Bo Wang, Koji Matsumaru, Jianfeng Yang and Kozo Ishizaki
Presented at the 2011 International Conference on Hot Isostatic Pressing Kobe, Japan, 12-14 April 2011
Submitted: 12 April 2011, Accepted: 24 May 2011
Topics CoveredAbstractKeywordsIntroductionExperimental ProcedureResults and Discussion Compressive StrengthConclusionsReferencesContact Details
Borosilicate foams were fabricated by melting borosilicate powder under a high pressure argon gas and subsequent isothermal heat treatment of molten glass. Borosilicate glass powder was melted at 1100 ºC for 1 h by capsule-free hot isostatic pressing (HIPing). Spherical pores filled with pressurized argon gas were introduced in the molten glass and argon molecules were dissolved in the glass network structure during HIPing process. By the isothermal heat treatment, argon-filled pores were expanded and dissolved Ar gas was released at 800 ºC under atmospheric pressure. Borosilicate foams have a high porosity up to 80 % with double distribution of micro-size and nano-size cells. The compressive strength was considerably increased from 15 to 56 MPa with increasing the total gas pressure from 10 to 70 MPa. It is conclude that the thickness of cell wall and the amount of small cells increase as the total gas pressure increases. The microstructure of glass foam is beneficial for the high strength.
Borosilicate Glass Foams, Porosity, Hot Isostatic Pressing, Gas Pressure, Compressive Strength
Cellular structure materials such as foams, reticulated and biomorphic ceramics are attractive for technological application, such as catalyst supports, filters for molten metals and hot gases, thermal insulators, refractory linings and implants [1-2]. Various processing routes have been proposed for ceramics foams, including foaming agent, polymeric sponge, or space holder method [3-6]. Glasses foams are promising materials due to their specific properties such as low density, low thermal conductivity, thermal stability, high surface area, good impact behavior and high permeability [7-10]. Cellular glasses are commonly foamed by the introduction of gases into molten glass, available from the decomposition or oxidation of specific powder additives [7-10]. However, the resulting foams have relatively low compressive strength due to the inhomogeneous microstructural including the coarse cells structure with linear dimensions approximately ranging from a few micrometers to a few millimeters (20 µm-5 mm) [7-10].
The fabrication method determines the range of porosity, the pore size distribution, and the pore morphology. An alternative method, based on expansion of pressurized pores trapped with a solid alloy compacted from powders, was first demonstrated by Kearns et al  for Ti-6Al-4V. In the first step, Ti-6Al-4V powders are compacted in the presence of argon gas by hot isostatic pressing (HIPing), resulting in a dense billet containing a small fraction of isolated, high-pressure, and micron-size pores. These argon-filled pores are then expanded through creep of the surrounding alloy during a high- temperature annealing step. This method has also been used to create high porosity CP-Ti [12-13] and the near-equiatomic Ni- Ti alloy (nitinol).
However, no work has dealt with the production of cellular glass ceramics by using HIPing. Under high gas pressure, besides the few amount of pressurized argon-filled pores introduced in the molten glass, the inert gas or nitrogen molecules can simply occupy the holes in the network structure, but do not react with the medium melt in which they are incorporated. It is suggested that the microstructure could be tailored and the pore diameter could be created from nanometer to micrometer by controlling gas absorption under high pressure and gas release under low pressure at elevated temperatures. Comparing to conventional methods, the tailored microstructure with smaller pore size by HIPing-foaming method will possess high strength. The present work reports on the influence of the gas pressure on the microstructure and compressive behavior of the borosilicate glass foams prepared by capsule-free hot isostatic pressing and subsequent isothermal heat treatment.
Amorphous borosilicate glass powder (Pyrex glass, Japan, code 7740) with a particle size 11 µm was used as starting powder. The chemical composition was as follow: 80 mass% SiO2, 13 % B2O3, 4 % Na2O, 2 % Al2O3 and 1 % K2O. The theoretical density of the glass used is 2.23 g/cm3. The glass powder was put into a graphite crucible coated by BN and replaced in HIP equipment (O2-Dr. HIP, Kobe Steel Co., Ltd.). Before increasing the temperature, the HIP chamber was vacuum purged and increased the HIP pressure to the final pressure backfilling with 99.995 % high purity argon. The glass powder was melted in HIP chamber at 1100 °C for 1 h with a heating rate of 400 K/h under pressures of 10, 20, 30, 40, 60 and 70 MPa. Expanded products were obtained in pebble form after isothermal heat treatment at 800 °C for 10 min under air atmosphere in an electrical furnace with a heating rate of 300 K/h.
The bulk density and closed porosity of the foam glasses were measured by the Archimedes displacement method. The porosity was calculated from the relative density and theoretical density, which was calculated by the rule of mixtures. The pore size distribution was determined by the mercury porosimetry (AutoPore IV, SHIMADIU). The pore surface and pore morphologies of the foam glasses were investigated by scanning electron microscopy (SEM, VE-7800, KEYENCE, Corp.). Square foam glass samples of 12.5 mm length and 5 × 5 mm section area having a porous structure were subjected to uniaxial compressive loading. The tests were conducted at a cross-head speed of 0.5 mm/min using universal testing machine. The edges of each traction surface were chamfered using 600-grit SiC paper before testing. Compressive strength was obtained by dividing the peak load by the cross- sectional area of the sample. Each final value was averaged over 5 measurements.
Results and Discussion
Fig. 1 shows the influence of HIPing pressure on the porosity and bulk density of borosilicate glass foams. A wide range of porosity is reached, 62 %-80 %. Under low gas pressure of 10-20 MPa, the glass foams have high porosity ~80 %. With increasing HIPing pressure from 20 to 70 MPa, the porosity of the samples decreases to ~ 60%.
Figure 1. Dependence of the bulk density and the total porosity on HIPing pressure, samples were HIPed at 1100 °C for 1h and annealing at 800 °C for 10min under different HIPing pressure.
Fig. 2 (a)-(d) show representative morphologies resulting form isothermal foaming at 800 ºC for 10 min. Under low HIPing pressure (10 MPa, as seen in Fig. 2(a) and (b)), nano-scale to micro-scale thin-wall cellular structure with uniform cell size and distribution can be obtained. Each bubble is isolated and has nearly spherical shape. With increasing the HIPing pressure to 70 MPa, pores have merged with each other, several coalesced cells in the 50-75 µm range are visible. The thickness of cell wall increase with the HIPing pressure increase. A number of smaller nano- or micro-cells is located in the cell walls (Fig. 2(d))
Specimens: one is about 8 µm, and another is in the 30~60 nm µm range. The amount of large cells ~8 µm decrease with increasing the HIPing pressure. The nano-pore size diameter decrease from 60 to 30 nm and the total amount of nano-pores increased with increasing the HIPing pressure. The pore sizes measured with mercury porosimetry are not the true sizes of the pores but of their interconnections. The average pore sizes are significantly smaller than those of the pores observed in the micrographs.
The final porosity and the microstructure evolution of cellular glass under different HIPing pressure were determined by the pore formation mechanism. As shown in Fig.4, under low HIPing pressure, it was dominated by the pore expansion by the decrease in pore pressure. Under higher HIPing pressure, pore created by dissolved gas release and pressurized argon-filled pore expansion.
The porosity was decided by the foaming rate and the surface tension of HIPed glass. It was assumed that the surface tension after HIPing process was same under different HIPing pressure. Under low HIPing pressure, the expansion of pore with lower inner pressure would result in slow foaming rates. Under higher HIPing pressure, supersaturated dissolved argon gas in the network structure was released, on the other hand, the pressure of pressurized argon bubble trapped in the glass was higher, which led to high foaming rate can make cause sufficient viscous flow deformation. However, premature foaming cessation could occur if large amount of argon gas molecule release coinstantaneous, which results in a fugacity of the physically dissolved gas connect with the specimen surface, so that the gas escaped to the ambient atmosphere and the driving fore for further pore expansion disappeared. This was in qualitative agreement with the surface roughness of the glass foams after annealing at 800 ºC. The HIPed-70 MPa glass foam was showed obviously much rougher surface than the HIPed-10 MPa glass foam. Finally, for HIPed at high gas pressure, specimens showed higher density.
Figure 2. Micrographs of the borosilicate glass foams by treated HIPed samples with different gas pressure at 800 ºC for 10min. (a)(b)10 MPa; (c)(d)70 MPa
Figure 3. Pore size distribution of borosilicate glass foams annealing at 800 ºC for 10 min under different HIPing gas pressure.
The microstructure evolution of cellular glass under different HIPing pressure was show in the Fig. 4. Under low HIPing pressure, the appropriate gas pressure in the pores decided the moderate driving force for pore expansion by viscous flow deformations of its surrounding molten glass, however, no including coalescence of pores, which could result in pores uniformly distributed with nano-scale thin cell wall. Under higher HIPing pressure, argon release was responsible for microstructure evolution. It can be assumed that, at evaluated temperature, with releasing the large amount of supersaturated argon, the fugacious gas molecule concentrate and form to nano-scale pores. And then, some of the nano-pores near the glass surface was escaped, some of them were connected to the nearest neighbor micro-cells making very larger open cells, the others nono-pores was act as nucleus absorb the gas molecule and expand to small micro-cells (as seen in Fig.4 (b)). It was in qualitative agreement with the pore size distribution.
The compressive strength is typically a function of the relative density. Fig. 5 illustrates the correlation between the compressive strength and the relative density for the prepared glass foams. The higher density results in the higher compressive strength. Usually, when foams were made from liquid components, surface tension could draw the material into the cell edges, leaving only a thin membrane across the faces of the cells, which ruptured easily. Then, although the foam had cells which are initially closed, its stiffness derived entirely from that of the cell edges, and its moduli were identical with those of open-cell foams . However, in the present closed-cell foams with gas pressure pores, the compressive behavior was decided by i) cell-wall bending; ii) edge contraction and membrane stretching; iii) enclosed gas pressure. When the enclosed gas pressure was much larger than atmospheric pressure, it modified the shape of the elastic collapse plateau especially contribute to the stiffness of the closed-cell foams . The pressure difference put the cell edges and faces in tension. They could not buckle and collapse until the applied stress has overcome both this tension and the buckling load of cell edges. Under higher HIPing pressure, large amount of nano-cells with high enclosed gas pressure were contributed to the high compressive strength. On the other hand, the expanded pores also remained gas pressure larger than atmospheric pressure, which also lead to high strength.
Figure 4. Schematic of pore formation models of cellular glass under different gas pressure.
Figure 5. Relationship between compressive strength and porosity. Samples were HIPed at 1100 °C for 1h and annealing at 800 °C for 10 min under different HIPing pressure.
Borosilicate foams with porosity of 62 %~80 % were fabricated by melting borosilicate powder under a high pressure argon gas and subsequent isothermal heat treatment of molten glass. The glass foams exhibits double size distribution: micro-pores by pressurized argon pores expansion and nano-pores by dissolved gas molecule release. The compressive strength increases with the decrease in porosity, and is between 15 and 56 MPa. The samples sintered under higher gas pressure have relatively high strength due to the large cell wall and nano-pores with enclosed high pressure in the microstructure.
1. P. Sepulveda, “Gelcasting Foams for Porous Ceramics,” Am. Ceram. Soc. Bull., 76 (1997) 61–65.
2. L. Montanaro, Y. Jorand, G. Fantozzi and A. Negro, “Ceramic Foams by Powder Processing,” J. Eur. Ceram. Soc., 18 (1998) 1339–1350.
3. K. Schwartzwalder and A. V. Somers, “Method of Making Porous Ceramics”, US Patent No. 3090094, May 21, 1963.
4. J. Saggio-Woyansky and C. E. Scottetal, “Processing of Porous Ceramics,” Am. Ceram. Soc. Bull., 71 (1992) 1674– 1682.
5. P. Sepulveda and J. G. P. Binner, “Processing of Cellular Ceramics by Foamingand In Situ Polymerisation of Organic Monomers,” J. Eur. Ceram. Soc., 19 (1999) 2059–2066.
6. H. X. Peng, Z. Fan, J. R. G. Evans and J. J. C. Busfield, “Microstructure of Ceramic Foams,” J. Eur. Ceram. Soc., 20 (2000) 807–813.
7. C. R. Rambo, E. Sousa, A. P. N. Oliveira and D. Hotza, “Processing of Cellular Glass Ceramic,” J. Am. Ceram. Soc., 89 (2006) 3373-3378.
8. H. R. Fernandes, D. U. Tulyaganov and J. M. F. Ferreira, “Preparation and characterization of foams from sheet glass and fly ash using carbonates as foaming agents” Ceram. int., 33 (2007) 963-968.
9. H. W. Guo, Y. X. Gong and S. Y. Gao “Preparation of high strength foam glass–ceramics from waste cathode ray tube” Material Letter 64 (2010) 997–999.
10. D. U. Tulyaganov, H. R. Fernandes, S. Agathophoulos and J. M. F. Ferreira, “Preparation and characterization of high compressive strength foams from sheet glass,” J. Porous. Mater., 13 (2006) 133-139.
11. M. W. Kearns, P. A. Blenkinsop, A. C. Barber and T. W. Farthing, “Manufacture of a Novel Porous Metal,” Met. Mater., 3 (1987) 85-88.
12. N. G. D. Murray and D. C. Dunand, “Effect of thermal history on the superplastic expansion of argon-filled pores in titanium: Part I kinetics and microstructure” Acta Mater., 52 (2004) 2269-2278.
13. N. G. D. Murray and D. C. Dunand, “Effect of thermal history on the superplastic expansion of argon-filled pores in titanium: Part II modeling of kinetics” Acta Mater., 52 (2004) 2279-2291.
14. S. Wu, C. Y. Chung, X. Liu, P. K. Chu, J. P. Y. Ho, C. L. Chu, Y. L. Chan, K. W. K. Yeung, W. W. Lu, K. M. C. Cheung and K. D. K. Luk, “Pore formation mechanism and characterization of porous NiTi shape memory alloys synthesized by capsule- free hot isostatic pressing,” Acta Mater., 55 (2007) 3437–3451.
15. L. J. Gibson and M. F. Ashby, “Cellular solids, structure and properties,” Cambridge University Press, Cambrige,( 1999) 175-231.
Bo Wang1, Koji Matsumaru1, Jianfeng Yang2 and Kozo Ishizaki1
1Department of Mechanical Engineering, Nagaoka University of Technology, Nagaoka, 940-2188, Japan
2State Key Laboratory for Mechanical Behavior of Materials, Xi’an Jiaotong University, Xianning West Road No. 28, 710049 Xi’an City, Shaanxi Province, China
This paper was also published in print form in "Advances in Technology of Materials and Materials Processing", 13 (2011) 24-29.