(377i) Single and Binary Component Adsorption of 1-Alcohols from an Alkane Using Various Activated Alumina Adsorbents

Introduction

Surfactants are
compounds used to reduce surface tension between two or more immiscible mediums.
Surfactants usually consist of a hydrophobic backbone and a hydrophilic tail.
The first step in the production of surfactants is to graft a hydroxyl
functional group onto an alkane resulting in an alcohol molecule. Complete
conversion is not achieved which brings about the necessity of alcohol-alkane
separation to enhance efficiency of surfactant production. Distillation and
other techniques are used to separate these materials; however, some alcohol
contaminants remain in the alkane stream after distillation impeding recycling
of the alkane stream. The purification of an alkane stream through the process
of adsorption with activated alumina has been shown to be a technically viable
process [1]. A study by Groenewald investigated the single component adsorption
of 1-alcohols from an alkane using activated alumina. There is however little
knowledge on the binary component adsorption behaviour of an alcohol-alkane
system using activated alumina as adsorbent.

The aim of this study
is therefore to extend the knowledge on the single component adsorption and
extend the study to binary component adsorption of alcohol contaminants from an
alkane stream through (i) the investigation of single component adsorption
behaviour of alcohols (1-hexanol, 1-octanol and 1-decanol) from an alkane
(n-decane) using various activated alumina adsorbents in a wider temperature
range; (ii) the investigation of binary adsorption behaviour of alcohols
(1-hexanol, 1-octanol, 1-decanol) from an alkane (n-decane) using various
activated alumina adsorbents; and, (iii) the comparison of single and binary
component adsorption behaviour.

Methodology

Adsorption data was
obtained using a bench-scale batch adsorption system. The set-up consists of beakers
mounted inside a water bath at constant temperature. Beakers containing a
previously determined quantity of alcohol-alkane solutions were immersed in the
water bath and allowed to heat up to the specified temperature. A known amount
of activated alumina adsorbent (Activated Alumina F220, Selexsorb CDx® or
Selexsorb CD®) was then added to the solutions and intermittently sampled. The alcohol-alkane
solutions comprised of n-decane with a known concentration of either 1-hexanol,
1-octanol or 1‑decanol, or a binary mixture of two of the mentioned
alcohols. The initial alcohol concentration was in the range of 0.3 to 3 mass%
and was varied to determine the effect of initial adsorbate concentration. The
effect of temperature was investigated by varying the temperature from 25^oC
to 45^oC and comparing the adsorption behaviour at the different
temperatures. In addition, the effect of alcohol chain length in a single
component system, and combination of different alcohols in a binary component
system were also investigated. The effect of the different variables was
investigated based on adsorption kinetic behaviour as well as adsorption
equilibrium behaviour.

Results
and Discussion

The three different
adsorbents that were investigated exhibited relatively similar adsorption
behaviour both with regards to kinetics as well as equilibrium, with Selexsorb
CDx® and Selexsorb CD® achieving slightly greater removal efficiencies (Figure
1).

Figure 1:: Concentration decay plots for
the adsorption of 1-hexanol (initial concentration of 1 mass%) onto various
activated alumina adsorbents, at 45^oC

Selexsorb CDx® and
Selexsorb CD® achieving slightly greater removal efficiencies was to be expected
as the only major difference between the three adsorbents is that Selexsorb
CDx® and Selexsorb CD® contains micropores while Activated Alumina F220 does not
contain micropores. Selexsorb CDx® and Selexsorb CD® had micropore volumes of 0.080 and 0.046 cm³/g respectively, as
determined by Brunauer-Emmett-Teller (BET) surface area analysis. Activated
Alumina F220, Selexsorb CDx® and Selexsorb CD® had BET surface areas of
approximately 334.34, 466.01 and 408.28 m²/g, respectively. The
adsorption capacities of the various adsorbents are depicted in Figure 2.

Figure 2: Equilibrium adsorption capacity
(total alcohol) for different adsorbate systems with a total initial alcohol
concentration of approximately 1 mass%, at 45^oC

Temperature had a
positive effect on the adsorption kinetics of all systems investigated, thereby
increasing the adsorption rate with an increase in temperature. This is
expected and is widely reported in various adsorption studies. Temperature did
not prove to have a notable effect on equilibrium adsorption capacity with only
a few cases exhibiting a slight increase in the capacity with an increase in
temperature. This could be attributed to a possible combination of chemisorption
and physisorption.

Initial alcohol
concentration and alcohol chain length had no noticeable effect on the kinetic
behaviour of the systems for the ranges measured here. Both parameters did
however prove to be significant in the equilibrium behaviour. In the range investigated,
initial alcohol concentration had the greatest effect on equilibrium adsorption
capacity. For an increase in initial alcohol concentration from approximately
0.3 to 1 mass%, equilibrium adsorption capacity increased notably. However, for
initial concentrations greater than 1 mass% the capacity remained relatively
constant and it is believed that the saturation capacity of the adsorbent was reached.
Since the chain lengths of the different alcohols investigated were relatively
close to one another, there was at times a slight decrease in equilibrium
adsorption capacity for the longer chain length alcohols in the single
component systems. In the binary component systems, the alcohols almost adsorbed
in the same ratio as initially in the solution, with all three adsorbents
exhibiting a slight selectivity for the shorter chain alcohol.

The kinetics of
alcohols in both a single component system as well as a binary component system
were investigated and compared. It was found that the rate of adsorption of a
single component remains relatively unchanged in a binary component system when
compared to a single component system. The initial alcohol uptake in the single
component system however was notably faster than in the binary component system
and may be ascribed to possible adsorbate-adsorbate interaction.

Similar to the
adsorption kinetics, the equilibrium behaviour of the single and binary
component systems was also investigated and compared. The most significant
difference between the single and binary component systems was found to be the
equilibrium adsorption capacity. For equal initial concentrations of an
alcohol, the uptake of that specific alcohol in a single component system
proved to be significantly greater than that of the same alcohol in a mixture
of two different alcohols (Figure 3).  This suggested antagonistic
adsorbate-adsorbate interaction.

Figure 3: Comparison of the kinetic
adsorption behaviour of 1-hexanol from n-decane in a single component system
and in a binary component system consisting of 1-hexanol and 1-decanol (50:50
by mass), onto Selexsorb CDx® at 45^oC (initial concentration for
1-hexanol alone in both the single and binary component systems was 1 mass%)

Conclusions

In conclusion, adsorption
by activated alumina was shown to be a viable solution for the purification of
alkane streams in the surfactant production process. The kinetic and
equilibrium behaviour of both single and binary component adsorption systems
were investigated, and it was found that initial alcohol concentration had the
greatest effect on the adsorption behaviour within the alcohol concentration
range investigated, with antagonistic adsorbate-adsorbate interaction present
in the binary alcohol systems.

The next step in the
study would be to model the adsorption kinetics and equilibrium behaviour by
use of kinetic models such as the pseudo-second-order model and the
intra-particle diffusion model, and equilibrium isotherm models such as the
single and multicomponent Langmuir and Freundlich isotherms. This will allow
for further investigation of the adsorption behaviour of these systems.

References

[1]       J.
Groenewald, “Evaluation and Comparison of three industrially relevant
adsorbents of their ability to remove alcohol contaminants from an alkane
solvent,” University of Stellenbosch, 2019.