Introduction
Addressing climate change requires significant efforts to curtail
greenhouse gas emissions, in particular anthropogenic
CO2 as well as acidic SO2 and NO, to
keep the average increase in global temperature below 2 °C and to
achieve negative emissions by the year 2100. Although a significant
amount of recent research has developed innovative technologies for
SO2, NO, and CO2 capture from
large-point sources, such as affluent post-combustion
streams,1–3 the amine scrubbing method, which is
plagued by inherent drawbacks including poor recyclability,
deterioration of equipment, energy requirements as high as 4.2 GJ/tonne
CO2, and poor performance in the presence of
SO2 or NO,4–7 remains the benchmark
technology. In that regard, developing cost-effective and more efficient
capture strategies is essential to addressing the impending issue of
climate change.
In recent years, adsorption, a technique wherein target gases are
physically (physisorption) or chemically (chemisorption) adhered to a
selective medium, has cemented itself as an alternative to amine
scrubbing technology for CO2capture.8,9 To bring adsorptive CO2capture technology one step closer to large-scale implementation,
adsorbent materials are required to be shaped into practical contactors
that not only demonstrate low mass and heat transfer resistances but
also address the operational challenges such as attrition and pressure
drop. Honeycomb monoliths with uniform channels and thin walls are
considered suitable configurations for this purpose. To date various
CO2 adsorbents including zeolite 13X, activated carbon,
and metal-organic frameworks (MOFs) have been shaped into monolithic
structures and their CO2 capture performance has been
investigated.10–12
While much is known about the performance of CO2adsorbents under binary feed conditions (dry
CO2/N2 feed), the multicomponent data is
quite scarce, especially under realistic feed conditions, where trace
contaminants like SOx and NOx are
present. To date, most studies have focused solely on examining the
zeolite stability in the presence of these species, or have focused only
on equilibrium adsorption of each species, but have not compared clean
and contaminated feeds to one another under dynamic
conditions.13,14 For example, Deng et al.13 showed that zeolite 13X has a high affinity towards
SO2 on account of its dipole moment, and predicted that
this would lead to reduced CO2 adsorption in
multicomponent streams, however, it has yet to be studied how the
presence of multiple species impacts CO2 adsorption on
zeolite 13X. Moreover, these effects are likely to be further
complicated by the presence of water, which is known to co-adsorb
alongside CO2 under clean-feed conditions and reduce the
adsorption capacity. For example, our recent study12demonstrated that pre-humidifying zeolite 13X monoliths before exposure
to CO2 leads to competitive adsorption between the two
gases and a reduction in CO2 uptake of 60-80%. Other
studies have demonstrated similar effects of humidity on
CO2 adsorption in zeolite 13X materials,15,16 however, no study yet exists which isolates the
combined influences of SOx, NOx, and
H2O on CO2 adsorption in zeolite 13X.
Aiming at addressing this question, we embarked on a study to
systematically investigate the CO2 adsorptive
performance of zeolite 13X in both honeycomb and bead forms under
various simulated flue gas conditions. We primarily addressed relative
CO2 adsorption changes imparted by
competitive/cooperative adsorption of flue gas impurities. Two
self-standing commercial zeolite 13X monoliths, with 600 and 800 cells
per square inch (cpsi) cell density and beads with particle size of 1.6
mm were used in this work. Multicomponent breakthrough runs were
conducted with a simulated flue gas consisting of
10%CO2/1000 ppm SO2/1000 ppm NO/balance
with N2 under dry and humid conditions and the effect of
feed conditions on CO2 capture performance of 13X
samples were investigated accordingly.
Experimental Section
Materials Synthesis and Characterization
All ultra-high pure and mixed gases were purchased from Airgas. The
crystallinity of zeolite samples was investigated by X-ray diffraction
(XRD) measurements on a PANanalytical X’Pert multipurpose X-ray
diffractometer with a scan step size of 0.02°/step at the rate of 147.4
s/step. The textural properties were evaluated by N2physisorption isotherms at 77 K on a Micromeritics (3Flex) instrument.
Prior to the measurements, the adsorbents were degassed under vacuum for
6 h at 350 ºC on a Micromeritics Smart VacPrep system. The
Brunauer-Emmet-Teller (BET) and non-local density functional theory
(NLDFT) methods were used to estimate the surface area and pore size
distribution (PSD), respectively. Thermogravimetric analysis (TGA)
experiments were carried out to determine the amount of zeolite in each
sample. Therein, the temperature was ramped at 15 °C/min from 25 °C-700
°C in 60 mL/min air.
CO2 and N2 Adsorption Isotherm
Measurements
Pure gas adsorption isotherms for CO2 and
N2 were obtained on a volumetric gas analyzer (3Flex) at
25 ºC. Prior to measurement, the samples were degassed using the
conditions from N2 physisorption. The isotherms were
then used to estimate the ideal selectivity values using ideal-adsorbed
solution theory (IAST). Additionally, TGA was also used to measure the
CO2 uptake capacity of fresh samples and used samples
(after breakthrough tests) using a 10%
CO2/N2 gas mixture at 25 ºC and flow
rate of 60 mL/min. Before the uptake runs, samples were degassed at 350
ºC under N2 with a flow rate of 40 mL/min for 1 hr.
Multicomponent Breakthrough Tests
The breakthrough experiments were performed using the setup described in
our recent work. 12 The multicomponent adsorption
tests were carried out at 1 bar and 25°C in a stainless-steel
breakthrough column with an inner diameter of 3.2 cm and a length of 15
cm. Four experimental conditions were investigated: dry-clean,
humid-clean, dry-contaminated, humid-contaminated. Under dry conditions,
10% CO2/N2 (dry-clean) or
10%CO2/1000 ppm SO2/1000 ppm NO/
N2 (dry-contaminated) gas mixtures were sent through the
bed with a flow rate of 100 mL/min whereas, under humid conditions, the
sample was first pre-humidified by sending a water–saturated He flow at
100 mL/min to the bed for 30 min. The He stream was humidified by
passing the gas through a bubbler before entering the column. Before the
breakthrough runs, the adsorbent was regenerated in-situ at 350
°C for 3-6 h under He flow at 100 mL/min and then cooled down to 25 °C
to start the test. The concentration of gas mixture in the outlet stream
was measured using a mass spectrometer (BELMass). Table 1summarizes the geometric characteristics of the adsorbents for the three
samples. Notably, the weights were not held constant across sample runs.
This was necessary to completely pack the bed. Because the samples
contained different infill percentages, the amount necessary to achieve
7.5 cm of bed packing changed between samples.
Table 1. Characteristics of the adsorption column for three
samples.