Comparison of electromagnetic absorption characteristics in the head of adult and a children for 1800 MHz mobile phones

Abstract

The specific absorption rate (SAR) produced by mobile phones in the head of children is simulated using an algorithm based in the finite difference time domain (FDTD) method. A new model based on a 10 year old child computed tomographic images was used. The electromagnetic parameters were fitted to this age. The results are compared to the SAR calculated in the head of adults. Comparison also were made with SAR calculated in the children model when using adult human electromagnetic parameters values. It is shown that in similar conditions, the SAR calculated for the children is higher than that for the adults. When using the 10 years old child model, values around 80% higher than those for adults were obtained

Full text

Comparison of Electromagnetic Absorption Characteristics in the
Head of Adult and a Children for 1800 MHz Mobile Phones
Claudio R. Fernández1,2, Giovani Bulla1, A. C. Pedra1 and Alvaro. A. A. de Salles1
1Electrical Engineering Department, Federal University of Rio Grande do Sul (UFRGS),
Porto Alegre, RS, 90035-190, Brazil
2Federal Center for Technological Education of Rio Grande do Sul (CEFET-RS), Pelotas,
RS, 96015-360, Brazil
Abstract — The Specific Absorption Rate (SAR) produced
by mobile phones in the head of children is simulated using
an algorithm based in the Finite Difference Time Domain
(FDTD) method. A new model based on a 10 year old child
computed tomographic images was used. The
electromagnetic parameters were fitted to this age. The
results are compared to the SAR calculated in the head of
adults. Comparison also were made with SAR calculated in
the children model when using adult human electromagnetic
parameters values. It is shown that in similar conditions, the
SAR calculated for the children is higher than that for the
adults. When using the 10 years old child model, values
around 80% higher than those for adults were obtained.
Index Terms — Children, Mobile Phones, Cell Phones,
Specific Absorption Rate – SAR, Finite Difference Time
Domain – FDTD.
I. INTRODUCTION
T
HE use of mobile phone by adults, young people,
children and the elderly has grown substantially in the
last years. In parallel with this, an increased concern by
the scientific community, the authorities and the
population regarding the safety of these phones has arised.
Several authors have used the Finite Difference Time-
Domain (FDTD) method to simulate the Specific
Absorption Rate (SAR) in the cell phone user’s head [1-
10]. It is currently the most appropriate choice when
highly non-homogeneous structures are involved for
which boundary techniques have fundamental limitations.
The SAR results estimated and measured show exposure
levels close to (or even above) the limits of the available
recommendations [11, 12].
Recently the use of cell phones by young and children
has been strongly stimulated. Some authors have focused
this, and different results were presented [10, 13-16]. In
[10], the model of the children head was based on a scaled
adult model and a SAR increase (compared with adult) of
around 120 % has been obtained. In [13], the head model
was based on MRI using similar electromagnetic
parameters as those for adults, and no significant
differences between adult and children SAR results were
observed. In [15], the head model was approximated by
spheres considering some variation of the electromagnetic
parameters, and an increase of around 20 % in the
calculated SAR was shown. In [16], using also scaled
model for the children’s head with adult electromagnetic
parameters, no significant variation for the average SAR
in the whole head was observed, and when considering
the brain, an increase of around 35% in the SAR was
calculated.
In this paper the FDTD method was used to simulate
the SAR in the head of a child, and compared with the
results simulated for the adults. A new model based on a
10 year old child computed tomographic images was used.
The electromagnetic parameters were fitted to this age.
Comparison also were made with SAR calculated in the
children model when using adult human electromagnetic
parameters values. Simulations were performed using
CRAY T 94 supercomputer from CESUP [17].
All the SAR results were calculated using planar
antennas.
The use of planar antennas with moderate directivity for
mobile phones has been suggested previously [1,5,6].
These antennas radiate more in the direction opposed to
the head, and they can be very low cost, resulting
therefore in an interesting alternative to this application.
II. ANTROPOMORPHIC NUMERICAL MODELS
The geometric and the electromagnetic parameter
differences between the adult and the child were
considered in the simulations.
SAR simulations for the adult human are performed
using a model based on [18] as in previous works [4-9]. In
order to compare the SAR due to cell phones in adults and
children a new model based on 102 computed
tomographic images (Fig.1) from a healthy 10 years old
child was developed.

Fig.1. Computed tomographic image of the 10 years old child:
coronal cut.
Both models were rotated to put the ear-to-mouth line
vertically (Fig.2). This facilitate the cell phones antennas
positioning.
Fig.2. Saggital cut of the 10 years old child rotated model.
Three different simulation cases were implemented. In
case A, the adult model and parameters [19] were used. In
case B, the child model with adult electromagnetic
parameters was used. In case C, the child model with
fitted electromagnetic parameters was used. These were
obtained from comparison with the results obtained for
rats [20]. The electromagnetic parameters for adult human
are well established, with an accuracy better than 5% [19].
However data for children are still not available. A study
with rats [20] shows that conductivity and permittivity
decrease with age. For 10 day and younger rats the values
are around 20% higher than for sexually mature (adult)
rats (e.g. 50 days). One reason for this could be the higher
water concentration in the tissues of the young. The
measured results for adult individuals in different animal
species show that there is a parameter variation lower than
5% from animal to animal when considering the same
type of tissue. This was the rationale, and using similar
correspondence between parameters values and age for
humans as for rats, we obtained the fitted parameters for
the children. These values are shown in Table I.
TABLE I – PHYSICAL PROPERTIES FOR THE ADULT MAN AND
FITTED FOR THE 10 YEARS OLD CHILD AT 1800 MHZ
Age adult 10 years old
Properties ρ(g/cm3) εr σ(Ω-1/m) εr σ(Ω-1/m)
Air 0.00 1.00 0.00 1.00 0.00
Average Skull 1.85 15.56 0.43 18.48 0.54
Skin (Wet) 1.01 43.85 1.23 54.63 1.53
Average
Muscle 1.04 54.44 1.39 61.68 1.57
Average Brain 1.03 43.54 1.15 52.52 1.44
Vitreous
Humour 1.01 68.57 2.03 81.81 2.47
Fat (Mean) 0.92 11.02 0.19 13.15 0.23
Eye Tissue
(Sclera) 1.17 53.57 1.60 63.91 1.95
Brain Spinal
Fluid 1.01 67.20 2.92 80.17 3.55
Nerve (Spinal
chord) 1.04 30.87 0.84 36.83 1.02
Lens Nucleus 1.10 34.65 0.79 41.34 0.96
The cranial perimeters in both models were
approximated from ellipsis. The calculated values are in
close agreement with those shown in [13]. To adjust for
the corresponding dimensions and in order to save
memory, FDTD simulation were performed having
different cell dimensions for each of the three cases
[21,22]. Then, the distance between the antenna and the
head are slightly different. These are shown in Table II.
TABLE II – SPATIAL DISCRETIZATION
Model Adult 10 years old
Cases A B and C
Scale factor 1 1
Cranial perimeter (mm) 563.5 523.9
∆x (mm) 0.764 0.946
∆y (mm) 1.920 1.892
∆z (mm) 1.920 2.044
Antenna distance (mm) 14.51 14.19
III. ANTENNA MODELS
The cell phones were modeled using patch antennas to
compare the SAR results with previous works [5,6]. The
patch antennas were designed following the cavity models
described by [23,24]. The antenna dimensions were
adjusted according to the cell grid shown in Table II.
Special care was taken to feed the antenna at the exact

resonance frequency. For example, in the simulated cases
B and C, the antenna dimensions were L = 55.19 mm, W =
32.17 mm, H = 1.89 mm, in accordance to the grid. Then
the resonance frequency estimated using the cavity model
should be f = 1832 MHz. However S11 simulations using
FDTD and FFT show that the resonance frequency is f =
1807 MHz (Fig.3).
Fig.3. Simulated S11 for the patch antenna used in cases B and C.
This was the frequency used to feed the antenna with a
normalized power harmonic signal. Then, even the
feeding frequencies are slightly different from case to
case, we ensure a high radiation efficiency in all the
simulated cases.
The main objective of this work is to compare the child
and adult exposures in three different situations (cases A,
B, and C). Then the obtained SAR results are normalized
for the adult (case A).
IV. RESULTS
In this section the peak SAR and average SAR (1 g and
10 g) are presented. The 1g-SAR and the 10g-SAR were
calculated as spatial averages of boxes with 14∆x × 5∆y ×
5∆z (1g-SAR, Case A), 28∆x × 11∆y × 11∆z (10g-SAR,
Case A), 10∆x × 5∆y × 5∆z (1g-SAR, Cases B and C),
21∆x × 12∆y × 10∆z (10g-SAR, Case B and C). Since
there is not a great variation in the densities of the
different tissues, this can be considered a reasonably
approximation.
Since the child and the adult head models were rotated
in order to a better positioning of the cell phone, then the
vertical and horizontal cuts do not correspond exactly to
coronal and frontal cuts. Fig. 4 shows the SAR calculated
in case A.
In order to make the necessary comparison, the values
are normalised to the exposure limit recommended by the
corresponding IEEE/ANSI standard [11], SAR = 1.6
W/kg for 1 g of tissue.
In Fig. 5, SAR results for the 10 years old child model
with adult parameters (Case B) for the same six vertical
and horizontal cuts and the same scale are shown.
FDTD/FFT simulation of the patch antenna ressonance
for cases B and C
In Fig. 6, SAR results for the 10 years old child model
with parameters fitted to this age (Case C) for the same six
vertical and horizontal cuts and the same scale are shown.
In Table III the maximum simulated SAR results in the
brain for the three cases are shown.
The results were compared with that obtained for the 1
g average SAR in the adult (case A) assumed to be in the
1.6 W/kg SAR recommended IEEE/ANSI exposure limit
[11].
It is observed that an increase of around 80% in the 1g-
SAR was obtained for the children model with the fitted
parameters (case C). Even when using adult parameters in
the child model (case B), an increase of around 60% is
observed. Hence, the results obtained for the child model
when using the fitted parameters were around 9% higher
than those when using adult parameters.
TABLE III – SAR
Model Adult 10 years old child
Electromagnetic
parameters
Adult
parameters
Adult
parameters
Children
parameters
Case A B C
Normalised values (W/kg)
Peak SAR
(one voxel) 1.704 3.356 3.636
1g-SAR 1.600 2.618 2.868
10g-SAR 1.319 1.984 2.119
Percentage difference
(increase/decrease)
Peak SAR
(one voxel) +6.5% +109.7% +127.3%
1g-SAR 0% +63.6% +79.2%
10g-SAR -18% +24.0% +32.5%

Fig.4. SAR for three vertical and horizontal cuts (Case A, adult). The color scale corresponds to 10 dB for each color. The same scale
is used in Fig.5 and Fig.6.
Fig.5. SAR for three vertical and horizontal cuts (Case B, child with adult parameters). The color scale corresponds to 10 dB for each
color. The same scale is used in Fig.4 and Fig.6.

Fig.6. SAR for three vertical and horizontal cuts (Case C, child with fitted parameters). The color scale corresponds to 10 dB for each
color. The same scale is used in Fig.4 and Fig.5.
V. DISCUSSIONS AND CONCLUSIONS
The SAR in the 10 years old child was calculated and
compared to the results obtained for adult. SAR results
around 80% higher than those for the adults were
observed for the children. This is expected to be due to
differences in dimensions and electromagnetic parameters,
and is in accordance with the results obtained by other
authors [10].
Due to the increase of the use of mobile phones by
children, and since compliance tests use head phantoms
based exclusively on adult data, the results shown in this
paper may suggest that further theoretical and
experimental research must be done in order to evaluate
these issues aiming to reduce risks for the children. This is
in accordance to the WHO – World Health Organization –
effort, included in the WHO Children’s EMF Research
Agenda, recommending research studies relevant to the
risk of adverse health effects in children from exposure to
electromagnetic fields (EMFs) [25].
ACKNOWLEDGMENT
The authors are grateful to MD Sonia Tozzi (from
Radicom) and Diego Mauricio Fernández Campos for
their collaboration to the 10 years old child computed
tomographic images, and also to Martin Elbern for his
contribution in the model for the head of the child.
REFERENCES
[1] M. A. Jensen and Y. Rahmat-Samii, “EM Interaction of Handset
Antennas and a Human in Personal Communications”, Proc. of the
IEEE, vol. 83, nº 1, pp. 7-17, January 1995.
[2] S. Watanabe et al., “Characteristics of the SAR Distributions in a
Head Exposed to Electromagnetic Fields Radiated by a Hand-Held
Portable Radio”, IEEE Trans Microwaves Theory Techniques, vol.
44, nº. 10, pp. 1874-1883, October 1996.
[3] M. F. Iskander et al., “Polarization and Human Body Effects on the
Microwave Absorption in a Human Head Exposed to Radiation
from Hand Held Devices”, IEEE T-MTT, vol. 48, nº. 11, pp. 1979-
1987, November 2000.
[4] A. A. de Salles, C. R. Fernández e M. Bonadiman, “Far Field, Near
Field and SAR Simulation for Cell Phones Operating Close to the
Head” IEEE – COMSOC International Telecommunications
Symposium (ITS2002), Natal – RN, September 2002.
[5] A. A. A. de Salles, C. R. Fernández and M. Bonadiman, “FDTD
Simulations and Measurements on Planar Antennas for Mobile
Phones”, Proc. SBMO/IEEE MTT-S IMOC 2003, pp. 1043-1048,
2003, ISBN 0-7803-7824-5/03.
[6] C. R. Fernández, M. Bonadiman and A A. A. de Salles, “FDTD
simulations and measurements for cell phone with planar
antennas”, Annales des Télécommunications, Tome 59, No. 9/10,
pp. 1012-1030, Sept/Oct. 2004

[7] A. A. de Salles, C. Fernández and M. Bonadiman, “Distância da
Antena e Potência Absorvida na Cabeça do Usuário de Telefone
Celular Portátil”, Rev. da Soc. Bras. de Telecomunicações, vol. 16,
no. 1, pp. 16-28, Junho 2001.
[8] A. A. de Salles, C. Fernández e M. Bonadiman, “Simulação do
Campo Distante e da SAR na Cabeça do Usuário do Telefone
Celular para Antenas Convencionais e Planares” X Simpósio
Brasileiro de Microondas e Optoeletrônica, Recife – PE, Agosto
2002.
[9] A . A . de Salles, C. R. Fernández e M. Bonadinan, “Simulações da
SAR na cabeça e antenas planares para telefones móveis”, Rev.
Bras. de Eng. Biomédica , vol. 19, no. 2. pp. 77-90, agosto 2003.
[10] O. P. Gandhi, G. Lazzi, and C. M. Furse, “Electromagnetic
absorption in the human head and neck for mobile telephones at
835 MHz and 1900 MHz”, IEEE Trans. Microwave Theory Tech.,
vol. 44, no. 10, pp. 1884-1897, Oct. 1996.
[11] American National Standards Institute (ANSI), “IEEE C95.1-1991:
IEEE Standard for Safety Levels with Respect to Human Exposure
to Radio Frequency Electromagnetic Fields, 3 kHz to 300 GHz”,
The Institute of Electrical and Electronics Engineers, Inc., 345 East
47 Street, New York, NY 10017-2394, USA.
[12] ICNIRP Guidelines, “Guidelines for Limiting Exposure to Time-
Varying Electric, Magnetic and Electromagnetic Fields (Up to 300
GHz), International Commission on Non-Ionizing Radiation
Protection”, Health Physics, vol. 74, nº. 4, pp. 494-522, April 1998.
[13] F. Schoenborn, M. Burkhardt, and N. Kuster, “Differences in
energy absorption between heads of adults and children in the near
field of sources”, Health Phys., vol. 74, no. 2, pp. 160-168, Feb.
1998.
[14] J. Wang and O. Fujiwara, “Comparison and Evaluation of
Electromagnetic Absorption Characteristics in Realistic Human
Head Models of Adult and Children for 900-MHz Mobile
Telephones”, IEEE Trans. Microwave Theory Tech., vol. 51, no. 3,
pp. 966-971, March 2003.
[15] V. Anderson, “Comparisons of peak SAR levels in concentric
sphere head models of children and adults for irradiation by a
dipole at 900 MHz”, Physics in Medicine and Biology, vol. 48, pp.
3263-3275, 2003.
[16] M. Martínez-Búrdalo, A. Martín, M. Anguiano and R. Villar,
“Comparison of FDTD-calculated specific absorption rate in adults
and children when using a mobile phone at 900 and 1800 MHz”,
Phys. Med. Biol., vol. 49, pp. 345-354, 2004.
[17] Centro Nacional de Supercomputação, at http://www.cesup.ufrgs.br
[18] The Visible Human Project, available at
,http://www.nlm.nih.gov/research/visible/visible_human.html
[19] C. Gabriel and S. Gabriel, “Compilation of the dielectric properties
of body tissues at RF and microwaves frequencies”, technical
report AL/OE-TR-1996-0037 available in the world wide web at
http://www.brooks.af.mil/AFRL/HED/hedr/reports/
[20] A. Peyman, A. Rezazadeh and C. Gabriel, “Changes in the
dielectric properties of rat tissue as function of age at microwaves
frequencies”, Phys Med Biol v.46 1617-1629
[21] Taflove, A., “Computational Electrodynamics- the Finite
Difference Time Domain Method”, Artech House 1995, ISBN 0-
89006-792-9.
[22] Taflove, A., “Advances in Computational Electrodynamics- the
Finite Difference Time Domain Method”, Artech House 1998,
ISBN 0-89006-834-8.
[23] R. Garg, P. Bhartia, I. Bahl and A. Ittipiboon, Microstrip Antenna
Design Handbook. Artech House, 2001.
[24] Balanis, C., “Antenna Theory Analysis and Design”, John Wiley &
Sons 1997, ISBN 0-471-59268-4.
[25] Who Children’s EMF Research Agenda, available at
http://www.who.int/peh-emf/research/children/en/print.html