2012 AIChE Annual Meeting

(178r) Probing the Adsorptive Behaviour of MIL-53(Al) Using Light Organics (C1-C4)

Authors

Lyubchyk, A. I., REQUIMTE/CQFB, Universidade Nova de Lisboa
Esteves, I. A. A. C., Universidade Nova de Lisboa


Probing the Adsorptive Behaviour
of MIL-53(Al) using light organics (C1-C4)

Fernando J.A.L. Cruz, Andriy
I. Lyubchyk, Isabel A.A.C. Esteves, Jos? P.B. Mota*

Requimte/CQFB, Dept. Chemistry, Universidade
Nova de Lisboa, 2829-516 Caparica,
Portugal.

fj.cruz@fct.unl.pt, pmota@fct.unl.pt


I. Introduction

Metal-organic
frameworks (MOFs)1 are three-dimensional
hybrid materials, formed by the coordination of metal ions with organic
linkers. They combine a high pore volume, a regular and geometrical reproducible
porosity, and the presence of tunable organic groups within the framework, thus
allowing an easy manipulation of the pore size.
Among this new
class of porous materials, the flexible MIL-53(Al, Cr)solids arise as particularly interesting substances, due to their
chemically "simple" structure and unique adsorption features.
Structurally, MIL-53(Al) is composed of chains of dicarboxylate
groups interconnected with AlO4(OH)2octahedra
as corner-sharing units. This 3D metal-organic framework contains 1D
diamond-shaped channels with pores of nanometer dimensions, and possess a
chemical formula of Al(OH)(O2-C-C6H4-CO2),
where Al denotes the trivalent cation. It has been
reported4,5 that the MIL-53 solids exhibit
a lattice breathing phenomenon over a certain temperature range, upon
adsorption of special molecules (e.g.,
H2O and CO2) that interact with the solid via hydrogen
bonding. This association between the adsorbate molecules and the framework
being responsible for the switching between a narrow-pore structure (np), in which the pores are slightly deformed due to
hydrogen bonding, and a large-pore form (lp),
characterized by a more open porosity. In the present work, the low- to
high-occupancy adsorption thermodynamics of light alkanes (C1-C4)
in the large-pore structure of MIL-53(Al) is predicted
from grand canonical Monte Carlo (GCMC) simulations and compared with
experimental gravimetric data. The adsorption experiments span a broad range of
pressures (0.01-7 MPa) and temperatures (303-353 K).
In our molecular simulation work, MIL-53lp(Al) is
assumed to have a perfect, rigid lattice, and both fluid-fluid and solid-fluid
interactions are modeled using the TraPPE-UA force
field.

II. Results and Discussion

The solid sample had been previously
characterized6 by standard physico-chemical techniques, namely
elemental analysis (C, H, N), powder XRD (λ = 1.5418 ), TGA, FTIR, and
solid state NMR (1H,13C, 27Al),
indicating a chemically pure substance (C = 43.38%, H = 2.25%) with a dry
structure identical to the one proposed in the literature.2 The
adsorption isotherms predicted by GCMC simulation, without any reparameterization of the TraPPE-UA
force field parameters, are in good agreement with the experimental
measurements within the whole temperature domain (Fig.1); particularly in the
low-medium pressure range where simulation results coincide almost exactly with
experimental data. For the C2-C4 molecules, the amount of
adsorbed fluid increases rapidly with pressure, until reaching an approximately
constant plateau and thus exhibiting type-I isotherms. The exact location of
that plateau depends on the molecular nature of the adsorbate, being reached
earlier for the lighter molecules.

A previously unobserved anisotropic
distribution of the confined CH4 molecules (Fig.2) is interpreted in
terms of a symmetry annihilation in the pseudo one-dimensional nanopores; this
fact arises from antiparallel alignments of the OH groups in the inorganic
octahedra.6 The total potential energy of the adsorbent/adsorbate
system is decoupled into fluid-fluid and solid-fluid interactions and analyzed
as a function of adsorbate loading and temperature. Macroscopic thermodynamical
properties, such as the Henry constant, H,
and the isosteric heat of adsorption, qst, are calculated and compared to
experimentally obtained values. The agreement between simulation and
experimental data is generally quite satisfactory.

FIG.1
Experimental (filled symbols) and simulated (open symbols)
excess adsorption isotherms for C1–C4 alkanes

in
MIL--‐53lp(Al),
at 303K, 323 K and 353 K; symbols: (♦/×) methane, (■/□)
ethane, (▲/Δ) propane, (●/○) n-butane.

For clarity, the adsorption data of ethane, propane and butane are displaced by
1, 2 and 3 mol/kg, respectively.

       

FIG.2 Molecular density field (lighter colors represent larger values of  r) and zero-potential hypersurface (ZPH)

for condensed C2H4
at 152.5 K and μ/kB
= –1623 K inside an even-index, diamond-shaped channel of MIL-53lp(Al).

(a) Front view of the channel; the colored lines show the perimeter  of the
ZPH at different axial positions along the

channel: x/a = 0 (yellow), x/a =
0.25 (red), x/a = 0.375 (green), and x/a = 0.5 (white). (b) Lateral view of the channel; the

solid line represents the ZPH for y/b = 0.5, and the dashed line shows the
hypersurface extended by a sphere of diameter σff/4.

Grants from Funda?o para a Cincia e a Tecnologia
(FCT/MCTES, Portugal) are gratefully acknowledged by F.J.A.L. Cruz
(SFRH/BPD/45064/2008), A.I. Lyubchyk
(SFRH/BD/45477/2008) and I.A.A.C. Esteves (PTDC/CTM/104782/2008)
.

References

1.    O. M. Yaghi,
M. O'Keeffe, N. W. Ockwig, H. K. Chae,
M. Eddaoudi, J. Kim, Nature 423, 705 (2003)
.

2.    T.
Loiseau, C. Serre, C. Huguenard, G. Fink, F. Taulelle,
M. Henry, T. Bataille, G. F?rey,
Chem. Eur. J. 10, 137 (2004).

3.    S.
Bourrelly, P. L. Llewellyn, C. Serre,
F. Millange, T. Loiseau, G.
F?rey, J. Am. Chem. Soc. 127, 13519 (2005).

4.    G.
F?rey, M. Latroche, C. Serre, F. Millange, T. Loiseau, A. Percheron- Guegan, Chem. Commun. 2976 (2003).

5.    N.
A. Ramsahye, G. Maurin, S. Bourrelly, P. L. Llewellyn, T. Loiseau,
C. Serre, G. F?rey, Chem.
Commun.
3261 (2007).

6.  
A. Lyubchyk, I.A.A.C. Esteves,
F.J.A.L. Cruz and J.P.B. Mota, J. Phys. Chem. C
115, 20628 (2011).