Porous Endothermic Article

Abstract

The present disclosure relates to a shaped article for an energy storage device comprising greater than 60.0 wt % of an inorganic endothermic material and having an open porosity of greater than 10% v/v and less than 60% v/v, wherein the inorganic endothermic material comprises particles of inorganic endothermic material coated with a binder.

Claims

1. A freestanding shaped article for an energy storage device comprising greater than 60.0 wt % of an inorganic endothermic material and having an open porosity of greater than 10% v/v and less than 60% v/v, wherein the inorganic endothermic material comprises particles of inorganic endothermic material coated with a carbonaceous binder.

2. The article according to claim 1, wherein the carbonaceous binder comprises at least 1.0 atomic % carbonyl groups.

3. The article according to claim 1, wherein the carbonaceous binder comprises an atomic ratio of oxygen to carbon of at least 1:15.

4. The article according to claim 1, wherein the carbonaceous binder does not have a melting point or has a melting point above an onset decomposition temperature of the endothermic material.

5. The article according to claim 1, wherein the article comprises at least 95 wt % inorganic endothermic material, and the inorganic endothermic material density is greater than 60% and less than 90% of a theoretical maximum density of the endothermic material.

6. The article according to claim 1, wherein the article has an open porosity in the range of 20% to 60% v/v.

7. The article according to claim 1, wherein the inorganic endothermic material density of the article is in a range of 60% to 80% of the maximum theoretical density of the inorganic endothermic material.

8. The article according to claim 1, wherein the article comprises at least 90 wt % of inorganic endothermic material.

9. The article according to claim 1, wherein the inorganic endothermic material comprises particles with a bimodal particle size distribution.

10. The article according to claim 9, wherein peaks of the bimodal distribution are between 30 and 200 microns apart.

11. The article according to claim 1, wherein the binder loading is no more than 1 g per 20 m.sup.2 of surface area of the endothermic material particles.

12. The article according to claim 1, wherein the article does not deform greater than 5% of an original dimension of the article when subjected to a pressure of 74.4 kPa over a temperature range of room temperature to 500° C.

13. The article according to claim 1, wherein a modulus of rupture is at least 400 psi measured in accordance to ASTM C203 Method I.

14. The article according to claim 1, wherein the article has a moisture weight gain of less than 5 wt % when tested in accordance to ISO 1716 standards.

15. The article according to claim 1, wherein the article is a housing comprising a plurality of recesses shaped to receive a plurality of electrochemical cells.

16. The article according to claim 1, wherein the article has a thermal conductivity (measured at 40° C.) of less than 5.0 W/m.Math.K.

17. A process for the production of a freestanding shaped article, the process comprising: (a) mixing together a formulation comprising: (i) particles of an inorganic endothermic material; (ii) a fugitive thermoplastic binder with a melting point below an endothermic decomposition temperature of the inorganic endothermic material; and optionally (iii) one or more additives to form a mixture; (b) heating the mixture above the melting point of the thermoplastic binder and below the endothermic decomposition temperature of the inorganic endothermic material; (c) shaping the mixture into a shaped article; and (d) removing at least part of the thermoplastic binder from the shaped article to leave a carbonaceous binder coating the particles of inorganic endothermic material.

18. The process according to claim 17, wherein the carbonaceous binder comprises products resulting from decomposition or crosslinking of the thermoplastic binder, the optional additives, or both.

19. The process according to claim 17, wherein the thermoplastic binder is removed at a temperature above the melting point of the thermoplastic binder and below the endothermic decomposition temperature of the inorganic endothermic material.

20. The process according to claim 18, wherein the process of removing part of the thermoplastic binder results in decomposition of the thermoplastic binder.

21. The process according to claim 18, wherein the thermoplastic binder is removed at a temperature at or above the decomposition temperature of the thermoplastic binder.

22. The process according to claim 17, wherein the thermoplastic binder comprises a polymer or wax and has a melting point in the range of 30° C. to 100° C.

23. The process according to claim 17, wherein the one or more additives comprise a surfactant.

24. The process according to claim 23, wherein the surfactant comprises a carbonyl group.

25. The process according to claim 23, wherein the formulation comprises between 10 wt % and 30 wt % surfactant relative to the thermoplastic binder.

26. The process according to claim 21, wherein the surfactant comprises or consists of a fatty acid.

27. The process according to claim 17, wherein the thermoplastic binder and additives comprise a paraffin wax and a fatty acid.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0145] FIG. 1 is a graph of the particle size distribution of Aluminium Trihydrate (ATH) in a preferred embodiment of the present disclosure.

[0146] FIG. 2 is a SEM image of an article comprising the endothermic material having the particle size distribution of FIG. 1, prior to dewaxing.

[0147] FIG. 3 is a SEM image of an article of FIG. 2 after dewaxing.

[0148] FIG. 4 is an Energy Dispersive X-ray spectroscopy (EDS) display of the SEM image of FIG. 3, with the light colour representing the presence of carbon.

[0149] FIG. 5 is a SEM image of another article after dewaxing.

[0150] FIG. 6 is an EDS display of the SEM image of FIG. 5, with the light colour representing the presence of carbon.

[0151] FIG. 7 is a perspective view of a battery housing design moulded using the ATH particles of FIG. 1.

[0152] FIG. 8 is a photograph of the test equipment used in determining the gas generation of the endothermic composition forming the housing.

[0153] FIG. 9 is a graph of the thermal mechanical analysis of a dewaxed endothermic article.

[0154] FIG. 10 is a graph of the thermal expansion of a sample upon heating and cooling of an endothermic article comprising the endothermic material.

[0155] FIG. 11 is a graph of the Thermal Gravimetric Analysis (TGA) and Differential Scanning calorimetry (DSC) of the endothermic material.

DETAILED DESCRIPTION OF A PREFERRED EMBODIMENT

[0156] An approximately 25 Kg batch was prepared by measuring out 21.25 Kg of Aluminium Trihydrate (ATH); 3.00-3.19 Kg of paraffin wax (melting point: 52° C.); and the remaining 0.56-0.75 Kg of stearic acid. The material in its injected state is comprised of 82-87 wt % Aluminium Trihydrate (ATH) (containing a maximum of 0.3 wt % Na as Na.sub.2O).

[0157] The remaining 13-18 wt % is comprised of organics which can be broken down further to 15-20 wt % stearic acid and 80-85 wt % paraffin wax. The stearic acid has two functions in the formulation, acting as a wetting agent for the mix and a lubricant for injection into the moulds. If the stearic acid drops below 1.5 total wt % then the material does not mix well and as a result there are issues filling a desired mould. The paraffin wax acts as a fugitive binder and, when the mix is heated above the melting point of the paraffin wax (e.g. about 52° C.), the mixture's viscosity is sufficiently reduced to fill the desired mould.

[0158] The ATH can be characterized further by particle size distribution with the final mixture having a d10 of 4.46 micron; a d50 of 30.7 micron; and a d90 of 148 micron. The particle size distribution is illustrated in FIG. 1. A bimodal particle size distribution was obtained through blending two separate batches of ATH, each having a different d50; one about 100 micron and the other just over 10 micron. The surface area of the ATH was 1.07 m.sup.2/g.

[0159] An exemplary bimodal distribution might have a first peak in the range 5-30 micron and a second peak in the range 50-300 micron. The present disclosure does not require such a distribution, and does not exclude distributions having only one, or having more than two peaks.

[0160] FIG. 2 further illustrates the bimodal distribution of particles within the wax based organic component (darker phase).

[0161] The materials are added to a heated mixer and mixed for about 6 to 10 hours at a heat setting to melt the wax and mixing time to obtain a homogeneous mixture. The mixture is transferred to an injection moulding press with a heated cavity. The cavity is heated to between 54-65° C. to maintain the desired low viscosity of the mixture.

[0162] The desired mould is placed on the machine and the mould cavity is sprayed with a silicon lubricant for ease of ejection. The mixture is injected to the desired mould at 2200-6650 kPa; the pressure is determined by the mould complexity and size and will require trials to optimize the settings for each specific mould. The cycle time per part will vary anywhere from 1-5 minutes depending on part size and geometry.

[0163] The mould is allowed to rest for a short period of time (e.g. 10 to 30 minutes) to allow for solidification of the wax. The resulting material is now in a solid state in the desired shape from the mould with the addition of the sprue from injecting. The sprue is removed and discarded, and the material is transferred to a setter with a powdery packing media used to draw out and absorb the organic components (wax and stearic acid).

[0164] Once the setter is filled with parts the parts are covered with the powdery packing media. The setters are then transferred to a tunnel kiln. The kiln is progressively heated from 110° C. to about 200° C. to wick out the wax (in air); this process typically takes 18-36 hours. It is thought that the extended period the organic material is exposed to high temperatures in an oxidative environment results in the residual organic material beginning to decompose and react and bond the ATH particles together. While the processing temperature does not reach the thermal decomposition of stearic acid, it does reach the decomposition of the wax. It is thought that reactive carbon-oxygen bonds, either created in the oxidative environment and/or the carbonyl groups in the surfactant, result in reaction products which bond the endothermic particles together.

[0165] The upper temperature limit is selected to avoid decomposition of the endothermic material. The parts are then removed from the packing media and any traces of the powdery media are removed by brush or air.

[0166] FIGS. 3 & 5 illustrates the material after the dewaxing process with void space occupying where the organic matter previously was. However, through EDS analysis (FIGS. 4 & 6), residual organic matter may be detected through the detection of carbon on the surfaces and congregating at the interfaces between particles, thereby functioning as a binder. The carbonaceous binder appears to concentrate around the smaller particles (i.e. areas with relatively high surface area), indicating that the bimodal particle size distribution may possess a synergistic benefit in terms of packing density, mechanical strength and adhesion between particles.

[0167] The resulting material (sample 7) is comprised of about 98 wt % ATH with the remaining weight percentage being the residual decomposed organic content (wax and stearic acid) and inorganic impurities including silica, calcia, magnesia, sodium oxide, iron oxide, and zirconia. The remaining organic content contributes to the overall MOR strength of the material while not affecting the flow of material during a thermal event and/or normal operating temperatures.

[0168] On the basis of there being 2 wt % residual carbonaceous binder, then 2 g of residual carbonaceous binder coated (98 g×1.07 m.sup.2/g) 104.9 m.sup.2 of ATH particles (1 gram per 52.4 m.sup.2 of ATH particles).

[0169] The residual organic content was determined by a mass balance of materials used and material removed in the dewaxing process.

[0170] After complete thermal decomposition the remaining material comprises>99 wt % alumina.

[0171] The process has some similarities with investment or lost-wax casting, with the distinction that the wax is not used to form wax patterns, but to act as a carrier and binder for the endothermic material. However, like investment casting it is able to produce components with high accuracy, repeatability and versatility.

[0172] As illustrated in FIG. 7, the moulded battery housing 10 is a hexagonal shape comprising seven cavities 20 each for receiving a 21700 size cylindrical cell. It will be appreciated that the weight of the housing could be further reduced by providing one or more open cavities to reduce the mass of material used.

[0173] The housing 10 had a hexagonal close packed design with seven cells. The cylindrical cavity diameter is between 21.2 mm and 21.6 mm, whilst the distance between adjacent central axes is 22.9 mm, resulting in a minimum wall thickness 30 of between 1.3 mm and 1.5 mm. The hexagonal shaped housing also has regions of greater wall thickness 40, adjacent the outer perimeter of the housing, thereby contributing to the structural stability of the housing.

[0174] For destructive battery testing the housings were wrapped with a 3.2 mm Superwool® Plus paper (for cushion) and placed inside an aluminium shell to simulate the protective cover article in operation (not shown).

[0175] In use, the battery housing holds seven 21700 size cylindrical batteries, with one end of each battery interfacing with a positive side current collector interfacing with battery connectors protruding from the cavities 20; and a negative side current collector interfacing with battery connectors protruding from the cavities 20 at the opposing ends (not shown). Other components, such as insulating plates, energy management system circuitry and sensors may also interface with the batteries and/or battery holders.

[0176] In some embodiments, the housing may contain cavities for the insertion of sensors into the housing to monitor the conditions, such as temperature. The use of PIM enables narrow conduits (e.g. less than 5.0 mm, preferably less than 2.5 mm and even more preferably less than 1.0 mm diameter) to be pre-formed into the housing, thereby avoiding the need to machine such design features in a separate operation.

Experiments

[0177] Comparative example (CE1) is a test sample made from a composite material comprising inorganic fibre and ATH (approximately 62 wt %). In comparison a composition (sample 7; 98 wt % ATH) under the scope of the present disclosure is able to provide between 3.24 to 4.28 times more endothermic absorptive capacity per unit volume of material. This is due to an increase in density of endothermic material and an increase in concentration of endothermic material (i.e. lower content of non-endothermic material). Samples 1 to 6; and 8 to 13 referred to below were produced using the same methodology as with sample 7, except for the differences noted in Table 4 (particle size distribution); loading of organic material in Table 5; and fillers in Table 8.

Relative Gas Generation

[0178]

TABLE-US-00002 TABLE 1 Theoretical Total Gas Collected Gas/Generation Expansion water Vapor Expansion ml/g of ml/g of ml/g of Sample material material material CE1 41.0 0.173 335.3 7 44.0 0.233 440.3

[0179] The gas expansion and gas generation data were obtained through the methodology as follows:

[0180] 1. The condensation chamber is weighed and clamped in place, 10 grams of the desired test material is weighed out +/−0.05 grams and placed in the 250 ml beaker, and the plugs were put in place to seal the system.

[0181] 2. The water chamber is filled to equilibrium for the pressure and temperature of the room.

[0182] 3. The Bunsen burner is lit and set at a distance so that the tip of the inner blue flame is at the base of the beaker.

[0183] 4. The test is allowed to run for 20 minutes before removing the Bunsen burner.

[0184] 5. The collected water in the graduated cylinder is measured and reported.

[0185] 6. The condensation chamber is weighed. Assuming pure water (density of 1 g/cc) the ml of water is recorded.

[0186] 7. Using the Ideal Gas Law (PV=nRT) the theoretical expansion of the water from liquid to vapour is calculated and reported.

[0187] With reference to FIG. 8, (from left to right), there is illustrated a propane powered Bunsen burner (1) is placed below a 250 ml beaker (2) containing the test material. The beaker is hermetically connected to a condensation chamber (3) which in turn is hermetically connected to a water chamber (4). The water chamber deposits water into a graduated cylinder (5) for measurement.

Relative Density, MOR, Hydrophobicity; Flammability and Combustibility

[0188] The residual coating of organic material (carbonaceous binder) on the surface of the article resulted in a very low level of moisture absorption. Additionally, despite the composition not containing fillers, the MOR of the article under the present disclosure is significantly higher than the comparative example comprising inorganic fibre.

TABLE-US-00003 TABLE 2 Hydrophobicity % Sample Density (Pcf) MOR (psi) wt gain (ISO 1716) CE1 43 268 74%  7 102 769 1%

[0189] The density was determined by weighing a sample of known volume.

[0190] Modulus of Rupture (MOR) was determined according to ASTM C203 Method I.

[0191] The hydrophobicity was determined according to ISO 1716.

[0192] Sample E-1 did not propagate a flame according to UL 94, a V-0 rating recorded (i.e. no glowing after 30 seconds, no flame or combustion after being exposed to the flame).

[0193] The LOI test procedure (900° C. hold for 30 minutes) resulted in an LOI of 35%. This is mostly due to the conversion of the chemically bound water from the ATH.

[0194] E-1 also passed ASTM136 (Standard Test Method for Behaviour of Materials in a Vertical Tube Furnace at 750° C.) as non-combustible.

Surface Analysis (XPS)

[0195] XPS experiments were performed using a Physical Electronics VersaProbe II instrument equipped with a monochromatic Al ka x-ray source (hv=1,486.7 eV) and a concentric hemispherical analyzer. Charge neutralization was performed using both low energy electrons (<5 eV) and argon ions. The binding energy axis was calibrated using sputter cleaned Cu (Cu 2p.sub.3/2=932.62 eV, Cu 3p.sub.3/2=75.1 eV) and Au foils (Au 4f.sub.7/2=83.96 eV).

[0196] Peaks were charge referenced to CH.sub.x band in the carbon 1s spectra at 284.8 eV. Measurements were made at a takeoff angle of 45° with respect to the sample surface plane. This resulted in a typical sampling depth of 3-6 nm (95% of the signal originated from this depth or shallower). Quantification was done using instrumental relative sensitivity factors that account for the x-ray cross section and inelastic mean free path of the electrons.

[0197] Ground material was analysed of sample 7 before (green) and after the de-waxing step (final product). The results (Table 3) indicate that the surface of both samples was coated with a carbonaceous material. This was confirmed with EDS analysis which confirmed that a surface layer of carbon at least substantially, if not entirely, covered the surface of the ATH particles. While only a small amount of carbon in the green sample was present as carbonyl groups, likely to be derived from the stearic acid, the carbonaceous material in the final product had a higher atomic ratio of oxygen to carbon, indicating a likely higher concentration of stearic acid and the formation of thermal oxidative degradation products during the dewaxing process. Two possible spectra curve fits were used to calculate the proportion of functional groups, with the carbon/oxygen functional groups determined to be O═C—O, with the possibility of the presence of a C═O functional group.

[0198] On the basis that Al is present as Al.sub.2O.sub.3 and Na is present as Na.sub.2O, then the remaining oxygen is 19.5 atomic %. This compares with 4.5 atomic % in the green sample, with 1 atomic % of this attributable to the O═C—O group. The final product has a ratio of oxygen to carbon of 19.5 to 41.8 (1:2.1). In contrast, the green product has a ratio of oxygen to carbon of 4.5 to 85.0 (1:18.9).

TABLE-US-00004 TABLE 3 (atomic %) C as C as C as CH.sub.x/ Sample Al Na O C.sup.total CH.sub.x O═C—O C═O COO 7 (green) 4.1 0.2 10.7 85.0 84.5 0.5 — 178 7 (final 14.8 1.1 42.2 41.8 35.6 3.0 3.3 12 product) Fit 1 7 (final 14.8 1.1 42.2 41.8 36.3 5.6 — 6 product) Fit 2

Thermal Mechanical Analysis (TMA)

[0199] TMA (Thermal Mechanical Analysis) was performed on sample 7 (green) and sample 7 (final product). The test methodology is based on ASTM E228, but with the application of an applied load. A Netzsch TMA 402 F3 Hyperion machine was used. The sample size was ¼″×¼″×1. The sample was placed vertically into the test chamber and a force of 3N was applied to the ¼″×¼″ face (or 74.4 Mpa of pressure) of the sample. During the test, the samples were heated at a rate of 0.5° C./min and subjected to 74.4 kPa of pressure. The testing equipment measures material displacement as a function of temperature. The displacement or distortion was measured as a % of the original sample dimension in the direction of the applied force (i.e. 100% displacement corresponds to a 1″ displacement).

[0200] For sample 7 (green), the sample failed (e.g. deformed greater than 100% of its original length) at about 50° C., corresponding to the softening/melting temperature of the wax. This indicated that the mechanical strength of the sample was limited to the mechanical strength of the binder (wax) at elevated temperatures.

[0201] For sample 7 (final product), the material initially expanded before beginning to deform against the subjected force at about 212° C., corresponding to the temperature at which the ATH commences degradation. The sample deformed to a maximum of 2.5% of its original length up to a temperature of 500° C. (FIG. 9). This indicated that the residual binder was able to maintain the mechanical integrity of the sample up until at least the endothermic on-set temperature. This indicates that the residual organic material has an increased melting temperature or is no longer a thermoplastic material (e.g. a bonded to ATH, crossed-linked and/or gelled compound).

Thermal Expansion

[0202] Thermal Expansion was determined in accordance to ASTM E228. As illustrated on FIG. 10, the linear expansion of a sample of the material increases to almost 0.4% at amount 250° C. before shrinking to −1.8% up to 500° C. corresponding to the release of ATH decomposition products (H.sub.2O and CO.sub.2). The shrinkage resulted a minor cracking although the sample maintained sufficient mechanical integrity to provide an insulative barrier after the endothermic material had decomposed.

Thermal Gravimetric Analysis and Differential Scanning Calorimetry

[0203] The Thermal Gravimetric Analysis (TGA) and Differential Scanning calorimetry (DSC) was measured using Simultaneous Thermal Analysis (STA). The data reported was provided by a third party (The National Brick Research Center, Clemson University) who collected the data in the following conditions: ambient temperature (20° C.) to 1000° C. at a ramp rate of 20° C. per minute in a standard air atmosphere with the sample contained within an alumina crucible. The results (FIG. 11) indicate an initial endothermic peak with an onset temperature between 200° C. and 250° C., absorbing greater than 800 J/cm.sup.3 as an individual peak combined with a secondary endothermic peak with the onset temperature between 400° C. and 450° C., absorbing greater than 100 J/cm.sup.3 as an individual peak yielding a total endothermic value greater than 800 J/cm.sup.3 (and >1000 J/g) for the entirety of the material.

[0204] TGA analysis of the sample prior to dewaxing revealed a peak at 227° C. which corresponded to a release of CO.sub.2, indicating the combustion of organic material (e.g. wax). No such peak was observed on the dewaxed sample.

Open Porosity

[0205] Open porosity was calculated using density measurements obtained by an autopycnometer, specifically a Micromeritics Autopycnometer (Model 1320 Serial #208), utilising helium gas. A sample of at least 3 cm.sup.3 is analysed, with a standard steel ball used as a reference check before each run.

[0206] The absolute density (also known as true, real, apparent or skeletal density) measures the volume of a sample excluding the pores and open void spaces between bound particles. (i.e. the pycnometer negates all open porosity). Therefore by taking the difference between the absolute density and bulk density of the same sample prior to the test, the open porosity can be determined.

[0207] The open porosity of sample 7 was determined to be approximately 30% v/v. As the density of ATH is approximately 2.4 g/cm.sup.3 and the density of binder is approximately 0.9 g/cm.sup.3, this equates to up to 87% of the added organic material being removed. This result is consistent with the calculated density of the sample 7 (1.63 kg/m.sup.3) which is about 68% of the maximum theoretical density of ATH, noting that the difference may be due to the presence of a small proportion of closed pores in the sample.

Effect of Particle Size Distribution

[0208] It was found that having a bimodal particle size distribution (for example as exemplified in FIG. 1) provides the best injection quality (while maintaining organics loading) for the hexagonal housing (FIG. 7), the highest MOR strength, and an acceptable linear shrinkage. It will be appreciated that other shapes and designs may result in different optimised particle size distributions.

TABLE-US-00005 TABLE 4 Injection MOR Sample d10 d50 d90 quality Shrinkage % (psi) 1 4.62 54.2 132 Poor — 750 2 4.32 36.7 147 Fair 0.35% 625 3 4.46 30.7 148 Good 0.60% 820

[0209] The Particle Size Distribution was measured using a Malvern Mastersizer 3000. This tool utilizes laser diffraction measurement by which a laser beam passes through a dispersed particulate sample and the angular variation in intensity of the scattered light is measured.

Effect of Organics Loading

[0210] An increase in organics (wax and stearic acid) resulted in a decline in mechanical strength as measured through the MOR and an increase in shrinkage.

TABLE-US-00006 TABLE 5 Sample Organics Shrinkage % MOR (psi) 4 14.0 wt % 0.42 895 5 15.0 wt % 0.60 820 6 17.5 wt % 0.93 640

[0211] Shrinkage was determined through measuring the difference in a known dimension in the green state after moulding and again after the de-waxing step performed at about 190° C. for 18 hours.

Effect of Fillers

[0212] A variety of fillers (fibrous and particulate) were added in small amounts (e.g. <5 wt %) to the sample 7 formulation. The additional fillers were found to generally increase density and mechanical strength (MOR), although at the detriment of the insulation properties of the material, as indicated with higher Cold Face Temperatures being recorded.

TABLE-US-00007 TABLE 6 Cold Face Additive Temperature Formula package Density (pcf) MOR (psi) (° C.) 7 A 102 769 215 [1.63 kg/m.sup.3] 8-13 B-G 100-106 967-1118 236-276

[0213] Additive package:

[0214] A: 2.5-3% stearic acid

[0215] B-G: 2.5*3% stearic acid and 1 to 4 wt % filler

Flame Screening (Cold Face Test)

[0216] This methodology tests the resistance to a lithium ion battery fire with direct flame impingement. The method includes using a Bernzomatic™ propane torch set 89 mm away from the test sample. The sample is subjected to the flame for 5 minutes while the cold face is monitored.

[0217] Samples were 8-inch (203 mm) discs clamped at the bottom 1 inch (or 25 mm) to secure the sample during testing. The optimal sample thickness was 0.25-0.28 inches (6.5 mm-7 mm). The flame was applied to the centre of the disc face or 4 inches (101 mm) from the edge of the sample perpendicular to the disc surface.

Destructive Battery Housing Testing

[0218] Housing (FIG. 7) with a minimum wall thickness of 1.5 mm made from CE-1 and Sample 7, and fitted with Samsung 50E 21700 cells.

[0219] The “Control” example separated the batteries by an equivalent distance to the other examples with an air gap.

Thermal Runaway Initiation Mechanism (TRIM)

[0220] The method consists of applying a high-powered heat pulse to a small area on the cell's external surface. A resistive heating element was provided in thermal contact with an outer edge of the battery cell. A section of the outer wall was removed to enable the heating element to provide the required thermal contact.

[0221] An energy source is provided to the resistive heating element and the target cell heated at 50° C./s until 500° C. or until thermal runaway obtained.

[0222] Further details on the procedure and resistive heating element used may be found in WO2018132911, which is incorporated herein by reference.

TABLE-US-00008 TABLE 7 Time to thermal Max. temp, of Material runway (s) adjacent cell (° C.) Control (air gap) 7.4 122.5 CE-1 7.8 116.8 Sample 7 67.7 88.5

[0223] The results highlight that the housing under the present invention significantly delays the onset of a thermal runaway event and once initiated the thermal event is less severe, as indicated by the lower maximum temperature of an adjacent cell within the housing.

Nail Penetration Test

[0224] The nail penetration procedure that was followed was based on SAE J2464:

[0225] 1. Start with fully charged cell (100% SOC).

[0226] 2. Mild steel nail, ∅3 mm, length adjusted such that penetration depth is through cell

[0227] 3. Propel nail at ≥8 cm/second (Radial penetration selected);

[0228] 4. An edge cell is selected for the target cell (FIG. 2).

TABLE-US-00009 TABLE 8 Time to thermal runaway (voltage definition) from Nail Penetration (seconds) Cell # 1 2 3 4 5 6 7 Control 0 0.2 1.1 1.6 2.2 2.8 3.3 Housing 0 0.2 0.8 2.9 17.3 140 209.5 CE-1 Housing 7 0 0.2 1.2 3.4 149.4 193.5 326.2

[0229] The results illustrate that that the housing (made from sample 7 material) under the scope of the present disclosure delays thermal runaway by greater than 50% compared to the comparative example.

Thermal Properties

[0230] The thermal properties of a 1.01 mm thick segment of sample 7 were determined over a temperature range of −40 to 85° C. The thermal properties (specific heat, diffusivity and conductivity) of the sample were determined using a NETZSCH LFA 467 HyperFlash™ instrument in accordance with ASTM E1461. The results are provided in Table 9 below:

TABLE-US-00010 TABLE 9 Thermal properties Temperature Specific heat Diffusivity Conductivity (° C.) Cp (J/g-K) α (mm.sup.2/s) λ (W/m-K) −40 0.897 1.66 2.61 0 1.07 1.30 2.43 40 1.23 1.06 2.28 85 1.33 8.98 2.10

[0231] Many variants, product forms, uses, and applications of the present disclosure will be apparent to the person skilled in the art and are intended to be encompassed by this disclosure.