New device to measure dynamic intrusion/extrusion cycles of lyophobic heterogeneous systems Ludivine Guillemot,1, ∗ Anne Galarneau,2 G´erard ´ Vigier,3 Thierry Abensur,4 and Elisabeth Charlaix1, † 1
Laboratoire de Physique de la Mati`ere Condens´ee et Nanostructures,
Universit´e Claude Bernard, 6 rue Amp`ere, 69622 Villeurbanne Cedex, France. 2
Laboratoire Mat´eriaux Catalytiques et Catalyse en Chimie Organique, ´ Institut C. Gerhardt FR 1878, Ecole Nationale Sup´erieure de Chimie de Montpellier, ´ 8 rue de lEcole Normale, 34296 Montpellier Cedex 5, France. 3
Laboratoire Mat´eriaux: Ing´enierie et Science, INSA de Lyon, 7 av. Jean Capelle, 69621 Villeurbanne Cedex, France. 4
Astrium Space Transportation SAS,
66 rte de Verneuil, 78133 Les Mureaux Cedex, France. (Dated: February 22, 2011)
∗
[email protected]
†
[email protected]; Corresponding author
1
A new technique, that can be applied for energy storage or energy dissipation, using a mesoporous material immersed in a non-wetting liquid, was proposed by Eroshenko [1]. This new technique could be more performant than traditional techniques of energy absorption. Wet got interested in these lyophobic heterogeneous systems (LHS) to use them for high frequencies vibrations damping. When a LHS is subjected to an external pressure, the non-wetting liquid is forced to enter into the pores till the mesoporous material is saturated. Then, when the pressure is lowered, the liquid goes out of the pores at a pressure that can be lower than the intrusion pressure. The corresponding pressure hysteresis is characteristic of a dissipation energy that can serve to absorb vibrations. The different behaviors of LHS can be characterized by measuring cycles P − ∆V where P is the external pressure applied to the system and ∆V , the volume change due to the intrusion of the liquid into the material pores and its possible extrusion.Several LHS were well-characterized in quasi-static conditions by Lefevre [2] but few experiments were conducted to evaluate their dynamic characteristics. For instance, Surani et al. calculated the energy dissipation of a LHS using Hopkinson bars [3] and Iwatsubo et al. conducted dynamic tests with LHS inserted in a pistoncylinder unit [4]. They showed that dissipated energy varies little with the solicitation frequency. Although these experiments proved the good working of LHS in dynamic, it still lacks many experimental data to valide a theoretical model that could predict the behavior of a heterogeneous system at different velocities and temperatures. That is why we designed a new apparatus that allows us to carry out intrusion/extrusion cycles with different systems liquid/porous material, controlling the temperature from ambient to 80o C 80C and the velocity from 0.08 to 80 mm/s. In the first part, the apparatus will be detailed as well as the analyse of the measures. Then, different tests will be presented.
2
I.
TEST DEVICE AND EXPERIMENTAL PROTOCOLES A.
Introduction
First, the lyophobic heterogeneous system, composed of the liquid and the mesoporous material, is inserted into a thermosealed shrinkable polymer container. To make this sample, both material and liquid are outgassed for several hours and are not in contact with the ambient air till the polymer container is closed. The sample making process is detailed in [2]. In order to apply pressure on the LHS sample in dynamic conditions, an original test apparatus was created. The technological difficulty of a such device lies in the realization of the dynamic sealing of the chamber at high velocities (∼100mm/s) and high pressures (60MPa) conditions. To overcome this obstacle, a metal bellows was used. With this latter, a static sealing of the chamber is sufficient. The bellows is the main part of the prototype : a rod attached to the bellows and axially activated by a tension/compression machine allows us to vary the length of the bellows and then a volume change of this one. As a result, the pressure inside the chamber, filled with a few compressible liquid glycerine, is increased or decreased and these pressure variations are transmitted to the LHS sample placed inside the chamber. Finally, intrusion/extrusion cycles are carried out by extending or compressing the metal bellows. A scheme of the test prototype is presented figure 1. The geometrical characteristics of the metal bellows allow us to reach a maximum pressure of 600 bar. The bellows stroke is ±3 mm, which corresponds to a volume change of about 2 cm3 . Finally, heater bands can be placed around the chamber to control the temperature up to 80C. They are regulated by a Pt100 probe inserted in the chamber wall. To measure the intrusion/extrusion cycles, it is necessary to know the pressure and volume change (graph P − ∆V ). The pressure is obtained by a miniature pressure sensor inserted at the base of the chamber and close to the SHL sample. The uncertainty on the value of the relative pressure is of ± 2 bar. The volume change is deducted from the displacement of the crossbeam of the tension/compression machine, which is measured by the machine software itself. The inside chamber temperature is known thanks to a thermocouple inserted through a sealed passage, near the pressure sensor. The device is screwed on a hydraulic tension/compression machine which allow us to
3
Rod activated by the tension machine Metal bellows Glycerol HLS sample Heater bands with temperature regulation
Thermocouple Grips screwed on the tension machine
Pressure transducer
FIG. 1: Transversal cut of the test device.
do intrusion/extrusion cycles up to 230 mm/s. The test apparatus, placed on the tension/compression machine, is showed figure 2.
FIG. 2: Device placed on a hydraulic tension/compression machine.
B.
Velocities qualification of the test device
The intrusion/extrusion cycle were conducted at speeds ranging from 0.08 to 160 mm/s. In quasi-static conditions, elongation and compression of the bellows are made at constant speed and therefore zero acceleration. However, when the speed increases, the deformation of the bellows is no longer constant during the phase of elongation or compression. From
4
a speed limit of 80 mm/s, acceleration of the machine is never constant. Therefore, the corrugations of the bellows are not deformed in the same manner and at the same time. Measurements of pressure and volume during a cycle performed at high speed are more uncertain. However, these high speed tests have shown the intrusion and extrusion of the liquid are still performed at 160mm/s and the hysteresis cycle is reproducible.
C.
Experimental data analyzes
For one test defined by one velocity and one temperature, three successive intrusion/extrusion cycles, with a one-second pause between each one, were carried out. In general, the first cycle differs slightly from the followings (≤ 0.6%), which is probably due to machine clearances. The second and third cycles are superimposed, which shows the stability of the HLS (see figure 3). 50
P (MPa)
40 30 20 10 0 -2000
-1500
-1000
-500
0
3
Volume change (mm /g)
FIG. 3: Three successive intrusion/extrusion cycles. The pressure is measured by a sensor and the volume change is deducted from the displacement of the piston of the traction machine (experimental raw data). The volume change is given per gram of mesoporous material.
The second cycle was selected for each test and it had to be corrected in order to get the real volume change corresponding to the intrusion/extrusion cycle. Indeed, the measured volume includes the volume change due to the water intrusion/extrusion, but also the deformations of the traction machine and the volume change due to the compressibility of the system and in particular that of the glycerol, which filled the chamber. Two methods were used to correct the raw data, a numerical one and a calibration one. We obtain the same results with the both methods, which confirm the volume is accurate, even if the volume 5
correction seems to be very huge. It is also important to notice that the correction only changes the volume, and thus, the pressure, the studied variable, is not modified at all. The figure 4 shows raw cycle and a corrected one. raw cycle corrected cycle
60
P (MPa)
50 40 30 20 10 0 -3000
-2000
-1000
0 3
Volume change (mm /g)
FIG. 4: Correction applied to study intrusion/extrusion cycle.
II.
LYOPHOBIC HETEROGENEOUS SYSTEMS TESTED
We have been able to studied several heterogeneous systems, for temperature from ambient to 80C and velocities to 0.08 to 80mm/s.
A.
Galinstan/CPG
Galinstan is a gallium-indium-tin eutectic alloy melting at -19C. As a liquid metal, it is naturally non-wetting on glass surface. To ensure this result, we carried out contact angle measures on a pyrex plan and we measured a static contact angle of about 130. The controlled pore glass (CPG) used for the intrusion/extrusion tests presents interconnected pores, a narrow pore sizes distribution and a mean pore diameter of 50nm. It was supplied by Sigma-Aldrich. The figure 5 shows five successive intrusion/extrusion cycles performed at 50C and low velocity (0.08mm/s). The first cycle presents a very flat intrusion stage, significant of a narrow pore sizes distribution. When we apply the Laplace capillarity law, we find a contact angle of 130, what was found by the static contact angle measures. This proves that Laplace law can be applied to describe metal alloy intrusion as well as water intrusion [5]. Besides, the extrusion pressure
6
1st
2nd
4th
5th
FIG. 5: Five successif intrusion/extrusion cycles performed with Galinstan/CPG at 50C and low velocity.
and the intruded volume decrease during the following cycles. The phenomenon is due to a chemical reaction between the mesoporous glass and the Galinstan, and especially the gallium. Indeed, another intrusion/extrusion test with only gallium as intrusion liquid with the same CPG. The same phenomenon happened faster than with Galinstan. It confirms first the intruded volume drop is not due to a segregation of the alloy when Galinstan is used and second, gallium is the element that reacts with the porous glass to produce a gaz, probably oxygen.
B.
Water/MCM-41
MCM-41 [6] is a model material : it presents independent cylindrical pores that are hexagonally ordered and a narrow pore sizes distribution. We synthesized it according to the protocol given in the article [7] and grafted with chlorodimethyloctylsilane [8]. The figures 6 present intrusions and extrusions of water in a sililated MCM-41, performed at 60C and various velocities (0.08 - 80 mm/s). It is possible to follow very well the intrusion and extrusion level depending on the speed and to measure the intrusion/extrusion pressures. Thus, we see the intrusion pressure increases slightly with the velocity, by 2%/velocity decade, and the extrusion pressure decreases by 4%/velocity decade. The velocity influences a few the hysteresis cycle, and thus the energy dissipation.
7
V↑ V↑
Intrusion
Extrusion
FIG. 6: Intrusions (left) and extrusions (right) measured for water/C8-grafted MCM-41 at 60C for various velocities between 0.08 and 80mm/s. III.
CONCLUSION
This apparatus allows us to study the dynamic characteristics of several lyophobic heterogeneous systems at different temperature and over three decades of time. Hysteresis cycles were measured at velocities reaching 80 mm/s, which correspond to a 10 Hz frequency. The large experimental data obtained can be used to valid a theoretical model that could predict the behavior of a lyophobic heterogeneous system at any temperature and frequnecy, only by knowing the porous material and liquid characteristics. Acknowledgements: We thank Thierry Abensur and EADS-Astrium for supporting this research.
[1] V. Eroshenko, Heterogeneous structure for accumulating or dissipating energy, methods of using such a structure and associated devices (2000), URL http://www.google.com/patents? hl=en\&lr=\&vid=USPAT6052992\&id=2zYDAAAAEBAJ\&oi=fnd\&dq= Heterogeneous+structure+for+accumulating+or+dissipating+energy,+methods+of+ using+such+a+structure+and+associated+devices\&printsec=abstract. [2] B. Lef`evre, Thesis, Institut National des Sciences Appliqu´ees de Lyon (2002). [3] F. Surani, X. Kong, D. Panchal, and Y. Qiao, Applied Physics Letters 87, 163111 (2005), ISSN 0003-6951, URL http://ieeexplore.ieee.org/xpls/abs\_all.jsp?arnumber=4817038.
8
[4] T. Iwatsubo, C. Suciu, M. Ikenagao, and K. Yaguchio, Journal of Sound and Vibration 308, 579 (2007), ISSN 0022460X, URL http://linkinghub.elsevier.com/retrieve/pii/ S0022460X0700452X. [5] B. Lef`evre, A. Saugey, J. L. Barrat, L. Bocquet, E. Charlaix, P. F. Gobin, and G. Vigier, The Journal of chemical physics 120, 4927 (2004), ISSN 0021-9606, URL http://www.ncbi.nlm. nih.gov/pubmed/15267355. [6] J. Beck, J. Vartuli, W. Roth, M. Leonowicz, C. Kresge, K. Schmitt, C. Chu, D. Olson, and E. Sheppard, Journal of the American Chemical Society 114, 10834 (1992), ISSN 0002-7863, URL http://pubs.acs.org/doi/abs/10.1021/ja00053a020. [7] T. Martin, A. Galarneau, F. Di Renzo, D. Brunel, and F. Fajula, Chemistry of Materials 16, 1725 (2004), URL http://pubs.acs.org/doi/abs/10.1021/cm030443c. [8] T. Martin, A. Galarneau, D. Brunel, V. Izard, V. Hulea, A. C. Blanc, S. Abramson, F. Di Renzo, and F. Fajula, Stud. Surf. Sci. Catal. 135, 4621 (2001), URL http://linkinghub. elsevier.com/retrieve/pii/S0167299101813306.
9
