Bonnet Fabien, Melich Mathieu, Puech Laurent, Anglès d'Auriac Jean-Christian, Wolf Pierre-Etienne
Univ. Grenoble Alpes, CNRS, Institut Néel , 38000 Grenoble , France.
Langmuir. 2019 Apr 16;35(15):5140-5150. doi: 10.1021/acs.langmuir.8b04275. Epub 2019 Apr 4.
Sorption isotherm measurement is a standard method for characterizing porous materials. However, such isotherms are generally hysteretic, differing between condensation and evaporation. Quantitative measurement of pore diameter distributions requires proper identification of the mechanisms at play, a topic which has been and remains the subject of intensive studies. In this paper, we compare high-precision measurements of condensation and evaporation of helium in Vycor, a prototypical disordered porous glass, to a model incorporating mechanisms on the single pore level through a semimacroscopic description and collective effects through lattice simulations. Our experiment determines both the average of the fluid density through volumetric measurements and its spatial fluctuations through light scattering. We show that the model consistently accounts for the temperature dependence of the isotherm shape and of the optical signal over a wide temperature range as well as for the existence of thermally activated relaxation effects. This demonstrates that the evaporation mechanism evolves from pure invasion percolation from the sample's surfaces at the lowest temperature to percolation from bulk cavitated sites at larger temperatures. The model also shows that the experimental lack of optical signals during condensation does not imply that condensation is unaffected by network effects. In fact, these effects are strong enough to make most pores to fill at their equilibrium pressure, a situation deeply contrasting the behavior for isolated pores. This implies that, for disordered porous materials, the classical Barrett-Joyner-Halenda approach, when applied to the condensation branch using an extended version of the Kelvin equation, should properly measure the true pore diameter distribution. Our experimental results support this conclusion.
吸附等温线测量是表征多孔材料的标准方法。然而,这类等温线通常具有滞后性,在冷凝和蒸发过程中有所不同。孔径分布的定量测量需要正确识别起作用的机制,这一主题一直是并仍然是深入研究的对象。在本文中,我们将原型无序多孔玻璃Vycor中氦气冷凝和蒸发的高精度测量结果与一个模型进行了比较,该模型通过半宏观描述在单孔水平上纳入机制,并通过晶格模拟纳入集体效应。我们的实验既通过体积测量确定了流体密度的平均值,又通过光散射确定了其空间涨落。我们表明,该模型在很宽的温度范围内一致地解释了等温线形状和光信号的温度依赖性,以及热激活弛豫效应的存在。这表明蒸发机制从最低温度下样品表面的纯侵入渗流演变为较高温度下大量空化位点的渗流。该模型还表明,冷凝过程中实验上缺乏光信号并不意味着冷凝不受网络效应的影响。事实上,这些效应足够强,使得大多数孔在其平衡压力下填充,这种情况与孤立孔的行为形成了鲜明对比。这意味着,对于无序多孔材料,当使用开尔文方程的扩展版本应用于冷凝分支时,经典的巴雷特 - 乔伊纳 - 哈伦达方法应该能够正确测量真实的孔径分布。我们的实验结果支持这一结论。