SEARCH

SEARCH BY CITATION

Keywords:

  • microtomography;
  • colloids;
  • porous media

Abstract

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Methods
  5. 3. Results
  6. 4. Conclusions and Perspectives
  7. Acknowledgments
  8. References

[1] High resolution synchrotron-based X-ray computed microtomography (X-CMT) was used to identify the morphology of colloidal deposits formed in porous media. We show that difference microtomography - whereby a tomographic reconstruction is performed across an absorption edge - provides valuable information on the nature and location of the aggregates formed by the deposition of colloidal particles. Column experiments were performed using an idealized porous medium consisting of glass beads through which colloidal ZrO2 particles were transported. Tomographic reconstructions of the porous medium and of the aggregate structure provide an unique opportunity to observe colloidal particle deposits and of their morphology. These results show that the local pore geometry controls particle deposition and that deposits tend to form in a rather heterogeneous manner in the porous medium.

1. Introduction

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Methods
  5. 3. Results
  6. 4. Conclusions and Perspectives
  7. Acknowledgments
  8. References

[2] The deposition of colloidal particles in a porous medium is relevant in a broad range of environmental and geophysical problems, such as oil extraction [Moghadasi et al., 2004], subsurface contaminant transport [Kretzschmar et al., 1999], siltation of streambeds [Packman et al., 2000], transmission of pathogens in surface and subsurface waters [Searcy et al., 2006], and water treatment involving filtration in both granular beds and membranes [Elimelech et al., 1995]. Filtration models, summarized by [McDowell-Boyer et al., 1986; Elimelech et al., 1995; Tufenkji and Elimelech, 2004], normally assume fixed pore geometry and parameterize deposition rates in terms of the size ratio between mobile suspended particles and pore spaces. Recent investigations have shown that the internal variability of the porous medium plays an important role in the overall filtration process, and that enhanced deposition can be caused by straining at grain contacts [Johnson et al., 1996; Bradford et al., 2004; Li et al., 2006; Xu et al., 2006]. In addition, numerical simulations suggest that depositing particles form complex structures [Wiesner, 1999], and accumulation of deposited particles will eventually alter the pore geometry to a significant extent, leading to a reduction of permeability and producing clogging of the porous medium [Mays and Hunt, 2005]. Clogging has been observed to be important for a number of applications, such as causing head loss in water treatment filters, reducing pore water flows important to benthic and hyporheic organisms, restricting delivery of treatment agents for remediation of subsurface contamination, and hindering brine reinjection after crude oil extraction.

[3] Prior efforts to understand the particle deposition process have normally been based on column experiments that rely on observations of the decrease of colloid concentration between the column inflow and outflow [Elimelech et al., 1995]. Some information is available on distributions of accumulated particle mass within columns, but these observations are still aggregated to a macroscopic scale, e.g., 1 cm slices (see compilation of available data by Mays and Hunt [2005]). As a result, available clogging models have high uncertainty, rely on extensive system-specific empirical data, and do not provide detailed descriptions of the internal variations in deposit structure [Mays and Hunt, 2005]. More detailed modeling approaches have also been used to investigate the development of deposit morphologies [e.g., Wiesner, 1999], but these efforts have been limited by a lack of information on the pore geometry and have therefore assumed highly simplified pore structures.

[4] In order to improve understanding of the dynamics of the clogging process and to facilitate the development of more general, mechanistic models of particle deposition, direct observations are needed to provide detailed information on the evolution of pore structure changes as a result of colloidal deposition. The objective of this paper is to show that Z-contrast (absorption-edge) X-ray difference microtomography provides an unique capability to observe the structure of colloidal deposits within a porous medium.

2. Methods

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Methods
  5. 3. Results
  6. 4. Conclusions and Perspectives
  7. Acknowledgments
  8. References

2.1. X-Ray Difference Microtomography

[5] X-ray computed microtomography (X-CMT) performed at a synchrotron research center provides the ability to obtain three-dimensional, in situ observations of the internal structure of materials with micron-scale resolution [Auzerais et al., 1996; Flannery et al., 1987; Li et al., 2006; Spanne et al., 1994; Wildenschild et al., 2005]. The distributions of individual elements can be determined by performing difference tomography, where the X-rays used to image the sample are selected to be just below and above the absorption edge of the element of interest. The resulting abrupt change in the X-ray absorption coefficient is used to obtain an element-specific 3-D distribution [McNear et al., 2005; Tsuchiyama et al., 2005; Gaillard, 2007]. We term this X-ray difference microtomography (X-DMT). This analysis must be performed using a monochromatic X-ray beam from a synchrotron light source.

[6] Sample imaging is performed by placing a suitable specimen, here a small column of a granular porous medium containing deposited colloids, on a rotation stage and illuminating it with a monochromatic X-ray beam. A phosphor plate is placed behind the object and is used to convert the pattern of X-ray intensity transmitted through the specimen to a visible light image. This image is then magnified using a microscope objective lens onto the surface of a CCD camera. Attenuation images or shadowgrams are then collected from many angles by rotating the specimen through small and regularly spaced angular increments. The collection of shadowgrams is then used to reconstruct the tomographic image, usually on a computer cluster. Approximately Nπ/2 images must be collected to allow successful reconstruction of the sample, where N is the width of the object in CCD pixels.

[7] In the ideal case, the specimen should absorb about 90% of the incident radiation across its most opaque section in order to produce images with enough contrast that will lead to the best signal-to-noise ratio for the reconstruction process. The absorption is measured by monitoring the ratio I/I0, where I is the transmitted intensity and I0 is the intensity of the incident beam, given that I/I0 = eμ(λ)ρx, where μ(λ) is the mass attenuation coefficient of the sample, ρ is the specimen density, and x is its thickness. The absorption of 90% of the incident beam corresponds to τ = μ(λ)ρx ≃ 2. Since in a difference tomogram the wavelength of the X-ray beam is dictated by the absorption edge of the element of interest, the sample size must be selected to achieve a τ that is as close as possible to the optimum value of 2. It is not always possible to work under these ideal conditions and relatively good tomographic reconstructions can be achieved with smaller τ values. However, when τ ≤ 0.5 the signal-to-noise ratio becomes too low to yield valuable shadowgrams.

[8] In this study, we observed the deposition of colloidal ZrO2 aggregates in a porous medium composed of glass beads. The optics used provide a spatial resolution of ∼6 μm. Zr (Z = 40) has an absorption edge corresponding to an X-ray wavelength of λ = 0.68887Å. This provides an optimum Z-contrast with sand or glass collector particles composed primarily of SiO2 (Si, Z = 14). In addition, the incoming X-ray photon energy required for difference tomography of Zr was sufficient to allow use of small columns (∼3 mm diameter) packed with the granular material comprising the porous medium.

2.2. Column Experiments

[9] Thin wall columns were made of high-density polyethylene (HDPE) rods (McMaster-Carr) machined to an internal diameter of 3.2 mm and a length of approximately 30 mm. The columns were packed with glass beads having a diameter between 0.18 and 0.25 mm (Spheriglass P-0100, Potters Industries Inc.) The columns were positioned vertically and cleaned by circulating as background electrolyte a solution of 1 M NaCl at pH = 6.66. After passing through the column at least 100 pore volumes, a stable suspension of colloidal ZrO2 particles (A-Grain Zirconia, Zircoa Inc.) at a concentration of 200 mg/L dispersed in the same background electrolyte was passed through the column for 1 hour at an upflow velocity of 1.24 cm/s. Under these experimental conditions, the particles are about neutrally charged (zeta potential: −1.06 ± 3.54 mV) and their average diameter is 1.5 ± 0.3 μm as determined by BIC-ZetaPALS. The resolution was not sufficient to image individual particles but cluster aggregates were well resolved. The columns were sealed at both ends, maintained in a vertical position, and transported to the synchrotron research facility the next day for analysis by X-ray computed microtomography. Additional experiments were also performed on site to verify that sample relocation did not change particle deposition patterns.

2.3. Difference Microtomography Measurements

[10] X-DMT analysis was performed at Northwestern's Synchrotron Research Center, Sector 5 of the Advanced Photon Source, Argonne National Laboratory. Transmission tomograms were recorded below and above the Zr K edge at 17.998 keV using a monochromatized beam obtained using a narrow gap, Si(111) double-crystal monochromator. The sample was positioned on a rotating Newport stage and was turned by 0.2 or 0.15 degree angle increment and illuminated by the highly collimated X-ray beam. The phosphor screen was imaged using an optical lens for magnification (X10) coupled to a Roper scientific (Photometrics) camera - 1300 × 1340, 24 micron pixels, with 16 bit readout - that was thermoelectrically cooled (Cryotiger). The first tomogram was collected at 30 eV below the Zr K edge whereas the second one was collected at 40 eV above. Between 900 to 1200 absorbance maps were collected per session and processed on a 15 node Linux computer cluster using the parallelized version of the CBREC code written by Paul Thomas of the Mayo Foundation, using parallel beam filtered back-projection with a Shepp-Logan kernel. The tomographic reconstructions were then processed using IDL (ITT Visual Information Solutions) and the Blob3D package [Ketcham, 2005].

3. Results

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Methods
  5. 3. Results
  6. 4. Conclusions and Perspectives
  7. Acknowledgments
  8. References

[11] Tomographic reconstructions of the porous medium containing zirconia colloidal deposits are presented in Figures 1, 2, and 3. Figure 1 shows a 3-D representation of the glass beads together with the colloidal deposits of ZrO2 that are present in the column. This reconstruction shows that the colloidal particles deposit upstream, at the bottom of the glass spheres, as expected from filtration theory. However, it can readily be observed that the particle aggregates have various size and morphology, as opposed to the homogeneous deposition predicted by the theory.

image

Figure 1. Three-dimensional representation of a column internal volume of dimension 0.6 × 0.6 × 1.2 mm showing the glass spheres (blue) and the zirconia deposits (red). This image was obtained by combining the tomogram collected below the Zr K-edge and the difference tomogram that provides the distribution of zirconia particles solely.

Download figure to PowerPoint

image

Figure 2. Thin section through the porous medium column showing the distribution of aggregates and colloidal size particles. The distribution of aggregates is heterogeneous within the porous medium.

Download figure to PowerPoint

image

Figure 3. Two examples of colloidal zirconia aggregates that formed around spherical collector grains. Scale bars are 50 μm in each of three orthogonal directions. Colloid deposition generally begins at the upstream point on the collector grain which is towards the bottom of the glass beads. The left panel shows the deposit around a single grain, and the right panel shows the deposit that formed around the contact between two adjacent grains.

Download figure to PowerPoint

[12] The heterogeneity in colloid deposition can be seen more clearly in Figure 2, which shows a single image plane corresponding to a thin section within the column. Neither the size distribution of the colloidal aggregates nor the spatial distribution of the aggregates are uniform across the horizontal plane. Further, inspection of many of these slices indicates that the geometry of the grain-to-grain connections (pore throats) affects the size of the colloidal aggregate that forms at each grain contact. This suggests that the colloid deposition process is influenced by slight differences in the packing of the spheres, i.e., the pore geometry and the connectivity between collector particles. Such internal variations are not currently considered in any macroscopic model for particle deposition in granular porous media.

[13] The colloid to collector particle diameter ratio here is about 10−2. Under this condition, straining - that is the physical entrapment of the particles by constrictions in the porous medium - is usually considered to be insignificant. However, recently [Bradford et al., 2004; Xu et al., 2006] have suggested that this mode of deposition should occur at a ratio of as low as 5.10−3. Our results confirm that deposition at grain contacts plays a key role in the clogging process, and that microstructural information is needed to provide a detailed description of the formation of particle aggregates within pores. This is exemplified in Figure 3, which shows the morphologies of two different colloidal aggregates. These images were extracted from the microtomographic data for the distribution of Zr within the sample.

[14] This volumetric information can be further analyzed using a dimensionless parameter: the sphere normalized surface to volume ratio, SNSVR = (SA/SAsphere)1.5, where SA is the surface area of the object and SAsphere is the surface area of a sphere that has the same volume of the object [Ketcham, 2005]. When this quantity is plotted as a function of the partial aggregate volume (Figure 4), one observes that a significant number of clusters tend to present a more spherical geometry. These deposits contain more than 100 primary particles and their volumes range from 10−6 to 10−4 mm3. For a smaller number of larger volume aggregates the morphology departs significantly from spheres. These types of deposits are characteristically represented in Figure 3.

image

Figure 4. (top) Sphere normalized surface to volume ratio as a function of the aggregate's volume and (bottom) histogram representing the number of aggregates with respect to their volume. Only aggregates that contain more than about 100 primary particles are considered, since smaller aggregate clusters are not resolved.

Download figure to PowerPoint

4. Conclusions and Perspectives

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Methods
  5. 3. Results
  6. 4. Conclusions and Perspectives
  7. Acknowledgments
  8. References

[15] In this study, we have demonstrated that synchrotron-based X-ray difference microtomography can be used to obtain detailed structural information on the formation of particle deposits within a granular porous medium. The main advance here is the use of Z-contrast microtomography combined with colloids having a suitable elemental composition to independently resolve the structure of colloidal deposits and the surrounding matrix of collector grains. This approach allows the pre- and post-deposition pore structure to be observed at the micron scale and by means of a single non-invasive measurement.

[16] This method, while powerful, is also subject to significant limitations. Principally, use of difference tomography requires the sample be imaged at a suitable X-ray absorption edge for the element of interest. Careful attention must then be paid to the extent of attenuation of the X-ray beam by the experimental sample, including the attenuation due to both the element of interest and the surrounding matrix. Here, experiments were conducted using zirconia colloids because zirconium provides a favorable K-absorption edge at λ = 0.68887Å. Even so, column diameters were restricted to just over 3 mm with a granular porous medium composed of small sand grains or glass beads (<300 microns in diameter). The use of higher magnification optics can provide better spatial resolution, c.a. 1 micron, but the diameter of the column would have to be reduced correspondingly. In addition, the requirement that many individual X-ray absorption images (shadowgrams) be obtained for tomographic reconstruction means that the image acquisition process is slow, therefore dynamic processes cannot be observed in real time. Instead, what the method offers is a very detailed three-dimensional picture of the bulk structure and static distribution of an element of interest within a small, fixed sample volume.

[17] Overall, difference tomography provides unprecedented potential for obtaining 3-D microstructural and chemical information in natural sedimentary environments. For example, this approach could be very useful for investigating contaminant distributions within sedimentary deposits, for observing the (re)distribution of particular sediment particles following perturbation events, or for examining the incorporation of particular elements in authigenic mineral deposits or lithifying microbial mats.

Acknowledgments

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Methods
  5. 3. Results
  6. 4. Conclusions and Perspectives
  7. Acknowledgments
  8. References

[18] (The authors gratefully acknowledge the support of the National Science Foundation through grant HS-0310657. This work was performed at the DuPont-Northwestern-Dow Collaborative Access Team (DND-CAT) Synchrotron Research Center located at Sector 5 of the Advanced Photon Source. DND-CAT is supported by the E.I. DuPont de Nemours & Co., The Dow Chemical Company, the U.S. National Science Foundation through grant DMR-9304725 and the State of Illinois through the Department of Commerce and the Board of Higher Education Grant IBHE HECA NWU 96. Use of the Advanced Photon Source was supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under contract W-31-109-Eng-38.)

References

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Methods
  5. 3. Results
  6. 4. Conclusions and Perspectives
  7. Acknowledgments
  8. References
  • Auzerais, F. M., J. Dunsmuir, B. B. Ferreol, N. Martys, J. Olson, T. S. Ramakrishnan, D. H. Rothman, and L. M. Schwartz (1996), Transport in sandstone: A study based on three dimensional microtomography, Geophys. Res. Lett., 23, 705708.
  • Bradford, S. A., M. Bettahar, J. Simunek, and M. T. van Genuchten (2004), Straining and attachment of colloids in physically heterogeneous porous media, Vadose Zone J., 3, 384394.
  • Elimelech, M., J. Gregory, X. Jia, and R. A. Williams (1995), Particle Deposition and Aggregation: Measurement, Modeling, and Simulation, Butterworth-Heinemann, Oxford, U. K.
  • Flannery, B. P., H. W. Deckman, W. G. Roberge, and K. L. Damico (1987), 3-dimensional X-ray microtomography, Science, 237, 14391444.
  • Gaillard, J.-F. (2007), Probing environmental particles and colloids with X-rays, in Environmental Particles and Colloids: Behaviour, Structure, and Characterisation, edited by K. J. Wilkinson, and J. R. Lead, pp. 613666, John Wiley, Chichester, U. K.
  • Johnson, P. R., N. Sun, and M. Elimelech (1996), Colloid transport in geochemically heterogeneous porous media: Modeling and measurements, Environ. Sci. Technol., 30, 32843293.
  • Ketcham, R. A. (2005), Computational methods for quantitative analysis of three-dimensional features in geological systems, Geosphere, 1, 3241.
  • Kretzschmar, R., M. Borkovec, D. Grolimund, and M. Elimelech (1999), Mobile subsurface colloids and their role in contaminant transport, Adv. Agron., 66, 121193.
  • Li, X. Q., C. L. Lin, J. D. Miller, and W. P. Johnson (2006), Pore-scale observation of microsphere deposition at grain-to-grain contacts over assemblage-scale porous media domains using X-ray microtomography, Environ. Sci. Technol., 40, 37623768.
  • Mays, D. C., and J. R. Hunt (2005), Hydrodynamic aspects of particle clogging in porous media, Environ. Sci. Technol., 39, 577584.
  • McDowell-Boyer, L. M., J. R. Hunt, and N. Sitar (1986), Particle transport through porous media, Water Resour. Res., 22, 19011921.
  • McNear, D. H., E. Peltier, J. Everhart, R. L. Chaney, S. Sutton, M. Newville, M. Rivers, and D. L. Sparks (2005), Application of quantitative fluorescence and absorption-edge computed microtomography to image metal compartmentalization in Alyssum murale, Environ. Sci. Technol., 39, 22102218.
  • Moghadasi, J., H. Muller-Steinhagen, M. Jamialahmadi, and A. Sharif (2004), Theoretical and experimental study of particle movement and deposition in porous media during water injection, J. Pet. Sci. Eng., 43(3–4), 163181.
  • Packman, A. I., N. H. Brooks, and J. J. Morgan (2000), Kaolinite exchange between a stream and streambed: Laboratory experiments and validation of a colloid transport model, Water Resour. Res., 36, 23632372.
  • Searcy, K. E., A. I. Packman, E. R. Atwill, and T. Harter (2006), Deposition of Cryptosporidium oocysts by stream-subsurface exchange, Appl. Environ. Microbiol., 72, 18101816.
  • Spanne, P., J. F. Thovert, C. J. Jacquin, W. B. Lindquist, K. W. Jones, and P. M. Adler (1994), Synchrotron computed microtomography of porous-media—Topology and transports, Phys. Rev. Lett., 73, 20012004.
  • Tsuchiyama, A., K. Uesugi, T. Nakano, and S. Ikeda (2005), Quantitative evaluation of attenuation contrast of X-ray computed tomography images using monochromatized beams, Am. Mineral., 90, 132142.
  • Tufenkji, N., and M. Elimelech (2004), Correlation equation for predicting single-collector efficiency in physicochemical filtration in saturated porous media, Environ. Sci. Technol., 38, 529536.
  • Wiesner, M. R. (1999), Morphology of particle deposits, J. Environ. Eng., 125, 11241132.
  • Wildenschild, D., J. W. Hopmans, M. L. Rivers, and A. J. R. Kent (2005), Quantitative analysis of flow processes in a sand using synchrotron-based X-ray microtomography, Vadose Zone J., 4, 112126.
  • Xu, S., B. Gao, and J. E. Saiers (2006), Straining of colloidal particles in saturated porous media, Water Resour. Res., 42, W12S16, doi:10.1029/2006WR004948.