Article pubs.acs.org/Langmuir
Water Adsorption on Clay Minerals As a Function of Relative Humidity: Application of BET and Freundlich Adsorption Models Courtney D. Hatch,*,† Jadon S. Wiese,† Cameron C. Crane,† Kenneth J. Harris,† Hannah G. Kloss,† and Jonas Baltrusaitis‡ †
Department of Chemistry, Hendrix College, 1600 Washington Avenue, Conway, Arkansas 72032, United States Department of Chemistry and Central Microscopy Research Facility, University of Iowa, EMRB 76, Iowa City, Iowa 52242, United States
‡
ABSTRACT: Water adsorption on kaolinite, illite, and montmorillonite clays was studied as a function of relative humidity (RH) at room temperature (298 K) using horizontal attenuated total reflectance (HATR) Fourier transform infrared (FTIR) spectroscopy equipped with a flow cell. The water content as a function of RH was modeled using the Brunauer, Emmett, and Teller (BET) and Freundlich adsorption isotherm models to provide complementary multilayer adsorption analysis of water uptake on the clays. A detailed analysis of model fit integrity is reported. From the BET fit to the experimental data, the water content on each of the three clays at monolayer (ML) water coverage was determined and found to agree with previously reported gravimetric data. However, BET analysis failed to adequately describe adsorption phenomena at RH values greater than 80%, 50%, and 70% RH for kaolinite, illite, and montmorillonite clays, respectively. The Freundlich adsorption model was found to fit the data well over the entire range of RH values studied and revealed two distinct water adsorption regimes. Data obtained from the Freundlich model showed that montmorillonite has the highest water adsorption strength and highest adsorption capacity at RH values greater than 19% (i.e., above ML water adsorption) relative to the kaolinite and illite clays. The difference in the observed water adsorption behavior between the three clays was attributed to different water uptake mechanisms based on a distribution of available adsorption sites. It is suggested that different properties drive water adsorption under different adsorption regimes resulting in the broad variability of water uptake mechanisms.
1. INTRODUCTION Clay minerals are ubiquitous in the natural environment and participate in important environmental fate and transport pathways throughout the geosphere,1 hydrosphere,2 and atmosphere.3−5 Clays are also used in many industrial applications due to their interesting swelling,6−8 adsorptive,9−12 and catalytic13−15 properties. In the United States, an estimated 50 million tons of clay materials are used in a single year for industrial applications.13 The widespread uses of clay minerals in such diverse applications and their important role in the environment affirm the need for fundamental studies of water adsorption. The effectiveness of clays as an adsorbent or catalyst in both industrial applications and the environment depends on the accessibility of adsorption sites and the interaction strength of the adsorbate or reactant, respectively, with the surface environment. It follows that the surface environment must be known under a broad range of conditions, including clay composition, structure, surface area, and relative humidity (RH). © 2011 American Chemical Society
An example of the importance of water adsorption on clay minerals in the environment is the role of mineral dust aerosol in the Earth’s atmosphere. Mineral aerosol makes up the largest fraction of naturally occurring aerosol by mass globally,16 constituting ∼45% of the total atmospheric aerosol load.17 The composition of mineral aerosol is typically dominated by quartz and clay minerals (aluminosilicates), with smaller amounts of other minerals, including feldspar, carbonates (calcite and dolomite), sulfates (gypsum), and metal oxides.18−20 Reid and Maring21 found that 70% of all mineral aerosol mass originating from Africa consisted of layered aluminosilicates, such as illite, kaolinite, and montmorillonite clays. Additionally, Asian dust storms (ADS) are dominated by the presence of clay minerals.22 Shi et al.22 found the clay mineral fraction collected from ADS to be predominantly mixed-layer illite/smectite (78%) particles, followed by illite (9%), chlorite (7%), and Received: November 1, 2011 Revised: December 15, 2011 Published: December 19, 2011 1790
dx.doi.org/10.1021/la2042873 | Langmuir 2012, 28, 1790−1803
Langmuir
Article
widely among models. The input values, typically obtained from laboratory studies, for modeling water uptake on atmospheric aerosols are important for simulation of dust particle size, density, mobility, and the pathway and rate of deposition. Quantification of mineral aerosol water content is considered a “high priority”37 with respect to obtaining accurate estimates of the impacts of mineral aerosol on atmospheric chemistry, climate, and biogeochemistry.38
kaolinite (6%). The prevalence of mineral aerosol has a significant yet poorly understood effect on the environment.23 Largely unresolved implications for mineral aerosol include their role in the indirect climate affect in which aerosol particles take up water and behave as cloud condensation nuclei (CCN)24−26 and ice nuclei (IN).27−30 Incorporation of mineral aerosol into cloud droplets can alter the radiative properties and lifetimes of clouds31 also indirectly impacting climate.32 The current consensus is that mineral aerosol plays a minor role in the indirect aerosol effect based on experimental hygroscopicity and CCN activation measurements.24,26,33 However, inconsistencies still remain among the data reported in the literature regarding the effectiveness of mineral aerosol as CCN. Herich et al.33 recently indicated significant discrepancies in CCN activation and hygroscopicity parameters for clay minerals using different aerosol generation techniques. Specifically, wetgenerated mineral aerosol nucleated cloud droplets at much lower critical supersaturation values compared to dry-generated aerosol. Additionally, the hygroscopicity parameter, κ, derived from CCN and hygroscopicity measurements lacked agreement between wet- and dry-generated mineral aerosol and between different approaches for determining κ. Herich et al.33 suggested that this discrepancy is due to rearrangement of soluble material on the surface of the clays upon wetting. Alternatively, these previous results may be an artifact of the extent of water coverage of wet- versus dry-generated aerosols prior to CCN and hygroscopicity measurements. Incidentally, the finding that wetted dust particles exhibit superior CCN activation compared to dry-generated aerosols33 suggests that water adsorption at subsaturated water vapor conditions may play a significant role in mineral aerosol CCN activation. Knowledge of the adsorption and desorption time scales and mechanisms and adsorption model parameters could aid in addressing these discrepancies in previous studies. More recent support for this theory comes from a number of theoretical studies which suggest that insoluble particles, such as mineral aerosol, nucleate cloud droplets by multilayer adsorption.34−36 These same authors formulated a theoretical framework, adsorption activation theory (AT), for modeling CCN activation of insoluble aerosol components based on the balance between the Kelvin effect and multilayer water adsorption using adsorption parameters from multilayer adsorption models. However, these previous studies used adsorption parameters to test this new theory based on a range of previously reported parameters for surfaces other than mineral particles. Water adsorption measurements and analysis by multilayer adsorption models reported here aim to address current uncertainties in aerosol measurements and provide adsorption parameters based on experimental results of water adsorption on clay minerals for analyzing the effects of multilayer adsorption on mineral aerosol hygroscopicity and CCN activation. In addition to implications for furthering our understanding of mineral aerosol contributions to the indirect aerosol effect on climate, quantification of mineral aerosol water content is important for studying atmospheric impacts on ocean processes. Mineral aerosol emission and deposition models are integral in determining the biogeochemical effects of mineral deposition to remote oceans. Unfortunately, there is currently disagreement regarding deposition pathways and atmospheric residence times of mineral species among many different global transport models.37 Textor et al.37 found that the input values for the uptake of water on mineral surfaces vary
2. BACKGROUND Geologically, clay minerals are defined as terrestrial soil particles having grain sizes of less than 2 μm in diameter and thus can easily be lofted into the atmosphere by strong winds and transported over long distances.39−42 Aluminosilicate clay minerals have complex chemical structures that give rise to interesting physiochemical properties, including structural expansion in some cases, which vary dramatically from their individual building blocks (i.e., silica and alumina layers) and among individual clay families. The study reported here focuses on three of the most abundant types of clay minerals found in atmospheric dust plumes, including kaolinite, illite, and montmorillonite. All three minerals are phyllosilicates composed of plate-like structures characterized by varying arrangements of alumina and silica sheets stacked in repeating layers. Kaolinite, which has a generalized formula of [Si4]Al4O10(OH)8 (elements enclosed in brackets indicate tetrahedral coordination),43 is a 1:1 type clay in which each aluminosilicate layer contains one tetrahedral (silica) sheet and one octahedral (alumina) sheet held together by oxygen anions shared by the Si and Al in neighboring sheets. The external surfaces of the alumina octahedra contain structural hydroxyl groups, and thus, individual kaolinite layers are held together by hydrogen bonds. As a result, the swelling capacity (i.e., ability to adsorb water molecules between aluminosilicate layers) is very low and kaolinite is considered a nonexpansive clay mineral.44 Characteristics of kaolinite clay include small cation exchange capacities (CEC) due to ionization of surface hydroxyls, particularly at basic pH values, relatively low surface areas, and large particle sizes. Illite has a generalized formula of M x [Si 6.8 Al 1.2 ]Al3Fe0.25Mg0.75O20(OH)4,43 where M is a monovalent interlamellar cation, and is considered to be part of the 2:1 clay family. 2:1 clays are characterized by aluminosilicate layers containing two tetrahedral (silica) sheets sandwiching one octahedral (alumina) sheet. Montmorillonite, a smectite clay with general formula Mx[Si8]Al3.2Fe0.2Mg0.6O20(OH)4,43 is also a 2:1 clay and has a similar structural backbone as illite. However, despite structural similarities, illite and montmorillonite clays differ significantly in their physical and chemical properties, including isomorphic substitution sites, surface area, CEC, particle size, and swelling capacity.43−45 In general, illite is considered to be nonexpansive due to the presence of strong ionic interactions between interlamellar potassium ions, which have relatively low hydration energies and a net negative structural charge arising from tetrahedral isomorphic substitutions (montmorillonite exhibits isomorphic substitution mainly in the octahedral layers). The nonexpansive nature of illite is verified by X-ray diffraction (XRD) measurements showing no observable increase in the interlamellar distance (d001) upon water adsorption.46 Additionally, illite is characterized by lower surface areas, lower CECs, and larger particle sizes relative to montmorillonite clay.44 For comparison, illite 1791
dx.doi.org/10.1021/la2042873 | Langmuir 2012, 28, 1790−1803
Langmuir
Article
Fit integrity was evaluated upon calculation of the average percent relative deviation error (%RDE) between the experimentally and theoretically derived adsorbed water content values. Use of multiple adsorption models allows for identification of different water adsorption characteristics for the three clays over the entire range of RH values studied (0− 94% RH) and reveals subtle differences between water adsorption on nonswelling and swelling clay samples.
typically has larger surface areas, higher CECs, and smaller particle sizes than kaolinite clays. Montmorillonite is unique relative to kaolinite and illite because it has the ability to structurally expand upon adsorption of water, driven in large part by hydration of exchangeable cations, such as Li+, K+, Na+, Ca2+, or Mg2+, in the interlamellar space of the clay layers.45,47,48 Details of clay layer expansion upon water adsorption have been reported previously in the literature.49−52 In general, water adsorption on montmorillonite is multidimensional, that is, water can adsorb on the external surfaces, within the interlamellar space (within the particles on a microscopic scale
