Deposition of electromagnetic energy in animals andin models of manwith andwithout grounding and reflector effects Om P. Gandhi Departments of Electrical Engineering andBioengineering, University of Utah,SaltLake City, Utah84112 Edward L. Hunt Department of Microwave Research, Walter Reed Army Institute of Research, Washington, D.C. 20012 John A. D'Andrea Departments of Electrical Engineering andBioengineering, University of Utah,Salt Lake City, Utah84112 Generalized curvesare given for rates of whole-body absorption of electromagnetic energy by models of human beings as a functionof frequency. Exposures were made in free space, or whenthe model made electricalcontact with the ground. The effects of the presence of reflecting surfaces also were analyzed. Peaksof absorptionwith and without a ground, are projected, respectively, to be (31 to 34 MHz) and (62 to 68 MHz) X (1.75/heightof modelin meters). Rates of energy deposition are given formodels of man and for animals subjected toradiation atapower density of 10 mW/cm 2 for the various conditions of exposure.At resonance, values of whole-bodyabsorptionas high as 4,077 to 8,154 watts for the adult human being are predicted. The times-to-convulsion of • 100-g rats at power densities of 3 to 20 mW/cm 2 confirmed predictions of extremely high rates of absorp- tion in the presence of reflecting surfaces. 1. INTRODUCTION We have previously reported on the observation of a strong resonance in the whole-body absorption of elec- tromagnetic waves by biological bodies[Gandhi,1974, 1975]. For plane-wave irradiation in free space,the highest rate of energy deposition occurs in fieldsthat are polarized along the longest dimension of the body (Ell œ)and for frequencies such that the major length (L) is approximately 0.36 to 0.40 timesthe free-space wavelength (X) of radiation. Peaksof whole-body ab- sorption for two other configurations (major length oriented along the direction of wave propagation or along the vector of the magnetic field) were also re- ported [Gandhi, 1974, 1975] for L/X on the orderof L/4•rbwhere 2rrb is theweighted averaged circumference of the animals. A handicapof past measurements has been that com- mercially available dolls that have been used as molds for models are not properlyproportioned. Usingcare- fully sculptured, accurately scaled figurines [Dreyfuss, 1967], measurements have now been extended to cover the region0.2 < L/X < 3.3. Specific absorption rates (SARs) are givenin this paperunderthree configura- tions in free sp•ace for the frequency range 0.5 to 8.7 times the Ell L resonance frequency fr, which, for man, is 65 X (1.75/length in meters) MHz. From these results, antenna theoryis used to de•lop an empirical equation for the SAR, for the Ell L orientation of models,at resonantand supraresonant frequencies. A comparison with the lethality data of $chrot and 39 Hawkins [ 1975] confirms the validity of the equation for ratsandmice of differing sizes. Results are also presented for a simulated human being with its feet in electrical contact with the ground. In such an exposure, a resonant frequency one-half as high with a peak absorption twice that of free-space irradia- tion is observed. Energy deposition rates are given for modelsof man and for animals that wereexposed to radiation at power densities of 10 mW/cm 2 under various conditions of exposure. 2. ELECTROMAGNETIC ENERGY ABSORPTION IN MAN IN FREE SPACE The procedure for obtaining the mass-normalized rate of electromagnetic energy deposition (SAR)through the use of reduced-scale models has been detailed in earlier publications [Gandhi, 1975; Gandhi et al., 1975]. The SAR in W/kg is calculated from the temperature rise, /XT, of 7.6, 10.2, 12.7, 15.2, 20.3, 25.4, 33.0, and 40.6 cm high, saline-filled (0.9% NaC1) figurines from the expression (4,180 /XT X specificheat of the medium in calories/g/øC)/irradiation timein seconds. TheSARs so calculated are divided by the model's scaling factor (height of the human/length of the figurine)to obtain the value for a human being.At least three measure- mentswere taken to calculate the averaged SARs that are given in this paper. Unless otherwise stated, the standard deviation of the measurements is less than