1028-3358/01/4607- $21.00 © 2001 MAIK “Nauka/Interperiodica” 0459 Doklady Physics, Vol. 46, No. 7, 2001, pp. 459–462. Translated from Doklady Akademii Nauk, Vol. 379, No. 1, 2001, pp. 38–41. Original Russian Text Copyright © 2001 by Belobrov, Gordeev, Petrakovskaya, Falaleev. In nanodiamonds obtained by the explosion method [1–3], the existence of diamond molecular forms [4] is based on the thermodynamic stability of 5-nm-sized diamond particles [5, 6]. There are several basic char- acteristics of molecular diamond which only quantita- tively differ for diamond and nanodiamond, namely, the region of coherent X-ray scattering, the specific surface area, the stable-particle size, etc. It is of importance that the exact reproduction of conditions for synthesis and extraction results in a yield of nanodiamonds with the same characteristics [7]. The absence of a physical parameter defining the diamond molecule is the basic cause of underestimating the fact that any ultradisperse diamond obtained by the explosion method consists of the spherical molecules and edged particles of nanoc- rystalline diamond. In this paper, the radiospectroscopic methods of electron paramagnetic resonance (EPR) and nuclear magnetic resonance (NMR) are used to experimentally prove the existence of diamond macromolecules. The chemical activity of nanodiamond depends on surface states traditionally associated with dangling bonds, i.e., with unpaired electrons. Therefore, nanodiamonds with exactly determined surface states were chosen for the experiment. These nanodiamonds, obtained using vari- ous methods, were subjected to different purification stages and had strongly distinguished values of incom- bustible remnants or ash content. For EPR-spectrum measurements of the same sam- ples, EPR-3, SE/x-2544, EMX EPR Bruker (9 GHz), and RE08 (36 GHz) spectrometers were used (Table 1). The measurements were carried out at temperatures of 290 and 77 K. While determining the absolute con- centration of unpaired spins, CuCl 2 · 5H 2 O served as a standard. Three standards were used in the measure- ment of the Lande g-value: α-diphenyl-β-picrylhydra- zyl (DPPG, g = 2.0036), standard samples of Mn 2+ ions (g = 1.9803) in MgO, and Li (g = 2.0023). The mean values of the g-value and line widths ∆H for the sam- ples under study are given in Table 1, where the stan- dard deviation for the last significant digit is indicated in brackets. Each sampling relates to the spectra of the same sample. The calculated values of ∆H were aver- aged by interpolating both the shape of the absorption line by the Lorentzian curve and the values of g-values obtained with various standards, scans, and modula- tions. The temperature-independent behavior of the para- magnetic-absorption line shape for nanodiamond was first observed in [8]. There, it was shown that the con- centration of unpaired electrons attained a value on the order of 10 19 spin/g within the temperature range from 77 to 600 K [8]. The g-values at temperatures of 77 and 290 K also did not differ within the experimental error. In addition, by varying the microwave-field amplitude from 1 to 40 dB, only the EPR-signal intensity changed, whereas the g-value and the line width remained invariable. Therefore, data obtained with dif- ferent equipment and for different temperatures, stan- dards, and attenuation magnitudes were used for increasing the sampling volume and reducing the mea- surement error. An example of a nanodiamond EPR spectrum is shown in Fig. 1. Parameters of paramagnetic absorption for nanodiamond particles with various surface states are given in Table 1. As can be seen, the nanodiamond EPR spectrum virtually does not change with signifi- cant variation of the particle-surface structure. The cor- responding mean g-values and ∆H for powder nanodi- amond (samples nos. 1–8) are, respectively, 2.0027(5) and 0.87(6) mT. The DPPG and manganese standards do not allow us to determine the g-value with sufficient accuracy. Using a Li standard, it is possible to increase the accuracy by an order of magnitude: for example, the g-value for samples no. 9 and no. 12 is equal to 2.00271(3) and 2.00266(2), respectively. In nanocomposite samples (nos. 9–15), the ratio γ for the pyrocarbon mass to that of the nanodiamond varies from 0 to 40%. In this case, the EPR spectrum virtually does not change. Consequently, the nanocom- posite spectrum represents the EPR spectrum of nano- PHYSICS Paramagnetic Properties of Nanodiamond P. I. Belobrov*, S. K. Gordeev**, É. A. Petrakovskaya***, and O. V. Falaleev*** Presented by Academician K.S. Aleksandrov December 25, 2000 Received January 19, 2001 * Institute of Biophysics, Siberian Division, Russian Academy of Sciences, Akademgorodok, Krasnoyarsk, 660036 Russia ** Central Research Institute of Materials, ul. Paradnaya 8, St. Petersburg, 191014 Russia *** Kirenskiœ Institute of Physics, Siberian Division, Russian Academy of Sciences, Akademgorodok, Krasnoyarsk, 660036 Russia