Determination of coefficient thermal expansion of glass microballoon/vinyl ester syntactic foams Zulzamri Salleh, Md Mainul Islam, and Jayantha Ananda Epaarachchi
Citation: AIP Conference Proceedings 1901, 030012 (2017); View online: https://doi.org/10.1063/1.5010477 View Table of Contents: http://aip.scitation.org/toc/apc/1901/1 Published by the American Institute of Physics
Determination of Coefficient Thermal Expansion of Glass Microballoon/Vinyl Ester Syntactic Foams Zulzamri Salleh1, 2, a), Md Mainul Islam1, b) and Jayantha Ananda Epaarachchi1, c) 1
Centre for Future Materials (CFM) and School of Mechanical and Electrical Engineering, Faculty of Health, Engineering and Sciences, University of Southern Queensland, Toowoomba, Queensland 4350, Australia. 2 Universiti Kuala Lumpur, Malaysian Institute of Marine Engineering Technology, 32200, Lumut, Perak, Malaysia. a)
Corresponding author:
[email protected] b)
[email protected] c)
[email protected]
Abstract. The coefficient thermal expansion, (CTE) of glass microballoon/vinyl ester syntactic foam was determined using dimensional changes of a temperature gradient plot. The CTE was measured and found to be up to 53-63 % lower than the vinyl ester resin matrix when mixing with different weight percentages of the glass microballoon ranging from 2 wt.% to 10 wt.% using a thermomechanical analyzer (TMA). The results of CTE showed that it has a strong relationship with the syntactic foam density (), radius ration (cavity porosity and matrix porosity . Experimental results showed that the CTE decreases when glass microballoons are added into the composites measured at different temperatures ranging from 30 to 70 °C. The CTE from the experimental results were also compared with Turner’s modification model for composites for its suitability for thermal expansion of syntactic foams. The results indicate that Turner’s modification model exhibits a close correlation with the reduction up to 80 % of CTE based on experiment.
INTRODUCTION Hollow glass microballoon particles when mixed with a resin matrix can be considered as a closed cell foam or syntactic foam. Previously a lot of works that has been carried out on particular syntactic foam characterized the mechanical properties, and also physical properties such as density, volume fraction, wall thickness and porosity [14]. Physical properties such as density syntactic foam, including a radial wall thickness ration, porosity and voids are most likely to be very effective in tailoring mechanical properties of syntactic foam [5]. This type of lightweight materials is useful for aerospace [6], marine, thermal insulation and packaging [7] because of their good mechanical properties for the environment. It will lead to a concern subject to high temperatures and thus it is important to determine the coefficient of thermal expansion, (CTE) [8]. Determination of CTE between the insulation and the substrate is important for use in electronic packaging and thermal insulation because it is necessary to reduce thermal stress and possible failure between two surfaces [8]. Meanwhile, the CTE has also been studied extensively in the aerospace shuttle particularly in the fuel tank to reduce the possibility of failure due to thermal stress [8]. An extensive study was also conducted on the characteristics of CTE of PU foam used as insulation for pipes and cooling gas cooler vehicles [9]. Meanwhile the report on other fillers such as glass microballoon did not include many findings on CTE in the literature review. In the recent study, Shunmugasamy et al. (2012) [8] reported that CTE is decreased for three different types of glass microballoon. Another study by Park et al. (2009) [10] also showed that CTE is lower than neat epoxy resins for different glass microballoon weight percentage (0-2 % by weight). Yung et al. (2008) [11] also revealed that the reduction of CTE was reported for different volume fractions of glass microballoon mixed with epoxy resin in syntactic foam. Therefore, these findings were not reported clearly about CTE relationship with the physical characteristics of glass microballoon especially porosity and voids are likely to have the same results with ceramic microballoon. In this study, the gap will be investigated to understand more deeply the effect of CTE of the Advanced Materials for Sustainability and Growth AIP Conf. Proc. 1901, 030012-1–030012-5; https://doi.org/10.1063/1.5010477 Published by AIP Publishing. 978-0-7354-1589-8/$30.00
030012-1
glass microballoon. In order to improve the understanding of this relationship, several models have been introduced for the purposes of this study. Shunmugasamy et al. (2012) [8] have found that Kerner and Turner model is suitable to be used for comparison with the experimental results of CTE. In his literature, he also revealed that some models can estimate a CTE of the particulate composites. Gunes et al. (1946) [12] also reported that several shape memory polymer (SMP) composites exhibited their CTE are comparable with theoretical model. Turner as the founder of this model also assumes that the filler is considered as an isotropic material and thus CTE will not depend on the physical properties of the filler itself [13]. The report also found the changes in the dimensions of the constituent material having the same bulk specimens when associated with temperature, while Kerner with other co-founder of the polymer composite model estimated the CTE considering for shear and isostatic stress occurred in particulate phase composites [14]. Generally, both models used the thermosetting polymer composite matrix resin where CTE is closed to the experimental results within the scope of the wall thickness parameter. Thus, the theoretical and experimental approaches will be used in this study to link the CTE with different weight percentages of glass microballoon (2-10 % by weight) in syntactic foam. All physical parameters of glass microballoon will be taken into account, such as syntactic foam density, wall thickness, radius ration, porosity and voids content towards developing the use of lightweight materials for the application of thermal stability.
EXPERIMENTAL Materials Fused borosilicate glass microballoons were procured from Potters Industries Inc., Australia under the trade name of Q-CEL Spherical (R) Hollow Microspheres. The vinyl ester resin with scientific name as diglicidyl ether of bisphenol A-based (DEBA) resin and methyl ethyl ketone peroxide (MEKP) catalyst, used as the matrix materials, are supplied by NOROX Australia for this study.
Fabrication of Syntactic Foams The different weight percentages (2 – 10 wt.%) of glass microballoon were mixed into the vinyl ester resin matrix. Glass rod was used as a mixer for the materials and then the constituent materials were poured into a plastic container. It took 5 minutes to ensure it was homogenized mixing to complete. Then the MEKP as hardener was added with the ratio of 5 % and stirred until styrene gas was evaporated with bubbles observed during this chemical reaction process. This mixture was poured into a mold to cure at room temperature for 24 hours and the finally cured in an oven at 70 °C for 3 hours. Finally, it was cut in the form of a solid cubic size of 3 mm x 3 mm x 1 mm.
Thermomechanical Analyser (TMA) Linear dimensions at different temperatures for the thermal expansion of the specimen was studied by using thermomechanical analysis apparatus (TMA) TA Instrument (Model TGA Q500). An expansion type probe was used for measuring the temperature-dependent dimensional changes. A preloading of 0.02 N was applied for all the tests. Minimum of 3 coupons were prepared for each of the composition. The use of the air compressor was connected to the system unit TA for cooling process after completion of the test. The heating rate in each run was kept at 3 oC/min and temperature range was ambient to 80 oC. Time, temperature and change in specimen height were recorded by data acquisition system during the test. The slope of tangent also called as coefficient of thermal expansion (CTE) between dimension change-temperature plot was predicted elsewhere [8-9].
Coefficient of Thermal Expansion (CTE) Model The rule of mixture (ROM) for composite materials is commonly used to obtain upper bounds of various properties. Shunmugasamy et al. (2012) [8] duplicated the relationship equation between ROM and CTE as shown in equation (1) below; α = 𝛼𝑚 𝛽𝑚 + 𝛼𝑔 𝛽𝑔
(1)
030012-2
where 𝛼𝑚 , 𝛼𝑔 are CTE for matrix resin and glass microballoon, respectively. The mixture of specimen in this study was used with weight percentages (wt.%), 𝛽𝑚 , 𝛽𝑔 of matrix resin and glass microballoon, instead of using volume fraction, which is introduced by Shunmugasamy et al. (2012) [8]. Kerner’s and Turner’s model have been modified from previous report by Shunmugasamy et al. (2012) [8] to include the physical parameter of glass microballoon such wall thickness and radius ration and excluded the porosity and voids. Tagliavia et al. [15] proposed that 𝐸𝑔 can be used at 60 GPa. The final modified Turner’s model can be divided and written by three equations as below in Equations (2) - (4); 1 + 𝑔 3 ) ∅𝑔 ] + 𝛼𝑔 𝛽𝑔 𝐸𝑔 (1 − 2∅𝑔3 )(1 − 2𝑚 ) 2 1 + 𝑔 3 ) ∅𝑔 ] + 𝛽𝑔 𝐸𝑔 (1 − 2∅𝑔3 )(1 − 2𝑚 ) 𝛽𝑚 𝐸𝑚 [(1 − 2𝑔 ) + ( 2
𝛼𝑚 𝛽𝑚 𝐸𝑚 [(1 − 2𝑔 ) + ( Cavity porosity, α∅𝑔 =
1 + 𝑔 3 ) ∅𝑚 ] + 𝛼𝑔 𝛽𝑔 𝐸𝑔 (1 − 2∅3𝑚 )(1 − 2𝑚 ) 2 1 + 𝑔 𝛽𝑚 𝐸𝑚 [(1 − 2𝑔 ) + ( 2 ) ∅3𝑚 ] + 𝛽𝑔 𝐸𝑔 (1 − 2∅3𝑚 )(1 − 2𝑚 )
(2)
𝛼𝑚 𝛽𝑚 𝐸𝑚 [(1 − 2𝑔 ) + ( Matrix porosity, α∅𝑚 =
1 + 𝑔 3 ) ∅𝑣 ] + 𝛼𝑔 𝛽𝑔 𝐸𝑔 (1 − 2∅3𝑣 )(1 − 2𝑚 ) 2 1+ 𝛽𝑚 𝐸𝑚 [(1 − 2𝑔 ) + ( 2 𝑔 ) ∅3𝑣 ] + 𝛽𝑔 𝐸𝑔 (1 − 2∅3𝑣 )(1 − 2𝑚 )
(3)
𝛼𝑚 𝛽𝑚 𝐸𝑚 [(1 − 2𝑔 ) + ( Voids, α∅𝑣 =
(4)
where 𝑔 is Poisson’s ratio of glass microballoon that can be used as 0.21 [4], while the modulus of elasticity, Em vinyl ester matrix resin is used as 22.82 GPa [4] and 𝑚 is Poisson’s ratio as 0.35 [16]. In this prediction, the void content can be ignored from the modelling by Turner because air/gas trap in syntactic foam does not require higher percentage ( 5 %) to contribute for evaluation. Thus, the CTE value investigation only focused on radius ratio’s (), cavity porosity (g) and matrix porosity (m). The linear dimensions for thermal expansion characteristic of the prepared specimens at different temperatures were evaluated by thermomechanical analyser.
RESULTS AND DISCUSSION Comparative study on CTE using Turner’s model Turner’s prediction model related to porosity in this study revealed that CTE’s trend for both cavity porosity and matrix porosity are closed to each other as shown in Fig.1. At 30 oC, Turner’s model predicted lower CTE compared with the experimental result. Here decreased from 28 µmoC-1 to 19 µmoC-1 but still below the experimental value which is 38 µmoC-1. It is also observed that in this model, many CTE values are much higher than experimental value at the elevated temperature namely 50 oC-70 oC. This model can predict that CTE value is closed to the experimental value for composition of glass microballoon between 2 - 4 wt.%. Therefore, the higher composition of glass microballoon has contributed to more porosity. As a result, a large gap occurred between experimental CTE and that from Turner’s model, especially at 70 oC (48 -57 µmoC-1). Using this model, reduction of CTE can be calculated from 78 % to 44 %, and 80 % to 45 % of the porosity of the cavity porosity and matrix porosity, respectively. From the results obtained, this model predicts higher CTE values at high temperatures when glass microballoon content decreases in syntactic foam has a porosity.
030012-3
1.2
90
CTE, Experimental
T : 30oC,
0.6
T : 50 oC T : 70oC
0.4
(µmoC-1)
/ m (µmoC-1)
(b)
70
0.8
(c)
60 50 40
(a)
30 20
0.2
(d)
10
0 0
2 4 6 8 Syntactic Foam (wt.%)
0
10
2
90
70
(µmoC-1)
40 (d)
50
(a)
40
(d)
30 20
10
10
0
4 6 8 Syntactic Foam (wt.%)
(c)
60
20
2
(b)
70
(a)
30
Ta: 70oC
80
60 50
4 6 8 10 Syntactic Foam (wt.%)
90
Ta: 50oC (b) (c)
80
(µmoC-1)
Ta: 30 oC
80
1
0
10
2
4 6 8 Syntactic Foam (wt.%)
10
FIGURE 1. Typical comparison results of CTE values using Turner’s model (a) Experimental (b) Radius ratio (), (c) Cavity porosity (g), (d) Matrix porosity (m), at 30 o C, 50 o C and 70 o C.
In this study, the average diameter of glass microballoon is ranging between 72 to 75 µm has been used to estimate the average outer and inner radius, ro and ri ranging between 36-38 µm size, respectively. The wall thickness of this microballoon was calculated using equation introduced by Tien et al. (2008) [17]. The thermal stability of the syntactic foam varied when referred to the wall thickness as the glass microballoon has different inner and outer radius. Similar result had also been detected from the previous study where the thin wall thickness (1-)
