It is unclear whether all intercalation techniques truly lead to the insertion of atoms between the graphite layers, or also lead to other effects which contribute to expansion. The objective of this project is to better understand the effects caused by different intercalation methods. Three intercalation methods were explored to determine the method which incurs the least damage to the surface and microstructure of the graphite intercalated compounds, yet achieves the best intercalation and therefore expansion. All the main findings are summarised below:  The gas phase sample had virtually no mass loss at the point where expansion took place. Therefore the intercalation was very efficient, producing large expansion without significant mass loss.  The mass loss that only occurs at the sublimation of iron chloride (320 ºC) indicates the excessive "un-intercalated" or residual iron chloride.  After oxidation, before purification, the gas phase sample has 25 % residual mass; this also proves the presence of impurities and residual iron chloride in the exfoliated sample. For the Hummers and electrochemical samples, expansion and mass loss occur over a wide temperature range, this indicates that graphite oxide was formed rather than the theoretically expected "insertion of atoms between the sheets".  The mass losses before 200 ˚C of the samples of the Hummers and electrochemical methods are more evidence that graphite oxide and graphite surface complexes with oxygen were produced.  The Hummers and electrochemical intercalation methods show similar expansion and mass loss curves, therefore it can be concluded that the reaction mechanism for both these methods is alike.  The gas phase method yields the best expansion of 250 % using the TMA, whereas both the other methods deliver approximately 220 %.  Using microwave expansion the electrochemical intercalation method yields the best bulk volume expansion of 1500 %, with the gas phase sample delivering a volume expansion of 1450 %. The Hummers samples are extremely damaged. This is clear from the several and deep oxidation pits visible throughout the basal plane of these samples. The basal plane and the edges are even eroded before purification and oxidation. This intercalation technique employs oxidisers in the preparation method which additionally oxidises the samples. This explains why the Hummers method renders the most damage. The residual material on the gas phase sample acts as catalysts making the sample very reactive and consequently damaging the surface during oxidation. The partially oxidised purified gas phase sample visibly shows the pits and roughened edges. There are two “types” of intercalation. The first intercalation “type” is the actual insertion of atoms or molecules between the graphite layers, whereas the other “type” of intercalation is the production of graphite oxide. The compound comprises carbon, oxygen and hydrogen, obtained by treating graphite with strong oxidisers. The functional groups usually found in graphite oxide are carbonyl (C=O), hydroxyl (-OH), phenol amongst others and also some impurities of sulphur when sulphuric acid is used. Both these intercalation types lead to expansion. It is recommended that a more efficient method for removal of residual material in the gas phase samples be explored. It is also recommended that more research be done to determine the reaction mechanisms during the three different intercalation methods. The graphite surface complexes of the intercalated compounds and the evolved gases during expansion should be analysed. / Dissertation (MEng)--University of Pretoria, 2015. / tm2015 / Chemical Engineering / MEng / Unrestricted

Intercalation is known to be an efficient method for tuning the band structure of layered materials to bring out superconductivity, without significantly altering the crystal structure of the host material. Graphite intercalation compounds and intercalated transitional metal dichalcogenides (TMDs) are two most studied representatives. This thesis presents an experimental study of several new superconductors obtained by intercalation of layered materials, including MoS2, black phosphorus and a topological insulator Bi2Se3. Polymorphism is an essential feature of MoS2. While, superconductivity in doped 2H-MoS2 has been extensively studied. Superconductivity in its 1T and 1T' counterparts has been neither observed, nor even predicted theoretically. In this thesis, we have investigated potassium (K)-intercalated MoS2 and found that doping with K induces both structural and superconducting phase transitions. We demonstrate that all three phases of MoS2 - 2H, 1T and 1T'- become superconducting as a result of intercalation, with critical temperature Tc of 6.9 K, 2.8 K and 4.6 K, respectively. Black phosphorus has been 'rediscovered' in the last few years due to its layered structure and unique electronic properties. This thesis describes successful intercalation of black phosphorus with several alkali metals (Li, K, Rb, Cs) and alkaline earth metal Ca, with all five compounds showing superconductivity. Importantly, and very unexpectedly, the found superconductivity of intercalated black phosphorus is independent of the intercalant, with all five compounds having exactly the same superconducting characteristics (Tc, critical fields, anisotropy). We suggest that the superconductivity is due to heavily doped phosphorene layers, with alkali metal atoms acting mainly as charge reservoirs. Superconducting topological insulators, such as Bi2Se3, are regarded as the most promising candidates for topological superconductivity. However, the nature of superconductivity in doped Bi2Se3, such as CuxBi2Se3, SrxBi2Se3 and NbxBi2Se3, remains controversial and so far no convincing evidence of topological superconductivity has been reported for these materials. In this thesis, we report superconductivity in a new family of superconductors derived from Bi2Se3, by intercalation with K, Rb and Cs metals. All three superconductors exhibit qualitatively identical but highly anomalous behaviour of magnetisation, with several new features consistent with the properties of topological superconductors. Specifically, the new materials exhibit a highly unusual extra diamagnetic screening in the Meissner state and two coexisting superconducting phase, including surface superconductivity that we attribute to heavily doped surface states of the original topological insulator (Bi2Se3). This work provides a new platform in the study of the interplay between the topological and superconducting orders. In conclusion, superconductivity has been induced in MoS2, black phosphorus and Bi2Se3 through alkali or alkaline earth metal intercalation. The study of these new superconducting materials has been summarised in the thesis.

Sepiolite is a clay mineral, a complex magnesium silicate, a typical formula for which is (OH2)4(OH)4Mg8Si12O30•8H2O. It is formed by blocks and cavities (tunnels) growing in the direction of the fibres. The tunnels, 3.7 x 10.6 Å in cross-section, are responsible for the high specific surface area and sorptive properties of sepiolite. The co-intercalation of 3-methyl cyclohex-2-en-1-one (MCH), the Douglas-Fir beetle anti-aggregation pheromone, with methanol, ethanol, acetone, or benzene into sepiolite tunnels was studied. The resulting nanohybrid materials were characterized by means of various techniques, such as multinuclear solid-state NMR spectroscopy, porosity studies and Thermal Gravimetric Analysis (TGA). This was done in the hope of obtaining slow and controlled release of MCH from the sepiolite tunnels. It was demonstrated by 13C MAS NMR (carbon-13 magic angle spinning nuclear magnetic resonance) that at room temperature there are two different MCH molecules: one MCH inside the tunnels and the other one outside the tunnels of the sepiolite. Heating nanohybrid materials at 60˚C for 20 hours removes the external MCH molecules from the sepiolite. 13C MAS NMR showed that by further heating nanohybrid materials at 120˚C for 20 hours, methanol, ethanol, or acetone peaks were greatly reduced; however, the benzene peak was not reduced. To better understand how benzene acts inside sepiolite, intercalation of d6-benzene, and co-intercalations of d6-benzene with MCH and d6-benzene with pyridine into sepiolite tunnels were carried out, and these samples were studied by the same techniques. Another technique was used in order to see whether the slow and controlled release of MCH from the sepiolite tunnels could be obtained: sepiolite-MCH nanohybrids were treated with 20 ml of 0.5 M HCl solution. It was found that when 1 gram of MCH-sepiolite sample was acid treated at room temperature, about 35% of intercalated MCH was removed from the sepiolite. The role of sepiolite clay was also studied in Maya-Blue representative structure sepiolite-indigo adduct. It is known that upon heating the sepiolite and indigo mixture, the stability that is present in Maya-Blue is achieved. It is still a mystery, however, how exactly indigo and sepiolite interact with each other.

Sepiolite is a clay mineral, a complex magnesium silicate, a typical formula for which is (OH2)4(OH)4Mg8Si12O30•8H2O. It is formed by blocks and cavities (tunnels) growing in the direction of the fibres. The tunnels, 3.7 x 10.6 Å in cross-section, are responsible for the high specific surface area and sorptive properties of sepiolite. The co-intercalation of 3-methyl cyclohex-2-en-1-one (MCH), the Douglas-Fir beetle anti-aggregation pheromone, with methanol, ethanol, acetone, or benzene into sepiolite tunnels was studied. The resulting nanohybrid materials were characterized by means of various techniques, such as multinuclear solid-state NMR spectroscopy, porosity studies and Thermal Gravimetric Analysis (TGA). This was done in the hope of obtaining slow and controlled release of MCH from the sepiolite tunnels. It was demonstrated by 13C MAS NMR (carbon-13 magic angle spinning nuclear magnetic resonance) that at room temperature there are two different MCH molecules: one MCH inside the tunnels and the other one outside the tunnels of the sepiolite. Heating nanohybrid materials at 60˚C for 20 hours removes the external MCH molecules from the sepiolite. 13C MAS NMR showed that by further heating nanohybrid materials at 120˚C for 20 hours, methanol, ethanol, or acetone peaks were greatly reduced; however, the benzene peak was not reduced. To better understand how benzene acts inside sepiolite, intercalation of d6-benzene, and co-intercalations of d6-benzene with MCH and d6-benzene with pyridine into sepiolite tunnels were carried out, and these samples were studied by the same techniques. Another technique was used in order to see whether the slow and controlled release of MCH from the sepiolite tunnels could be obtained: sepiolite-MCH nanohybrids were treated with 20 ml of 0.5 M HCl solution. It was found that when 1 gram of MCH-sepiolite sample was acid treated at room temperature, about 35% of intercalated MCH was removed from the sepiolite. The role of sepiolite clay was also studied in Maya-Blue representative structure sepiolite-indigo adduct. It is known that upon heating the sepiolite and indigo mixture, the stability that is present in Maya-Blue is achieved. It is still a mystery, however, how exactly indigo and sepiolite interact with each other.

Sepiolite is a clay mineral, a complex magnesium silicate, a typical formula for which is (OH2)4(OH)4Mg8Si12O30•8H2O. It is formed by blocks and cavities (tunnels) growing in the direction of the fibres. The tunnels, 3.7 x 10.6 Å in cross-section, are responsible for the high specific surface area and sorptive properties of sepiolite. The co-intercalation of 3-methyl cyclohex-2-en-1-one (MCH), the Douglas-Fir beetle anti-aggregation pheromone, with methanol, ethanol, acetone, or benzene into sepiolite tunnels was studied. The resulting nanohybrid materials were characterized by means of various techniques, such as multinuclear solid-state NMR spectroscopy, porosity studies and Thermal Gravimetric Analysis (TGA). This was done in the hope of obtaining slow and controlled release of MCH from the sepiolite tunnels. It was demonstrated by 13C MAS NMR (carbon-13 magic angle spinning nuclear magnetic resonance) that at room temperature there are two different MCH molecules: one MCH inside the tunnels and the other one outside the tunnels of the sepiolite. Heating nanohybrid materials at 60˚C for 20 hours removes the external MCH molecules from the sepiolite. 13C MAS NMR showed that by further heating nanohybrid materials at 120˚C for 20 hours, methanol, ethanol, or acetone peaks were greatly reduced; however, the benzene peak was not reduced. To better understand how benzene acts inside sepiolite, intercalation of d6-benzene, and co-intercalations of d6-benzene with MCH and d6-benzene with pyridine into sepiolite tunnels were carried out, and these samples were studied by the same techniques. Another technique was used in order to see whether the slow and controlled release of MCH from the sepiolite tunnels could be obtained: sepiolite-MCH nanohybrids were treated with 20 ml of 0.5 M HCl solution. It was found that when 1 gram of MCH-sepiolite sample was acid treated at room temperature, about 35% of intercalated MCH was removed from the sepiolite. The role of sepiolite clay was also studied in Maya-Blue representative structure sepiolite-indigo adduct. It is known that upon heating the sepiolite and indigo mixture, the stability that is present in Maya-Blue is achieved. It is still a mystery, however, how exactly indigo and sepiolite interact with each other.

Increased public awareness, stricter legislation standards, and environmental and health effects associated with water pollution are driving the development of improved wastewater treatment techniques. In order to meet these challenges, a novel and cost effective process has been developed at the University of Manchester to treat water contaminated with dissolved organics by exploiting a combination of adsorption and electrochemical regeneration. Adsorption of organics takes place on the surface of a non-porous and highly electrically conductive graphite adsorbent, followed by anodic electrochemical regeneration leading to oxidation of the adsorbed organic contaminants. The mechanism of degradation of adsorbed organics during electrochemical regeneration is particularly important from the point of view of the breakdown products. Ideally, complete oxidation of the adsorbed organics to CO2 and H2O should occur, but it is also possible that intermediate by-products may be formed. These breakdown products could be released into the water, be released as gases during the regeneration process or may remain adsorbed on the surface of the adsorbent. Information about the breakdown products is an important requirement for the commercial application of the process. This PhD project focused on an investigation of the formation of intermediate oxidation products released into the water (liquid phase) and with the regeneration gases. Phenol was chosen as a model pollutant and a graphite intercalation compound (GIC) adsorbent, Nyex®1000 (Arvia® Technology Ltd) was used. The main oxidation products formed during both batch and continuous adsorption with electrochemical regeneration were 1,4-benzoquinone, maleic acid, oxalic acid, 4-chlorophenol and 2,4-dichlorphenol. These compounds were detected in small concentrations compared to the overall concentration of the phenol removed. Two mechanisms of organic oxidation during electrochemical regeneration of the GIC adsorbents were identified. The first was the complete oxidation of the adsorbed species on the surface of the adsorbent and the second involved the indirect electrochemical oxidation of organics in solution. Breakdown products were found to be formed due the indirect oxidation of organics in solution. The formation of (chlorinated and non-chlorinated) breakdown products was found to be dependant on current density, pH, initial concentration, chloride content and the electrolyte used in the cathode compartment. The concentrations of chlorinated breakdown products can be minimized by using low current density, low initial concentrations, a chloride-free environment and/or treating the water over a number of adsorptions and regeneration cycles. On the other hand, non-chlorinated breakdown products can be minimized by applying higher current density and treating the solution over several cycles of adsorption and regeneration. Therefore, selection of optimum conditions is important to reduce the formation of undesirable breakdown products. The formation of free chlorine during batch electrochemical regeneration was also investigated under a range of operating conditions including the initial concentration of chloride ions, current density and pH. The outcomes of this study have important implications in optimising the conditions for the formation of chlorinated breakdown products and in exploring the role of electrochlorination for water disinfection. Analysis of the regeneration gases has revealed that the main components of the gases generated during the electrochemical regeneration of GIC adsorbents were CO2 and H2O. A preliminary mass balance has suggested that about 60% of the adsorbed phenol was oxidised completely to CO2. However, further work is needed to determine the fate of the remaining phenol. The surface characterization of the GIC adsorbent during adsorption and electrochemical regeneration was carried out using surface techniques including Fourier transform infrared spectroscopy (FTIR), Raman spectroscopy, Energy dispersive X-ray spectroscopy (EDS) and Boehm titration. FTIR and Raman spectroscopy were found to be unsuitable for determining the concentration changes at the surface of the adsorbent during adsorption and regeneration. However, Boehm titration has shown that the GIC adsorbent has phenolic, carboxylic and lactonic groups. The concentrations of phenolic groups were found to be higher after phenol adsorption and to decrease during electrochemical regeneration. The results of EDS analysis gave results which were consistent with these observations. Another important aspect of this PhD project was to explore the potential application of adsorption and electrochemical regeneration using GIC adsorbents to water disinfection. A model microorganism E. coli was selected for adsorption and electrochemical regeneration studies under a range of experimental conditions. This study has provided evidence that the process of adsorption and electrochemical regeneration using GIC adsorbents can be used for disinfection of water. Disinfection of water was found to be a combination of two processes: the adsorption of microorganisms followed by their deactivation on the surface; and electrochemical disinfection in solution due to indirect oxidation. The possible disinfection mechanisms involved in these processes include electrochlorination, pH changes and deactivation by direct oxidation of microorganisms. Scanning electron microscopy was found to be a useful tool for investigating changes in surface morphology of microorganisms during adsorption and electrochemical regeneration. The disinfection of a variety of bacteria, fungi and yeasts was tested and evaluated. However, disinfection of protozoa including C. parvum was not demonstrated successfully. It was also demonstrated that the process of adsorption with electrochemical regeneration using GIC adsorbents can be used to simultaneously remove organics and to disinfect microorganisms.

Sepiolite is a clay mineral, a complex magnesium silicate, a typical formula for which is (OH2)4(OH)4Mg8Si12O30•8H2O. It is formed by blocks and cavities (tunnels) growing in the direction of the fibres. The tunnels, 3.7 x 10.6 Å in cross-section, are responsible for the high specific surface area and sorptive properties of sepiolite. The co-intercalation of 3-methyl cyclohex-2-en-1-one (MCH), the Douglas-Fir beetle anti-aggregation pheromone, with methanol, ethanol, acetone, or benzene into sepiolite tunnels was studied. The resulting nanohybrid materials were characterized by means of various techniques, such as multinuclear solid-state NMR spectroscopy, porosity studies and Thermal Gravimetric Analysis (TGA). This was done in the hope of obtaining slow and controlled release of MCH from the sepiolite tunnels. It was demonstrated by 13C MAS NMR (carbon-13 magic angle spinning nuclear magnetic resonance) that at room temperature there are two different MCH molecules: one MCH inside the tunnels and the other one outside the tunnels of the sepiolite. Heating nanohybrid materials at 60˚C for 20 hours removes the external MCH molecules from the sepiolite. 13C MAS NMR showed that by further heating nanohybrid materials at 120˚C for 20 hours, methanol, ethanol, or acetone peaks were greatly reduced; however, the benzene peak was not reduced. To better understand how benzene acts inside sepiolite, intercalation of d6-benzene, and co-intercalations of d6-benzene with MCH and d6-benzene with pyridine into sepiolite tunnels were carried out, and these samples were studied by the same techniques. Another technique was used in order to see whether the slow and controlled release of MCH from the sepiolite tunnels could be obtained: sepiolite-MCH nanohybrids were treated with 20 ml of 0.5 M HCl solution. It was found that when 1 gram of MCH-sepiolite sample was acid treated at room temperature, about 35% of intercalated MCH was removed from the sepiolite. The role of sepiolite clay was also studied in Maya-Blue representative structure sepiolite-indigo adduct. It is known that upon heating the sepiolite and indigo mixture, the stability that is present in Maya-Blue is achieved. It is still a mystery, however, how exactly indigo and sepiolite interact with each other.

La méthode solide-liquide en milieu alliage fondu à base de lithium a permis ces dernières années la synthèse de plusieurs composés d'intercalation du graphite (CIG) insérés à coeur au sein des systèmes graphite-lithium-alcalino-terreux. Dans le cadre de cette thèse, cette méthode de synthèse a été étendue aux systèmes graphite-lithium-lanthanoïde, avec une difficulté supplémentaire qui est la méconnaissance des diagrammes de phases binaires lithium-lanthanoïde dont les données sont capitales pour déterminer les domaines de température et de composition chimique des alliages susceptibles de conduire à des CIG. L'immersion de plaquettes de pyrographite dans certains alliages lithium-europium judicieusement choisis a mené à un composé binaire EuC6 ainsi qu'à un composé ternaire graphite-lithium-europium de premier stade.La cinétique de formation de EuC6 a été suivie par diffraction des rayons X ex situ afin de comprendre les différentes étapes de la réaction et d'identifier les phases intermédiaires menant au composé final thermodynamiquement stable. Ce mécanisme révèle un processus réactionnel plus « coopératif » que celui menant au composé CaC6 et a été décrit par une succession d'étapes contribuant à l'insertion à coeur de l'europium.La composition élémentaire du composé ternaire a été déterminée grâce à une analyse par faisceau d'ions qui a permis de doser simultanément les trois éléments lithium, carbone et europium. Le résultat de cette analyse a conduit à la formule chimique Li0,25Eu1,95C6. EuC6 a également été étudié par microsonde nucléaire, le rapport atomique C/Eu de 6 a ainsi notamment pu être confirmé.Des études structurales ont été menées pour les composés binaires et ternaires. D'une part, il a été possible d'effectuer la résolution structurale complète du binaire EuC6, qui cristallise dans une maille hexagonale de groupe d'espace P63/mmc. D'autre part pour le ternaire Li0,25Eu1,95C6, la séquence d'empilement poly-couche selon l'axe c du feuillet inséré a été modélisée, par combinaison des données structurales avec les informations issues de l'analyse par faisceau d'ions.Les composés d'intercalation du graphite sont des solides de basse dimensionnalité qui se prêtent idéalement à l'étude des relations structure-propriétés. Ainsi dans le système graphite-lithium-calcium, le caractère supraconducteur des composés CaC6 et Li3Ca2C6 a été étudié par spectroscopie de spin de muon ([mu]SR). Pour le système graphite-lithium-europium, des mesures magnétiques réalisées préalablement à ce travail ont été poursuivies et complétées par des analyses [mu]SR (pour Li0,25Eu1,95C6 et EuC6) ainsi que par spectrométrie Mössbauer de 151Eu (pour Li0,25Eu1,95C6) à basse température. / The molten alloy solid-liquid method containing lithium has recently enabled the synthesis of several bulk graphite intercalation compounds (GICs) in graphite-lithium-alkaline earth metal systems. As part of this thesis, this synthesis method was extended to graphite-lithium-lanthanide systems, with an additional difficulty which is the lack of knowledge of lithium-lanthanide binary phase diagrams whose data are crucial for determining the temperature range and chemical composition of alloys that may lead to GICs.The immersion of pyrographite platelets in some europium-lithium alloys wisely chosen led to a binary EuC6 compound as well as a graphite-lithium-europium first stage ternary compound.Kinetics study of EuC6 compound was followed by ex situ X-ray diffraction in order to understand the different reaction steps and identify intermediate phases leading to the thermodynamically stable final compound. This mechanism revealed a reaction process more "cooperative" than that leading to CaC6 binary compound and was described by a succession of steps that contribute to the bulk insertion of europium.The elementary composition of the ternary compound was determined by ions beam analysis allowing the simultaneous quantification of the three elements lithium, carbon and europium. The refinement of these analyses led to the chemical formula Li0,25Eu1,95C6 for the ternary compound. EuC6 has also been studied by nuclear microprobe analysis, and especially the C/Eu atomic ratio equal to 6 has been confirmed.Structural studies have been undertaken for binary and ternary compounds. On one hand, it was possible to fully resolve the three-dimensional structure of the binary EuC6, which crystallizes in a hexagonal unit cell with P63/mmc space group. On the other hand, the c axis stacking sequence of the poly-layered intercalated sheet of the ternary compound was modeled by combining structural data with information from the ions beam analysis. The graphite intercalation compounds are low-dimensional solids that are ideal for the study of structure-properties relations. Thus in graphite-lithium-calcium system, superconducting character has been studied for CaC6 and Li3Ca2C6 compounds by muons spin spectroscopy ([mu]SR). For the graphite-lithium-europium system, previous magnetic measurements have been continued and supplemented by [mu]SR analysis (for Li0,25Eu1,95C6 and EuC6) and by low temperature 151Eu Mössbauer spectroscopy (for Li0,25Eu1,95C6).

Hydrotalcite (HTC) was intercalated with different aromatic carboxylic acids via two synthesis methods: reconstruction and co-precipitation. The reconstruction method involves the rehydration of the products of LDH calcination. The co-precipitation method involves the addition of a base to solutions containing a mixture of the MII and MIII ions found in the metallic layers of an LDH. The intercalated compounds were then compounded with flexible grade PVC to see if these compounds had any effect on the heat stability of the PVC. Complete intercalation of these stabilisers did not occur; however layered double hydroxides did form for almost every synthesis. The organic acids that were to be intercalated were also present in every synthesised stabiliser. Neat hydrotalcite was the best overall stabiliser with an early stability time of 32.40 min and a final or long term stability time of 106.51 min. The best modified layered double hydroxide (LDH) in terms of early stability was 4-hydroxybenzoic acid + HTC synthesised with the reconstruction method. This stabiliser had an early stability time of 25.40 min. The best performing modified stabiliser in terms of late stability was salicylic acid + HTC synthesised with the co-precipitation method. This stabiliser had a late stability time of 71.32 min. The highly activating nature of the hydroxyl substituent group should make hydrotalcites intercalated with hydroxybenzoic acids good free radical scavengers. The substituent group positions that give the best PVC heat stability are the ortho and para positions. The pKa 2 value for an organic acid may be used as a selection parameter for intercalation into hydrotalcite. If a high pKa 2 value organic acid is intercalated into hydrotalcite, the resulting compound will have good PVC heat stabilisation properties. / Dissertation (MEng)--University of Pretoria, 2017. / Chemical Engineering / MEng / Unrestricted

no / Cryptolepine, a naturally occurring indoloquinoline alkaloid used as an antimalarial drug in Central and Western Africa, has been found to bind to DNA in a formerly unknown intercalation mode. Evidence from competition dialysis assays demonstrates that cryptolepine is able to bind CG-rich sequences containing nonalternating CC sites. Here we show that cryptolepine interacts with the CC sites of the DNA fragment d(CCTAGG)2 in a base-stacking intercalation mode.