Porous properties of mesoporous silicas from two silica sources (acid-leached kaolinite and Si-alkoxide)

Two different mesoporous silicas (MesoPS) were synthesized by hydrothermal treatment in NaOH solution of two silica sources. One of these starting silicas was derived from selectively acid leached metakaolinite, and the other was from tetraethylorthosilicate (TEOS). The syntheses used a surfactant, cethyltrimethylammonium bromide (CTABr) and were carried out at different CTABr/Si, NaOH/Si and H₂O/Si ratios. In the MesoPS from kaolinite, the specific surface areas (S _BET) of the products were >1500 m²/g when prepared with 0.2 ≤ CTABr/Si ≤ 0.4, 0.3 ≤ NaOH/Si ≤ 0.6 and H₂O/Si = 150. These S _BET values are higher than those obtained from TEOS (ca. 1300 m²/g). The XRD patterns of these products contain a hexagonal (10-) peak with a lattice parameter a ₀ = 4.2–5.2 nm in the MesoPS derived from kaolinite and a ₀ = 4.0–4.6 nm in the products from TEOS. The regularity of the hexagonal structure is higher in the MesoPS derived from TEOS than from kaolinite. The pore size distributions of the products show a sharp peak at 2.8 nm in the MesoPS from kaolinite and at 2.4–2.5 nm in those from TEOS. The meso-structure is found to be formed during the stirring step of the synthesis and becomes more regular after hydrothermal treatment. The differences in the porous properties of the two MesoPSs from kaolinite and TEOS are attributed to differences in the dissolution rates and silica concentrations in the synthesis solutions. The ²⁹Si MAS NMR spectra show a higher number of Q ³ (Si(OSi)₃OH) units in the MesoPS from kaolinite and this is suggested to be related to the difference in their porous properties.     

Characterization of sepiolite as a support of silver catalyst in soot combustion

Turkish sepiolite–zirconium oxide mixtures were applied as a support for the silver catalyst in a soot combustion. Sepiolite–Zr–K–Ag–O catalyst was characterized by XRD, N₂ adsorption, SEM, TPR-H₂ and EGA-MS. The combustion of soot was studied with a thermobalance (TG-DTA). The modification resulted in a partial degradation of the sepiolite structure, however, the morphology was preserved. The adsorption of N₂ of the modified sepiolite is a characteristic for mesoporous materials with a wide distribution of pores. The specific surface area S_BET equals 83 m²/g and the pores volume is 0.23 cm³/g. The basic character of the surface centers of sepiolite is indicated by CO₂ desorption (TPD-MS) at 170 °C and at about 620 °C due to a surface carbonates decomposition. The thermodesorption of oxygen at 650–850 °C indicates the decomposition of AgO_x phases at the surface. The presence of AgO_x phases is also confirmed by TPR-H₂ spectrum (low temperature reduction peak at 130 and 180 °C). The high-temperature reduction at about 570 °C is probably related to Ag–O–M phases on the support.The soot combustion takes place at T₅₀ = 575 °C. Without silver (sepiolite–Zr–K–O) T₅₀ = 560 °C but sepiolite modified with silver (sepiolite–Zr–K–Ag–O) undergoes the same process at T₅₀ = 490 °C.     

Adsorption and photocatalytic properties of TiO<Subscript>2</Subscript>/mesoporous silica composites from two silica sources (acid-leached kaolinite and Si-alkoxide)

Two different mesoporous silicas (MPS) were synthesized by hydrothermal treatment in NaOH solution of two SiO₂ sources. These were microporous silica (MicroPS) derived from selectively acid leached metakaolinite and tetraethylorthosilicate (TEOS). The hydrothermal syntheses of the MPSs were performed at a ratio of SiO₂/cetyltrimethyl- ammonium bromide (CTABr)/NaOH/H₂O = 1/0.1/0.3/150. The specific surface areas (S _BET) of the MPSs from MicroPS (MPS(M)) and TEOS (MPS(T)) were 1070 and 1020 m²/g, respectively. Composites of MPS (75 mass%) with TiO₂ (25 mass%) were prepared using both SiO₂ and two commercial TiO₂ powders, P25 and ST-01. The adsorption–desorption behavior of methylene blue (MB) by the four resulting composites and the two MPSs alone was unique in showing partially reversible behavior. The maximum MB adsorption, observed in the composite of ST-01 with MPS(M), designated (S/M), was 0.034 mmol/g. The rates of MB adsorption in the dark and photodecomposition under UV illumination were considerably different for the four composites and two TiO₂ powers, and followed the order ST-01 < S/T < P25 < P/T ≈ P/M ≪ S/M. The removal rate of MB by the composite S/M by adsorption and photodecomposition was further enhanced by heating at 700 °C. Direct photodecomposition of MB without adsorption in the dark was greatly enhanced in the composites, especially in that composed of MPS(M) and ST-01.     

Examination of the surface properties of kaolinites by inverse gas chromatography: Dispersive properties

The application of inverse gas chromatography (IGC) for the examination of the surface properties of untreated kaolinite and kaolinites surface-treated with titanate coupling agents is discussed. This paper presents and discusses the dispersive properties expressed by γ_S^D, the dispersive component of the surface free energy, as determined at various temperatures. The γ_S^D values of all the samples were negatively correlated with the temperature of IGC measurement. At the same temperature of IGC measurement, the values of γ_S^D determined by IGC were significantly lower for the surface-treated kaolinites than for the untreated ones. The γ_S^D value of the kaolinite surface-treated with isopropyl triisostearoyl titanate (ITT) at 150°C exhibited a surface energy close to that (28.3 mJm^-2) of polyethylene with no surface polarity.     

Mineralogical and morphological changes of calcined paper sludge at different temperatures and retention in furnace

This work reports the project of thermal effects in mineralogy and texture of the paper sludge to employ them as pozzolanic material.For this reason, the chemistry and mineralogical composition has been studied, as well as its morphology by XRD, SEM and EDX.The initial sludge has been treated to 700, 750 and 800 °C during 2 and 5 h being observed that initial kaolinite becomes metakaolinite and that its pozzolanic activity with a paper sludge treated to 700 °C for 2 h is comparable to that of a commercial metakaolinite. The transformation of kaolinite after the dehydroxylation is to convert in amorphous metakaolinite. At the temperature mentioned above, calcite from the initial sludge is maintained active.It is concluded that the pozzolanic activity of metakaolinite is strongly related to the crystallinity of the original kaolinite. Well-ordered kaolinite is transformed into more reactive metakaolinite.     

Algerian kaolinite used for mullite formation

In the present study, mullite was synthesized through reaction sintering of Algerian kaolinite and high purity alumina. The raw powders were wet ball milled in a planetary ball mill. Powders' morphology and the microstructure of the sintered samples were characterized by means of a scanning electron microscope. An X-ray diffractometer equipped with a heating facility and a differential thermal analyzer were used to follow mullite formation. Cylindrical specimens were produced by uniaxial cold compaction at a pressure of 75 MPa and sintered at different sintering temperatures for different sintering times. The heating rate was 10 °C/min. It was found that Algerian kaolinite was suitable for mullite production through reaction sintering with pure Al₂O₃. Formation of complete mullite occurred at 1550 °C. A relative density of 94% (of the theoretical density) was achieved at a relatively low sintering temperature of 1600 °C and a sintering time of 4 h.     

Optimization of mesoporous alumina sol-gel synthesis by Taguchi and response surface methodology

Taguchi method (TM) and response surface methodology (RSM) have been employed to optimize three parameters, including the amounts of P123, the amounts of nitric acid and calcination temperature, in order to define an optimal setting for sol-gel synthesis of high surface area mesoporous alumina powder (MA). Herein, the comparison of the both statistical approaches has been examined and discussed considering the nitrogen adsorption as the response variable because this important character for mesoporous materials is exceedingly sensitive to the synthesis parameters. The BET surface area (S_BET) and pore volume of MA under Taguchi optimal condition were 323.5 m² g⁻¹ and 0.551 cm³ g⁻¹, respectively, by conducting confirmation test. Furthermore, the confirmation test showed high S_BET of MA (363.4 m² g⁻¹), which was in a good agreement with calculated S_BET result (431.25 m² g⁻¹) by a quadratic model under RSM optimal condition. Moreover, 3D response surface plots and 2D contour plots of desirability have been discussed to visualize the influence of input factors on response variable. It is also concluded that RSM shows more appropriate (12.33% higher S_BET than TM) and efficient optimal condition with determining a quadratic function as the relationship between S_BET and synthesis parameters.     

Dissolution kinetics of dehydroxylated nickeliferous goethite from limonitic lateritic nickel ore

The dissolution kinetics in 2 M H₂SO₄ of variously dehydroxylated nickeliferous goethites was investigated for five oxide-type lateritic nickel deposits. Goethite was the main constituent with minor amounts of quartz, talc, kaolinite and Mn oxides. Dissolution of Fe from heated materials followed the Kabai equation. There was a 9–34-fold increase in the Kabai dissolution rate constant (k) for samples heated at 340–400 °C due to both the increased surface area (1.5–2.6 fold) and higher density of structural defects (5–10 fold) in the variously dehydroxylated products. The presence of structural Al and Cr in goethite appears to reduce dissolution rate possibly through the greater M³⁺–OH, O bond strength relative to Fe³⁺, Ni²⁺–OH, O. Nickel showed congruent dissolution with Fe indicating that Ni was uniformly incorporated in the goethite structure. Pre-heating goethite to 600–800 °C for 30 min resulted in incongruent dissolution of Fe and Ni. It is postulated that some Ni is ejected from the neo-formed hematite structure and resides on the crystal surface or in voids. These results may contribute to the development of more efficient procedures for Ni extraction including heap leaching of lateritic nickel ores.     

Hydrothermal synthesis (200 °C) of Co–kaolinite and Al–Co–serpentine

In the systems CoO–Al₂O₃–SiO₂–H₂O and CoO–Al₂O₃–SiO₂–HCl–H₂O, at initial pH between 5.5 and 8.1 and temperature of 200 °C, kaolinite is unstable and the following phases form through a dissolution-precipitation process: a) kaolinite and Co-bearing kaolinite; b) Al–Co–serpentine; and c) poorly crystalline phases. Identification of the several phases was carried out from a combination of X-ray diffraction and transmission/analytical electron microscopy.Co–kaolinite shows variable morphologies: a) Platy lath-shaped particles with very low Co content; b) Spherical particles, with relatively constant Co contents (in the order of 0.10 apfu); c) Kaolinite stacks with very variable Co contents (up to 0.25 apfu). Analytical data indicate that the presence of Co(OH)₂ in the system favors the dissolution process as well as serpentine formation but it leads to the parallel formation of abundant poorly crystalline phases. The Co-content in kaolinite increased as a function of the Co(OH)₂/CoCl₂ ratio in the initial systems, and it is reflected by a parallel increase of the b-cell parameter of kaolinite. The average composition of the coexisting Al–Co–serpentine is: (Al_1.20Fe_0.11Co_1.27)(Si_1.64Al_0.36)O₅(OH,Cl)₂, with Cl contents in the order of 0.14 apfu.The assemblage Co–kaolinite + Al–Co–serpentine, which appears to be stable at 200 °C, has not been described in natural environments, probably because it requires unusual Al- and Co-rich chemical systems.     

Preparation of porous material from talc by mechanochemical treatment and subsequent leaching

Synthesis of porous silica via mechanochemical treatment of talc and subsequent acid leaching was investigated by X-ray diffraction (XRD), Fourier transformation infrared spectroscopy (FTIR), scanning electron microscopy (SEM) and N₂ adsorption techniques. Raw talc was ground for different times and then leached with 4 M hydrochloric acid (HCl) at 80 °C. Grinding for 6 h and subsequent leaching for 2 h of raw talc produced the porous silica with a specific surface area of 133 m²/g and total pore volume of 0.22 ml/g. The increase in specific surface area (S_BET) of the porous silica reflected the formation of micropores of 1.21.8 nm and mesopores of 4.05.5 nm in diameter inside the porous structure. The number of micropores decreased with prolonged leaching time, which can be attributed to a condensation reaction. The characteristic of hysteresis loop indicated mainly slit-shaped pores. The apparent activation energy for the leaching process was calculated to be about 21.6 kJ/mol, indicating that the kinetic process of talc leaching was the diffusion-controlled reaction. Mechanochemical treatment may promote the amorphization of talc, being greatly favorable to the subsequent leaching.     

Influence of surfactants on porous properties of carbon cryogels prepared by sol–gel polycondensation of resorcinol and formaldehyde

The role of surfactants on carbon cryogels is investigated by using three different surfactants, nonionic (SPAN80), cationic (trimethylstearylammonium chloride; C18) and nonionic polymeric fluorinated (FC4430) surfactants. By using different SPAN80 concentrations (10.0, 5.0, 2.5, 1.0 and 0.5 vol.%), double-structure carbon microspheres with S_BET (630–700 m²/g) and V_mes (0.51–0.93 cm³/g) are obtained. Mesoporous carbon cryogels with different S_BET and V_mes are prepared by using C18 with different volume ratios of cyclohexane to water in a C18/water/cyclohexane mixture. Carbon cryogels with S_BET (690–810 m²/g) and V_mes (0.83–1.74 cm³/g) are obtained when cyclohexane is contained in the mixture, on the contrary, when there is no cyclohexane in the mixture, a water-based carbon cryogel with low S_BET (480 m²/g) and V_mes (0.29 cm³/g) is obtained. Carbon cryogels prepared by using C18 have larger mesopore size and broader mesopore size distribution compared with carbon cryogels prepared by using other surfactants. Microcellular (sponge-like) carbon cryogels with mesoporous surface, S_BET (210–660 m²/g) and V_mes (0.37–0.92 cm³/g), are obtained by introducing FC4430 (two concentrations) to two starting RF solutions (C/W=6,45). Low FC4430 concentration leads to carbon cryogels with higher S_BET (610 and 660 m²/g) and narrower mesopore size distributions compared to the high concentration counterpart. Hence, it is found that different surfactant types have interesting effects on morphologies and porous properties of RF carbon cryogels.     

Influence of the type of sepiolite on the modification of the pore-size distribution in γ-Al2O3 supports

γ-Al₂O₃ modified supports with bimodal pore-size distributions were prepared by the addition of different types of natural sepiolites (α or β) into alumina. The supports were characterized by nitrogen physisorption, mercury porosimetry, X-ray diffraction, HRTEM and DTA techniques. A wide range of S_BET (94–238 m² g^− 1), pore volumes (0.3–0.82 cm³ g^− 1), and pore sizes were obtained in the supports depending on the type of sepiolite and its concentration added into alumina. The pore sizes were distributed as follows: mesopores around 1.8 nm in radius, mesopores in the radius range 3.0–25 nm and macropores between 25 and 300 nm in radius. The shape of the pore-size distributions depended on the type of sepiolite: the modal peak for pores larger than 3.0 nm was broad with β-type sepiolites and narrow with α-type sepiolites. The mesopore and macropore sizes can be controlled by the type of sepiolite as well as its concentration added to alumina.     

Morphology and structure of YSZ powders: Comparison between xerogel and aerogel

Yttria-stabilized zirconia (YSZ) fine powders were prepared via sol–gel route in order to shape thermal barrier coatings (TBCs) from these powders. The main objective is to develop new undirectional coatings to allow best thermo-mechanical accommodations compared to conventional processes. To reach this aim, powders have to be able first to be highly dispersed into a sol (non-agglomeration, high specific surface area, etc.) and second to crystallize in the required metastable phase t′. Two routes have been used to dry gels: the conventional way which consists of simple evaporation of the solvent is compared to drying in supercritical conditions. Both YSZ powders after calcination at 950 °C of xerogel (Ex-xero-YSZ powder) and aerogel (Ex-aero-YSZ powder) crystallize in the tetragonal form. N₂ adsorption/desorption analysis of the Ex-xero-YSZ powder indicates an S_w of 2.8 m²/g. For the Ex-aero-YSZ powder, the S_w (26 m²/g) is much higher than of Ex-xero-YSZ, leading to a better sintering capability. This high S_w is correlated to the small crystallite size (26 nm) and the alveolar morphology of Ex-aero-YSZ powders compared to Ex-xero-powder (49 nm). By reducing particles size and increasing the S_w of the powders, supercritical drying appears as a promising way to prepare stable slurries or loaded sols from fine YSZ particles for TBC applications. Indeed, after preparing nanometric powders, they are dispersed into a sol before shaping on superalloys substrates. After thermal treatment at 950 °C for 2 h which corresponds to the working temperature of TBC, the final aim will be to prepare ceramic YSZ coatings.     

Preparation of activated carbon derived from Jatropha curcas fruit shell by simple thermo-chemical activation and characterization of their physico-chemical properties

High surface area activated carbons were prepared by simple thermo-chemical activation of Jatropha curcas fruit shell with NaOH as a chemical activating agent. The effects of the preparation variables, which were impregnation ratio (NaOH:char), activation temperature and activation time, on the adsorption capacity of iodine and methylene blue solution were investigated. The activated carbon which had the highest iodine and methylene blue numbers was obtained by these conditions as follows: 4:1 (w/w) NaOH to char ratio, 800 °C activation temperature and 120 min activation time. Characterization of the activated carbon obtained was performed by using scanning electron microscopy (SEM), X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FT-IR) and nitrogen adsorption isotherm as BET. The results present that the activated carbon possesses a large apparent surface area (S_BET = 1873 m²/g) and high total pore volume (1.312 cm³/g) with average pore size diameter of 28.0 Å.     

Synthesis of onion-like mesoporous silica from sodium silicate in the presence of α,ω-diamine surfactant

Mesoporous silicas with vesicular and onion-like morphologies were assembled through hydrogen-bonding pathway from sodium silicate as silica source and electrically neutral α,ω-diamine, Jeffamine D2000 surfactant (H₂NCH(CH₃)CH₂[OCH₂CH(CH₃)]₃₃NH₂) as template in aqueous media at different synthesis temperatures (25, 60 and 100 °C). Assembling the material at 100 °C afforded onion-like core shell mesoporous silica, while at relatively lower temperature, e.g. 25 and 60 °C, multilamellar vesicles were obtained. Mesoporous silica with onion-like morphology was also obtained by a two-step synthesis involving an aging period of 20 h at room temperature followed by a hydrothermal stage (1–12 h) at 100 °C. The heavily cross-linked (Q⁴/Q³ ratio of 4.43) onion-like mesophase silica exhibited high hydrothermal stability. The BET surface area, pore volume and KJS (Kruk-Jaroniec-Sayari) pore diameter of the onion-like mesoporous silica were found to be 464 m² g⁻¹, 1.16 m³ g⁻¹ and 7.2 nm, respectively.     

Chitosan intercalated montmorillonite: Preparation, characterization and cationic dye adsorption

Chitosan intercalated montmorillonite (Chi-MMT) was prepared by dispersing sodium montmorillonite (Na⁺-MMT) into chitosan solution at 60 °C for 24 h. The Chi-MMT was characterized by XRD, XRF and FT-IR. The intercalation was accomplished via the ion-exchange of Na⁺ ions with –NH₃⁺ of chitosan, resulting in the expansion of d₀₀₁ from 1.42 nm of Na⁺-MMT to 2.21 nm of Chi-MMT. The chitosan content in the Chi-MMT measured by TGA was about 17 mass%. The adsorption capacity of Chi-MMT was investigated in comparison with the starting Na⁺-MMT and chitosan using three different cationic dyes, i.e. basic blue 9 (BB9), basic blue 66 (BB66) and basic yellow 1 (BY1). The Chi-MMT showed the highest adsorption capacity in the range of 46–49 mg/g when the initial dye concentration was 500 mg/L, being equivalent to 92–99 wt.% of dye removal. The adsorption capacities of Chi-MMT for all basic dyes increased with an increase of initial dye concentration. An increase of adsorption capability of Chi-MMT was attributed to the existence of intercalate-chitosan. It could enlarge the pore structure of Chi-MMT, facilitating the penetration of macromolecular dyes, and also electrostatically interact with the applied dyes. These results indicated the competency of Chi-MMT adsorbent for basic dye adsorption.     

Chlorobenzene total oxidation over palladium supported on ZrO2, TiO2 nanostructured supports

Two series of supported Pd catalysts were synthesized on new mesoporous–macroporous supports (ZrO₂, TiO₂) labelled M (Zr and Ti). The deposition of palladium was carried out by wet impregnation on the calcined TiO₂ and ZrO₂ supports at 400 °C (Pd/Zr₄, Pd/Ti₄) and 600 °C (Pd/Zr₆, Pd/Ti₆) and followed by a calcination at 400 °C for 4 h. The pre-reduced Pd/MX catalysts were investigated for the chlorobenzene total oxidation and their catalytic properties where compared to those of a reference catalyst Pd/Ti-Ref (TiO₂ from Huntsman Tioxide recalcined at 500 °C) and of a palladium supported on the fresh mesoporous–macroporous TiO₂ (Pd/Ti). Based on the activity determined by T₅₀, the Pd/Ti and Pd/Ti₄ catalysts have been found to be more active than the reference one. Moreover activity decreased owing to the sequence: Pd/TiX  Pd/ZrX and in each series when the temperature of calcination of the support was raised. The overall results clearly showed that the activity was dependant on the nature of the support. The better activity of Pd/TiX compared to Pd/ZrX was likely due to a better reducibility of the TiO₂ support (Ti⁴⁺ into Ti³⁺) leading to an enhancement of the oxygen mobility. Production of polychlorinated benzenes PhCl_x (x = 2–6) and of Cl₂ was also observed. Nevertheless at 500 °C the selectivity in HCl was higher than 90% for the best catalysts.     

A study of the mineralogical alteration of bentonite in saline water

The mineralogical alteration of bentonite was studied in saline water at 60 and 90 °C. To obtain the altered bentonite, a crude bentonite (Kunigel V1) was dispersed into simulated sea-water, and maintained at 60 or 90 °C over 6000 h. Magnesium was accumulated into bentonite during the experiments at levels over the cation exchange capacity (CEC; 0.78 meq/g for the original bentonite) at only 90 °C. The excess Mg was not replaced by ammonium ions, suggesting that the precipitation of Mg solid in bentonite occurred. The IR spectra and the thermogravimetric/differential Thermal Analysis (TG-DTA) profiles show the neo-formation of magnesium hydroxide in the altered bentonite. The CEC decreased from 0.78 to 0.45 meq/g as the amount of accumulated Mg increased. The distribution coefficient K_d for Cs in the altered bentonite was half of that in the original bentonite. It is conjectured that the thermal alteration of bentonite in saline water affects the sorption capacity of Cs onto smectite.     

Origins of the Rio Capim kaolinites (northern Brazil) revealed by δO and δD analyses

Deuterium (δD) and oxygen (δ¹⁸O) isotope data from the Rio Capim kaolin, northern Brazil, were combined with optical studies in order to better understand the genesis and evolution of the kaolinites. The results show that δ¹⁸O values from a lower soft kaolin unit range from 6.0‰ to 19.2‰ for Ka (size ranging from 1 to 3 μm) and Kb (size ranging from 10 to 30 μm) kaolinites, and from 15.4‰ to 24.9‰ for Kc (size < 200 nm) kaolinites. The δD values range from − 63.1‰ to 79.5‰ for the Ka + Kb kaolinites, and from − 68.8‰ to − 244.35‰ for the Kc kaolinites. An upper semi flint kaolin unit, dominated by Kc kaolinites, displays δ¹⁸O and δD values ranging from 15.1‰ to 21.8‰, and − 71.3‰ to − 87.4‰, respectively. Based on these data, and on the δ¹⁸O and δD values obtained for the surface meteoric water and groundwater, it can be concluded that the kaolinites are not in equilibrium with the modern weathering environment, but they reflect isotopic compositions of the formation time, probably due to the interaction with fossil groundwater. However, mineralogical contaminations derived from replacements of framework grains also had great influence in the isotopic composition of these kaolinites. In addition, the isotope values of the Kc kaolinites from the semi-flint kaolin unit is variable, which is due to the presence of Kc kaolinites of different origins, including kaolinites derived from the underlying soft kaolin unit, kaolinites formed during different phases of paleoweathering, as well as later phases of coarse-grained kaolinites formed along fractures. Due to these complexities, binary diagrams contrasting δ¹⁸O and δD values, worldwide applied for distinguishing supergenic from hypogenic kaolinites, as well as those formed under weathering conditions, can not be applied to interpret the origin of the kaolinites in the Rio Capim Kaolin.     

Adsorption and binding of the transgenic plant proteins, human serum albumin, β-glucuronidase, and Cry3Bb1, on montmorillonite and kaolinite: Microbial utilization and enzymatic activity of free and clay-bound proteins

Human serum albumin (HSA), β-glucuronidase (GUS), and the Cry3Bb1 protein from Bacillus thuringiensis subsp. kumamotoensis are expressed by genetically-modified plants. Commercial samples of these proteins adsorbed and bound rapidly on the clay minerals, kaolinite (K) and montmorillonite (M). Adsorption increased as the concentration of protein increased and then reached a plateau. The greatest amount of adsorption and binding occurred with the Cry3Bb1 protein, of which there was no desorption: 6.7 ±0.21 μg adsorbed and bound μg^− 1 of M; 2.1 ± 0.39 μg adsorbed and bound μg^− 1 of K. With GUS, 2.2 ± 0.29 μg adsorbed and 1.7 ±0.21 μg bound μg^− 1 of M; 1.5 ± 0.28 μg adsorbed and 1.0 ± 0.03 μg bound μg^− 1 of K. HSA was adsorbed and bound the least: 1.2 ±0.04 μg adsorbed and 0.8 ± 0.05 μg bound μg^− 1 of M; 0.4 ± 0.05 μg adsorbed and 0.4 ± 0.03 μg bound μg^− 1 of K. However, X-ray diffraction analyses indicated that only HSA intercalated M, and none of the proteins intercalated K, a nonswelling clay. When bound, the proteins were not utilized for growth by mixed cultures of soil microorganisms, whereas the cultures readily utilized the free (i.e., not adsorbed or bound) proteins as sources of carbon and energy. The enzymatic activity of GUS was significantly enhanced when bound on the clay minerals. These results indicated that recombinant proteins expressed by transgenic plants could persist and function in soil after release in root exudates and from decaying plant residues as the result of the protection provided against biodegradation by binding on clay minerals.