The Anatomy of Amorphous, Heterogeneous Catalyst Pellets.

This review focuses on disordered, or amorphous, porous heterogeneous catalysts, especially those in the forms of pellets and monoliths. It considers the structural characterisation and representation of the void space of these porous media. It discusses the latest developments in the determination of key void space descriptors, such as porosity, pore size, and tortuosity. In particular, it discusses the contributions that can be made by various imaging modalities in both direct and indirect characterisations and their limitations. The second part of the review considers the various types of representations of the void space of porous catalysts. It was found that these come in three main types, which are dependent on the level of idealisation of the representation and the final purpose of the model. It was found that the limitations on the resolution and field of view for direct imaging methods mean that hybrid methods, combined with indirect porosimetry methods that can bridge the many length scales of structural heterogeneity and provide more statistically representative parameters, deliver the best basis for model construction for understanding mass transport in highly heterogeneous media.

Use of activated Chromolaena odorata biomass for the removal of crystal violet from aqueous solution: kinetic, equilibrium, and thermodynamic study.

In the present study, biomass from the Chromolaena odorata plant's stem was activated using sulfuric acid to adsorb crystal violet (CV) dye. The adsorption operation of CV dye was studied considering the effect of variables like pH, initial dye concentration, time, adsorbent dosage, and temperature. The pseudo-second-order equation best fitted the kinetic study. The thermodynamic parameters such as activation energy (9.56 kJ/mol), change in Gibbs energy (81.43 to 96.7 kJ/mol), enthalpy change (6.89 kJ/mol), and entropy change (-254.4 J/mol K) were calculated. Response surface methodology estimated that at pH (4.902), adsorbent dosage (8.33 g/L), dye concentration (82.30 ppm), and temperature (300.13 K) dye removal of 97.53% is possible. FTIR, SEM, XRD, BJH, and BET confirmed adsorption operation. The adsorbent can be reused for 3 cycles effectively. Langmuir isotherm which best fitted the adsorption operation was used for designing a theoretical single-stage batch adsorber for large-scale operation.

Assessing the quality of RNA isolated from human breast tissue after ambient room temperature exposure.

High quality human tissue is essential for molecular research, but pre-analytical conditions encountered during tissue collection could degrade tissue RNA. We evaluated how prolonged exposure of non-diseased breast tissue to ambient room temperature (22±1°C) impacted RNA quality. Breast tissue received between 70 to 190 minutes after excision was immediately flash frozen (FF) or embedded in Optimal Cutting Temperature (OCT) compound upon receipt (T0). Additional breast tissue pieces were further exposed to increments of 60 (T1 = T0+60 mins), 120 (T2 = T0+120 mins) and 180 (T3 = T0+180 mins) minutes of ambient room temperature before processing into FF and OCT. Total exposure, T3 (T0+180 mins) ranged from 250 minutes to 370 minutes. All samples (FF and OCT) were stored at -80°C before RNA isolation. The RNA quality assessment based on RNA Integrity Number (RIN) showed RINs for both FF and OCT samples were within the generally acceptable range (mean 7.88±0.90 to 8.52±0.66). No significant difference was observed when RIN at T0 was compared to RIN at T1, T2 and T3 (FF samples, p = 0.43, 0.56, 0.44; OCT samples, p = 0.25, 0.82, 1.0), or when RIN was compared between T1, T2 and T3. RNA quality assessed by quantitative real-time PCR (qRT-PCR) analysis of beta-actin (ACTB), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), cyclophilin A (CYPA), and porphobilinogen deaminase (PBGD) transcripts showed threshold values (Ct) that indicate abundant and intact target nucleic acid in all samples (mean ranging from 14.1 to 25.3). The study shows that higher RIN values were obtained for non-diseased breast tissue up to 190 minutes after resection and prior to stabilization. Further experimental exposure up to 180 minutes had no significant effect on RIN values. This study strengthens the rationale for assessing RIN and specific gene transcript levels as an objective method for determining how suitable RNA will be for a specific research purpose ("fit-for purpose").

In-situ microwave-assisted catalytic upgrading of heavy oil: Experimental validation and effect of catalyst pore structure on activity.

In-situ combustion alone may not provide sufficient heating for downhole, catalytic upgrading of heavy oil in the Toe-to-Heel Air Injection (THAI) process. In this study, a new microwave heating technique has been proposed as a strategy to provide the requisite heating. Microwave technology is alone able to provide rapid heating which can be targeted at the catalyst packing and/or the incoming oil in its immediate vicinity. It was demonstrated, contrary to previous assertions, that heavy oil can be heated directly with microwaves to 425 °C, which is the temperature needed for successful catalytic upgrading, without the need for an additional microwave susceptor. Upgrading of >3.2° API points, a reduction in viscosity to less than 100 cP, and >12% reduction in sulfur content was achieved using commercially available hydrodesulfurization (HDS) catalyst. The HDS catalyst induced dehydrogenation, with nearly 20% hydrogen detected in the gas product. Hence, in THAI field settings, part of the oil-in-place could be sacrificed for dehydrogenation, with the produced hydrogen directed to aid hydrodesulfurization and improve upgrading. Further, this could provide a route for downhole hydrogen production, which can contribute to the efforts towards the hydrogen economy. A single, unified model of evolving catalyst structure was developed. The model incorporated the unusual gas sorption data, computerized x-ray tomography and electron microprobe characterization, as well as the reaction behavior. The proposed model also highlighted the significant impact of the particular catalyst fabrication process on the catalytic activity.

Molecular hydrogen and catalytic combustion in the production of hyperpolarized 83Kr and 129Xe MRI contrast agents.

Hyperpolarized (hp) (83)Kr is a promising MRI contrast agent for the diagnosis of pulmonary diseases affecting the surface of the respiratory zone. However, the distinct physical properties of (83)Kr that enable unique MRI contrast also complicate the production of hp (83)Kr. This work presents a previously unexplored approach in the generation of hp (83)Kr that can likewise be used for the production of hp (129)Xe. Molecular nitrogen, typically used as buffer gas in spin-exchange optical pumping (SEOP), was replaced by molecular hydrogen without penalty for the achievable hyperpolarization. In this particular study, the highest obtained nuclear spin polarizations were P =29% for(83)Kr and P= 63% for (129)Xe. The results were reproduced over many SEOP cycles despite the laser-induced on-resonance formation of rubidium hydride (RbH). Following SEOP, the H2 was reactively removed via catalytic combustion without measurable losses in hyperpolarized spin state of either (83)Kr or (129)Xe. Highly spin-polarized (83)Kr can now be purified for the first time, to our knowledge, to provide high signal intensity for the advancement of in vivo hp (83)Kr MRI. More generally, a chemical reaction appears as a viable alternative to the cryogenic separation process, the primary purification method of hp(129)Xe for the past 2 1/2 decades. The inherent simplicity of the combustion process will facilitate hp (129)Xe production and should allow for on-demand continuous flow of purified and highly spin-polarized (129)Xe.

Techniques for direct experimental evaluation of structure-transport relationships in disordered porous solids.

Determining structure-transport relationships is critical to optimising the activity and selectivity performance of porous pellets acting as heterogeneous catalysts for diffusion-limited reactions. For amorphous porous systems determining the impact of particular aspects of the void space on mass transport often requires complex characterization and modelling steps to deconvolve the specific influence of the feature in question. These characterization and modelling steps often have limited accuracy and precision. It is the purpose of this work to present a case-study demonstrating the use of a more direct experimental evaluation of the impact of pore network features on mass transport. The case study evaluated the efficacy of the macropores of a bidisperse porous foam structure on improving mass transport over a purely mesoporous system. The method presented involved extending the novel integrated gas sorption and mercury porosimetry method to include uptake kinetics. Results for the new method were compared with those obtained by the alternative NMR cryodiffusometry technique, and found to lead to similar conclusions. It was found that the experimentally-determined degree of influence of the foam macropores was in line with expectations from a simple resistance model for a disconnected macropore network.

Insights into the influence of the cooling profile on the reconstitution times of amorphous lyophilized protein formulations.

Lyophilized protein formulations must be reconstituted back into solution prior to patient administration and in this regard long reconstitution times are not ideal. The factors that govern reconstitution time remain poorly understood. The aim of this research was to understand the influence of the lyophilization cooling profile (including annealing) on the resulting cake structure and reconstitution time. Three protein formulations (BSA 50mg/ml, BSA 200mg/ml and IgG1 40mg/ml, all in 7% w/v sucrose) were investigated after cooling at either 0.5°C/min, or quench cooling with liquid nitrogen with/without annealing. Significantly longer reconstitution times were observed for the lower protein concentration formulations following quench cool. Porosity measurements found concomitant increases in the surface area of the porous cake structure but a reduction in total pore volume. We propose that slow reconstitution results from either closed pores or small pores impeding the penetration of water into the lyophilized cake.

Combining mercury thermoporometry with integrated gas sorption and mercury porosimetry to improve accuracy of pore-size distributions for disordered solids.

The typical approach to analysing raw data, from common pore characterization methods such as gas sorption and mercury porosimetry, to obtain pore size distributions for disordered porous solids generally makes several critical assumptions that impact the accuracy of the void space descriptors thereby obtained. These assumptions can lead to errors in pore size of as much as 500%. In this work, we eliminated these assumptions by employing novel experiments involving fully integrated gas sorption, mercury porosimetry and mercury thermoporometry techniques. The entrapment of mercury following porosimetry allowed the isolation (for study) of a particular subset of pores within a much larger interconnected network. Hence, a degree of specificity of findings to particular pores, more commonly associated with use of templated, model porous solids, can also be achieved for disordered materials. Gas sorption experiments were conducted in series, both before and after mercury porosimetry, on the same sample, and the mercury entrapped following porosimetry was used as the probe fluid for theromporometry. Hence, even if one technique, on its own, is indirect, requiring unsubstantiated assumptions, the fully integrated combination of techniques described here permits the validation of assumptions used in one technique by another. Using controlled-pore glasses as model materials, mercury porosimetry scanning curves were used to establish the correct correspondence between the appropriate Gibbs-Thomson parameter, and the nature of the meniscus geometry in melting, for thermoporometry measurements on entrapped mercury. Mercury thermoporometry has been used to validate the pore sizes, for a series of sol-gel silica materials, obtained from mercury porosimetry data using the independently-calibrated Kloubek correlations. The pore sizes obtained for sol-gel silicas from porosimetry and thermoporometry have been shown to differ substantially from those obtained via gas sorption and NLDFT analysis. DRIFTS data for the samples studied has suggested that the cause of this discrepancy may arise from significant differences in the surface chemistries between the samples studied here and that used to calibrate the NLDFT potentials.

NMR cryoporometry characterisation studies of the relation between drug release profile and pore structural evolution of polymeric nanoparticles.

PLGA/PLA polymeric nanoparticles could potentially enhance the effectiveness of convective delivery of drugs, such as carboplatin, to the brain, by enabling a more sustained dosage over a longer time than otherwise possible. However, the link between the controlled release nanoparticle synthesis route, and the subsequent drug release profile obtained, is not well-understood, which hinders design of synthesis routes and availability of suitable nanoparticles. In particular, despite pore structure evolution often forming a key aspect of past theories of the physical mechanism by which a particular drug release profile is obtained, these theories have not been independently tested and validated against pore structural information. Such validation is required for intelligent synthesis design, and NMR cryoporometry can supply the requisite information. Unlike conventional pore characterisation techniques, NMR cryoporometry permits the investigation of porous particles in the wet state. NMR cryoporometry has thus enabled the detailed study of the evolving, nanoscale structure of nanoparticles during drug release, and thus related pore structure to drug release profile in a way not done previously for nanoparticles. Nanoparticles with different types of carboplatin drug release profiles were compared, including burst release, and various forms of delayed release. ESEM and TEM images of these nanoparticles also provided supporting data showing the rapid initial evolution of some nanoparticles. Different stages, within a complex, varying drug release profile, were found to be associated with particular types of changes in the nanostructure which could be distinguished by NMR. For a core-coat nanoparticle formulation, the development of smaller nanopores, following an extended induction period with no structural change, was associated with the onset of substantial drug release. This information could be used to independently validate the rationale for a particular synthesis method. Hence, the specific reasons for the effectiveness of the synthesis route, for obtaining core-coat nanoparticles with delayed release, have been elucidated.

Improving sensitivity and accuracy of pore structural characterisation using scanning curves in integrated gas sorption and mercury porosimetry experiments.

Gas sorption scanning curves are increasingly used as a means to supplement the pore structural information implicit in boundary adsorption and desorption isotherms to obtain more detailed pore space descriptors for disordered solids. However, co-operative adsorption phenomena set fundamental limits to the level of information that conventional scanning curve experiments can deliver. In this work, we use the novel integrated gas sorption and mercury porosimetry technique to show that crossing scanning curves are obtained for some through ink-bottle pores within a disordered solid, thence demonstrating that their shielded pore bodies are undetectable using conventional scanning experiments. While gas sorption alone was not sensitive enough to detect these pore features, the integrated technique was, and, thence, this synergistic method is more powerful than the two individual techniques applied separately. The integrated method also showed how the appropriate filling mechanism equation (e.g. meniscus geometry for capillary condensation equations), to use to convert filling pressure to pore size, varied with position along the adsorption branch, thereby enabling avoidance of the further systematic error introduced into PSDs by assuming a single filling mechanism for disordered solids.

Probing hysteresis during sorption of cyclohexane within mesoporous silica using NMR cryoporometry and relaxometry.

The conventional data analysis methods for obtaining a pore size distribution (PSD) from gas sorption data make several critical assumptions that impact significantly on the accuracy of the PSD thereby obtained. In particular, assumptions must be made concerning the nature of the pore-filling or emptying process in adsorption, or desorption, respectively. The possibility of pore-pore interactions is also generally neglected. In this work, NMR cryoporometry and relaxometry have been used to study the adsorption and desorption of cyclohexane within a mesoporous, sol-gel silica catalyst support pellet with the aim of assessing the impact of the aforementioned problems for gas sorption PSDs and developing solutions. The advanced melting effect makes cryoporometry a particularly sensitive probe of adsorbate ganglia spatial distribution. It has been demonstrated that utilising gas sorption scanning curves provided insufficient additional information to alleviate the aforementioned problems with interpreting gas sorption data. The NMR data has shown how the nature of the sorption hysteresis changed with amount adsorbed, due to detectable variations in the mechanisms of pore-filling and emptying along the isotherm. Hence, relating a particular condensation or evaporation pressure to a specific characteristic pore size is not as straightforward as assumed in typical pore size analysis software. However, the NMR techniques reveal the additional information required to improve pore size estimates from gas sorption for disordered solids.

Investigation of the problems with using gas adsorption to probe catalyst pore structure evolution during coking.

A common approach to try to understand the mechanism of coking in heterogeneous catalysts is to monitor the evolution of the pore structure using gas adsorption analysis of discharged pellets. However, the standard methods of analysis of gas adsorption data, to obtain pore-size distributions, make the key assumption of thermodynamically-independent pores. This assumption neglects the possibility of co-operative adsorption phenomena, which will shown to be a critical problem when looking at coking catalysts. In this work the serial adsorption technique has been used to detect and assess the extent of co-operative effects in adsorption within coking catalysts. The reaction of decane over a hydroprocessing catalyst was used as a case study. It has been shown that the conventional analysis method would lead to a flawed picture of the pore structure changes during the coking process. For the case-study considered in this work, it was found that co-operative adsorption effects meant that 26% of the measured adsorption was occurring in pores up to three times larger than the size conventional analysis would presume. The serial adsorption technique was thus shown to provide important additional information on pore structure evolution during coking. A study of the kinetics of adsorption has been used to infer information about the general spatial location of the coking process within a pellet.

Tissue imprints: assessing their potential for routine biobanking collection.

Biomedical research depends on the availability of good quality biospecimens. Unfortunately, certain specimens are scarce due to disease rarity or size restrictions of surgical materials. To increase access to limited surgical specimens, Biobanks need to reassess and adjust their collection programs. We evaluated the feasibility of adapting "touch imprints" to gain access to limited surgical specimens as well as to maximize the use of "precious" specimens. We utilized 12 kidney samples for touch imprints on defined areas of microscope glass slides and FTA paper. DNA was isolated from glass slides on the day of preparation, Day 0, and from glass slide and FTA paper preparations after two weeks of storage at room temperature and -80°C. Yield and purity of DNA from reference kidney samples were compared to DNA from the touch imprints and the quality determined by real-time PCR using the amplification of Cyclophilin A (Cyc A) as an index. DNA quality for glass slides at Day 0 was not significantly different from DNA after two weeks at room temperature (glass at room temperature; p=0.111 and 0.097, yield and purity, respectively) and after two weeks at -80°C (glass -80°C; p=0.358 and 0.281, yield and purity, respectively). Glass slide DNA at room temperature and -80°C were not significantly different (p=0.795 and 0.146 for yield and purity, respectively). DNA from FTA paper at room temperature and from FTA paper at -80°C were significantly different from glass at room temperature and glass at -80°C (p=0.002, respectively). Threshold values for Cyc A were ≤28 for the reference DNA and ≤32 for DNA from glass and FTA paper. This study demonstrates that touch preparations on microscope glass slides and FTA paper can provide sufficient and good quality DNA suitable for PCR. Touch imprints could therefore be adopted by biobanks to collect and bank biological materials from limited surgical specimens.

Probing the impact of advanced melting and advanced adsorption phenomena on the accuracy of pore size distributions from cryoporometry and adsorption using NMR relaxometry and diffusometry.

The accuracy of pore size distributions (PSD) obtained from gas adsorption and cryoporometry is compromised by the presence of advanced adsorption and advanced melting effects, respectively. In order to improve PSD accuracy, it is necessary to know the extent of such effects. In this work cryoporometry and adsorption have been combined to study the onset of advanced melting effects in a sample partially-saturated with different volumes of condensate, in turn, by pre-equilibration with different vapour pressures of adsorbate. NMR relaxometry and diffusometry have been used to independently study the size and connectivity of adsorbed liquid ganglia at different molten fractions. It has been found that the onset of significant advanced melting coincided with abrupt changes in levels of individual pore-filling and ganglia inter-connections determined by NMR. These findings also highlighted where significant advanced adsorption processes were occurring, where larger pores were being filled with condensate before smaller pores. These studies have enabled the critical pores governing the advanced processes to be identified, and the likely errors in PSDs arising from advanced effects to be quantified.

Determination of the location of coke in catalysts by a novel NMR-based, liquid-porosimetry approach.

In this work, a new technique, suitable for chemically-heterogeneous materials, has been used to characterise the structural properties of porous heterogeneous catalysts. A liquid-liquid exchange (LLE) process within nanoporous catalysts has been followed using NMR relaxometry and NMR diffusometry. In order to validate the new technique, two model materials were used. First, a chemically-pure, sol-gel silica, with a simple, mono-disperse pore-space, was studied. The second model material was a bidisperse, eggshell Pt-alumina catalyst. The Pt-alumina catalyst was studied both fresh, and coked following chemical reaction. The degree of structural and chemical complexity added by coking was restricted by the localisation of the coke deposition to the Pt-eggshell layer. Under so-called 'metered' supply conditions, when a high affinity liquid (water) displaced a low affinity liquid (cyclohexane) from the sol-gel silica, entrapment of the low affinity liquid was observed which was similar to that observed in mercury porosimetry. In a similar experiment, comparing LLE in fresh and coked samples of the Pt-alumina catalyst pellets, it was found, for the fresh sample, that water initially displaced cyclohexane from a sub-set of the most accessible, smallest pores, as might expected under metered conditions, but this did not occur for coked catalysts. This finding suggested coking had removed some smaller pores located close to the surface of the pellet, in agreement with where the Pt-metal was preferentially located and coking was known to have occurred.

Altered strand transfer activity of a multiple-drug-resistant human immunodeficiency virus type 1 reverse transcriptase mutant with a dipeptide fingers domain insertion.

Prolonged highly active anti-retroviral therapy with multiple nucleoside reverse transcriptase inhibitors for the treatment of patients infected with human immunodeficiency virus type 1 (HIV-1) can induce the development of an HIV-1 reverse transcriptase (RT) harboring a dipeptide insertion at the RT fingers domain with a background thymidine analog mutation. This mutation renders viral resistance to multiple nucleoside reverse transcriptase inhibitors. We investigated the effect of the dipeptide fingers domain insertion mutation on strand transfer activity using two clinical RT variants isolated during the pre-treatment and post-treatment of an infected patient, termed pre-drug RT without dipeptide insertion and post-drug RT with Ser-Gly insertion, respectively. First, the post-drug RT displayed elevated strand transfer activity compared to the pre-drug RT, with two different RNA templates. Second, the post-drug RT exhibited less RNA template degradation than the pre-drug RT but higher polymerization-dependent RNase H activity. Third, the post-drug RT had a faster association rate (k(on)) for template binding and a lower equilibrium binding constant K(d) for the template, leading to a template binding affinity tighter than that of the pre-drug RT. The k(off) values for the pre-drug RT and the post-drug RT were similar. Finally, the removal of the dipeptide insertion from the post-drug RT abolished the elevated strand transfer activity and RNase H activity, in addition to the loss of azidothymidine resistance. These biochemical data suggest that the dipeptide insertion elevates strand transfer activity by increasing the interaction of the RT with the RNA donor template, promoting cleavage that generates more invasion sites for the acceptor template during DNA synthesis.

HIV-1 reverse transcriptase dissociates during strand transfer.

Steps in the replication of human immunodeficiency virus type 1 (HIV-1) occurring in the virus but not in the host are preferred targets of antiretroviral therapy. Strand transfer is unique; the DNA strand being made by viral reverse transcriptase (RT) is moved from one RNA template position to another. Understanding the mechanism requires knowing whether the RT directly mediates the template exchange or dissociates during the exchange, so that it occurs by polymer dynamics. Earlier work in vitro showed that the presence of an RT-trapping polymer would allow synthesis on the original or donor template but completely block transfer and subsequent synthesis on the second or acceptor template. One interpretation is that the RT must dissociate during transfer, but an alternative is that sequestration of non-polymerizing RTs prevents polymerization-independent ribonuclease H (RNase H) cleavages of the donor template necessary for strand exchange. To resolve this ambiguity, we designed a primer-template system that allows strand transfer without RNase H activity. Using an RNase H negative mutant RT, we showed that a polymer trap still prevented strand transfer. This confirms that RT dissociates during strand transfer. The presence of HIV-1 nucleocapsid protein, which promotes strand exchange, had little effect on this outcome. Additional assays showed that both the wild-type RT and a multiple nucleoside RT inhibitor-resistant HIV-1 RT containing an extended fingers domain, which is characterized by its enhanced primer-template binding affinity, were unable to transfer with the trapping polymer. This implies that common sequence variations among RTs are unlikely to alter dissociation feature.

Selective incorporation of functional dicarboxylates into zinc metal-organic frameworks.

Zinc(II) nitrate reacts with different ratios of 1,4-benzenedicarboxylic acid (H(2)bdc) and 2-halo-1,4-benzenedicarboxylic acid (H(2)bdc-X, X = Br or I) to give [Zn(4)O(bdc)(3-x)(bdc-X)(x)], in which preferential incorporation of bdc is observed. The selective incorporation is related to crystal growth rates, and the proportion of incorporated bdc-X rises with increasing reaction time.

NMR studies of cooperative effects in adsorption.

The conversion of gas adsorption isotherms into pore size distributions generally relies upon the assumption of thermodynamically independent pores. Hence, pore-pore cooperative adsorption effects, which might result in a significantly skewed pore size distribution, are neglected. In this work, cooperative adsorption effects in water adsorption on a real, amorphous, mesoporous silica material have been studied using magnetic resonance imaging (MRI) and pulsed-gradient stimulated-echo (PGSE) NMR techniques. Evidence for advanced adsorption can be seen directly using relaxation time weighted MRI. The number and spatial distributions of pixels containing pores of different sizes filled with condensate have been analyzed. The spatial distribution of filled pores has been found to be highly nonrandom. Pixels containing the largest pores present in the material have been observed to fill in conjunction with pixels containing much smaller pores. PGSE NMR has confirmed the spatially extensive nature of the adsorbed ganglia. Thus, long-range (≥40 µm) cooperative adsorption effects, between larger pores associated with smaller pores, occur within mesoporous materials. The NMR findings have also suggested particular types of pore filling mechanisms occur within the porous solid studied.

MF-DFT and experimental investigations of the origins of hysteresis in mercury porosimetry of silica materials.

In order to be able to make a proper interpretation of mercury porosimetry data, to obtain a structural characterization of a porous solid, a full understanding of the causes of hysteresis in mercury porosimetry is required. Several different theories have previously been proposed, but it is still difficult to make a priori predictions of the level of hysteresis anticipated. In this work, the effect of the degree of smaller scale surface roughness on the hysteresis width has been studied using mean-field density functional simulations and the results obtained confirmed by experiments on silica materials. It has been found that the hysteresis width decreases with increased degree of surface roughness, as characterized experimentally by the surface fractal dimension.