G Love: Results
Sample 1: Paraffin Wax
Description of Paraffin Wax from MSDS: Waxy, white, odorless, solid. Freezing point: 47° to 65°C. Chemical formula not available.
Mass: 16.11mg
Pan used: 50m L, vented (with holes)
Baseline filename: D:\Program Files\Pyris\Data\Pyris 1 DSC\Glove\baseline50holes.dcd
Data filename: D:\Program Files\Pyris\Data\Pyris 1 DSC\Glove\paraffin.dcd
Method Steps:
1) Hold for 3.0min at -20° C
2) Heat from –20 to 200° C at 10° C/min
Figure 1. DSC measurement of heat flow versus temperature for parrafin wax. Peaks 1 and 2 mark the melting points of two different crystalline components of the paraffin wax, 24.9° C and 50.6° C, respectively.
To calculate the latent heat of fusion at a melting point, start by calculating the area of the peak. Then divide the area by the heating rate, 10° C/min in this case.
For Peak 1, latent heat of fusion = 32.7 J/g
For Peak 2, latent heat of fusion = 154.1 J/g
The feature in Figure 1 labeled by Number 3 likely marks the glass transition of water. This feature is not expected to be present if a measurement of heat flow versus decreasing temperature is collected. Feature 3 could also be an artifact of the DSC instrument, as well as the decrease in heat flow in the high temperature range.
Sample 2: Bodiheat Powder from a packet that provides heat by chemical reaction. The product is called Bodiheat and was bought from CVS Pharmacy store. The powder is contained within a vented packet, which is sealed in plastic.
Estimated Mass: 40.5mg
Mass after DSC measurement: 40.18mg
Pan used: 50m L, vented (with holes)
Baseline filename: no baseline used
Data filename: D:\Program Files\Pyris\Data\Pyris 1 DSC\Glove\bodiheat.dcd
Method Steps:
1) Hold for 120min at 23° C (actual temperature = 1.7° C)
Figure 2. Plastic envelope of Bodiheat product.
Ingredients of Bodiheat, as listed on product: Iron Powder, water, activated carbon, vermiculite, salt, super absorbent polymer, purge natural mineral powder
Reaction begins by exposure of packet to air.
In order to begin measuring the sample as soon as possible after the reaction begins, the strategy was to eliminate the amount of time between exposing the packet to air and loading the sample into the DSC. Making an estimate of the sample mass eliminates the time of measuring its mass. The estimate was obtained by taking a Bodiheat packet out of the plastic, which exposes it to air, then cutting the packet with a blade in order to access the powder and fill the pan. The mass of this sample was measured. The sample that was actually measured in the DSC came from a newly exposed Bodiheat packet and was prepared in the same manner. Sample mass was measured after DSC measurement. However, it is possible that the mass of the sample changed during the chemical reaction.
The software was programmed to maintain a constant temperature = room temperature. However, a temperature offset caused the actual sample temperature to be maintained at 1.7° C. Therefore, another Bodiheat sample (Sample 3) was measured after correcting for the temperature offset. The measurement results for Sample 2 are presented along with Sample 3 in Figure 3 below.
Sample 3: Another sample of Bodiheat powder
Estimated Mass: 40.5mg
Mass after DSC measurement: 53.0mg
Pan used: 50m L, vented (with holes)
Baseline filename: no baseline used
Data filename: D:\Program Files\Pyris\Data\Pyris 1 DSC\Glove\bodiheat23.dcd
Method Steps:
1) Hold for 120min at 34.4° C (actual temperature = 23.0° C)
Figure 3. Isothermal DSC measurement of Bodiheat powder. Sample 2 was held at 1.7° C. Sample 3 was held at 23.0° C.
Sample 2 reaches a constant heat flow 10 minutes after being exposed to air. However, after Sample 3 reaches a maximum negative heat flow, the heat flow begins to decrease at a constant rate. This difference in sample behavior is likely caused by the difference in the amount of oxygen to which each sample was exposed. The instructions on the Bodiheat product say that the packet begins to warm up when exposed to air. Even though ventilated pans were used for both samples, the amount of oxygen contained within the pan may have been different for the two samples. (The purpose of the holes in the pan is to prevent pressure from building up inside the pan as the sample is exposed to changing temperatures during measurement, not to promote air circulation around the sample.) Once the sample is placed in the DSC, it is no longer exposed to air. Taking note of the significant difference in mass of the two samples after the DSC measurement was performed, it can be assumed that due to the greater volume of the sample with the greater mass, Sample 3, less oxygen was contained within the pan of Sample 3. Therefore, it is possible that not enough oxygen was present to maintain the chemical reaction and the heat flow decreased. Less time was allowed for measurement of Sample 3 because the time reserved for use of the DSC by 3.082 had expired before the measurement could be completed.
Suggestions for further experimentation: Next time when doing a DSC measurement, after the temperature is increased, it should then be decreased in order to observe crystallization, which would be marked by sharp dips in the curve, signifying a release of latent heat, opposite to the melting peaks.
References: www.psrc.usm.edu/macrog/dsc.htm
E.M. Barrall, J.F. Johnson, "Differential Scanning Calorimetry: Theory and Applications," Thermal Characterization Techniques. P.E. Slade, Jr., L.T. Jenkins, Eds. New York: Marcel Dekker, Inc. (1970) pp.1-39
The following resistance (R) measurements were taken with an ohmmeter:
Sample: 80%Ni/20%Cr wire, diameter = 0.41mm
Expected R (from manufacturer): 2.571 ohms/ft
Measured R = 2.46 ohms/ft
R for wire wrapped around finger = 2.43 ohms/ft
R after work hardening = 2.42 ohms/ft
Sample: 60%Ni/16%Cr/24%Fe wire, diameter = 0.38mm
Expected R = 2.670 ohms/ft
Measured R = 2.53 ohms/ft
R for wire wrapped around finger = 2.53 ohms/ft
R after work hardening = 2.53 ohms/ft
Sample: Stainless Steel 70%Fe/19%Cr/11%Ni, diameter = 0.404mm, annealed
Expected R = 1.712 ohms/ft
Measured R = 1.98 ohms/ft
R for wire wrapped around finger = 2.00 ohms/ft
R after work hardening = 2.03 ohms/ft
As shown by these results, bending or work hardening of the wire did not change its resistance by any significant amount.
In order to calculate the heat transfer coefficient of commercially available fleece gloves, a stainless steel heating element (73 cm in length) was placed inside one glove. A power supply was attached to the heating element, and set to 3.6 V DC (to simulate a lithium polymer battery). One thermistor probe was placed inside the glove, in the region of the palm, while another thermistor was placed outside of the glove. To prevent air flow from inside the glove, the glove's opening was sealed with a weight. The power supply was then turned on, and the heating element was allowed to reach equilibrium for about 5 minutes. Temperature measurements were then recorded from both thermistors. The values are given below.
| Temp. inside glove (°F) | Temp. outside glove (°F) |
| 109.9 | 71.5 |
| 112.4 | 71.9 |
| 112.6 | 71.6 |
| 112.5 | 71.1 |
| 112.2 | 71.6 |
| 112.1 | 71.7 |
| Average T inside (°F) | Average T outside (°F) |
| 111.95 | 71.57 |
| 112.2 | 71.6 |
Using the steady-state equation below, the heat transfer coefficient of the fabric, H, was calculated.
I V = H (T [in] - T[out])
H = (3.6)(5.5) / (111.95 - 71.57)
H = 0.49 J/m2*K
Sample 4: PEG 900 (polyethelyne glycol, molecular weight (MW) = 900 g/mole)
Measured Mass: 25.87 mg
Pan used: 50mL, vented (with holes)
Baseline filename: D:\Program Files\Pyris\Data\Pyris 1 DSC\Glove\BL 3-25-02.dcd
Data filename: D:\Program Files\Pyris\Data\Pyris 1 DSC\Glove\PEG900.dcd
Method Steps:
1) Hold for 3.0 min at -40.00° C
2) Heat from -40.00° C to 110.00° C at 10.00° C/min
3) Hold for 3.0 min at 110.00° C
4) Cool from 110.00° C to -40.00° C at 10.00° C/min
Figure 4. DSC measurement of PEG 900.
Melting point (Tm) = 26.63°C
Latent heat of fusion (DHf) = 159.9 J/g
Crystallization temperature (Tc) = 9.8°C
Heat of Crystallization (DHc) = -150.8 J/g
Sample 5: Octadecane
Measured Mass: 21.3 mg
Pan used: 50mL, vented (with holes)
Baseline filename: D:\Program Files\Pyris\Data\Pyris 1 DSC\Glove\BL 3-25-02.dcd
Data filename: D:\Program Files\Pyris\Data\Pyris 1 DSC\Glove\octadecane.dcd
Method Steps:
1) Hold for 3.0 min at -40.00°C
2) Heat from -40.00°C to 110.00°C at 10.00°C/min
3) Hold for 3.0 min at 110.00°C
4) Cool from 110.00°C to -40.00°C at 10.00°C/min
Figure 5. DSC measurement of octadecane.
Tm = 27.2°C
DHf = 251.3 J/g
Tc = 16.5°C
DHc = -283.5 J/g
Three different wires of approximately 12 inches in length were tested for their tensile strength with the aid of Yin Lin in lab 13-5019. The setup of the instron is shown here. Because of this setup, approximately three inches of wire was used to allow the wire to be clamped on the top and bottom of the instron tensile grips. Thus, there was actually more like 9 inches of wire stretched in this experiment. This experiment was meant to give a general idea of the variation between the three wires. Each wire was as close to 9 inches lengthwise as could be done in with the apparatus available in the lab, but it was assumed that the results posted below give us a good enough result that the wire will withstand the inevitable bending it will undergo once placed in the glove.
The 12 inch wires were stretched with a 50lb load at 0.5 inch/min. at a low crosshead speed of 0.05 inch/min. The following equation was used to calculate the stress using the maximum load read off the following graphs.
Load (lbs.) / Area (inch2) = Stress (kpsi)
The extension lengths of the wires were calculated with the following equation:
Dist. of graph (inch) x Crosshead speed (in./min.) / Rate of stretch (in./min.) = Extension (inch)
The results were as follows:
| Ni:Cr 80:20 wt% | Ni:Cr:Fe 60:16:24 wt% | Stainless Steel Fe:Cr:Ni 70:19:11 wt% | |
| Starting Diameter (mm) | 0.41 | 0.40 | 0.375 |
| Area of Wire (inch2) | 0.00807 | 1.192 x 10-4 | 1.7 x 10-4 |
| Stress suggested by companies (ksi) | 120 | 74 - 130 | ~95 |
| Length of Graph (inch) | 19.5 | 21.7 | 35 |
| Extension (inch) | 1.95 | 2.17 | 3.5 |
| Maximum Load (lbs.) | 24.2 | 21 | 17.5 |
| Stress Calculated (ksi) | 121 | 107.8 | 102.2 |
From the graphs (see below), it is shown that the Ni:Cr 80:20 wt.% has the largest elastic region before changing to plastic deformation. The elastic region is the initial curve on the graph before it begins to level out. Thus, the Ni:Cr 80:20 wt.% would most likely be the best choice for use in our project. Although the Stainless Steel had the largest deformation (stretched the furthest distance) before breaking, it had a relatively small elastic region, meaning most of the deformation would take place in the plastic region. When the deformation takes place in the plastic region, the wear on the wire would be much faster and the wire would be more likely to break due to strain on the wire from bending-something which would happen a lot in the gloves. All are relatively close, however, the Ni:Cr 80:20wt% is the most optimal.
The graph below shows Stress vs. Strain curves for the three samples:
Result from Tensile Testing of High Resistance Wires.
Close up of Elastic Region From Previous Graph.
Sample 1: Paraffin Microspheres Fabricated at Homogenizer Speed 3.5
The average particle size was estimated to be approximately 200 microns.
Sample 2: Paraffin Microspheres Fabricated at Homogenizer Speed 5.0
The average particle size was estimated to be approximately 150 microns.
The paraffin microspheres appeared much flakier than expected. They actually looked much more like platelets than microspheres. However, the size of these platelets is on the right order of magnitude. Because this sample was given to us and its exact composition is unknown, it is hard to say what might have caused this platelet formation. This will be investigated further if it also occurs in the octadecane microspheres.
Sample 3: Octadecane Microspheres Fabricated at Homogenizer Speed 4.0
It appears that the octadecane may have melted and recrystallized during fabrication. A second problem may have been that the microspheres were removed from the lyophilizer before all the water had been removed. While they were refrozen afterwards, it is possible that this also affected the quality of the microspheres. One possible solution to the problem is to cool the PVA solution during microsphere fabrication.
Sample 4: Octadecane Microspheres Fabricated at Homogenizer Speed 5.0.
Another image of octadecane microspheres fabricated at homogenizer speed 5.0.
Closer View of Octadecane Microspheres.
Sample 7: PDMS resin with Octadecane Microspheres
Measured Mass: 26.59 mg
Pan used: 50mL, vented (with holes)
Baseline filename: D:\Program Files\Pyris\Data\Pyris 1 DSC\Glove\BLsilicone.dcd
Data filename: D:\Program Files\Pyris\Data\Pyris 1 DSC\Glove\BLsi-octsphere.dcd
Method Steps:
1) Hold for 3.0 min at -10.00°C
2) Heat from -10.00°C to 60.00°C at 5.00°C/min
3) Cool from 60.00°C to -10.00°C at 5.00°C/min
DSC measurement of PDMS resin (left) and PDMS resin with octadecane microspheres (right).
There are no peaks in the heat flow data for the PDMS sample containing octadecane microspheres. This indicates that no phase transition is occurring. This is due to the fact that a complication occurred while processing the microspheres, which appeared to have destroyed the microspheres so that none were left in the microspheres container. This sample was made by incorporating what was left of the microspheres, a white, flaky substance in the container, into the PDMS resin. This white, flaky substance made up no more than 5% of the sample. It is likely that there actually were no microspheres present in this PDMS/octadecane sample.
The spike of low heat flow seen below -20°C for both samples is generated during the first step of the test method, in which the DSC allows 3 minutes for equilibration of the system after the sample chamber is open and closed to allow loading of the sample.
Suggestions for further experimentation: Make a large quantity of octadecane microspheres, taking care not to let the microspheres melt and recrystallize during processing, as may have been the complication for the previous set of octadecane microspheres because the melting temperature is low. Incorporate a much higher percentage of microspheres into the PDMS resin.
The fleece fabric (100% polyester) showed the best insulating properties. It will therefore be used as the outer layer of the glove. Results were obtained by placing a heating element inside the fabric and measuring the temperature differential (using a thermistor) across the fabric as a function of time. It was then possible to calculate the heat transfer coefficient for each fabric using the equation
H = I*V/[A*(T1-T2)]
| 100% Polyester | 100% Polyester (fleece) | 20/80 Cotton/Polyester | 80/20 Cotton/Polyester | |
| H (J/m2*K) | 1.12 | 1.49 | 1.35 | 1.08 |
Experiment 1: Maximum Wire Temperature
The maximum surface temperature of the heating element is an important factor to consider when designing a safe product. Cotton and Polyester can withstand temperatures up to 250°
C, and the wire insulation is functional to a temperature of 260°
C. For this reason, we performed the following experiment to measure the maximum temperature of uninsulated Ni:Cr 80:20 wire carrying a voltage of 3.7 V.
A thermistor was covered with a paper electrical insulator and then wrapped with a 10 cm segment of a 100 cm wire. The wire-wrapped thermistor was then covered with 100% polyester fleece to decrease the energy loss to the environment. The wire was attached to a power source set to 3.7 V, the voltage of the lithium-polymer batteries. The power source and digital thermometer were then turned on. As the wire grew warmer, the temperature was watched to ensure it remained at a safe level. Once the temperature reached a maximum (fluctuating within +/- 0.2° C), the temperature was recorded.
Result:
Maximum Temperature: 60°C
The voltage remained constant at 3.7 V
The current remained constant at 0.43 A
Experiment 2: Temperature Dependence of Thermal Switch Resistance
Bimetallic thermal switches were purchased from Portage Electric Products, Inc. Unfortunately, the electrical characteristics of the switches were not available from the company. The following experiment was performed to measure the temperature dependence of the resistance.
A thermal switch was wrapped with a 10 cm segment of a 100 cm Nickel-Chromium wire. The wire-wrapped switch was then wrapped with notebook paper (for electrical insulation), and a thermistor was placed in contact with the paper-covered wire. 100% polyester fleece was then placed around the thermistor/wire/switch unit to prevent heat loss to the environment. The switch was connected to a voltmeter capable of recording resistance values, and the heating element was attached to a power source. The power source was set at 3.7 V, in order to keep the temperature below 60° C. The voltmeter and digital thermometer were turned on, and the initial resistance (base resistance) and temperature were recorded. The power source was then turned on to begin warming the heating element. As the temperature rose, the resistance was recorded at intervals of one degree Celsius.
| Temperature (C) | R (ohms, Trial 1) | R (ohms, Trial 2) |
| 26 | 0 | 0 |
| 27 | 0 | 0 |
| 28 | 0 | 0 |
| 29 | 0 | 0 |
| 30 | 0 | 0 |
| 31 | 0 | 0 |
| 32 | 0 | 0 |
| 33 | 0 | 0 |
| 34 | 0 | 0 |
| 35 | 0 | 5 |
| 36 | 31 | > Megaohms |
| 37 | > Megaohms | > Megaohms |
| 38 | > Megaohms | > Megaohms |
Result:
Resistance rose rapidly at the point of switching above the limits of the voltmeter. The other switches were tested in the same manner, with a performance similar to the data above. None of the switches switched at the exact temperature recorded by Portage Electric Products, but all were within +/- 3°
C. Once the switching temperature was reached, all of the circuits were completely opened within one degree Celsius. The base resistance of every switch was below the range of the voltmeter.
Conclusion
The maximum temperature of the wire surface is within a safe operating range, considering the materials to be used in our glove. The electrical characteristics of the thermal switches are precisely what we desired. Below the switching temperature, the base resistance is insignificant when compared with that of the wire, so the power drawn by the switch during operation is negligible. Upon switching, the switches open the circuit quickly (within one degree Celsius). At temperatures above the switching point, the current through the switches is near zero. The switches are therefore sufficient to regulate the operating temperature range of the gloves.
Results for Octadecane Spheres in PDMS are shown here.
DSC measurement from a 750mL sample of crushed octadecane in PDMS resin (Sample D).
Tm = 28.7°C
DHf = 34.5 J/g
Tc = 25.1°C
DHc = -42.8 J/g
Comparing this result to that of pure octadecane, the amount of heat stored/released by the octadecane in this sample is only about 15%! Octadecane in this sample is embedded in PDMS resin, which has a very low thermal conductivity, as described in the materials selection page. This means that the PDMS does not transport heat to and from the octadecane well enough so that the octadecane can store/release heat efficiently, or as well as it can on its own. Therefore, the heat of crystallization and the heat of fusion is significantly less.
In this sample, the octadecane powder was evenly distributed throughout the PDMS resin and there was no flaking away of the octadecane from the resin when the sample was cut.
Sample 9: High concentration of crushed octadecane embedded in PDMS resin. Processing of this sample
DSC measurement of a highly concentrated sample of crushed octadecane in PDMS resin (Sample B).
Tm = 29.3°C
DHf = 90.6 J/g
Tc = 26.4°C
DHc = -93.1 J/g
This sample is from 750mL solution of crushed octadecane in PDMS (Sample B). Of all samples processed, this one was the most concentrated with octadecane.
Suggestions for further experimentation: Since the low thermal conductivity of the PDMS is the problem in using this method of embedding microspheres, an alternative method should be tried. An alternative is to embed the other phase change material that we have studied, PEG, into polypropylene. Polypropylene has a much higher thermal conductivity, comparable to that of fabrics. We will try this method next.
This figure shows the discharge of the Li-Polymer Battery when attached to 100cm and allowed to run. The voltage was measured by observing the fluctuation of voltage on a voltage meter. The dip in the beginning of the wire may be due to a fluctuation of room temperature, change in the resistance of the wire or another variable that may have been in the room influencing the flux of the reading. When the battery is almost run down, the graph drops quickly as shown here. It was allowed to discharge slightly longer than it had been planned but luckily the voltage did not drop to zero. If the voltage were to drop to zero, it may harm the battery.
For this experiment, a power supply was used (not a battery). The voltage was set to 3.74, and fluctuated sporatically within the range of 3.71 to 3.78V. A wire length of 100cm was used for all subjects. The comfortable temperature reported by each subject while in a freezer at approximately zero degrees Farenheit is recorded below:
| Subject | T (F) |
| 1 | 91.3 |
| 2 | 90.4 |
| 3 | 89.4 |
| 4 | 93.1 |
| 5 | 89.8 |
| 6 | 92.0 |
| 7 | 84.7 |
| 8 | 91.7 |
| 9 | 91.6 |
| 10 | 90.9 |
| AVERAGE | 90.5 |
The average comfortable temperature was 90.5 F (32 C). It was therefore decided to use the thermal switch that will turn the circuit off at 32 C. While this is slightly above the comfortable temperature of some subjects, it is not extremely hot and should not be uncomfortable. In addition, the gloves ideally will have a manual switch allowing the user to turn them off if they feel they are too hot.
Results from Microscopic Imaging are here.
Optical microscope image of octadecane microspheres. It appears that in some places a number of microspheres clustered together, however the diameter of these individual particles is also on the order of 50 microns (scale is approximate).
Another optical microscope image of octadecane microspheres (scale is approximate).
While octadecane microspheres have finally been succesfully fabricated, the process remains extremely inefficient. The three grams of octadecane used initially resulted in less than 0.1g of microspheres. This is probably due to the inefficient filtering process. Because octadecane is lighter than water, standard centrifuge techniques to separate the microspheres from a solution cannot be used. When the microspheres are filtered a significant amount of material is lost on the filter. Therefore while microsphere fabrication is a very promising technology and has the potential to maximize the efficiency of phase change materials, it is not feasible for this project.
The temperature difference between the inside of the gloves and the freezer (measured by a thermistor placed several centimeters away from the glove) was recorded as a function of time. The glove stayed above the freezer temperature for approximately 40 minutes. This would obviously be much longer if a hand were actually in the glove.
The results are shown below:
|
Time |
Time |
Temp1 |
Temp2 |
|
(pm) |
(min) |
(° C) |
(° C) |
|
5:54 |
0 |
23.4 |
-20.7 |
|
5:59 |
5 |
4.6 |
-20.9 |
|
6:04 |
10 |
-5.8 |
-21.2 |
|
6:10 |
16 |
-12.0 |
-22.2 |
|
6:15 |
21 |
-15.5 |
-22.1 |
|
6:20 |
26 |
-16.7 |
-20.6 |
|
6:25 |
31 |
-18.4 |
-22.4 |
|
6:30 |
36 |
-19.6 |
-24.0 |
|
6:35 |
41 |
-20.1 |
-22.2 |
|
6:40 |
46 |
-20.0 |
-20.7 |
|
6:45 |
51 |
-19.4 |
-20.7 |
|
6:50 |
56 |
-19.2 |
-20.0 |
|
6:58 |
64 |
-18.7 |
-19.7 |
|
7:03 |
69 |
-18.4 |
-19.3 |
|
7:09 |
75 |
-17.6 |
-18.1 |
|
7:14 |
80 |
-16.9 |
-18.9 |
|
7:19^ |
85 |
-17.1 |
-20.0 |
|
7:24 |
90 |
-16.7 |
-20.9 |
|
7:29 |
95 |
-17.6 |
-21.2 |
|
7:37 |
103 |
-18.2 |
-21.6 |
|
7:42 |
108 |
-18.7 |
-20.6 |
|
7:47 |
113 |
-18.2 |
-18.6 |
The PCM gloves were each heated to 60 C using a blow dryer. The temperature differential between the inside of the glove and a thermistor placed a few centimeters away from the glove was measured as a function of time. As shown in the graph below, both gloves containing PCM stayed above room temperature significantly longer than the control glove (no PCM incorporated):
Sample 10: Polyethylene Glycol Heat Sealed in Polyethylene
DSC measurement of heating PEG in PE sample.
Tm = 35.5°C
DHf = 62.13 J/g
DSC measurement of cooling PEG in PE sample.
Tc = 31.6°C
DHc = -61.2 J/g
PEG in Polyethylene:
4.4mg of PEG was incorporated into this sample (total weight=7mg)
Total heat released during freezing = 61.2J/g x 7mg = 0.43J
Without encapsulant, 4.4mg of PEG would have produced 151J/g x 4.4mg = 0.66J
(from initial DSC testing of pure PEG)
Percent Loss Due to Polyethylene = (0.66-0.43)/0.66 = 36%
Octadecane in PDMS:
8.3mg of octadecane was incorporated into this sample (total weight=12.5mg)
Total heat released during freezing = 93.1J/g x 12.5mg = 1.16J
Without encapsulant, 8.3mg of octadecane would have produced 283.5J/g x 8.3mg = 2.36J
(from initial DSC testing of pure octadecane)
Percent Loss Due to PDMS = (2.36-1.16)/2.36 = 51%
Therefore the polyethylene is a much more efficient encapsulant than PDMS. This result was expected because the thermal conductivity of PDMS is much lower than the thermal conductivity of polyethylene.
However, overall the octadecane sample still gave off more heat during crystallization. This is due to the fact that it was possible to embed a higher fraction of octadecane in PDMS than PEG in polyethylene and also the higher latent heat capaciy of the octadecane. Therefore the octadecane in PDMS will be used as the final PCM in the gloves.
In the future, it would be very interesting to try heat sealing the octadecane in polyethylene (or use another method to combine the two materials). This has the potential to allow even more heat out and therefore create an even more efficient glove.