Studies on the adsorption of gases on iron-kieselguhr powder

Abstract

The main aspects of the study on the adsorption of argon, nitrogen, carbon monoxide, carbon dioxide and hydrogen have been carried out on iron?kieselguhr powders in the temperature range of ?191 to 450°C. They are summarised below:

The iron?kieselguhr samples were prepared by the method of impregnation and also precipitation. In all about ten samples were prepared in the present study. In the impregnation method, the kieselguhr was impregnated with a solution of ferric nitrate and the nitrate on kieselguhr was decomposed to the oxide at 400–450°C. The metal on kieselguhr was then obtained by reduction of the oxide on kieselguhr in a stream of hydrogen at 450°C. In the precipitation method, the hydroxide of the metal was precipitated in the presence of kieselguhr by the addition of ammonia. After filtration and washing, the wet mass was dried and the metal oxide reduced to the metal by hydrogen at 450°C.

The surface areas of the various unreduced and the reduced catalysts and also the supports were determined by measurement of the adsorption of nitrogen, argon, and carbon monoxide at ?191°C and of carbon dioxide at ?70°C and application of the BET equation to the data obtained. The surface areas of the various samples prepared by different methods were compared on the basis of the nitrogen areas. These studies have shown that with impregnated catalysts there will be a decrease in the surface area of kieselguhr, whereas for the precipitated catalysts, there will be an increase in the area. Further, the surface area of an impregnated catalyst increases on reduction due to channelling and formation of cracks whereas the area of a precipitated catalyst decreases due to sintering.

The surface areas of the iron?kieselguhr samples determined by employing nitrogen and argon have been compared. It has been shown that the higher values for the area obtained by nitrogen adsorption is not due to the chemisorption of nitrogen on iron at ?191°C, but is due to the differences in the adsorbent–adsorbate interactions. It is suggested that nitrogen undergoes greater polarisation than argon near the surface of a metal resulting in a closer packing of the adsorbate molecules or larger adsorption.

The surface areas obtained by the application of the Harkins and Jura equation to the adsorption of nitrogen compare very well with the areas obtained by the BET method for several samples. However, differences of 20–30% between the two methods have been noticed with some samples, which cannot be explained easily.

The differences in the adsorption isotherms of CO and H? on iron?kieselguhr samples represents the chemisorption of CO and gives an estimate of the fraction of surface occupied by metal atoms. It has been shown that these values do not differ to a large extent from the values obtained by taking into consideration the suppression of the physical adsorption of CO by a chemisorbed layer of the same gas (CO).

Techniques like differential thermal analysis and electron microscopy were employed to study the properties of the unreduced iron?kieselguhr samples. These studies have shown that the fine pore structure of natural kieselguhr is completely lost on its impregnation with ferric nitrate solution. Further the precipitation method employed in the preparation of some ferric oxide?kieselguhr samples does not give rise to any iron hydro?silicates as suggested by earlier workers.

The rate of adsorption of hydrogen has been measured at different temperatures under conditions of constant pressure and constant volume. The kinetic data have been examined in the light of equations proposed from time to time. The Elovich equation represents the kinetic data over a considerable range of surface coverage. Elovich plots reveal multiple kinetic stages and explanations have been offered to account for these.

The effect of temperature and pressure on the Elovich constants revealed certain interesting results. The plot of the Elovich constant (?) against the initial pressure showed a break around 15 cm Hg. In the pressure range beyond 15 cm Hg, ? varied relatively slowly with the initial pressure. The adsorption–initial pressure plots also indicated a break at 15 cm Hg, indicating a change from one type of adsorption sites to another. The temperature coefficients of ? and log ? are found to be either positive or negative depending on the iron?kieselguhr sample employed and other experimental conditions—constant pressure and constant volume.

The kinetic data can also be represented by the application of the equation proposed by Polanyi and co?workers. It has been shown that the Elovich equation and the equation of Polanyi et al. reduce to the same form after expansion, which means that both equations are equally applicable for any gas–solid system but in view of the various mechanisms proposed for the Elovich equation, the latter should be preferred.

The power rate law of Kwan could not be applied to the present experimental data. This is partly due to the uncertainty involved in deriving momentary rates, which are necessary to apply the power rate law.

The various mechanisms put forward to explain the slow adsorption of hydrogen (the rate of which follows the Elovich equation) have been classified into those based on the concept of site generation followed by site decay and those based on surface heterogeneity. A suitable mechanism has been proposed to explain the kinetics of hydrogen adsorption on iron?kieselguhr.

Adsorption isotherms of hydrogen have been determined on iron?kieselguhr in the temperature range of ?191 to 350°C. At ?191°C, there was considerable adsorption on both kieselguhr and on iron?kieselguhr. It was noticed that the hydrogen adsorbed at ?191°C could be completely desorbed at the same temperature with two samples of IK–Kg (II and IV), whereas with another sample, IK–Kg I, about 1.5 c.c. of hydrogen was retained at ?191°C. Hydrogen retained by IK–Kg I at ?191°C may be chemisorbed on the small iron particles situated in the pores of the kieselguhr. The adsorption on IK–Kg II and IV at ?191°C is mostly due to physical adsorption. In order to get an idea about the distribution of adsorbed hydrogen between the metal and the support, it is convenient to employ the ratio (VCO ? VH?) / VT which gives an estimate of the fraction of surface occupied by the metal.

At temperatures of ?70°C and above, there is no adsorption of hydrogen on the kieselguhr samples, so that any adsorption noticed with iron?kieselguhr samples must only take place on the iron surface. There is considerable adsorption of hydrogen at ?70°C and 0°C on the samples IK–Kg I and IV, whereas the adsorption on the low surface area sample IK–Kg V is negligible.

At temperatures of 97°C and above, the retention of hydrogen is determined with a Töpler pump on the samples IK–Kg I and IV. The H??adsorption isotherm on the powder having adsorbed hydrogen has been determined at all the temperatures studied. Based on the studies on retention and readsorption, the adsorption of hydrogen on IK–Kg I and IV above 97°C can be classified into groups: (1) Irreversibly chemisorbed hydrogen which is not desorbed at the temperature of adsorption and (2) Reversibly chemisorbed hydrogen which comes off when evacuated at the temperature of adsorption.

The effect of temperature on the adsorption of hydrogen on iron?kieselguhr is different from that noticed with other iron catalysts. The adsorption isobar of hydrogen on iron?kieselguhr shows a minimum at 0°C and an ascending region in the temperature range of 100 to 350°C, whereas on pure iron powder and other iron catalysts, the adsorption maximum has been reported around 100 or 200°C. The anomalous behaviour of the iron?kieselguhr system has been attributed to the presence of the support, which can influence the adsorption on account of the following reasons: (1) Because of the distribution of metal on kieselguhr, the physical condition of the metal may correspond more to a film than to a cube catalyst. (2) The support may give rise to an interface between the metal and the support and hydrogen can slowly enter into these portions only at higher temperatures. (3) There may be a slow diffusion or solution of hydrogen at the metal surface which always increases with temperature.

With increase in temperature above 97°C, there is an increase in the retention and readsorption of hydrogen on iron?kieselguhr. The increase in retention with temperature shows it to be an activated process and this has been attributed to an increase in the number of active sites with temperature.

The adsorption of carbon monoxide has been investigated in the temperature range of ?191°C to 450°C on iron?kieselguhr powders. At ?191°C, the carbon monoxide is chemisorbed at the iron surface on all the iron?kieselguhr samples.

At 0°C and 25°C, the iron?kieselguhr samples prepared from the high surface area kieselguhr did not form any iron carbonyl (as with pure iron powder investigated in this laboratory), whereas the low surface area iron?kieselguhr gave small amounts of iron carbonyl. The composition of the iron carbonyl could not be determined on account of the minute amounts formed. The difference in the behaviour of iron powder and iron?kieselguhr has again been attributed to the presence of the kieselguhr support.

In the temperature range of 0 to 97°C, there was adsorption of carbon monoxide on the high surface samples IK–Kg I and IV and therefore adsorption isotherms were determined at this temperature. The calculation of the differential heats of adsorption in this temperature range showed that the adsorption of carbon monoxide was mostly physical in nature.

The catalytic conversion of carbon monoxide to carbon dioxide has been noticed at temperatures of 200°C and above. It was found that the rate of catalytic conversion increased with increasing temperature. The solid products formed during this reaction between carbon monoxide and iron?kieselguhr are iron carbide, iron oxide (Fe?O?) and free carbon (in small amounts).

At 400°C and above, it was found that the iron? kieselguhr powder took up large quantities of carbon monoxide without producing any gaseous products of reaction. When the rate of uptake of carbon monoxide became very slow, carbon dioxide was detected as one of the products in the gas phase. At these temperatures, iron oxide (Fe?O?) and free carbon formed in large amounts, and iron carbide was also one of the solid products. A mechanism has been proposed for the reaction of carbon monoxide with iron?kieselguhr in this temperature range.

The adsorption of carbon dioxide has been studied on iron?kieselguhr in the temperature range of ?78°C to 450°C. A slow uptake of carbon dioxide decreasing with increase in temperature has been noticed up to 200°C. At temperatures of 300°C and above, there is a continuous uptake of CO? by the iron?kieselguhr sample resulting in the formation of iron carbide, iron oxide, free carbon and different amounts of carbon monoxide. The formation of these products has been explained on the basis of the dissociation of carbon dioxide to carbon monoxide and oxygen at higher temperatures.

Adsorption isotherms of nitrogen have been determined on the iron?kieselguhr samples in the temperature range of 200 to 450°C. It is noticed that the adsorption decreases with temperature up to 350°C, above which it increases. The ascending region in the isobar above 350°C is due to an activated adsorption of nitrogen and involves probably a partial dissociation of the nitrogen molecule to atoms at the catalyst surface.