Andrew Knight Rob Hauser Joe Oshier
Chem 4411L
T.A. Darrin Bellert
The process of adsorption and desorption is studied on the compounds of alumina and silica during this experiment. The isotherms that display the behavior of N2 upon these compounds are represented as are the pertinent results that can calculated from them. From the BET theory of adsorption, it was possible to calculate the surface area of the adsorbent and the DHads for N2 for the adsorbent. The values differed slightly in the means by which they were calculated (adsorption or desorption). The value for the surface area of alumina is 4.96E+06 +/- 3.04E+05 m2/g (95% CI). For silica the experimental surface area is 6.51E+06 +/- 1.29E+06 m2/g . The value for the DHads (N2) for alumina is -5.58E+03 +/- 1.44E+02 Joules/mol and the DHads (N2) for silica -5.61E+03 +/- 2.61E+01 Joules/mol.
The purpose of this laboratory experiment is to study the process of adsorption. The process of adsorption should be set apart from the process of absorption. In adsorption, molecules of the adsorbate are binded to the surface whereas in absorption, the their just filling the spaces of the pores in the solid. This process of binding is generally weak and reversible (as seen in immediate desorption). Some of the best known and classic adsorbents are silica, alumina, and activated charcoal. Two of these compounds are used in the experiment. The reason for their adsorbing characteristics are their enormous surface area per unit weight. When the surface of the adsorbent is saturated by the adsorbate, a decrease in adsorbence will be observed. This is due to the limited number of surface sites available for chemisorption. Although, adsorption could occur beyond the initial monolayer of adsorbate according to BET theory. The measurement of adsorption is usually carried out at a constant temperature (77K for this experiment). The gas generally used for this is Nitrogen. However, argon and krypton are used in special cases. The opposite process is called desorption. The sample must be saturated with the gas before an accurate desorption isotherm can be constructed. The path of the desorption isotherm may be different from that of the adsorption isotherm. This may be due to hystersis effects. The area of the adsorbent can be calculated from the isotherms. Different values corresponding to this are probably due to the effects previously mentioned.
The measurements were taken from combinations of the ideal gas laws and by variations in calculated values. An instrument known as the Omnisorb 360 was used for the experiment. This instrument consists of vacuum pumps and "plumbing" along with the sample containers. Gas expansions throughout the "plumbing" and sample containers as well as a known flask gave the needed volumes of each necessary component. The gas He was used during volume determinations because it is not adsorbed at this temperature. The variations upon the gas expansions were due to either adsorption or desorption which ever was relevant. The manifold of the Omnisorb was filled with a certain pressure of N2 and expanded into the sample container. Some of the gas was adsorbed by the sample. After the sample was saturated with the gas, desorption runs could take place. In the desorption runs, the gas from the sample container was expanded into the manifold. As before, the difference from expected values were due to the gas being desorbed. Seen in figure 3 is the "plumbing" and sample containers of the instrument. The sample container that was used was submerged in liquid N2. This correspond to circle 5 in figure 2.
Figure 1- Complete view of the Omnisorb 360.
The adsorbents used for this experiment were alumina and silica. The adsorbate used was N2. The first step in the experiment was to determine the necessary volumes of the relevant plumbing and containers. The table below summarizes the results from these determinations. The process used for these calculations was Boyle's Law for ideal gases.Table 1 - Determinations of the average volumes of the relevant pieces of the instrument and containers. Al subscript denotes Alumina and Si subscript denotes Silica.
Section Average Std. Volume Dev. (mL) (mL) Manifold 26.068 0.077 Container 27.636 0.147 Al Container 28.652 0.224 Si
From the processes described in the introduction and method sections, an isotherm for both the adsorption and desorption processes were constructed for both adsorbents. The complete data for Alumina will be presented first.
Figure 4 - An Isotherm constructed from the data acquired for the Adsorption run of Alumina (series1). The desorption data is plotted as series2.
Figure 5 - The BET treatment of the Adsorption for Alumina.
Summary of Adsorption and Desorption on values using the BET theory
c Std. nmono Std. DHads Std. Specific Std. Dev. Dev. Dev. Area Dev (mol) (mol) (Joules/mol) (Joules/mol)(m2/g) (m2/g) Alumina 69.25 6.98 4.24E-03 2.31E-04 -5505.87 64.55 5.11E+06 2.81E+05 Ad Alumina 87.09 13.58 3.95E-03 3.61E-04 -5652.58 99.80 4.80E+06 4.39E+05 De Silica 79.62 73.99 1.79E-03 6.26E-04 -5595.23 594.89 5.85E+06 2.05E+06 Ad Silica 83.00 10.31 2.19E-03 1.16E-04 -5621.83 79.51 7.17E+06 3.79E+05 De
The values in table 2 were calculated using the BET theory. For this treatment to be valid, a few assumptions have to be made. The assumptions for this are:
With these assumptions, the data in the table has the corresponding values. There is a slight difference in the adsorption and desorption runs. This is due to the differences seen in the adsorption and desorption isotherms. The main reason for this difference may be very slight hysteresis effects. For alumina there is not much evidence of this, but the path is more pronounced in the silica isotherm. A large std. dev is associated with the Adsorption BET treatments with leads to high uncertainties in the results. A significant lower error is in the desorption runs, but the results agree. This is evident in the area calculations in which the adsorption areas are greater than the desorption areas. A reason for this is that the pressure taken for the desorption runs was not fully at equilibrium. The estimate used by the group may have been slightly off. An important note that must be made when dealing with BET theory is that it only valid for a specific range of (p/p0) or z values. The values are reflected in the linear plots included above. The region of validity must be linear for the BET theory. This is an intermediate region and the theory fails at low and high pressures. An alternative method to approach this adsorption problem is to use the Langmuir theory. The Langmuir theory differs from the BET theory slightly. The assumptions made in the Langmuir theory are as follows:
The ability of a molecule to adsorb at a given site is independent of the occupation of neighboring sites.
Figure 10 - The Langmuir isotherm for Alumina is linear through low and moderate surface coverage. But as the pressure increases, and the coverage is high, the line deviates from linear.
It is seen from the results obtained in this experiment that adsorption is a unique and useful process. One, for it can be used to determine the surface area of small particle when no other means is feasible or even possible. Also this experiment shows that the reason why the classic adsorbents are such, there high surface to mass ratio. The phenomenon of adsorption is put to practical use in everyday life. It is the process of adsorption that is used to clean the otherwise dirty water that is used by the great cities of the world everyday. Activated charcoal is used for this grungy process based on the facts that have been studied today. Surprisingly, a little bit a charcoal can clean a lot of water as seen in the gas study of this experiment. Another use of the property, is in gas masks. W.W.I gas masks, and gas masks of today, purify the air through adsorption. This shows that there may be some bit a practicality to what we study after all. Also, it shows that there may be more than one way, or a combination of ways to get what you need.
Atkins, P.W.; Physical Chemistry; 5th ed. W.H. Freeman and Co. New York, NY.; 1994.
Drake ,J.M.; Nitrogen Adsorption on Porous Silica: Model-Dependent Analysis; Journal of Physical
Chemistry; v.98, 1994, p. 380-2.
Williams, Kathryn; Statistical Analysis of Experimental Data. , University of Florida. 1995.
Williams, Kathryn; Laboratoray Instructions for Physical Chemistry. , University of Florida. 1995.