Ery was only tested up to 55°C, at which temperature crystallization could be easily initiated. Crystallization initiation experiments at lower temperatures were carried out on a larger scale. Sor-Man was stirrable at 30°C, but only partially crystallized. Mal-Man also could be easily initiated up to approximately 60°C, but this was no longer possible at lower temperatures. Iso-Man was completely vitrified at 30°C and could not be initiated. Sor-Dul was malleable at temperatures below 30°C, but crystallization could not be initiated, and the substance remained transparent. Mal-Dul and Iso-Dul were neither malleable nor initiable. It was striking that all mixtures with erythritol and D-sorbitol were still malleable at room temperature.Since DSC also failed to detect a glass transition point up to 20 °C, these materials are not vitrified and therefore theoretically soluble. Due to the soluble method chosen here, involving manual stirring, the amount of energy introduced is difficult to estimate. It can therefore be assumed that with a suitable stirring tool and sufficient energy, the materials could be crystallized in a targeted manner. The other mixtures were not deformable at 30 °C. This is consistent with the DSC data, which indicate a T. g at 44.1 °C and for Iso-Man a T g at 45.03 °C. Mal-Dul and Iso-Dul also showed a glass transition at 28.7 °C and 34.6 °C, respectively. Therefore, crystallization can only be successfully initiated at higher temperatures. However, these substances generally show good stability during supercooling. f) Non-eutectic mixture in a 10L storage tank

[0169] The non-eutectic mixture of 53 wt% maltitol and 47 wt% erythritol was investigated in a 10-liter storage tank. Unlike previous scales, this tank offers a comprehensive investigation of the system consisting of the storage tank and PCM. In addition to the storage tank structure, the temperature distribution within the tank is discussed. Subcooling and resolvability are also addressed.

[0170] Construction

[0171] To ensure the greatest possible subcooling, materials with low surface roughness were selected. The smoother the surface, the fewer points of attack for heterogeneous crystallization. The storage tank wall and base were made of glass. The heating coil was made of stainless steel 1.4301 with a surface roughness of Rz 4 and Ra 0.8, respectively. To provide as few points of attack for heterogeneous crystallization as possible, the first version of the 10 l storage tank used to investigate the air trigger method contained only three thermocouples (see Figure 20; gray points 180, 190, and 200). However, it was shown that the thermocouples did not represent points of attack for crystallization. Therefore, for the stirrer trigger method, the number of thermocouples was increased to 7 to gain a better understanding of the temperature distribution. The exact placement can be seen in Figure 20.The thermocouples were positioned on three concentric circles. The outermost red circle was located between the glass wall and the heating coil. The blue and green circles were located inside the heating coil. Since space was required for the triggering method there, the thermocouples could only be installed in the upper and middle areas. The stirrers and the glass frit were therefore located in the center of the tank to ensure the best possible mixing. The anchor stirrer was installed as shown in Figure 20 a) and was 50 mm above the bottom. The ViscoJet was installed in the center and located 200 mm above the bottom. The glass frit was placed directly on the bottom. Due to the cylindrical shape of the tank, which was intended to avoid dead spaces, and the triggering mechanisms in the center, the space for the heating coil was limited. The total length of the heating coil was 5 m with an inner tube diameter of 13 mm.To make optimal use of the space, it was laid out in eleven turns with a diameter of 127.5 mm. The heat transfer fluid entered the storage tank and flowed from top to bottom through the heating coil, before exiting again via a straight, upward-running pipe at the edge. The connection to the test bench was made via two 1 m long hoses. The temperature measurement in the HTF (heat transfer fluid) is located before DV-3 and after DV-4 using the PT100 thermocouples TIR130 and TIR140, see Figure 21 . Directly next to each one there is also a pressure measurement. The flow in the entire test bench is measured and controlled by the turbine flow meter FCR10. In addition, the variable area flow meter FIR20 is located in the return line. This setup allows subcooling and triggerability to be investigated as before. However, the thermocouples provide a more accurate picture of the temperature distribution in the PGM.In addition, the heating coil and test bench provide an opportunity to calculate the performance and efficiency. However, due to the small heat exchange surface and residence time, the temperature difference between the inlet and outlet of the storage tank is small, which is why detailed statements cannot be made and the values ​​should be viewed more quantitatively.

[0172] Hypothermia

[0173] Figure 22 shows the temperature profile in the PGM. The diagram starts with a completely melted storage device and shows the temperature profile as the PGM cools without triggering. The glass jacket of the storage device allows for precise visual differences in the PGM. Visual inspection of the storage device in the supercooled state revealed no crystallization or nucleation. The melt was transparent and showed few inclusions of gas bubbles.

[0174] Triggering

[0175] At a trigger temperature of 70 °C, the anchor stirrer, the ViscoJet, and the injection of air were compared. The focus here is initially on the temperature distribution during the trigger phase, which can be seen in Figure 23. During this time, the valves towards the test bench were closed so that the heat distribution could be observed without the influence of the HTF. The temperature distribution with the anchor stirrer was very uniform. Initially, all temperatures converged due to stirring. After 20 minutes, a first peak indicated the onset of crystallization, and after 25 minutes, crystallization occurred throughout the entire volume.

[0176] With the ViscoJet, the temperature at the bottom (u) of the reservoir deviated significantly from that in the middle (m) and top (o). Due to its location in the middle of the reservoir, this stirrer type mixed the lower area of ​​the reservoir significantly less effectively. Thus, TIR180 only showed temperatures of 40 °C throughout the entire triggering process. TIR210 also showed lower temperatures, but was still triggered. TIR180 was obviously located in an area that was difficult for the ViscoJet to reach, and mixing was insufficient in the lower area of ​​the reservoir. Nevertheless, it was noticeable that the time until the temperature jump was significantly shorter than with the anchor stirrer. Crystallization started after just 4 minutes and also exhibited significantly fewer temperature peaks.

[0177] The temperature fluctuated more significantly with the air trigger method in the middle (m) and top (o) than with the stirrers. Since the air used for triggering was at ambient temperature, this also caused temperature fluctuations as soon as air bubbles reached the thermocouples. Because the bubble distribution in the lower region was not uniform across the entire area, this temperature measurement was more uniform. Nevertheless, the temperature in the lower region approached that of the upper region after approximately 12 minutes. 18 minutes after the start of the triggering, the temperature jump caused by the phase change became apparent.

[0178] The fastest crystallization initiation was achieved with the ViscoJet. This type of stirrer is particularly suitable for mixing highly viscous media. This was advantageous here, as the mixture becomes more viscous as the temperature decreases and the number of nuclei increases. If a consistently high energy input from the stirrer is still possible, rapid crystallization occurs.

[0179] In principle, initiating crystallization at 70 °C has always been most successful. This is due to two opposing effects. When crystallization is initiated at 100 °C, the viscosity is low, allowing the stirrers to work most effectively. At the same time, supercooling is low, meaning the energetic pressure to adopt the thermodynamically more stable form is low. At 50 °C, this is reversed: the energetic pressure for crystallization is high, but the high viscosity prevents rapid crystallization and the introduction of mechanical energy through the initiation method. At 70 °C, both effects balance each other out, and simple crystallization is possible.

Claims

Claims 1. A composition for use in a long-term latent heat storage device, the composition comprising a non-eutectic mixture of a crystallizing phase change material and a vitrifying phase change material.

2. The composition according to claim 1, wherein the crystallizing phase change material and / or the vitrifying phase change material comprises, preferably consists of, a sugar alcohol.

3. The composition according to claim 2, wherein the sugar alcohol is selected from the group consisting of D-sorbitol, xylitol, erythritol, maltitol, isomalt, D-mannitol, dulcitol and myo-inositol.

4. The composition of claim 3, wherein the sugar alcohol for the crystallizing phase change material is selected from the group consisting of erythritol, D-mannitol, dulcitol, and myo-inositol.

5. The composition of claim 3, wherein the sugar alcohol for the vitrifying phase change material is selected from the group consisting of D-sorbitol, xylitol, maltitol and isomalt.

6. The composition according to any one of the preceding claims, wherein the crystallizing phase change material and the vitrifying phase change material are present in a weight ratio in the range of 60:40 to 40:

60.

7. The composition according to any one of the preceding claims, wherein the crystallizing phase change material and the vitrifying phase change material have a melting temperature T m in the range of 90 to 250 °C.

8. The composition according to the preceding claim 7, wherein the crystallizing phase change material has a melting temperature T m in the range of 105 to 250 °C.

9. The composition according to any one of the preceding claims 7 or 8, wherein the vitrifying phase change material has a melting temperature T m in the range of 90 to 150 °C.

10. The composition according to any one of the preceding claims, wherein the crystallizing phase change material and the vitrifying Phase change material a melting enthalpy Ah m in the range of 100 to 350 J / g.

11. The composition according to the preceding claim 10, wherein the crystallizing phase change material has a melting enthalpy Ah m in the range of 245 to 350 J / g.

12. The composition according to any one of the preceding claims 10 or 11, wherein the vitrifying phase change material has a melting enthalpy Ah m in the range of 100 to 245 J / g.

13. Use of a composition according to any one of the preceding claims 1 to 12 for long-term latent heat storage.

14. A long-term latent heat storage device comprising a composition according to any one of the preceding claims 1 to 12.

Images