ISSN 0020-1685, Inorganic Materials, 2014, Vol. 50, No. 10, pp. 992–996. © Pleiades Publishing, Ltd., 2014. Original Russian Text © Son Tung Luu, Hoo Van Nguyen, E.G. Rakov, 2014, published in Neorganicheskie Materialy, 2014, Vol. 50, No. 10, pp. 1074–1079. 992 INTRODUCTION Carbon nanotubes (CNTs) and carbon nanofibers (CNFs) have a relatively large specific surface area and readily undergo acid functionalization, with attach- ment of carboxyl, hydroxyl, and other functional groups capable of ion exchange with metal salts. Ion-exchange metal sorption processes with the participation of CNTs and CNFs can be used for a variety of purposes, in particular for the removal of toxic impurities from solutions (including natural and waste water), water softening, the separation and pre- concentration of metals in hydrometallurgical raw materials processing techniques and analytical deter- mination of metals, and the fabrication of composites and photo- and electrocatalysts. Metal ion sorption by CNTs and CNFs has been the subject of many reports and several review papers [1–4]. Most attention has been paid to lead, copper, and cadmium (20 to 30 papers devoted to each metal); less attention has been given to nickel and zinc; and several reports have been concerned with other chem- ical elements. Research effort has been concentrated primarily on dilute and very dilute solutions. Fe(III) sorption has only been described by Li et al. [5], who studied sorption by CNT and CNF sheets. They described the process kinetics by a first-order rate equation and the equilibrium state by the Langmuir equation and proposed using electrodesorption for sorbent regeneration. Fe(II) sorption by CNTs or CNFs was not considered. At the same time, CNTs decorated with Fe 3 O 4 nanoparticles are magnetic materials and (like CNTs with Fe 2 O 3 ) are of interest as sorbents [6–8] or starting materials for the fabrication of ceramic composites. The objectives of this work were to bridge the gap in this area of ion-exchange sorption and to investigate Fe(II) and Fe(III) sorption from aqueous solutions by various CNTs and CNFs. We studied relatively con- centrated solutions. EXPERIMENTAL We used CNFs and CNTs functionalized by a pro- cedure similar to that described previously [9]. The functionalization resulted in the formation of oxygen- containing surface groups, such as –C(O)OH, =СО, and ≡СОН, which imparted the ability to form stable dispersions and participate in ion exchange to the CNTs and CNFs. The concentration (“solubility”) of functionalized CNFs and CNTs in aqueous disper- sions was 4.0 (sample f-CNF-1), 3.7 (f-CNT-2), and 2.0 g/L (f-CNT-3). The specific surface area of the CNFs and CNTs before the functionalization was about 70 and 242 m 2 /g, respectively. The other chem- icals used were of reagent grade. Sorption experiments were carried out as follows: A weighed amount of CNFs and CNTs was dispersed in 50 mL of distilled water by a URAN-A horn-type son- icator, and 10 mL of a Fe(NO 3 ) 3 ⋅ 9H 2 O or FeSO 4 ⋅ 7H 2 O solution was added. After a predetermined time, the precipitate was collected on blue ribbon filter paper and dried. To determine the amount of sorbed iron, the dried precipitate was calcined in air at 750°C for 3 h and weighed in oxide form (the composition Fe 3 O 4 was taken as weighing forms). To examine the effect of pH, it was varied by adding an appropriate amount of HNO 3 , H 2 SO 4 , or NH 4 OH and measured by a Martini pH-56 tester. The microwave activation of the sorption process was performed for 3 min using a Samsung M1712NR oven at microwave powers in the range 100–800 W. Ion-Exchange Iron Sorption by Carbon Nanotubes and Nanofibers Son Tung Luu, Hoo Van Nguyen, and E. G. Rakov Mendeleev University of Chemical Technology, Miusskaya pl. 9, Moscow, 125047 Russia e-mail: eg_rakov@rctu.ru Received February 12, 2014 Abstract—We have studied room-temperature equilibrium in systems containing an aqueous Fe(II) or Fe(III) salt solution and carbon nanofibers or carbon nanotubes with various contents of functional groups. The sorption capacity of the sorbents has been determined as a function of contact time, sorbent weight to solution volume ratio, salt concentrations in solution, solution pH, and sorbent “solubility” (degree of func- tionalization). Equilibrium data have been described by the Langmuir and Freundlich equations, and the sorption kinetics have been represented by a first-order or pseudo-second-order equation. We have demon- strated that the sorption process can be accelerated by physical activation of the system. DOI: 10.1134/S0020168514100161