Creep and Stress Relaxation of Hybrid Organic-Inorganic Aerogels

Key Engineering Materials Vol. 423 (2010) pp 167-172 © (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/KEM.423.167 Creep and stress relaxation of hybrid organic-inorganic aerogels N. de la Rosa-Fox1, J. A. Toledo Fernández1, V. Morales-Flórez2, M. Piñero3, L. Esquivias4 1 Dpto. Física de la Materia Condensada, Universidad de Cádiz. España 2 Física de Materiales. ICMSE-CSIC. Sevilla. España. 3 Dpto. Física Aplicada, CASEM, Universidad de Cádiz, España. 4 Dpto Física de la Materia Condensada, Universidad de Sevilla. España nicolas.rosafox@uca.es, Keywords: organic-inorganic hybrid aerogel, uniaxial compression, nanoindentation, creep compliance, stress relaxation. Abstract. Organic/inorganic hybrids silica aerogels were synthesized by the classical sol-gel method with application of high power ultrasounds to the liquid mixture. Precursors were tetraethoxysilane (TEOS), as inorganic phase, and polydimethyl siloxane (PDMS), as organic one. These hybrid organic-inorganic materials are known as ORMOSIL (ORganically MOdified SILicates). Monolithic aerogels were obtained by supercritical drying in ethanol. Failure tests by uniaxial compression shows an increase of the rupture modulus as well as a decrease of the Young’s modulus with the polymer content, tuning from a brittle solid to a rubbery-like one. These hybrid aerogels behave as elastomers showing a decrease in the relaxation viscoelastic modulus. Nanoindentation tests have been performed in these hybrid aerogels: load/unload cycles about 1.5 mN of maximum load have shown a decreasing value of the reduced modulus, as well as both plastic and elastic work with the organic content, while hardness remains almost constant. Elastic recovery parameter rised with the increasing organic content. Results from creep tests made with uniaxial compression configuration are discussed and compared with nanoindentation. Viscoelastic behavior of these hybrids materials can be described by a rheological model. Introduction Organic/inorganic hybrid silica aerogels are nanostructured materials that combine mechanical and texture properties of a nanoporous silica matrix with those from the organic polymer phase embedded into it [1]. These kinds of materials gather characteristics that make them very attractive from a practical point of view because they show optical transparency as well as high porosity. They also possess some characteristics from organic polymers, such as flexibility, low density and formability, while the inorganic fraction contributes for increasing surface hardness and strength, and improving transparency and good optical properties [2]. From a mechanical point of view, their behaviour can be tuned from fragile to rubber, depending on the content of the organic polymer and the copolymerization degree. This qualitative change can be explained in terms of the percolation of the organic phase [3]Their structural applications are restricted due to one of the main drawbacks of silica aerogels: its fragility. However, a kind of aerogels, tens times stronger than those of typical aerogels, have been prepared by Leventis´group by means crosslinking silica with some organic polymers on the original gel skeletons. For this, they used 3-aminopropyltriethoxysilane as a co-precursor to obtain wet gels with amino groups on their surfaces. The reinforced of the network was achieved by connecting the All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of the publisher: Trans Tech Publications Ltd, Switzerland, www.ttp.net. (ID: 83.46.15.231-05/09/09,07:34:22) 168 Mechanical Properties of Solids XI terminal amino groups with polymerizing epoxides [4] or isocyanates [5]. This type of materials shows densification and inelastic hardening upon compression beyond the yield point. The viscoelastic behaviour of silica aerogels has been also characterized by other authors by means of acoustic techniques, based on the analysis of thickness resonances of air-surrounded aerogel plates at ultrasonic frequencies [6]. They reported that shear modulus (G) and shear loss (tan δG) exhibit a similar behavior to those reported for some kind of organic polymers for which the interpretation given is based on non-local cooperative interactions of large molecules which could also be applied to aerogels[7]. Non-linear viscoelasticity is based on the time evolution of the stress relaxation and the creep compliance [8, 9]. Several rheological models can be used to evaluate the mechanical behaviour [10], all of them based on the combination of spring and dashpot.In this way, viscoelastic behaviour of organically modified silica aerogels (ORMOSIL) synthesised with the organic polymer Polydimethyl siloxane (PDMS) has been analyzed and described elsewhere, by means of uniaxial compression and nanoindentation techniques [11]. It was reported that the inclusion of polymeric organic chains in inorganic silica aerogel matrix produces a rubbery-like material, whose macroscopic elastic modulus decreases with the polymer content and matches satisfactorily the values obtained by nanoindentation. The present work studies the viscoelastic behaviour of this type of organic-inorganic hybrid silica aerogels. Comparison between recovery during stress relaxation testing in compression and creep data from nanoindentation has been carried out for this insight. Experimental procedure Silica gels were synthesized by the classical sol–gel method by means of a two-step procedure. First TEOS (tetraethoxysiloxane), as inorganic phase, was partially hydrolyzed under stoichiometrically with acid water (pH =1) in a molar ratio of TEOS:H2O of 1:0.84. At this step the solution received an ultrasonic energy dose of 320 J cm-3 resulting a transparent and homogeneous solution. In the second step PDMS(polydimethylsiloxane), as organic phase, was added to complete the hydrolysis reaction with a molar ratio TEOS:H2O of 1:3.16 with an application of another 320 J cm-3 ultrasonic energy dose. Several PDMS content were used as the weight percent of the total silica content in the sample. The liquid sol was kept in hermetically close container until gelification took place, and the resulting gel was further immersed in an extra volume of ethanol in order to accelerate the aging process and to expel the residual water from the pores. This way provided aerogels with PDMS content up to 60 % by weight referred to the nominal content in silica. The supercritical drying, was conducted in an autoclave, with ethanol (260 º C and 90 bar), which allows the simultaneous drying of up to 12 cylindrical samples about 3 cm height and 0.7 cm radius. Several aerogel monolithic pieces were obtained, whose specific surfaces varied in a range of 400 to 900 m2 g-1. Thermogravimetric analysis were perfomed to elucidate if the organic chains were affected by drying temperature, showing that this supposition may be neglected. As a matter of fact, the hybrid organic-inorganic silica aerogels have been obtained without organic chain degradation by thermal decomposition. The structure is depicted by the organic chain cross linking bonding the porous silica clusters via the copolymerization between TEOS and PDMS silanol terminated.The whole thermal and textural and structural characterization process for these samples have been previously described [1] Nanoindentation creep test was carried out on a hybrid organic/inorganic aerogel with 40 wt.% of PDMS. A Nanotest automatic device from MicroMaterials Ltd. (UK) equipped with a diamond indenter with Berkovich pyramidal tip (100 nm diameter) was used for this study. Load– depth curves were recorded in 25 indentations in line at 20 µm distances, using maximum loads up to 4mN and load rates ranging between 5 to 75 µN/s. Maps were registered on 10×10 grid indentations with 10 µm apart each one, surface was polished at optical level. Creep data were collected along 3000 s, in a cabinet with controlled humidity around 40% and temperature of 28 ºC. The tests were run under different loads in the range from 0.3 to 1.5 mN. Key Engineering Materials Vol. 423 169 Pure silica aerogel and hybrid aerogel with 40% wt. of PDMS were also tested for failure with a Shimadzu universal machine equipped with a 5 kN load cell in the uniaxial compression, and the same configuration test was applied for the stress relaxation study. A series of test curves at different constant loads were recorded during 2000 s, at room temperature, on cylindrical samples 0.8 cm diameter, 1.4 cm height. All the stress relaxation experiments were performed on an aerogel sample with 50 wt % PDMS, preliminary strain recovery observed was highest. Results and discussion Failure tests by uniaxial compression shows an increase of the rupture modulus as well as a decrease of the Young’s modulus with the polymer content, tuning from a brittle solid to an elastomeric one (Fig. 1). 30 pure silica aerogel PDMS 40%/TEOS Fig. 1. Stress-strain plot for comparing the elastic behaviour and the maximum rupture stress between pure silica aerogel (solid circles) and an organic-inorganic 40 wt% PDMS hybrid silica aerogel sample (solid triangles). σ (MPa) 20 10 0 0,0 0,1 0,2 0,3 0,4 ε (%) Some plastic deformation is observed for the pure silica aerogel due to the collapse of the pore structure, whereas the hybrid aerogel behaves as an elastomer, showing large strains at low stresses. At high strain it is apparent the increases of the elastic modulus for the hybrid sample, transient from 40 MPa to 122 MPa, the highest value being similar to that obtained from the pure silica aerogel at lower strain (144 MPa). Under compression the PDMS chains behaves as the softer structure is folded by the chemical bonds Si-O-Si, which allows a maximum strain of 25%, just around the value in which an abrupt slope change is observed. Then we can suppose that starting from here, till failure, the mechanical behaviour is conducted by the silica clusters which behave as the stiffer structure. Creep strain in nanoindentation was analyzed on hybrid organic-inorganic aerogel with 40 wt.% of PDMS. This parameter can be related to some structural parameters [12]. In this way the creep compliance can be also studied. It shows how the material acts under a constant load, and it can be defined as J (t ) = ε (t ) σ0 Being impossible to define the classical strain in nanoindentation experiments, in the literature the expression for this purpose can be found [13]. For a Berkovich tip, the indenter is considered as cone-shaped of semi-apical effective angle θ=70.32º, bearing in mind that this fact implies an error lower than 3%. Then, the creep compliance can be expressed as follows: J (t ) = A( t ) tg θ P0 170 Mechanical Properties of Solids XI where A(t) is the depth-dependent contact area for the Berkovich tip A( t ) = 24.5h2 with h being the depth rate and P0 is the applied load. For this sample, the creep compliance data were obtained at four different constant loads of 0.3 mN, 0.5 mN, 1.0 mN, and 1.5 mN, and are depicted v.s time in Fig. 2a). It is clearly evident the existence of two linear regimes in the three curves corresponding to 0.3, 0.5 and 1.0 mN constant load , which are transformed into only one straigth for the load of 1.5 mN. The J(t) behaviour in the first linear regime, which ends at about 600 s, correspond to the response of the porous glass structure network, which deforms instantly. At higher time begins the second linear regime behaviour where the increase of J (t) values, for the curves at 0.3 and 0.5 mN, is related to the viscoelastic nature of the system. It is important to appreciate, in the curve of 0.3 mN load, a saturation at 3000 s that would indicate the start of a rubbery state of viscous flow. The increase of J (t) with the load (50 to 110 GPa-1) is indicative of the viscoelastic nature of these organic-inorganic hybrid aerogels. Fig. 2b) it allows to differentiate softer from stiffer surface sites and a pore site with a constant creep compliance and an increase when the tip touch the pore wall. 120 a) Aerogel/PDMS 40% 110 100 -1 90 80 J(t) GPa J(t) Gpa -1 140 130 120 110 100 90 70 1.5 mN 1.0 mN 0.5 mN 0.3 mN 60 50 b) Aerogel/PDMS 40% stiff site P 0=0.5 mN 80 pore site 70 60 soft site 50 40 40 10 100 10 1000 100 1000 t(s) t(s) Fig.2 a) Creep compliance curves dependence of load stress for nanoindentation experiments performed on PDMS 40 wt.% organic-inorganic hybrid silica aerogel Fig.2 b) Creep compliance curves performed at three different surface sites at constant load of 0.5 mN, revealing different mechanical behaviours at nanometric scale. Stress relaxation data from uniaxial compression at constant strain, give the relaxation modulus rate as: σ (t ) G (t ) = ε0 giving information about the general viscoelastic behaviour. The obtained results for the PDMS 50 wt % organic-inorganic hybrid aerogel at different constant strain are presented in Fig. 3a). The observed increase of the relaxation modulus run parallel with the deformation applied. All measured data were fitted to a bi-exponential equation (Fig. 3b), which provided good fit, visually and in terms of the correlation coefficients. These values obtained in the early stage corresponds with a fast response of the polymer chains which acts as an elastic spring within the range. Then the collapse of porous silica clusters under compression gives the viscoelastic response and, finally, the viscous flow behaviour of the organic chains acts as a damper. Under these conditions, the bulk presents a rheological behaviour described by Burger model which results from the combination of Kelvin and Maxwell parallel elements [14]. Also, considering the modelled bi-exponential equation, if y0=18 MPa as the relaxation modulus G (inset of Fig. 3b), the macroscopic result agrees with the microscopic interpretation of the mechanical behaviour given by nanoindentation techniques. Inasmuch as the creep compliance at t=0, known as Key Engineering Materials Vol. 423 171 instantaneous value, verifies that G=1/J for J=50 GPa-1 that correspond to G=20 MPa in agreement with the bi-exponential fit. PDMS 50 wt.% a) b) strain (ε %) 35 % 33 % 32 % 31 % 30 % 27 % 25 % 22 % 18 % 14 % 9% 4% 8 20 Chi^2/DoF = 0.00029 R^2 = 0.99702 σ (MPa) σ (MPa) 16 Data: aero43vc11b_A Model: ExpDec2 Equation: y = A1*exp(-x/t1) + A2*exp(-x/t2) + y0 Weighting: y No weighting 19 y0 A1 t1 A2 t2 17.91636 0.96747 47.99682 0.93511 735.49984 ±0.00358 ±0.00863 ±0.85277 ±0.00384 ±10.05391 1000 1500 2000 18 0 0 900 1800 time (s) 0 500 time (s) Fig. 3 a) Stress relaxation curves at several different constant strain for PDMS 50 % silica aerogel sample. Outlined strains on the right correspond to the reached values for the experiment. Fig. 3 b) Curve fitting to a bi-exponential model showing the parameters in the inset. Conclusions Organically modified PDMS-silica aerogels are homogeneous materials whose physical and mechanical properties, may vary according to the organic phase content. Failure tests by uniaxial compression shows an increase of the rupture modulus, from 8 MPa for pure silica aerogel to 28 MPa for a PDMS 40 wt % content. as well as a decrease of the Young’s modulus from 144 MPa for pure silica aerogel to 40 MPa for PDMS 50 wt % content. These hybrid aerogels behave as elastomers showing a evolution in the relaxation viscoelastic modulus rate. The model of microscopic mechanical behavior of hybrid allow us to describe the aerogel as viscoelastic material, in which there are two regimes of non-linear elastic behavior. Macroscopic interpretation provides that the Burger’s model is in good agreement with the microscopic level results. Acknowledgements This work has been financed by the projects MAT2005-01583, from the Ministerio de Educación y Ciencia (España), and by the financial support of the Junta de Andalucía to the team TEP-115. References [1] M. Piñero, V. Morales-Flórez, N. de la Rosa-Fox and L. Esquivias: Bol. Soc. Esp. Ceram. V, 44(5), (2005) 291–293 [2] J. D. Mackenzie: J. Sol-Gel Sci. and Tech. 26 (2003) 23-27. [3] V. Morales-Florez, J.A. Toledo-Fernandez, N. de la Rosa-Fox, M. Piñero, and L. Esquivias: J. Sol-Gel Sci. and Tech. 50 (2009) 170-175. [4] MAB Meador, EF Fabrizio, F Ilhan, A Dass, G Zhang, P Vassilaras, JC Johnston, N Leventis: Chem Mater 17 (2005) 1085 172 Mechanical Properties of Solids XI [5] A Katti, N Shimpi, S Roy, H Lu, EF Fabrizio, A Dass, LA Capadona, N Leventis: Chem Mater 18 (2006)285 [6] T. E. Gómez, F. R. Montero: Appl. Phys. Lett., Vol. 81 (2002) 348 [7] T. L. Szabo and J. Wu,: J. Acoust. Soc. Am. 107, (2000) 437 [8] H.Lu, B.Wang, J.Ma, G.Huang, H. Viswanathan, Mech. of time-depend. Mat. 7 (2003) 189-207 [9] M.Sakai, M.Sasaki, A.Matsuda, Acta Mater. 53 (2005) 4455 [10]I:M:Ward, J.Sweeney, in “An introduction to the mechanical Properties of solid polymers” (2nd Ed.), John Wiley & Sons, Ldt. (2004) Ch. 10 [11] N. de la Rosa-Fox, V. Morales-Flórez , J.A. Toledo-Fernández, M. Piñero, R. Mendoza-Serna, L. Esquivias, J. Eur. Cer.Soc. 27 (2007) 331 1–3316 [12] C. Tsou, C. Hsu, W. Fang, T. S. Lai and H.C. Li: In IEEE6th International Conference on Electronic Packaging Technology, (2005) 0-7803-9449-6/05 [13]M. R. VanLandingham. J. Res. atl. Inst. Stand. Technol. 108, (2003) 249-265 [14] B. Moginger: Rheologica Acta 32(4) (1993) 370-379 View publication stats