Table 7.17 A Summary of Available Experimental Data on Fields and SAR Measurements in Biological Phantoms and Test Animals Irradiated by Electromagnetic Fields

Object Exposed	Frequency Range	Radiation Source Characteristics	Polari- zation	Dosimetric Parameter	Measurement Technique	Refer- ence	Summary of Results
Saline-filled rectangular phantoms 45.7 x 78.8 x 25cm	10-30 MHz	Planewave, CW	KHE	Maximum and average SAR	Differential power measurement	a	Experimental data demonstrate the square-law dependence of SAR on frequency of RF source in the HF band, and linear rela- tionship between Pin and SAR.
* 190-560-g rats * Water-filled cylindrical phantoms 190 and 290 g (17-cm long by 3.8-cm diameter) 29-g mice	2450 MHz	Multimodal resonating cavity, pulse repetition rate = 120Hz, pulse width=2.5 ms	Unspecified	Mass absorption density (mJ/g), area absorption density (mJ/cm2) and latency to convulsion	Calorimetric technique	b	Absorbed dose measurements in phantoms are compared with those in rats. At a given power level, mean latency to convulsion is proportional to the weight of rats. Latency to convulsion in rats and mice as function of SAR is given.
* 190-560-g rats * Water-filled cylindrical phantoms 190 and 290 g	2800 MHz	Far field	Unspecified	Mass absorption density (mJ/g) and area absorption density (mJ/cm2)	Calorimetric technique	c	SAR measurements in phantoms are compared with those in rats.
* 96- 90 Wistar rats * 25-g mice Brain-phantom prolate spheroids (a/b up to and including 5.75)	285-4000 MHz	Planewave (two-plate stripline)	E, H, and K	SAR	Insertion loss measurement	d e f	For rats, mice, and spheroidal phantoms, E polarization is the most absorbing, H polarization the least, and K polarization slightly more absorbing than the H. For rats, in K polarization the frequencies of peak absorption and the maximum absorption at these values demonstrate W 1/3 and W 2/3 dependencies, r espectively, upon the weight of the animal. In phantom spheres the peak SAR occurs in a resonant region where ka = 0.55-0.59.
* Rhesus monkeys (anesthetized) * 3.5-kg prolate spheroid monkey phantoms * 70-kg prolate spheroid human phantoms	1-50 MHz	Far-field exposure chamber	E, H, and K	SAR	Differential power measurements	f g	Average SAR increases approximately as the frequency squared, for all phantoms as well as monkey exposures. Experimental data on spheroidal phantoms are in good agreement with theoretical data calculated by long-wavelength analysis (Durney et al. 1975). Comparisons are made between measured SAR data on monkeys and calculated data on ellipsoidal models (Massoudi et al., 1977a).

(Table 7.17 continued)

Object Exposed	Frequency Range	Radiation Source Characteristics	Polari- zation	Dosimetric Parameter	Measurement Technique	Refer- ence	Summary of Results
Tissue-equivalent materials simulating rats 1, 4, 10, 12, 30 & 60 d old	2450 MHz	Rectangular waveguide, TE10 mode, CW	Tail-on (0°) and head-on (180°) to direction of incident wave. Also, stepwise variations from 0° to 180°	SAR and SAR distribution	Thermography from SAR distribution and insertion-loss measurement for average SAR	h	For 180°orientation, the head area, including the brain, absorbs the most microwave energy. In the opposite orientation, the tail and abdominal regions absorb the maximum. Total SAR changes only slightly between 0° and 180° orientations. SAR is lowest at 90° orientation angle.
* 25-30-g mice (8.5 cm long) * 100-125-g rats (16 cm long) * 380-420-g rats (22.5 cm long, excluding tail)	710-3000 MHz	Standard-gain horns	E and H	Latency to tonic-clonic convulsion	Obsersvation via closed-circuit television and elapsed time recorded on a printout counter	i	E polarization produces consistently shorter convulsion latencies. An inverse relationship between frequency and animal size was observed over the range of frequencies and sizes tested.
* Rats (60-440 g) * Mice (35-50g)	2450 MHz	Horn antenna (free-field exposure)	---	SAR	Calorimetric technique	j	Simultaneous exposures of groups of animals in varying numbers and various configurations have been m ade. SAR data as a function of position and weight are presented. Position of animal in an exposure matrix significently affects the SAR.
Mice (4 wk old)	918 MHz	Rectangular waveguide (24.8 x 12.4 cm)	---	SAR	(Differential power measurements	k	During 8-h irradiation sessions, a reduction in SAR after the initial hour of each session was observed for both female and male groups of mice exposed to a 5-W forward power but not for a 2.5-W forward power.
Spheres of simulated muscle tissue (3.3-cm and 8-cm radii)	450, 915, & 2450 MHz	Planewave (horn antenna)	---	Spatial distribution of E-field	Miniature isotropic probes	e m	Spatial distribution of E-field within spheres is compared with theoretically predicted values. Some internal dosimetric techniques are compared. Measured data are obtained with implantable probe in living cat's brain.
Saline-filled and biological phantom figurines (dolls 12.1, 18.4, 22.3, & 23.5 cm tall)	500, 985, & 2450 MHz	TEM paprallel-plate chamber and free space	EHK, EKH, KHE, & KEH	SAR distribution	Temperature-rise measurement	n o p	Measured SAR data for dolls are scaled to obtain SAR values for humans as function of L/. For E polarization, (a) SAR observed for L/
* Saline-filled (0.9%) NaCl) figurines (7.6, 10.2, 12.7, 15.2, 20.3, 25.4, 33, & 40.6 cm high) * Saline-filled prolate spheroid (L/2b=6)	500, 710, 987, & 2450 MHz	Free space	EKH, KEH, HEK	SAR	Temperature-rise measurements	o q	For EHK polarization, SAR dependence on frequency (f) is (a) an f2 type for frequencies well below resonance (L/2.75 to f3 type at sub-resonant region (0.2>fres the relative absorpion coefficient asymptotically approaches the optical values (1-power reflection coefficient).
Saline-filled figurines (10.2, 12.7, 15.2, & 20.3 cm tall)	150-1500 MHz	Monopole-above-ground radiation chamber	EKH	SAR	Temperature-rise measurements	o q	Projected SARs for a human being of height L, with feet in conductive contact with ground, are given as a function of L/
Saline-filled figurines	987 and 2450 MHz	Monopole antenna in front of flat and corner reflectors	---	SAR	Temperatire-rise measurements	n q	Enhancement factor as a function of L/
Rats (averaged mass near 100g)	987 MHz	Monopole anatenna in a 90° corner reflector	---	Time to convulsion	---	q	Times to convulsion of rats for incident wave of 3- to 20-mW/cm2 power densities confirm some of the predictions of enhanced SAR in the presence of reflecting surfaces.

(Table 7.17 continued)

Object Exposed	Frequency Range	Radiation Source Characteristics	Polari- zation	Dosimetric Parameter	Measurement Technique	Refer- ence	Summary of Results
Prolate spheroid phantoms: 330g (17 x 6.1cm) 301g (13.2 x 6.6cm)	918 Mhz	Cylindrical wave-guide excited with circularly polarized TE11-mode field configuration	Major axis of the spheroid, both parallel and perpenciular to wave-guide axis	Average SAR, peak SAR, and SAR distribution	Thermographic methods and differential power measurements	r	Thermogoraphic SAR measurements on exposed spheroidal phantoms are presented. Average and peak SAR values show very little change with orientation of phantoms.
A sacrificed 338-g rat	918 MHz	Cylindrical wave-guide excited with circularly polarized TE11-mode field configuration	Axial head, axial tail, and trans-verse illuminations	SAR distribution	Thermographic methods	r	Computer-processed thermograms showing iso-SAR lines normalized for 1-W input in three orientations.
132-490-g anesthetized and freely moving rats	918 MHz	Cylindrical wave-guide excited with circularly polarized TE11-mode field configuration	---	Mean SAR	Differential power measurement	r	For both moving and anesthetized rats, the mean SAR remains relatively constant with low standard deviation.
Spherical phantoms, 6-11.1-cm diameters * Prolate spheroidal phantoms with semimajor axes of 6.82-19.6cm and aspect ratios (a/b) of 2-7.73 * 1/4.62 scale human, 37.7cm tall * 40-g mouse and its phantom	VHF 19 MHz 19 MHz 1600 MHz 2450 MHz	Resonant cavity Near-field synthesizer Stripline exposure apparatus Planewave Microwave brain deactivator	--- E, K, and H	SAR distribution, peak SAR	Thermographic techniques	s t	Thermograms show SAR distribution inside phantoms exposed to electric, magnetic, and combinations of electric and magnetic fields in the near- field synthesizer. Peak SAR data for phantoms exposed in near-field synthe- sizer are compared with those of the reson- ant cavity. Peak SAR values are also given for phantoms exposed in stripline chamber and microwave brain deactivator. For short exposure time and temperature rise of more than 2°, measurements made in near-field synthesizer on tissue- equivalent spheroids agree well with theory (Durney et al., 1975) and values measured in the resonant cavity.

(Table 7.17 continued)

Object Exposed	Frequency Range	Radiation Source Characteristics	Polari- zation	Dosimetric Parameter	Measurement Technique	Refer- ence	Summary of Results
Inhomogeneous saline-filled rec- tangular phantom (8 x 6 x 2cm)	2370 MHz	Planewave	KEH	SAR	E-field measurement	u v	SAR due to vertical component of induced E-field as a function of conductivity of the central region was measured and experi- mental data compared with theo- retical values. At 2370 MHz, an optimal conductivity exists for the local region to gain the most effective EM heating.
18 rates simult- neously exposed	2.6 GHz	Standard gain horn	E	Average SAR	Carlimetric technique	w	Power density levels at positions near the center of the array were higher than those near the edge.
Erythrocyte membrane	2.45 GHz and 12.5-18GHz	Horn antenna	---	SAR	Temperature- rise measurements	x	At all frequencies and power levels tested, increased loss of either hemoglobin or K+ from microwave-irradiated-rabbit RBCs should be ascribed to thermal effects on the stability and/or permeability of erythrocyte membrane.
25-g mice to 45-g mice	31-34 MHz, 62-68 MHz	90° corner reflector with quarter- wavelength monopole	---	SAR	Colonic temper- ature elevation and calrimet- ric method of average-SAR measurement	q	Extremely high rates of absorption in the presence of reflecting surfaces were confirmed.
Guinea pigs, rabbits, mice, cacti	X-band microwave frequency range	Rectangular horn antenna	---	Power density	Changes of power density	w	Significent changes in radiation pattern occurred as result of presence of biological specimen.
20.3-cm saline- filled dolls and variouss orientations to E	2450 and 987 MHz	---	E, K	SAR	---	y	Whole-body SARs varied smoothly frtom E- to K- orientation values as body orientation was altered between the two extreme positions ini this plane.
Frog sciatic nerves, cat saphenous nerves, rabbit vagus nerves, rat- diaphragm muscles	2450 MHz	S-band waveguide	H, K	SAR, and nerve conduction and contraction	Fiber-optic probe for tem- perature measurements	x	When the temperature of nerves and muscles exposed to EM fields was kept constant, no change in conduction characteristics or contraction was observed.
Scaled figurines 20.3, 25.4, 33, and 40.6 cm long and Long Evans rats	285 MHz, 355.6 MHz, 462.3 MHz, 569 MHz	---	E, K	SAR	Calorimetric technique and temperature-rise measurements	aa	Head resonance was observed in each.

(Table 7.17 continued)

Object Exposed	Frequency Range	Radiation Source Characteristics	Polari- zation	Dosimetric Parameter	Measurement Technique	Refer- ence	Summary of Results
Four 300-g male Walter Reed strain rats	2450 MHz, 987 MHz	Exposed rectan- gular waveguide	0°, 90°, 180°, 270°, within waveguide	Spatial distribution of heat loading and pattern of succinic dehydrogenase	Temperature- rise measurements and cytochemical measurement	bb	Data obtained suggests that further exploration of frequency parameter could lead to extremely rapid method of sacrifice with simultaneous inactivation of brain enzymes in unrestrained rats.
100-g rats placed 3/2 in front of 90° corner reflector	987 MHz, 2650 MHz	90° corner reflector	---	Time to convulsion	---	y	Departure of only 10% between projected and oserved values; may be even lower for heavier animals.
Eleven rats 420-450 g	200-700 MHz	Single brass monopole ( /4 long, 1.2-cm diameter)	E	Rise in temperature and change in behavior	Temperature- rise measurements	cc	Showed increasingly larger rise in body temperature and increasing tendency for behavioral disruption to occur in animals as sfrequency was increased from 200 to 500 MHz.
Adult male CFI mice	2450 MHz	Inside a rectan- gular waveguide	---	Specific heat dissipation rate	Calorimetric technique	dd	Indicated increased heat-dissipation rate of animals (compared to sham) for average absorbed dose rates above 12 mW/g.
Pupae of the insect Tenebrio molitor L	1.3, 5.95, and 10 GHz	Right-angle, waveguide E bends	E, K, and H	SAR distribution	Thermographic imaging system	ee	At X band, waveguide-irradaiated insects experienced localized SAR values significantly higher than whole-body average SAR.
Phantom model of man	0.5, 0.75, 2, & 2.5 GHz	EM standing wave	E	Electric-field distribution	Implantable electric- field probe	ff	Distribution of measured electric field is compared with distribution of theoretical results obtained numerically from the moment-method solution of electric-field integral equation with pulse basis functions. Agreement between the experiment and theory tends to deteriorate at lower frequencies.
Human volunteers	18.5 MHz	TEM cell	EKH, EHK, KEH, KHE, HEK, HKE	Average SAR	Insertion-loss technique	gg	First set of human whole-body average SAR is presented for three subjects exposed in free space to 11 hW/cm2 ata 18.5 MHz. Measured average SARs for the two principal E-orientations (EKH and EHK) are larger than the published predictions by a factor of 2 or 3.
Saline-filled reduced-scale figurines	610-795 MHz	Leakage fields of parallel- plate applicator	---	Average SAR internal electric field	Temperature- rise measurements and electric- field measurements	hh	Experimental results using reduced-scale figurines for whole-body average SAR agree well with results obtained from the empirical relationship (Chatterjee et al., 1980). For near-field leakage- type exposures, internal E-fields and average SARs are considerably smaller than for far-field exposure conditions.
3.3-cm-radius muscle-equivalent sphere, M. mulatta head	1.2 GHz	Far field of a standard-gain horn	E and H	Rate of temperate rise	Temperature- rise memasurements	ii	Thermal response model accurately predicts temperature distribution in muscle-equivalent spheres and in detached/attached M. mulatta heads. Orientation of M. mulatta cadaver body with respect to EM- field vectors significantly affects temperature distribution in the head.
Conducting models of swine and rat	60 MHz	Electric field of astength 9.5 kV/m	---	Induced current measurements	---	jj	Body shape has a major influence on fields actually experienced by a man or animal exposed to a 60-MHz electric field.
Muscle-equivalent model of sitting Rhesus monkey	1.29 GHz and 225 MHz	Rectangular horn antenna	E	SAR distribution and average SAR	Temperature- rise measurement, calorimetric and thermographic technique	kk ll	Measured average SAR is about three times that predicted for a prolate spheroidal model of similar mass, at 1.29 GHz. However, at 225 MHz, very close agreement between theory and expepriment is seen.
Scaled phantom models of humans and test animals	57.3 MHz	TE101 mode rectangular resonant cavity	---	SAR	Thermographic technique	mm	Instrumentation system is developed for determining relationship between the applied uniform electric-field intensity at 60 MHz and the pattern of internal electric currents induced in exposed models. The 60-Hz field exposures are simulated by exposing scale models to 57.3-MHz fields of high strength on a resonant cavity.
Muscle-equivalent full-sized model of man	2 GHz	Horn antenna	E	Average SAR and SAR distribution	Calorimetric and temperature- rise measurement	nn	Relatively high SAR values are found in the limbs compared to the axis of the trunk of the model. Measured average SAR values are about three times higher than that extimated theoretically for a prolate spheroidal model of men.
Full-size muscle- equivalent human model	1.29 GHz	Far field of a standard gain horn	E	SAR distribution	Temperature- rise measurement	oo	Temperature rise is measured at front surface and several depths within the phantom. Based on these measurements, an approximate average- SAR value is obtained which agrees well with theoretical predictions based on absorption in the prolate spheroidal model of man.
Saline-filled prolate spheroidal phantom	400 MHz	Neaer field of a dipole source	E	Average SAR	Temperature- rise measurement	pp	Although relative SAR increases as distance from antenna decreases, the rate of increase of SAR values is slower than (/d)2, which is the characteristic variation of the far-field absorption produced by an electrically short dipole.
Human volunteer	3 to 41 MHz	TEM cell	EKH and EHK	Average SAR	Insertion- loss technique	qq	Measured average SAR, at 10 MHz, exceeds the average of the standard model calculations by a factor of 3 for free space and a factor of 4 for ground conditions.

REFERENCES

a. Allen, 1975		aa. Hagmann et al., 1979b
b. Phillips et al., 1975		bb. Meyerhoff et al., 1979
c. Gandhi, 1974		cc. D'Andrea et al., 1977
d. Gandhi, 1975a		dd. Ho and McManaway, 1977
e. Gandhi, 1975c		ee. Olsen and Hammer, 1982
f. Allen et al., 1975		ff. Chen et al., 1982
g. Allen et al., 1976		gg. Hill, 1982
h. Leicher-Preka and Ho, 1976		hh. Chatterjee et al., 1982b
i. Schrot and Hawkins, 1976		ii. Burr and Krupp, 1980
j. Kinn, 1977		jj. Kaune and Phillips, 1980
k. Ho et al., 1977		kk. Olsen et al., 1980
l. Bassen, 1977		ll. Olsen and Griner, 1982
m. Bassen et al., 1977b		mm. Guy et al., 1982
n. Gandhi, 1975b		nn. Olsen, 1982
o. Gandhi and Hagmann, 1977a		oo. Olsen, 1979
p. Gandhi et al., 1976		pp. Iskander et al., 1982
q. Gandhi et al., 1977		qq. Hill, 1984
r. Guy and Chou, 1976
s. Chou and Guy, 1977
t. Guy et al., 1976
u. Chen and Guru, 1977a
v. Guru and Chen, 1977
w. Bigu del Blanco and Romero-Sierra, 1978
x. Peterson et al., 1979
y. Gandhi, 1980
z. Chou and Guy, 1978