JOURNAL OF MATERIALS SCIENCE: MATERIALS IN MEDICINE 12 (2001) 391±397 Characterization of copper corrosion products originated in simulated uterine ¯uids and on packaged intrauterine devices J. M. BASTIDAS, N. MORA, E. CANO, J. L. POLO Centro Nacional de Investigaciones Metalu Ârgicas (CSIC), Avda. Gregorio del Amo 8, 28040 Madrid, Spain This paper studies the characterization of corrosion products originated after 1 and 12 weeks' immersion of copper specimens in simulated uterine ¯uids at pH 6.3 and 8.0 and at 37  C temperature. The experimental techniques used were X-ray photo-electron spectroscopy, scanning electron microscopy, and energy dispersive X-ray. The compounds found were calcite (CaCO 3 ), calcium phosphate, cuprite (Cu 2 O) and copper hydroxide (Cu(OH) 2 ). The morphology of corrosion products was a non-uniform, layer showing some paths through which copper ions can be released. In parallel, corrosion products formed on packaged, unused copper-containing intrauterine devices (IUD) were analyzed. Cuprite (Cu 2 O) and chalcocite (Cu 2 S) were the main species identi®ed. # 2001 Kluwer Academic Publishers 1. Introduction It is known that the anti-fertility effect of an intrauterine device (IUD) can be improved if a copper wire is used to encase the stem of the device [1]. Though the mechanism of the contraceptive action of copper-containing IUDs is unclear, one hypothesis is that a copper-containing IUD may be effective as a result of the dissolution of the copper into uterine secretions and the formation of cupric ions, which lead to the inactivation of sperm and the suppression of myometrial contractions [2, 3]. A sterile in¯ammation that is enhanced by copper has also been hypothesized [1]. Available information seems to indicate an inverse relationship between the surface area of copper on IUDs and pregnancy rates [2±7]. Clinical studies have indi- cated that the average copper release rate after 3 years of use, depending on IUD type, is around 25 mg/day [8]. Circumstances that obstruct copper release may have a negative in¯uence on IUD effectiveness. Thus, the presence of extensive calcite (CaCO 3 ) deposits covering the copper wire on these devices has been associated with accidental pregnancies [9]. Despite this, the in¯uence of corrosion products formed on the copper surface in IUDs is a poorly studied subject. Little information is available about the acute and sub- acute toxicity of copper released from copper devices. Copper, as a heavy metal, is toxic to the developing blastocyst and to sperm [3]. However, IUDs containing copper wire do not seem to have a direct toxic effect on the blastocyst since proteins in uterine secretions may provide protection against the free copper ions [10]. In this sense, copper is preferable to other metals such as iron, nickel, zinc, gold, silver and stainless steel [7, 11]. The original red color of copper on an IUD can change during storage and/or transportation before the device is used. The red color of copper is associated with cuprite (Cu 2 O), which is an extremely thin ®lm that forms spontaneously when a clean copper surface is exposed to air at room temperature [12, 13]. Thus, to answer the question as to whether a copper IUD can be used when the copper surface is covered by a dark layer is of practical importance [8, 14]. The aim of this paper is to characterize the corrosion products formed at early stages (after 1 and 12 weeks) on copper specimens immersed in simulated uterine ¯uids at two pH values. Another aim of the paper is to characterize the thin dark colored layer formed on copper on IUDs after 30 months storage at room temperature. 2. Materials and methods Copper specimens with a chemical composition (wt %): 0.009 Sn, 5 0.001 As, 5 0.001 Bi, 0.003 Ni, 5 0.001 Fe, 0.015 Pb, 5 0.001 Mn, 0.019 P, 5 0.0005 Ag, 5 0.001 S, 5 0.005 C, 5 0.002 Sb, 5 0.001 Al, and balance copper were used. The copper was phosphorus- deoxidized with a low residual phosphorus content (Type Cu-DLP, ISO 1337). The specimens, of 2.0 6 2.0 cm 2 , were mechanically cut from sheets 1.0 mm thick, and their surfaces were hand-polished with different grades of emery paper down to grade 600, degreased with acetone, rubbed energetically with cotton wool soaked in ethanol and ®nally dried at room temperature. Table I indicates the composition of the simulated uterine ¯uid tested [15]. Reagent grade chemicals were used. Two pH values, 6.3 and 8.0, were tested. The pH value was adjusted by adding diluted reagent grade 0957±4530 # 2001 Kluwer Academic Publishers 391