Magnetic Properties of Synthetic Eumelanin-Preliminary Results[dagger]

ABSTRACT

We report an experimental and theoretical study of magnetic properties of synthetic eumelanin. The magnetization curves are determined by using both a vibrating sample magnetometer and a superconducting quantum interferometer device in an extended range of magnetic fields ranging from -10 kOe to 10 kOe at different temperatures. We find that the eumelanin magnetization can be qualitatively explained in terms of a simple model of dipolar spheres with an intrinsic magnetic moment. The latter one is experimentally measured by using X-band electron paramagnetic resonance. Our findings indicate that synthetic melanins are superparamagnetic.

INTRODUCTION

Melanin is a biologic pigment present in several human organs. In the eyes it is mainly concentrated in the iris, which takes a dark color depending on its amount. In skin and hair it can be classified into two groups-eumelanin (brown/black) and pheomelanin (red/yellow). In the skin it performs the essential function of preventing damage caused by solar ultraviolet radiation (1). In the brain, melanin is known as neuromelanin, a granular natural pigment of dark brown hue. Neuromelanin is very common in neurons of the substantia nigra, which is located in the human midbrain. This area of the human brain seems to suffer severe neuromelanin degradation during Parkinson's disease (2). Melanin is also present in some animals where it performs different physiologic functions (3). Melanin has the ability to capture and release metal ions without any apparent change in morphology; this has been interpreted as indicative of a structure which allows the efficient transport of metal ions without damage (4).

Recently, both X- and Q-band electron paramagnetic resonance (EPR) spectroscopies have been extensively used to investigate and understand the induced changes in the neuromelanin ferrodomains (5,6). On the other hand, preliminary results on mesencephalic neuromelanin to patients with Parkinson's disease showed a smaller total magnetization than the control group (7). Recent research into the electronic properties of eumelanin has indicated that it may consist of basic oligomers adhering to one another (4).

In this work, we are interested in the study of magnetic properties of synthetic eumelanin by combining experimental and computer simulations. Basically, the magnetic moment per particle is experimentally measured and it is used in a theoretical model which consists of magnetic dipolar spheres coupled to an external magnetic field. This model is used in a standard Monte Carlo (MC) simulation to numerically compute the magnetization curve. Our simulation data are compared with experimental results obtained by using a vibrating sample magnetometer (VSM) and a superconducting quantum interference device (SQUID) on synthetic eumelanins at different temperature conditions.

MATERIALS AND METHODS

Eumelanin is synthesized from a mixture of L-dopa with benzoyl peroxide in a dimethyl sulfoxide solvent. Complete details of the synthesis process can be found in Dezidério et al. (8). Three aliquots of 0.1 g of melanin are weighed using a Mettler Toledo analytic balance with 0.01 mg precision. The powder is placed in three small identical nonpermeable capsules made from gelatine normally used as a drug vial for oral medicine.

In addition, if we keep P low enough to avoid saturation, y is proportional to P^sup ½^, then any observed change in y is due only to a change in χ''. Then for a sample of the same material this change will be due to a variation in the number of spins in the sample. If we want to determine the number of spins in an unknown sample comparing it with a standard sample, it is convenient to use the same value of P to record the EPR spectrum of both samples. In this case, any change in y is due to a variation in the other three parameters involved in Eq. (4). If the EPR experiment is carried out with the same geometrical conditions for the two samples, e.g. same diameter, height, volume, tube, etc., a variation in y must be due to a change in χ'' and Q'^sub U^, η is assumed constant for both samples as it is proportional to the fraction of the cavity's microwave field energy that is concentrated at the sample position and proportional to the volume's ratio (sample volume/cavity volume) (9,12-14).

Finally, if the dielectric constants of both samples are similar, it can be assumed that the quality factor is about the same (see Poole and Charles [9], p. 176), but if a difference exists, Q'^sub U^ will change. In our case, the dielectric constant values for melanin and silicon were extrapolated from the literature (15) and are approximately 6.5 and 13.7, respectively, in the GHz region. We decide to use the methodology suggested by Yordanov and Lubenova (16) that recommended recording the EPR signal of the unknown sample (or the standard sample) together with a reference sample such as Mn^sup 2+^. Thus, the intensity ratio (unknown sample/Mn^sup 2+^ or standard Mn^sup 2+^) will be less affected by any difference in dielectric constants of both samples (unknown and standard). These ratios will represent the signal intensities y^sub 1^ and y^sub 2^ for the unknown and standard samples, respectively, in Eq. (3).