The adsorption of light hydrocarbons (C2–C5 olefins and paraffins, toluene) on HZSM-5, silicalite, and HY was studied for application in treatment of exhaust streams of the petrochemical industry and of vehicles under cold start conditions. At this aim the trapping capability was evaluated on hydrated zeolites by breakthrough curves at low hydrocarbon partial pressure (0-1 kPa), in the temperature range 298–523 K and at space velocity of 30000 h−1. The basic adsorption properties of materials were also verified for three selected hydrocarbons (ethylene, isobutene, and toluene) by equilibrium isotherms on dehydrated zeolites at 298 K. The role of physicochemical characteristics of adsorbent materials was discussed in relation with their trapping capability of different types of hydrocarbons.
Adsorption techniques have been studied in the past several years as methods for separation or capture of hydrocarbons (HC) in applications that require the recovery of these compounds in high purity (petrochemical industry) [
The three basic technologies that have been developed for the control of the cold start HC emissions are the close-coupled catalyst, the electrically heated catalyst, and the HC traps. An extensive review of these solutions, with a description of the related patents, is reported in [
The most suitable materials for HC trapping are carbon-based materials, silicoaluminophosphates and zeolites [
Other investigations have been conducted in real conditions, using a zeolite based HC trap located at the exhaust of a gasoline engine [
The aim of the present paper was to study the adsorption properties of various zeolitic materials towards different HCs, taking into consideration those compounds which need to be recovered in petrochemical applications, such as light paraffins, or removed from the engine exhaust of light-duty gasoline vehicles, typically C2–C5 paraffins and olefins, plus monoring aromatic compounds [
All adsorption measurements presented in this paper were performed on HZSM-5, silicalite, and HY in powder form (Zeolyst International). The main characteristics of these zeolitic materials are reported in Table
Zeolitic materials used for adsorption tests.
Sample | Form | Si/Al | Pore opening, Å | Surface areaa, m2/g | Pore volumeb, cm3/g |
---|---|---|---|---|---|
HZSM-5 | Powder | 15 | 5.3 × 5.6 | 405 | 0.24 |
5.1 × 5.5 | |||||
|
|||||
Silicalite | Powder |
|
5.3 × 5.6 | 400 | 0.24 |
5.1 × 5.5 | |||||
|
|||||
HY | Powder | 15 | 7.4 | 730 | 0.51 |
bData from [
While petrochemical applications require the removal of light paraffins and olefins, the HCs present in engine exhaust gases comprise paraffins, olefins, and aromatics of various molecular weights in the range C1–C11. In Table
Composition and sources of hydrocarbon mixtures used for adsorption tests (provided by Praxair).
Mixture HC/N2 | Composition of individual HC, ppm | Total concentration, C3 ppm |
---|---|---|
Ethylene/N2 | 1280 | 853 |
|
||
Isobutene/N2 | 643 | 857 |
|
||
Toluene/N2 | 358 | 835 |
|
||
C4-C5 alkanesa/N2 | n-Butane: 123 | 847 |
Isobutane: 128 | ||
n-Pentane: 105 | ||
Isopentane: 102 | ||
Neopentane: 100 | ||
C5: 102 | ||
|
||
Isobutane-isobuteneb/N2 | Isobutane: 323 | 858 |
Isobutene: 320 |
bEquimolar mixture (as C3) of isobutane and isobutene.
Adsorption isotherms at room temperature were obtained by a gravimetric technique and performed with the apparatus sketched in Figure
Scheme of the experimental apparatus used for measurements of equilibrium isotherms by gravimetric tests.
The gravimetric tests were affected at room temperature for the three selected HCs (ethylene, isobutene, and toluene) in order to obtain a basic characterization of the three zeolitic materials in terms of equilibrium adsorption capacity towards compounds representative of the main HC families that need to be captured in the applications considered.
The breakthrough curves were performed by the experimental apparatus shown in Figure
Scheme of the experimental apparatus used for measurements of adsorption properties by flow microreactor.
An online analyzer equipped with flame ionisation detector (Siemens Fidamat 5 E-AP) was used for hydrocarbon concentration analysis before and after the adsorbent bed.
These tests were effected at different temperature values (from 298 K to TWC light off values) and low hydrocarbon partial pressures (0-1 kPa), which are usually encountered in engine exhaust gases.
The breakthrough curves were constructed reporting hydrocarbon concentrations measured at the outlet of the adsorbent bed as function of time on stream. These concentrations were acquired in terms of C3, since the analyzer calibration was affected by using propane mixtures (see Section
The zeolitic materials were tested in all flow adsorption runs without any preliminary thermal treatment, in order to simulate the utilization conditions of a HC trap at the engine exhaust. In this way each material retained its intracrystalline water content when it was exposed to hydrocarbon gas stream.
In Table
Equilibrium adsorption capacities (wt%) obtained at 298 K by gravimetric tests at saturation and 0.1 kPa for different HCs on HZSM-5, silicalite, and HY.
Zeolite | Ethylene | Isobutene | Toluene | |||
---|---|---|---|---|---|---|
Saturation | 0.1 kPa | Saturation | 0.1 kPa | Saturation | 0.1 kPa | |
HZSM-5 | 7.2 | 3.8 | 8.1 | 5.7 | 12.8 | 8.0 |
Silicalite | 6.8 | 2.3 | 11.5 | 8.7 | 13.8 | 7.7 |
HY | 5.4 | 1.6 | 22.5 | 14.8 | 17.1 | 7.1 |
Adsorption capacities (wt%) derived by breakthrough curves at various temperature values for different HCs on HZSM-5, silicalite, and HY. HC partial pressure: 0.1 kPa, space velocity: 30000 h−1.
Zeolite |
|
Ethylene | Isobutene | Toluene | Isobutene | C4-C5 paraffins |
---|---|---|---|---|---|---|
Isobutane | ||||||
HZSM-5 | 298 | 0.5 | 2.2 | 4.0 | 4.1 | 1.7 |
348 | — | 3.7 | 2.9 | 5.4 | 0.8 | |
373 | 0 | 4.5 | — | 5.4 | 0.2 | |
398 | — | — | 1.4 | — | — | |
413 | — | 4.6 | — | 4.0 | — | |
423 | — | — | 1.0 | — | — | |
463 | — | 2.4 | — | 1.6 | — | |
523 | — | 0.6 | — | — | — | |
|
||||||
Silicalite | 298 | 0.2 | 1.1 | 5.0 | 0.3 | |
373 | 0 | 0.5 | 2.5 | — | — | |
423 | 0 | 0.1 | 1.6 | — | — | |
|
||||||
HY | 298 | 0.04 | 2.8 | 6.8 | — | 0.7 |
348 | — | — | — | — | 0.2 | |
353 | — | 6.3 | — | — | — | |
373 | 0 | — | 5.2 | — | 0.1 | |
403 | — | 4.5 | — | — | — | |
423 | 0 | — | 2.1 | — | — |
In Figure
Adsorption isotherms at 298 K of ethylene (
Increasing pressure gave rise to a sharp weight increase for toluene, because of the capillary condensation phenomena occurring when the saturation vapor pressure of toluene (3 kPa at 298 K) was approached [
The breakthrough curves were used to calculate the cumulative values of adsorption capacity as a function of time (equation (
Cumulative adsorption of ethylene, isobutene, and toluene on HZSM-5 adsorbent bed versus time on stream at 298 K and 30,000 h−1.
In the hypothesis that adsorption of the three hydrocarbons on HZSM-5 at 298 K is essentially due to physical interaction between gas molecule and solid surface (van der Waals interaction) the low adsorption capacity exhibited by HZSM-5 for ethylene can be attributed to the absence in the molecule of any permanent dipole moment that implies a weaker contribution of coulombic forces to the overall van der Waals interaction [
In order to investigate the influence of Si/Al ratio on adsorption capacities, measurements on silicalite, a zeolite isostructural of ZSM-5 and Al-free, were performed. The comparison between adsorption isotherms of ethylene, isobutene, and toluene at 298 K on silicalite is shown in Figure
Adsorption isotherms at 298 K of ethylene (
The lower adsorption capacity observed in gravimetric tests for isobutene on HZSM-5 with respect to silicalite (Figures
In measurements affected under dynamic conditions (Figure
Cumulative adsorption of ethylene, isobutene, and toluene on silicalite adsorbent bed versus time on stream at 298 K and 30,000 h−1.
It is possible to hypothesize that the presence of water intracrystalline inside the zeolites tested by dynamic experiments hindered the oligomerization phenomena and made prevalent the effect of the adsorption sites number associated with the presence of Al in the MFI framework, determining higher adsorption capacities of isobutene on HZSM-5. Saturation uptake for toluene was comparable to that obtained on HZSM-5, but the process turned out slower, with bed saturation being reached after about 150 minutes. On the other hand, the adsorption rate during the first minutes was lower on silicalite with respect to HZSM-5 for both isobutene and toluene (the maximum slope did not exceed 0.11
The influence of structure type on adsorption capacities was analyzed by adsorption measurements on HY, a faujasite-type zeolite, with the same Si/Al ratio of ZSM-5. Adsorption isotherms of ethylene, isobutene, and toluene at 298 K on HY are reported in Figure for HZSM-5, having pore opening of 5-6 Å, a pore-mouth blocking occurs because of the production of isobutene oligomerisation, with the subsequent reduction of adsorption capacity compared to silicalite, as reported above; in the case of HY, having a pore opening (7.4 Å) larger than HZSM-5, oligomerisation process of isobutene proceeds in parallel with the adsorption and stops only when the free volume inside the zeolite is filled with the oligomer products.
The results of flow experiments on HY at 298 K are reported in Figure
Adsorption isotherms at 298 K of ethylene (
Cumulative adsorption of ethylene, isobutene and toluene on HY adsorbent bed versus time on stream at 298 K and 30,000 h−1.
The flow tests also evidenced that on HY the adsorption rate was lower than on HZSM-5 during the first 5 minutes and comparable to that observed on silicalite (i.e., the slope of toluene curve did not exceed 0.10
The above results evidence that the adsorption capacities calculated by breakthrough curves were always lower with respect to the values obtained by gravimetric tests. Taking into account that the zeolite microchannels contained intracrystalline water molecules during flow adsorption runs (see Section
The breakthrough curves were determined at various temperatures in the range 298–523 K in order to determine adsorption properties in conditions simulating the engine warm-up phase. As regards ethylene the already scarce adsorption observed at 298 K on HZSM-5 became negligible at 373 K confirming that this material is not suitable for trapping this type of hydrocarbon at low partial pressure. Further, no adsorption was detected for ethylene on HY and silicalite also at higher temperatures.
The adsorption of toluene on HZSM-5 as function of temperature is shown in Figure
Cumulative adsorption of toluene on HZSM-5 adsorbent bed versus time on stream at 30,000 h−1 and different temperature values.
The trapping capability progressively decreased raising the temperature up to 423 K, but an appreciable adsorption was measured also at the highest temperature investigated (1% at 423 K). The decrease of the uptake capacity when temperature is raised is expected, being adsorption of an exothermic process, and indicates that interaction between toluene and HZSM-5 at low partial pressure is of physical nature. Similar behaviors were observed for toluene on silicalite (Figure
Cumulative adsorption of toluene on silicalite adsorbent bed versus time on stream at 30,000 h−1 and different temperature values.
Cumulative adsorption of toluene on HY adsorbent bed versus time on stream at 30,000 h−1 and different temperature values.
Regarding the adsorption rate no significant effect of temperature was detected, in particular an almost unitary efficiency (slope close to 0.18
A particular behavior was observed for isobutene adsorption when temperature was increased. Increased uptakes were found on HZSM-5 (Figure
Cumulative adsorption of isobutene on HZSM-5 adsorbent bed versus time on stream at 30,000 h−1 and different temperature values.
Activated physical adsorption can occur in microporous solids when the width of the pores is very close to the diameter of the adsorbate molecule, since in this case the molecule encounters an energy barrier to its passage through the constriction [
As regards silicalite (Figure
Cumulative adsorption of isobutene on silicalite adsorbent bed versus time on stream at 30,000 h−1 and different temperature values.
Cumulative adsorption of isobutene on HY adsorbent bed versus time on stream at 30,000 h−1 and different temperature values.
The adsorption affinity of HZSM-5 towards paraffins is shown by curves of Figure
Cumulative adsorption of C4-C5 paraffin mixture on HZSM-5 adsorbent bed versus time on stream at 30,000 h−1 and different temperature values. Reactor feed: n-butane, isobutane, n-pentane, isopentane, and neopentane equimolar mixture.
Cumulative adsorption of C4-C5 paraffin mixture on HY adsorbent bed versus time on stream at 30,000 h−1 and different temperature values. Reactor feed: n-butane, isobutane, n-pentane, isopentane, and neopentane equimolar mixture.
The role of olefin double bond and the low affinity towards paraffins was confirmed by the curves shown in Figure
Cumulative adsorption of equimolar isobutane/isobutene mixture on HZSM-5 adsorbent bed versus time on stream at 30,000 h−1 and different temperature values.
Three zeolitic materials (HZSM-5, silicalite, and HY) were studied as adsorbents for light hydrocarbon traps. Isotherm gravimetric tests at room temperature were performed up to saturation pressure for ethylene, isobutene, and toluene, while dynamic tests in flow microreactor were carried out at low partial pressure (0.1 kPa) and different temperature values (298–523 K) on the same three hydrocarbons plus two mixtures of C4-C5 paraffins and isobutane/isobutene.
The three adsorbents showed high affinity at room temperature for toluene and isobutene, but minor affinity for paraffins and ethylene. In particular the adsorption of ethylene was appreciable on HZSM-5, but resulted negligible for silicalite and HY. Equilibrium adsorption capacities obtained with breakthrough curves at 0.1 kPa resulted in being very lower with respect to values reachable by gravimetric tests at the same partial pressure, due to the presence of intracrystalline water in zeolite samples.
Gravimetric tests and breakthrough curves evidenced chemisorption phenomena as regards isobutene adsorption on Al containing materials (HZSM-5 and HY), with possible formation of oligomeric species inside zeolite microchannels.
Increasing temperature from 298 up to 423 K adsorption capacity measured by dynamic tests for toluene on the three materials decreased while it became negligible for ethylene already at 373 K also on HZSM-5. As regarding isobutene a maximum in adsorption capacity was reached for HZSM-5 at 413 K and for HY at 353 K. While the results obtained for ethylene and toluene could be interpreted in terms of physical adsorption, a chemical adsorption contribution was invoked to explain isobutene uptake increment with temperature.
Experiments affected with a C4-C5 paraffin mixture evidenced the minor affinity of the three materials towards saturated hydrocarbons. Breakthrough curves performed with isobutene-isobutane equimolar mixture evidenced the role of C=C double bond on adsorption capacity of HZSM-5 towards hydrocarbons and the possible occurring of alkylation reactions.
The authors declare that there is no conflict of interests regarding the publication of this paper.