Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
HIGH TEMPERATURE THERMO-ACOUSTIC BARRIER WITH LOW SMOKE AND ODOR
Document Type and Number:
WIPO Patent Application WO/2019/182561
Kind Code:
A1
Abstract:
A thermal barrier material is disclosed for applications such as heat shields in automobiles and other vehicles. The thermal barrier material is made on a Fordenier paper making machine from a slurry that is very low in organic compounds that can cause smoke and odor when exposed to high temperatures. A first thermal barrier material is disclosed that can withstand up to 650°C and a second thermal barrier material is disclosed that can withstand up to 1000°C, in each case, the barrier material withstands it max temperature specification while producing extremely iow quantities of smoke and extremely low levels of offensive odors. Prior art thermal barrier materials with similar temperature specifications on the other hand produce many times the smoke and offensive odors at the same temperatures and do so for a far longer time.

Inventors:
DEATS MARY (US)
Application Number:
US2018/023226
Publication Date:
September 26, 2019
Filing Date:
March 20, 2018
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
INTERFACE PERFORMANCE MAT INC (US)
International Classes:
F02B77/11; B60R13/08; F02B77/13
Foreign References:
US20080167407A12008-07-10
JPH07330360A1995-12-19
CN104176991A2014-12-03
US20100021718A12010-01-28
US6851506B22005-02-08
US20150260075A12015-09-17
Attorney, Agent or Firm:
O'NEILL, Janelle, A. (Moyles IP, LLC1 Enterprise Drive, Suite 42, Shelton CT, 06484, US)
Download PDF:
Claims:
CLAMS

What is claimed is:

1. A thermal barrier material for use in shielding components of a vehicle from hot surfaces such as exhaust system surfaces wherein a sample of the thermal barrier material, when exposed to a temperature of 400* Centigrade, produces smoke having a density less than 5 g/cm3 as measured according to the ASTM F1315 standard.

2. The thermal barrier material of claim 1 wherein a sample of the thermal barrier material, when exposed to a temperature of 400s Centigrade, produces smoke having a density' less than 2 g/cm3 as measured according to the ASTM F1315 standard.

3. The thermal barrier material of claim 1 wherein a sample of the thermal barrier material, when exposed to a temperature of 400* Centigrade, produces smoke having a density of about 0.88 g/cm3 as measured according to the ASTM F1315 standard.

4. The thermal barrier material of claim 1 wherein the sample of the thermal barrier material further produces less than 4 pp of 1-butanat gas as measured by Gas

Chromatography (GO) and Mass Spectrometry (MS).

5. The thermal barrier material of claim 1 wherein the sample of the thermal barrier material further produces less than 3 ppm of 1 -butana! gas as measured by Gas

Chromatography (GO) and Mass Spectrometry {MS).

6. The thermal barrier material of claim 1 wherein the sample of the thermal barrier material further produces less than 2 ppm of 1-butana! gas as measured by Gas

Chromatography (GC) and Mass Spectrometry (MS).

7. The thermal barrier material of claim 1 wherein the sample of the thermal barrier material further produces less than 1 ppm of 1-butanat gas as measured by Gas

Chromatography (GC) and Mass Spectrometry (MS).

8. The thermal barrier material of claim 1 wherein the sample of the thermal barrier material further produces no detected 1-butanaf gas as measured by Gas Chromatography

(GC) and Mass Spectrometry (MS).

9. The thermal barrier material of claim 8 wherein the sample of the thermal barrier material further produces less than 150 ppm Dimethoxymethane gas as measured by Gas

Chromatography (GC) and Mass Spectrometry (MS).

10. The thermal barrier material of claim 8 wherein the sample of the thermal barrier material further produces less than 100 ppm Dimethoxymethane gas as measured by Gas Chromatography (GC) and Mass Spectrometry (MS).

11 , The thermal barrier material of claim 10 wherein the sample of the thermal barrier material further produces about 55 ppm Dimethoxymethane gas as measured by Gas

12. The thermal barrier material of claim 8 wherein the sample of the thermal barrier material further produces less than 50 ppm Dimetboxy eihane gas as measured by Gas Chromatography (GC) and Mass Spectrometry (MS).

13. The thermal barrier material of claim 12 wherein the sample of the thermal barrier materia! further produces about 42.6 ppm Dimethoxymethane gas as measured by Gas Chromatography (GC) and Mass Spectrometry (MS). 14, The thermal barrier material of claim 1 wherein the sample of the thermal barrier material further produces less than 100 ppm CO gas as measured according to A STM 800 standards.

15. The thermal barrier material of claim 14 wherein the sample of the thermal barrier material further produces less than 50 ppm CO gas as measured according to ASTM

800 standards.

16. The thermal barrier materia! of claim 15 wherein "the sample of the thermal barrier material further produces about 46 ppm CO gas as measured according to ASTM 800 standards,

1 ?. The thermal barrier material of claim 1 wherein the sample of the thermal barrier material further produces less than 700 ppm CC½. gas as measured according to ASTM 800 standards.

18, The thermal barrier material of claim 17 wherein the sample of the thermal barrier material further produces less than 600 ppm C02 gas as measured according to ASTM 800 standards,

19. The thermal barrier material of claim 18 wherein the sample of the thermal baitier material further produces about 5SS ppm C(¾ gas as measured according to ASTM 800 standards.

20. The thermal barrier material of claim 1 wherein the sample of the thermal barrier materia! does not ignite hen exposed to a temperature of 650" Centigrade in a furnace.

IS

Description:
HIGH TEMPERATURE THERMO-ACOUSTIC BARRIER

WITH LOW SMOKE AND ODOR

TECHNICAL FIELD

This disclosure relates generally to heat shield barriers and more particularly to high temperature barriers for use as heat shields in the automotive and other industries. The disclosure also relates to high temperature barriers that also exhibit acoustic absorption properties, BACKGROUND

Heat shield material has long been used in automotive manufacturing to shield panels, electronics, wiring, and other components from the heat of adjacent hot surfaces such as an exhaust manifold or a catalytic converter in recent years, increasing engine efficiencies and increasing emissions standards have resulted in higher engine and exhaust system temperatures. As a result, certain components of engines, and exhaust systems in particular, in modern vehicles can be significantly hotter in operation than in the past. For example, un-burnt gasoline In an exhaust stream is sometimes intentionally burned in the catalytic converters thereby increasing the temperature of the converters’ outer surfaces compared to older technology. Surrounding panels and components must be protected from this heat.

Traditional heat shields and thermal barriers in vehicles typically have a three-layer construction comprising a thermal insulation materia! sandwiched between two aluminized steel plates. As temperatures have increased, these traditional heat shields have begun to exhibit various problems an shortcomings. For example, some original equipment manufacturers (OEMs) have received customer complaints of a campfire-like odor accompanied by smoke being defected in the passenger cabin during the initial operation of new vehicles. The root cause of the odor and smoke has often been determined to he the bum-ouf of organic components such as binders and celiuiosic fibers in the thermal barrier material of heat shields. The demand for quieter vehicles has resulted In requirements for better acoustic absorption as we!!, Much of the need for acoustic absorption is beneath the floor panels of vehicles where hot surfaces of exhaust systems exist. This poses a challenge because acoustic absorption materials are not always able to withstand high temperatures present near exhaust components of a vehicle. This is a related problem in need of a solution.

Accordingly, a need exists for a thermal barrier material that addresses and solves the problems of ignition, smoke, and unpleasant odors encountered with traditional prior art thermal barriers when exposed to high temperatures in modern vehicles. A further need exists for a thermal barrier that also exhibits acoustic absorption properties in regions of high temperatures. These thermal and acoustic absorption materials should be producible on traditional paper making machines and should be moldahle to desired shapes without losing their integrity. It is to the provision of a thermo-acoustic barrier material that addresses these and other needs that the present invention is primarily directed.

SUMMARY

Briefly described, a high temperature thermal barrier material is provided that Is able to withstand temperatures up to 1000 ® C without producing significant amounts of smoke and unpleasant odor. The material is made In sheets on a traditional paper making or Fordenier machine and may be formed into desired shapes and configurations before or after it is completely dry. In one embodiment for use in lower temperature environment, the barrier material has demonstrated the ability to withstand temperatures of 850 ® C (1112° F) for extended periods of time without burning, producing smoke, or emitting unpleasant odors. This embodiment will be referred to herein as the T1650 embodiment. In another embodiment, the barrier material has demonstrated the ability to withstand temperatures of 1000 ® C (1832 ® F) without these undesirable effects. This embodiment is referred to herein as the 711000 embodiment. in another embodiment, the thermal barrier materia! is bonded to one side of an acoustic absorption materia! to form a thermo-acoustic barrier, in a heat shield, the thermal barrier is oriented so that it faces a hot surface such as the surface of a catalytic converter with the acoustic absorption materia! facing away fro the hot surface. The thermal barrier has a low thermal conductivity so that heat does not pass easily through to the acoustic absorption materia!. The acoustic absorption material is thus protected from the heat and functions to absorb sound that might otherwise penetrate into the passenger compartment. The result is a quieter cooler vehicle in which panels, wiring, and other components are shielded from the high temperatures of the exhaust system.

A method of forming the high temperature thermo-acoustic shield also is disclosed. Briefly, the method comprises spreading a layer of thermal barrser material in the form of a siurry on the surface of an acoustic absorption materia! to form a layere thermo-acoustic composite. The acoustic absorption material may be perforated before the thermal barrier material is spread on its surface. The thermal barrier material flows into the perforations and bonds the two layers of material securely together. The thermal barrier material is then dewatered and dried in a paper making machine. Finally, the thermo-acoustic material may be formed into a specific desired configuration to fit in a designated area and sandwiched between aluminized metai plates for support, durability, and beat reflection.

These and other aspects, features, and advantages of the invention will be appreciated better upon review of the detailed description set forth below made in conjunction with the accompanying drawing figures, which are briefly described as follows.

8i¾iEf DESCRIPTION OF THE DRAWINGS

Fig. 1 is a perspective vie of a thermo-acoustic barrier in flat sheet form that embodies principles of the invention.

Fig. 2 is a side elevationai view of the barrier of Fig. 1 showing the layered construction of the barrier. Fig. 3 Is a perspective view of the thermo-acoustic barrier as seen from the opposite side.

Fig. 4 is a photograph of a test device designed to test the cold formality of the thermal barrier material of the thermo-acoustic barrier.

5 Fig, 5 is a photograph of a cup-shaped piece of the thermal barrier material formed according to the method of Fig. 3 and following heat and breakage test.

Fig, 8 Is a photograph showing the side-by-side testing of a prior art thermal barrier and a thermal barrier of the present invention for smoke and offensive odor emissions when heated.

10 Fig, 7 is a chart showing results of the smoke and offensive odor test shown in Fig. 5.

Fig. 8 is a photograph showing the side-by-side testing of a prior art thermal barrier and a thermal barrier of the present invention for flame ignition point when heated.

Fig, 9 shows charts and associated graphs illustrating the results of side-by-side testing of a prior art thermal barrier and a thermal barrier of the present invention for thermal conductivity (thermal mapping).

Fig. 10 is a chart showing results of the side-by-side testing of a prior art thermal barrier and a thermal barrier of the present invention for toxicity of gasses generated when heated.

Fig. 11 is a summary chart compiling the results of various tests conducted: on the

20 Til 000 thermal barrier of the present invention and a prior art thermal barrier with similar performance specifications.

Fig. 12 is a summary chart comping the results of various tests Gonducfed on the TI850 thermal barrier of the present invention and a prior art thermal barrier with similar performance specifications.

">

Reference wi now be made in more detail to the drawing figures, wherein like reference numerals indicate like parts throughout the several views. Fig. 1 illustrates a thermo-acoustic barrier that embodies principles of the invention in one preferred form . The thermo-acoustic barrier 18 comprises a high temperature thermal barrier layer 17 bonded to an acoustic absorption layer 18 The term“high temperature" as used herein means temperatures encountered adjacent hot surfaces of modern engines and exhaust systems. Such temperatures generally range from between 650" C and 1000° C (1112° F and 1832 ®

F) but can be somewhat lower or higher in specific cases. The high temperature thermal barrier layer 17 is formulated and fabricated as detailed below to withstand high

temperatures while generating very Sow {compared to the prior «r t) smoke and very low odor Intensity and offensiveness.

An acoustic absorption layer 18 is secure to the thermal barrier layer 17 on one side thereof. The acoustic absorption layer 18 can be secured to the thermal barrier layer by any appropriate means such as with an adhesive for example. One preferred method of securing the Sayers together is shown in Figs. 2 and 3. Fig. 2 shows the thermo-acoustic barrier 16 with the acoustic absorption layer facing up and Fig. 3 shows a cross section of the thermo-acoustic barrier. In this embodiment, the acoustic absorption layer 18 is punched to form a plurality of holes 19 that extend through the acoustic absorption layer.

The thermal barrier layer 17 is initially applied in the form of a slurry onto an upwardly facing surface of the acoustic absorption layer 18. The slurry flows partially Into the holes 19 as perhaps best illustrated in Fig. 2. As the slurry is dewatered and dried, preferably using a Fourdrinier or other type of paper making machine, the thermal barrier material in the holes 19 dries and locks the thermal barrier layer 18 and the acoustic absorption layer 17 together with a mechanical bond

The acoustic absorption layer 18 may be formed of any material that performs the function of absorbing sound before it enters the passenger compartment of a vehicle. In the preferred embodiment, the acoustic absorption layer 18 is made of a non-woven fiberglass sound absorbing materia! such as that available from Owens Corning Corporation of Toledo, Ohio and other suppliers. Other possible materials that may be suitable for the acoustic absorption layer include, without limitation, cotton and organic sound absorbing baits, silica fiber mats, and sound absorbing foa to mention a few.

While the thermo-acoustic barrier is shown as a flat sheet or tile in Figs. 1 and 2, it should be understood that in use the barrier often wi!l be shaped to fit in a specific tight space between a hot surface such as a catalytic converter and the floor panels of a vehicle. Furthermore, the thermo-acoustic panel may be adhered to one side of a shaped aluminized metal sheet or sandwiched between two metal sheets that can be pressed into a desired shape and also serve as thermal reflectors. So, the fiat sheet or tile shown in the Illustrative embodiment is not intended to limit the invention, but only to illustrate the layered construction of the barrier in a simple and easily understood form.

As detailed below, it has been found through experimentation that the smoke and unpleasant odors often produced by prior art thermal barriers (of which consumers complain) result from the burn-off of organic bidders and other organic compounds present in the material of these barriers in contrast, the materials from which the thermal barrier of the present invention is made are very low in organic compounds and binders compared to prior art thermal barriers, in one preferred embodiment, the thermal barrier of this invention may be made as follows:

MA ING THE THERMAL BARRIER

Table 1 below shows the ingredients used to make the thermal barrier of the present invention and, for each Ingredient, the percent-by-weight of the ingredient used In a slurry to be made into the thermal barrier in a paper making machine. T S50 TSiK

Table 1

Except for t e basalt fibers, the fibers and clays in Table 1 are combined with wafer (between 7 and 50°C) into a slurry using a puiper. To maintain the length of the basalt fibers, they are added directly to the mixing chest and homogenized into the mixing stock to avoid the shear forces generated in the piflper. The latex is then added and precipitated onto the fiber and fillers. The resulting slurry is spread onto the conveyor belt at the wet end of a traditional Fourdrinier paper making machine forming a wet web of fibers. If the thermal barrier is to be combined with an acoustic absorption barrier, the siurry may be spread onto a thin sheet of the acoustic absorption material, which may have been prepared with holes to facilitate binding the two layers together. In the machine, the wet web is dewatered and dried. The resulting web can then be cut into desired shapes and molded If desired to fit into areas where it is to be used. TESTSN©

Most of the tests described below were earned out according to the corresponding established industry method {typically an ASTM standard). However, due to the subjective nature of the offensive odor testing, interna! testing methods were developed that quantified the intensity and offensiveness of odors produced by prior art thermal: barrier materials and by thermal barrier materials of the present invention. An objective test for the presence of chemicals known to produce offensive odors also was carried out. In addition, the thermal masking tests were conducted using an internal testing method historically used to assess various thermal conduction properties of heat shield insulating material. These tests are detailed below.

1. Gold Formabiiity and Vibration Testing

Figs, 4 and 5 illustrate devices used to test the cold formabify of thermal barrier material made according to the above described process. The test was carried out according to test method Wi-TP-033_0. Both the TS650 an the 111000 embodiments of the thermal barrier materials were tested. For each material, a circular sample 23 of the material was die-cut from a sheet and positioned over a spherical depression on the anvil of a press 22. A spherical ram 24 was then pressed onto the sample until the sample was urged into the depression, thereby molding the sample into a bowl-shaped configuration.

After each sample was molded into a bowi shape as described, it was heated in a furnace to 400" C for 30 minutes and then placed in a tabletop shaker 26 {Fig, 5) for 5 minutes. Flat die cut specimens that had not been molded also were heated and shaken in this manner. This test sought to simulate the heat and vibration that might be experienced by the material when used in a vehicle. If the sample breaks into several pieces after heating and shaking or displays large separations, then it is likely that the material will crack or break up during crash forming of a commercial heat shield or during normal use. As shown in the summary test resuits chart of Fig. 11 , the test revealed for the T!OSO materiai that there were no cracks and the sample was in-taet after heating and vibration as described. The TS1000 material was observed to exhibit some small cracks, but the sampie was otherwise in-iact after heating and vibration. The conclusion is that the thermal barrier material of the present invention exhibits acoeptab!e cold formability properties.

To determine loss of mass due to dusting compared to prior art thermal barrier materials, die-cut and cold formed thermal barrier samples of the present invention and samples of prior art thermal barrier materials were tested, in each case, a sampie was weighed, heated to 400 s C for 30 minutes, placed in a table top shaker for 5 minutes, and then weight again. Any loss in weight is due to dusting of material away from the sample during the heating and shaking process. Fig, 5a shows the results of these tests. As can be seen, for the three prior art samples tested, total loss of weight due to dusting (loss from die- cut sampie plus loss from cold formed sampie) ranged between 0.85% and 1 ,04%, Loss from the die cut sample was significantly less than loss from the cold formed sampie fo each of these prior art thermal barrier material.

in stark contrast, the total loss of weight due to dusting for the Til 000 thermal barrier sample under the same test conditions was a mere 0,14% with about half of the loss (,06%) being due to die-cut sampie loss. For the TI650 sampie of the present invention, total toss of weight was still a mere 0.14% but the great majority of the loss (0.12%) was due to dusting losses from the die-cut sampie. The cold formed sample lost only 0.02% of its weight during the test. The conclusion is that heat barriers formed according to the present invention exhibit far less weight loss due to dusting than do the prior art heat barriers tested.

2. Smoke and Offensive Odor Test

Figs, 6 and 7 illustrate subjective testing of the thermal barrier material of this invention for the production of smoke and offensive odor at high temperatures. As discussed above, consumer complaints have focused on this unpleasant aspect of the prior art. For this test, a lab hot plate 29 was heated to 400 s C. A sample of prior art thermal barrier material 28 was placed on the hot plate 29 and held down by weights 33. The material was then observed by members of a pane! who focused on odor produced by the sample over time as its temperature rose. Members of the pane! rated odors produced by the sample for Intensity and offensiveness over a 5 minute period. Ail responses of the pane! members were then tabu!ated.

The same test was carded out with a sample of the TltOOO thermal barrier sample 29 made according to the present invention and a prior art thermal barrier with similar specifications. Again, the panel members rated the intensity and offensiveness of odors produced by the sample just as they had done with the prior art sample 28, The results of this test are shown in the chart of Fig, 7, which plots the results of the tests on a scale of rating vs. time. As can be seen, the intensity of odors produced by the prior art thermal barrier material (graph 38) was rated between 4 and 5 from .5 minutes until 2 minutes before slowly settling at a rating of about 1 after 2.5 minutes. The offensiveness of these odors was rated even higher at 6 until 2 minutes into the test before slowly falling to 1 at 4 minutes.

In contrast, the intensity of odors produced by the TI1O0D and TI650 samples made according to the present invention (chart 41 ) rated very low at just above zero for the full duration of the test. Offensiveness of these odors for these samples rated between 1 and 1.5 at the beginning, ramping down to just above zero at one minute into the test. Thus, thermal barriers made according to the present invention showed a drastic reduction in intensity and offensiveness of odors produced at high temperatures compared to those produce by the prior art thermal barrier.

In addition to these subjective tests, an objective odor evaluation was commissione by an outside laboratory. The laboratory tested liberated gases from a sample of prior art thermal barrier material and a sample of thermal barrier materia! made according to the present invention when heated as describe above. Gas Chromatography (GC) an Mass Spectrometry (MS) techniques were used to determine the presence of 1 -butanol and Dimethoxymethane, both deemed by most humans to be associated with and indicative of offensive odors. As can be seen from the summary chart of Fig. 11 , the prior art sample was determined to produce 4.58 part per million (ppm) of 1 -butanol while a sample of the present invention produced less than 3 ppm, less than 1 ppm and, in this test, no detectable 1 -butanol. As for Dimethoxymethane, the prior art sample produced 190 ppm while the sample of the present invention produced less than 100 ppm, less than 50 ppm and specifically 42,6 ppm , Such levels are considered indicative of low levels of offensive odors to humans.

Fig, 12 shows the same data for the TI650 thermal barrier sample vs the comparative prior art sample. The prior art sample produced 8.14 ppm of 1 -butanol white the TI650 sample produced les than 4 ppm, less than 2 ppm, and in this particular test, no 1 -butanol. The prior art sample produced 224 ppm Dirnethoxymethane while the T185t) sample of the present invention produced less than 150 ppm, less than 100 ppm and, in this particular test, 55 ppm. Such ranges are considered to be indicativ of low amounts of offensive odors.

This objective testing supports the conclusions of the subjective tests that a thermal barrier of the present invention pro ucesfar less offensive odors when heated than does the prior art.

The density of produced smoke for the barriers of the present invention also was measured according to the ASTM F1315 standard. The measured density for both the TI850 and the THQQQ samples was less than 5 g/crn 3 , less than 2 g/cnr 1 and more specifically measured to be 0.88 g/crri 3 , Such smoke densities are considered barely detectable. As can be seen in Fig. 6, which shows the prior art sample 28 and the TS1000 sample 29 slde-by-side on a 490" C hotplate, the density of smoke produced by the IT10DQ sample 29 is far less than the density of smoke produced by the prior art sample 28. The conclusion is that the high temperature thermal barrier material of the present invention produces negligible smoke when heated to high temperatures whereas the prior art produces significant smoke of which consumers complain. 3. Shock Flame Testing

A prior art thermal barrier material and the TI1000 ther al barrier material of the present invention were tested to determine their tendency to ignite at high temperatures. These two products have similar maximum temperature specifications of 1000" C. The test setup is shown in Fig. 8. A furnace 43 wa preheated to a temperature of 850" C before placing a 2 inch by 6 inch prior art sample and a 2 inch by 6 inch sample of the THQQQ thermal barrier in the furnace, A viewing port in the furnace wali allowed the flame point and smoke production, if any, to be visually determined. After a short time In the furnace, the prior art thermal barrier sample caught fire as seen in Fig. 8 and began to bom. This is considered a catastrophic failure of the barrier. The T11000 thermal barrier sample of the present invention did not Ignite at 650" C. In fact. Til 000 thermal barrier was subsequently tested at its design temperature of 1000° C and again did not ignite or fait.

Similarly, a prior art thermal barrier material and the TI850 thermal barrier material of the present invention were tested for ignition using the same procedure. These two products have similar maximum temperature specifications of 850" C. Again, the two samples were placed in a furnace pre-heated to 650" C and observed. As shown in the photograph of Fig. 12, the prior art sample ignited and failed at this temperature while the TI650 sample of the present invention did not.

4. Thermal Mapping Tests

Prior art thermal barriers and thermal barriers of the present invention were tested to determine the thermai conductivity of the material. This test is sometimes referred to as a thermal mapping test and was carried out according to AST tv! standard F433, I the test, a sample of interest was placed directly on a pre-heated 400" C hotplate. An infrared thermometer was used to map the rise in temperature of the top (exposed) side of the sample. The test was conducted for samples of thickness 0,8 mm and 1 ,0 mm for each of a prior art thermai barrier material, the TI650 barrier of the present invention, and the TS1000 barrier of the present invention. The resuits are shown in Fig, 7 where the charts on the right show graphically that the prior art thermal harrier conducted significantly more heat to its exposed face than did either the 71650 or the 171000 samples of the present invention. This is true for both the 0.8 and 1.0 mm thicknesses of the samples.

The test resuits are shown numerically on the left in Fig. 9. For the 0,8 m thick samples, the exposed face of the prior art sample rose to a temperature of 324 * C while the exposed faces of the 7I650 and 711000 samples rose to only 313 * C and 304 * C

respectively. Similarly for the 1.0 mm thick samples, the exposed face of the prior art sample rose to 323 * C while the exposed faces of the IT85G and TH 000 samples rose to temperatures of 312 * C and 289° C respectively. These results demonstrate a thermal conductivity (thermal K) for the prior art thermal barrier material of 0.188 while the conductivity of the thermal barriers of this invention were 0.114 for the 71650 materia! and 0,095 for the Tii 000 material. The conclusion is that thermai barriers of the present invention have significanttiy lower thermal conductivities than the prior art and trans it less heat from one surface to the opposite surface.

5. Toxicity of Generated Gases Test

A sample of prior art thermal barrier material and a sample of the TI1000 barrier material of the present invention were tested according to ASTM standard 800 relating to Measurement of Gases Present or Generated During Fires. Specifically, gasses produced by these samples when burned were collected and analyzed using Fourier Transform Infrare Spectroscopy (FTfR) for toxic compounds contained in the resulting smoke. The testing measured the presence of the following compounds: CO; CCT; HCL; HON; HBr; HP; NO; NO ; and SG 2, With the exception of carbon monoxide (CO) and carbon dioxide (C0 2 ), none of the toxic compounds were present. As for CO and CC½ levels in the gasses were determined and are presented in the table of Fig. 10. As can be seen, the Til 000 thermal barrier material produced more than 8 times less CO and more than 13 times less Ct¾ than the prior art thermai barrier material, a substantial an significant improvement. 8. Summary of Testing

Fig. 11 presents a table comparing results of the above described testing and other tests for a sample of the Til 000 thermal barrier material. Also shown are the results of the

5 same tests for a sample of prior art thermal barrier material with similar performance

specifications. Results for the prior art sample are shown in column 4 while results for the TI1000 sample of the present invention are shown in coiumn 5. First, the caliper (thickness} and density of each sample was measured using the indicated ASTM standards. The Caliper of the prior art sample was determined to be 0.80 mm and its density was

10 determined to be 1.15 g/cm 3 . This compares to the Til 000 sampie of the present invention, which had a caliper of 0,85? mm and a density of 0,90 g/cm' ' . The two samples were very similar in thickness and density.

A horizontal flame spread test was conducted on the two samples according to SAE J369 testing standards. In this test, each sampie was suspended in a horizontal orientation and a Bunsen burner was placed beneath one end of the sample if the sample ignited and the flame did not self -extinguish, a rate at which the flame was observed to spread would be tabulated in this test, the prior art sampie did not ignite (DNS) and the sampie of the Til 000 thermal barrier also did not ignite.

A compression/recovery test was conducted on both samples according to ASTM

20 F36K standards using an Armstrong Static Indentation Machine. This test measures the ability of the material to absorb compressive forces and, once compressed, how well the material returns to its original caliper. The thickness of the sample is measured and then the sampie Is subjected to an extreme loa for a specifie time sufficient to compress the material. The load is then removed and the material Is allowed to rebound partially to its

"> original thickness. The final thickness is then measured. The rebounded thickness divided by the original thickness represents the compresslon/rebound measurement expressed as a percentage. Greater rebound is desirable in these tests, the prior art sample rebounded by 16/28 or 57% while the sample of the present invention rebounded by 20/2? or 74%, Thus, a thermal barrier of the present invention is more tolerant of compressive loads than the thermal barrier of the prior art.

The thermal conductivity ot each sample was measured according to the procedure outlined above. The results are tabulated again in the summary chart of Fig, 11.

Fig, 12 presents a table comparing results of the above described testing and other tests for a sample of the TI650 thermal harrier material. Also shown are the results of the same tests for a sample of prior art thermal barrier materia! with similar performance specifications. As with tests for the T11000 sample, results for the prior art sample are shown in column 4 while results for the T1650 sample of the present invention are shown in column 5. As can be seen from Fig 12, the T1650 sample made according to the present invention performed significantly better than the prior art sample in virtually every test.

The invention has been described herein in terms of example embodiments considered by the inventor to represent the best modes of carrying out the invention. St will be understood by one of ski! in the art, however, that a wide gamut of additions, deletions, and modifications, both subtle and gross can be made to the illustrative embodiments without departing from the spirit and scope of the invention, which is delineated oniy by the claims.




 
Previous Patent: HEAT DISSIPATING ELEMENTS

Next Patent: SENSORS ACCESS CONTROL