hollow glass microspheres composites

hollow glass microspheres composites

Journal of Non-Crystalline Solids 352 (2006) 615–619 www.elsevier.com/locate/jnoncrysol Formation of low density polyethylene/hollow glass microspher...

352KB Sizes 6 Downloads 118 Views

Journal of Non-Crystalline Solids 352 (2006) 615–619 www.elsevier.com/locate/jnoncrysol

Formation of low density polyethylene/hollow glass microspheres composites M.M. Ashton-Patton, M.M. Hall, J.E. Shelby

*

Department of Ceramic Engineering, New York State College of Ceramics at Alfred University, 2 Pine Street, Alfred, NY 14802, USA Available online 21 February 2006

Abstract Hydrogen is used to absorb heavy particle radiation, which is the most damaging radiation in space for humans. Low density polyethylene/hollow glass microsphere composites have been suggested as a possible radiation shield because of the high concentration of hydrogen and the low gravimetric density of the microspheres. Composites pressed under 3.90 MPa (566 psi) and 120 C have the highest probability of success thus far compared to polymers pressed at higher pressures and lower temperatures. Hollow glass microspheres made of borosilicate glasses do not break as easily as hollow glass microspheres made of aluminosilicate glasses. A smooth microsphere surface is better than a rough surface because it distributes the force more evenly, resulting in a more hydrostatic stress environment.  2006 Elsevier B.V. All rights reserved. PACS: 81.05.Qk; 83.80.Ab; 81.05.Kf; 81.05.Lg Keywords: Glasses; Scanning electron microscopy; Polymers and organics; Processing; Radiation

1. Introduction President G.W. Bush recently outlined three goals for the United States space program. First, the moon will be explored as a possible landing and launching location for future deep space exploration. Second, development of a next generation spacecraft will be carried out. Finally, the most fundamental of the goals and the subject most related to this report, the development of radiation shielding to allow humans to spend extended periods of time outside of the Earth’s atmosphere without detrimental biological effects [1]. Space exploration is limited by the amount of radiation exposure astronauts are allowed; the current limit is 120 days. Currently space missions take place within the Earth’s protective magnetic field to maximize the number of days an astronaut can be in space [2,3]. While many types of radiation are encountered in space flights, heavy *

Corresponding author. Tel.: +1 607 871 2470. E-mail address: [email protected] (J.E. Shelby).

0022-3093/$ - see front matter  2006 Elsevier B.V. All rights reserved. doi:10.1016/j.jnoncrysol.2005.11.058

particle radiation (HZE) is one of the most harmful types of radiation in terms of damage to humans. These particles are normally blocked using material with a high specific density of protons. Polyethylene is currently the best know absorber of HZE particles [4]. This study involves development of a low density polyethylene (LDPE)/hollow glass microsphere (HGM) composite. Loading the LDPE with microspheres reduces its density and allows use of microspheres made from various compositions of glass. Since the bulk density of the hollow glass microspheres is much less than that of LDPE, the use of a composite reduces the weight of the material, thus decreasing the amount of fuel needed. The hollow glass microspheres can be filled to high pressures of hydrogen to maintain the proton concentration in the composite. The glass composition can be changed to absorb other types of radiation in addition to HZE particles, e.g., neutrons, allowing the number of layers of shielding to be decreased. Various types of glass microspheres were studied to observe resistance to breakage during hot pressing of the

616

M.M. Ashton-Patton et al. / Journal of Non-Crystalline Solids 352 (2006) 615–619

composite. Effects of pressing parameters on the properties of the LDPE/HGM composites, including density and percent survival of microspheres, are presented in this paper. 2. Experimental 2.1. Materials A LDPE polymer in the form of spherical particles with a particle size distribution of 5–50 lm was used. This material has a melt index of 23 g/10 min (measured using ASTM D 1238), a density of 930 kg m3, and a melting point of 114.5 C. Three types of hollow glass microspheres were used in this study, which are designated as glass 1, 2, or 3. Glass 1 is an aluminosilicate glass with a bulk density of 170 kg m3. Glass 2 is a soda lime borosilicate glass, with a bulk density of 150 kg m3. Glass 3 is a different borosilicate glass with a bulk density of 160 kg m3. The void inside the hollow glass microspheres is filled with a small amount of an unknown blowing agent, but predominantly filled with vacuum. Densities were measured using a helium pycnometer. Table 1 list the materials and their properties. Composites are termed composite 1, 2, or 3, with the number corresponding to the type of glass used to prepare the composite. 2.2. Sample preparation Ten grams of LDPE samples with 5 wt% microsphere additions were mixed by hand and compression molded using a hydraulic press. The mold produces a rectangular bar, with dimensions of 0.12750 ± 0.00005 m · 0.01340 ± 0.00005 m. The batch was added to the bottom half of the mold and heated to 110 ± 11 C or 120 ± 11 C, for 10 min; the top was then added. The mold was held for another 5 min to allow the top to equilibrate thermally. No pressure was applied for the first 15 min. The pressure was then increased to 6.51 MPa (944 psi) or 3.90 MPa (566 psi) and held for 15 min. The mold was cooled to room temperature and the sample was removed. Three different pressing conditions were used in this study. Condition A consists of pressing at 6.51 MPa (944 psi) and 110 ± 11 C. Condition B consists of pressing at 3.90 MPa (566 psi) and 110 ± 11 C. Condition C consists of pressing at 3.90 MPa (566 psi) and 120 ± 11 C. 2.3. Apparatus and methodology Density of the composites was measured using the Archimedes method with kerosene as the immersion fluid. Densities below the density of kerosene were estimated by cutting a rectangle from a bar, polishing the edges, and measuring and weighing the sample to calculate density. Kerosene and LDPE are non-reactive. The standard deviation of the density measurements was calculated to be 2 kg m3, based on a large number of repeated measure-

Table 1 Particle size and bulk density of LDPE and HGMS prior to composite formation Material

Particle size (lm)

Bulk density (kg m3)

LDPE Glass 1 Glass 2 Glass 3

5–50 1–130 1–130 1–130

924.5 170 150 160

ments. Scanning electron microscopy (SEM) images were taken using an environmental SEM, in low vacuum mode, with 20 kV accelerating potential. The secondary electron (SE) detector and the backscattered (BS) electron detectors were both used. No metallic coating was necessary. 3. Results Fig. 1 shows the LDPE and HGM prior to composite formation. The LDPE particles have a smooth surface, with a spherical microstructure, and a particle diameter of 5–50 lm. Glasses 2 and 3 are both borosilicates. The surfaces of these microspheres are smooth, the particle diameter ranges from 1 to 130 lm, and 50% by weight of the spheres have they mean particle diameter less than 60 lm. Glass 1 is an aluminosilicate material, similar to a flyash in composition. These particles are spherical, but differ from the other materials because the surface is rough. Many of the spheres contain smaller groups of bubbles. The particle diameter of these microspheres, approximated as true spheres, range from 1 to 130 lm, with a mean particle diameter of 60 lm. Rectangular LDPE bars, with no microspheres added, have a pressed density of 930 kg m3. Composites 1, 2, and 3 pressed under condition A have densities of 940, 920, and 940 kg m3, respectively. Use of condition B resulted in the following densities for composites 1, 2, and 3: 920, 850, and 860 kg m3, respectively, while use of the higher pressing temperature in condition C resulted in a densities of 910, 770, and 820 kg m3 for composites 1, 2, and 3, respectively. A theoretical density can be calculated using Eq. (1), where V is the volume, q is the density, the subscript g is for the HGMS, the subscript p is for the LDPE and the subscript c is for the composite. The theoretical densities for composites 1, 2, and 3 are calculated to be 760, 740, and 750 kg m3, respectively. Densities for all composites and pressing conditions are listed in Table 2. Fig. 2 compares the theoretical densities and the actual densities. V g qg þ V p qp ¼ qc .

ð1Þ

Scanning electron microscopy was performed on all the samples to observe possible reasons for the differences between theoretical and actual densities. It is obvious that broken microspheres cause the increase in densities relative

M.M. Ashton-Patton et al. / Journal of Non-Crystalline Solids 352 (2006) 615–619

617

Fig. 1. (a) LDPE image taken using SEM in secondary electron (SE) mode, with an accelerating potential of 20 kV, and a 985· magnification, (b) soda lime borosilicate (c) borosilicate (d) aluminosilicate hollow glass microspheres taken using a SEM in SE mode, with an accelerating potential of 20 kV, and a 283· (b–c) and 80· (d) magnification.

Table 2 Different processing conditions result in different densities for each composite

1 2 3

80

Density (kg m3) Theoretical

Condition A

Condition B

Condition C

760 740 750

940 920 940

920 850 860

910 770 820

The theoretical values are calculated; all other densities were measured. The standard deviation for the measured densities is 2 kg m3.

to the theoretical prediction. Fig. 3 shows composite 2 processed under condition B. Fig. 4 shows more successful composite 2 processed under condition C. Using the measured densities, it is possible to estimate the percentages of unbroken HGM in the composite. Using Eq. (1), calculated values indicate condition A yielded approximately 17% whole HGM for all three composites. Use of condition B yields 17.3%, 27.3%, and 25.3% unbroken HGM for composites 1, 2, and 3, respectively. Use of condition C, which yields the largest changes in density, results 18.3%,

Percent Survival

Composite #

100 Aluminosilicate Borosilicate Soda Lime Borosilicate

60 40 20 0

Condition A Condition B Condition C

Fig. 2. Effect of pressing conditions on percent of surviving HGMS. Lines were added to guide the eye.

66.7%, and 32.8% unbroken HGM for composites 1, 2, and 3, respectively. Table 3 lists the percent survival of the HGMS under various processing conditions. Fig. 5 shows these results in graphic form.

618

M.M. Ashton-Patton et al. / Journal of Non-Crystalline Solids 352 (2006) 615–619

Theoretical Density (g cm-3)

1.00

Theoretical Condition C Condition B Condition A

0.80

0.60

0.40

0.20

0.00 Glass 1

Fig. 3. SEM image of Glass 2/LDPE processed under condition B. Image was taken at 283· with an accelerating potential of 20 kV. The image was taken in BSE. An unbroken HGM is indicated by the solid head arrow and a broken one is indicated by the open head arrow.

Fig. 4. SEM image of Glass 2/LDPE processed under condition C. Image was taken at 283· with an accelerating potential of 20 kV. The image is 60% SE and 40% BSE. An unbroken HGM is indicated by the solid head arrow and a broken HGM is indicated by a open head arrow. The broken HGMS have an O shaped appearance.

Table 3 Percent survival of HGM as a result of the processing condition Composite #

Condition A

Condition B

Condition C

1 2 3

17.7 17.6 17.0

17.3 27.3 25.3

18.3 66.7 32.8

Glass 2

Glass 3

Fig. 5. Experimental and theoretical densities of the composites resulting from the various processing conditions.

sure. If the LDPE is fluid enough, it will act like a fluid surrounding the HGM and the compressive stresses on the HGM in the mold will be hydrostatic. The SEM images show that failure is due to force acting in compression, causing a tensile stress to exist in the glass, leading to failure. Fig. 3 shows microsphere failure from the top surface of the composite, which is the direction the force was applied. No microspheres appear to fail by fracture in the lateral direction. Fig. 4 shows a number of HGM, which imploded due to an excessive force. More microspheres survive at lower pressures and higher temperatures, because the compressive stress does not exceed the materials intrinsic strength, and the LDPE is fluid enough to act as a liquid. The increase in density in composites where the microspheres are crushed is a result of the replacement of the empty interior of HGM by polymer, i.e., the HGM are no longer acting like bubbles in the composite. The HGM in this study have bulk densities ranging from 150 to 170 kg m3. A solid glass of the same composition has a density in the range of 2500 kg m3. Since the air is removed when the glass fails, the crushed glass increases the local density by over an order of magnitude, causing the density of the composite to increase. The experimental apparatus does not allow for pressing at lower pressures. To obtain a higher percentage of HGM that do not fail, the temperature at which the compression takes place should be increased in increments of 10 C. If experiments using higher temperatures are not sufficient to obtain densities closer to the predicted values, then HGM with thicker walls should be used. 5. Conclusion

4. Discussion HGM exhibit their greatest strength and, hence, have the greatest probably of survival, under hydrostatic pres-

Composites pressed under lower pressures and higher temperatures have the highest probability of success. HGM made out of borosilicate glasses do not break as easily as HGM made of an aluminosilicate material. A smooth

M.M. Ashton-Patton et al. / Journal of Non-Crystalline Solids 352 (2006) 615–619

surface is better than a rough surface for the HGM because it distributes the force more evenly, resulting in a more hydrostatic stress environment. Acknowledgement This work is funded by NASA Grant No. 03-OBPR0000-0074.

619

References [1] G.W. Bush, Office of the Press Secretary, ed., January 14, 2004. [2] S.G. Mashnik, K.K. Gudima, I.V. Moskalenko, Adv. Space Res. 34 (2004) 1288. [3] J.K. Poudrier, Space Res. 3 (2004) 24. [4] R.K. Tripathi, J.W. Wilson, F.A. Cucinotta, B.M. Anderson, L.C. Simonsen, Adv. Space Res. 31 (2003) 2383.