See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/268340762

EFFICIENCY OF SOLITARY-WAVES RADIATED BY THE DISCHARGE IN A

CONFINED PLASMA COLUMN

Article

CITATIONS READS

0 18

1 author:

Gérard Dubost
Université de Rennes 1
132 PUBLICATIONS 470 CITATIONS

SEE PROFILE

All content following this page was uploaded by Gérard Dubost on 23 September 2015.

The user has requested enhancement of the downloaded file.

 EFFICIENCY OF SOLITARY-WAVES RADIATED BY THE
DISCHARGE IN A CONFINED PLASMA COLUMN

Gérard DUBOST
Institut d’Electronique et de Télécommunications de Rennes
UMR CNRS 6164, Université de Rennes 1,
Avenue du Général Leclerc, Campus de Beaulieu, 35042 – Rennes Cedex – France

André BELLOSSI
Villa Gabrielle, Chemin du Goh Vras, 56730 – Saint Gildas de Rhuys – France

James BARE
8005 Marble Avenue NE Albuquerque, NM87110, USA

ABSTRACT
To day it is obvious to consider the radiation efficiency of the « Priore machine » which was reported to cure
tumors in rats essentially imputable to the discharge in a plasma tube (1964). The ultra-sound radiation
hypothesis has to be rejected because the huge transmission attenuation. We explain now the propagation of
solitary-waves or solitons by the non linearity effects produced inside a confined plasma tube excited by a high
frequency modulated by a low frequency square signal [10]. The soliton measured with success is the single
solution of the Kortweg-de-Vries non linear differential equation. These non-scattered waves are called “pseudo
sonorous” owing to the slow speed of the argon ions. The radiation is effective because the plasma column acts
as a small dipole antenna in the near field area. The ion current distributions and the radiated electromagnetic
fields have been measured inside two different laboratories (France and USA). The soliton theoretical
magnitude has been expressed in terms of the Landau length, the mean distance between the two argon ions, and
the Debye screen wavelength of the plasma. The TM wave radiated by the plasma antenna has been measured
with success in a good agreement with the theoretical one. The spectral analysis showed a radiation for the low
modulation frequencies comprised between 2 and 20 KHz associated with their odd harmonics. The deduced
endogenous field has been recently used to calculate the ionic current induced along a nervous fiber which can
be considered as an individualized electrical unit. The diffracted electric field of low frequency can reach
several dozens of kV/m in a near environment [11]. The interaction between solitary waves and dielectric
matter, for example the sea water, can give rise to Zenneck waves, practically propagating without attenuation
[10].

INTRODUCTION
Around 1970 the analysis of the radiation of the « Priore machine » has been done as well as can be expected
because of its extraordinary complexity. Nevertheless the absence of the X radiation has been proved and
recently the ultrasound hypothesis has to be rejected because the huge transmission attenuation through the
glass and the air medium. Most studies related to interactions between millimeter waves and living come from
Russia and Eastern European countries. The Russian safety is stricter than that of most western countries, whose
standards are only essentially based on the calculated thermal load that would be produced on people exposed to
R.F. radiations, especially during experiments that are necessary to develop high power military systems. In fact
it was important to discover the possible biological fields subtle effects. We demonstrated [1], [2], [3] that the
endogenous field inside a mouse radiated by a millimeter waves device may produce only subtle biological
effects. So for f = 60 GHz an endogenous electric field comprised between 3 and 18 V/m has given rise to a
direct or indirect effect on promotion of malignant tumors and an increased activity. In cellular membrane
 G. Dubost and A. Bellossi, J. Bare

electric field of some V/m at a R.F frequency of several dozen MHz can create a strong agility of different ions.
It is not easy to be on the safe side of the radiation protection limits as it is shown on an example at 900 MHz
[12]. Furthermore different electromagnetic fields led to various biological effects when considering separately
electric and magnetic induction fields of low frequencies [4].

We thought to analyse the low frequency radiation due to the ionic current discharge of a confined plasma [10].
We suppose a plasma without collisions. A gas at a partial pressure P and at the temperature To is introduced
inside a quasi cylindrical column of a Vp vacuum volume. The discharge is lighted by means of two conductors
located at each outside end of the column (collars or spirals) which are connected to an impedance adapter. This
one is linked to a f high frequency generator modulated by square signals of fr modulation frequency. Let us V
be the difference of potential applied between the two collars. The theoretical maximum speed of the electrons
inside the discharge is given by (1) :
(v e ) M = (2.V.e )
me 1/ 2 (1).

With V = 430 Volts measured at f = 27 MHz and e me = 1.76 10

11
, we deduced : v e ( )M = 1.2 107m / s .

The plasma argon corresponds to the following parameters : P = 5T = 6.65 102 Pascals, Ti ≅ To =300 degrees
Kelvin, Te= 3.104 , α = n e ( n e + n o ) : (2). The ionization degree is α with n o and n e the argon and
electron densities. For a neutral gas : n e ≅ n i where n i is the ionic density. The mean distance d e between two
electrons, or two ions, or one ion and one electron is roughly given by : d e = ( n e )−1 / 3 . With α = 10−4 and

n o = P KTo (3) where K is the Boltzmann constant we deduced from (3) : no = 1.6.1023 per m3 , and from
(2) : n e ≅ n i = 1.6.1019 / m3 . The Debye potential around argon ion is : Φ = ( e 4 π εo r ) exp − r / λ De ( ) (4),

with the Debye electronic wavelength : (

λ De = εo KTe / n e.e2 1 / 2 ) (5).
2
The Landau length is : ro = e 4πεo KTe (6) and the Debye screen wavelength : λS = λ De 2 for a single
ionized gas. We deduced : λ De = 3µm, ro = 5.6.10−4 µm, λS = 2.1µm, d e = 0.4µm. In that case the
plasma is a cinetic classic one with : ro << d e << λS . It can be given as a perfect gaz faintly ionized. The
discharged column is equivalent to a dipole of L length. That is the problem of plasma antennas [6]. Let us
assess vi the ion speed. Three assessments are given :

- Impact ion-electron : ( vi )c # ( 2me / mi ) ( ve )M (7) , with mi = 6.63.10-26 Kg

- Thermic speed : ( vi )th = ( 2KTi / mi )1 / 2 (8), with K = 1.38.10-23 J/deg.
- Laplace formula : ( vi )L = ( γ KTo / mi )1 / 2 (9), with γ = 1.402.

We found with the previous values : ( vi )c = 324m / s, ( vi )th = 352m / s, ( vi )L = 296 m/s. The phase
speed CS is given by : CS = ( KTe / mi ) 1/ 2
(10). It is a no dispersive wave [7] and CS equal to : 2500 m/s.
The radiation of the ionic current is explained because of the discharge length L = 0.30 m is in some idea with
the wavelength CS / f r . In effect for 1 ≤ f r ≤ 10KHz, we have 0.25 ≤ CS / f r ≤ 2.5m . The radiated wave
kind is of transversal electromagnetic. The theoretical electric field ( Eθ ) f r radiated at the distance r from the

ionic discharge current : If r is given by :

( Eθ )f r = L. If r r 3 .4 π εo.ωr (11)

The relation (11) is available if : (µoεo )1 / 2ωr r << 1 and r >> L.

If r is given by : If r = 2 π f re ( ∆ni ) f r.VP (12)

2
 Solitary-waves radiated by a confined plasma column

( ∆n i )f r is the argon ion density used in the discharge, in non correlation with the non linear effects. From (11)
and (12) we deduced the useful argon ions :

( ∆ni )f r .Vp = ( E θ )f r 34 π εo / eL
r
(13)

For various frequency f r (0.5, 1, 4, 6, 8 and 10 KHz) the measured field ( Eθ )f r was practically constant. So
with (13) we can conclude that the useful argon ions are independent of f r . We will put : ∆n i = ( ∆n i )f (14).
r
The table I shows for f r =2KHz the measured electric field and the theoretical one calculated for If r =0.14 10-6
A, expressed in terms of r (11).

r(m) 0.90 1.10 1.30 1.50 1.70 1.90 2.20 2.35

Measured 33 27 13 10 6 3.3 2.6 1.9
Theory (11) 40 22 13 8.7 6.0 4.3 2.8 2.3

Table I : Electric field (mV/m) E θ at f r =2 kHz [10]

We put : ΨM = ∆n i / n i (15) which is the solution of Kortweg-de-Vries non linear equation, called solition [9].
With E θ = 4.3.10-3 V/m (Table I) and from (13) for r = 1.90m, fr = 2.103Hz, Vp = 10-3m3, and L = 0.3m , we

ions / m 3 and ΨM = 4.2 10 .

10 -9
find : ∆n i = 6.8 10

The purity of the spectra for the most characteristic rays is shown in Table II and III measured at fr = 2 and 10
KHz, with r = 0.75m.

f(k Hz) 0.45 0.8 1.2 1.6 2 2.45 2.8 3.2 3.55 4 4.45
Eθ m V/m 3.3 3.3 7.5 1.7 70 1.3 10 2.6 3.3 3.3 2.6

Table II - Purity of the electrical field radiated : fr = 2 kHz r = 0.75m [10]

f(k Hz) 6.5 7.5 10 11.5 12.1 13.5 14.5

Eθ m V/m 1.3 2.7 107 13 3.4 1.7 0.8

Table III - Purity of the electrical field radiated : fr = 10 kHz, r = 0.75m [10]

The induction magnetic field deduced from (11) is equal to : (Hφ )fr = εo ωr r (Eθ ) fr (16).
3 −15 -16
So for 0.9 < r < 2.35m et fr = 2.10 Hz, we deduced the very weak values : 5.10 ≤ βφ < 7.5.10 Telsa. The
experimental results shown in Tables I, II, III, have been obtained at the University of Rennes 1.

The following measurements of the discharge ionic current were carried out in Albuquerque (USA) using an
equivalent device shown in fig 1, which is more powerful. In effect the plasma is started from a 50 Telsa
pressure, that is 6.65.103 pascals, of argon 80 %, neon 20 % mixed with a small amount of mercury added into
the tube. Tube length, without electrodes, is approximately 60cm. Tube outer diameter is 25mm. Inside diameter
is about 22 mm (fig 1). Antenna tuner with a balun allows a correct matching between the tube electrodes and

3
 G. Dubost and A. Bellossi, J. Bare

the amplifier output. The S.W.R indicates the optimum functioning. Typical power levels for all frequencies
were around 295 to 300 watts (instead of 100 watts at University of Rennes). A square wave generator modulate
at the frequency fr the high frequency signal of 27 MHz : - The higher pressure makes the tube get too hot. As
the mercury ionizes a much greater diameter plasma will form. The high temperature ( Ti )H of mercury ions
g
explains their reradiation at (λ )Hg comprised between 0.4 and 0.7 µm in the visible range with
(λ )Hg = hc / K( Ti )Hg (17). (h is the Planck constant). With 0.4 ≤ ( λ )Hg ≤ 0.7µm, we deduced from (17) :

2.1.10 ≤ ( Ti )Hg ≤ 3.6.10

4 4
(Kelvin). Absorbed progressive waves due to ionized mercury atoms appears
inside the plasma tube (fig. 2). For fr = 8 kHz we find with λ Hg =5.9 cm after λ Hg = ( CS ) / f r (18) the phase
speed ( CS )Hg : ( CS )Hg =472 m/s. The table IV shows the λ Hg for different frequencies f r .

fr( kHz ) 2 3 4 5 6 7 8 9 10 Origin

λ Hg (measures) 23.8 16.0 11.9 9.5 8.3 7.0 5.9 5.1 4.6 Albuquerque
λ Hg (18)
23.6 15.7 11.8 9.4 7.9 6.7 5.9 5.2 4.7 Rennes
(theory)

Table IV : Wavelengths/cm of progressive mitigated waves of mercury ions

The measurements confirm the wavelength law in inverse proportion of the modulation frequency fr (18). This
progressive wave is non-dispersive wave for CS is independent of fr.

Non linear waves or solitons theory. An hydrodynamic theory can explain their formation. The theory is
unidimensional : [9]. Its knowledge will permit to find the argon ion density used in the discharge. The
normalized potential φ = eΦ / KTe << 1 (4) respond to a second order differential equation, the maximum of
which φM being given by the equation :

1/ 2
⎞ ⎡⎛ V ⎞ ⎤
2
⎛
exp ( φM ) + ⎜ V ⎟ ⎜⎢ ⎟ − 2φM ⎥ =1+ V (19).
⎝ CS ⎠ ⎢⎣ ⎝ CS ⎠ ⎥⎦ CS

Let V / CS be equal to 1 + ν o . For a solitary impulse single solution, we have to respect the inequality :
φM < 1 (1 + ν o )2 (20). A limit development of (19) gives : ν o = φM / 3 (21). The table V shows ν o (19)
3
2
and ν o (21) in terms of φM .

φM 10-3 5.10-3 10-2 0.025 0.05 0.10 0.15 0.30 0.50

3
νo = φM / 3 3.3 10-10 4.1 10-8 3.3 10-7 5.2 10-6 4.2 10-5 3.3 10-4 1.1 10-3 0.9 10-2 4.2 10-2
(21)
ν o (19) 3.3 10-10 4.2 10-8 3.4 10-7 5.4 10-6 4.5 10-5 3.9 10-4 1.4 10-3 1.4 10-2 7.6 10-2

Table V : φM Solution of (19) and (21)

The normalized potential of the first kind such as : ψ = εφ1 follows the non linear equation of Kortweg-de-
3 3
Vries (22) : ∂ψ / ∂τ + ψ ∂ ψ / ∂ξ + (1 / 2). ∂ ψ / ∂ξ =0 (22). The normalized variables are :

τ = ε3 / 2.ωpi t (
ξ = ε1 / 2 x / λ De − ωpi t ) (23) with : ωpi = e ( ni / εomi ) 1 / 2

4
 Solitary-waves radiated by a confined plasma column

Plasma tube

Antenna tuner with S.W.R.

indicator and balun

Power amplifier (300 watts)

Transmitter (55 watts)

Square wave generator

figure1 : Equipment (USA)

3
figure 2 : Progressive mitigated waves due to ionized mercury atoms ( f r = 8.10 Hz) (USA)

5
 G. Dubost and A. Bellossi, J. Bare

ωpi is the pulsation of ionic plasma : ωpi = e ( ni / εo mi )1 / 2 , and λ De the Debye electronic wavelength : (5).
The single solution of (22) called “soliton” is given by :

ψ = 3νo sec h
2
[( γo / 2 )1 / 2 (ξ − νoτ )] (24)

With (23), we can write (24) like that (25) :

ψ ( x, t ) = 3γ o sec h 2 { (νoε / 2 )1 / 2 [x / λDe − ωp t (1 + ενo )]}

i
(25)

ψ is the fluctuation of the ions density following the long time scale which is normalized to the whole ions.

The function sech is lower than one. So we can write with (15) : ψ M = 3νo = ε ( φM )1 = ∆n i / n i (26)
The soliton speed is equal to :

VS = ( KTe / mi )1 / 2 . (1 + εν o ) (27)

From (26) we deduced : νoε = ψ 2M / 3 ( φM )1 and with (21) :

(28) ν oε = [(φM )1 ] 5 3 , ψM = [(φM )1 ] 3 (29), ε = ( φM )12 (30)

With the mean distance d e between two ions : d e = ( n e )−1 / 3 , (4), (6) we deduced for the normalized
potential :
r ⎛ d ⎞
(φ M )1 = o . exp ⎜⎜ − e ⎟⎟ (31).
de λ De
⎝ ⎠

From (26) and (29) we have :

[ (
ψ M = ∆n i n i = (ro / d e ) exp − d e λ De )]3 (32)

The found experimental value obtained at the University of Rennes 1 : ∆n i / n i = 4.2.10−9 (26), and related

in our paper, can be compared with the theoretical one : (31). We have already calculated : ro = 5.6.10 −4 µm ,

d e = 0.4 µm, λ De = 3µm . We find ψ M = 1.8.10 −9 and from (29) or (30) : (φ M )1 = 1.2 10 −3 and
Φ M = 3.1 mV. Besides with (4) and r = d e , which is the mean distance between two ions, we calculate :
−3
Φ = 3.8 mV and φ = 1.5.10 . So we proved the correlation between the hydrodynamic theory and the
experiments and confirm the judicious choice of the plasma parameters. We have to note that the soliton speed
(27) with : (10) depends of its amplitude 3ν o but is near C S . The Table VI shows the variations of the
parameters ε : (30), ν o ε : (28), and ψ M : (29) in terms of Debye’s normalized potential (φ M )1 (31).

6
 Solitary-waves radiated by a confined plasma column

(φ M )1 (31) 10 −3 5. 10 −3 10 −2 0.025 0.05 0.1 0.15 0.30

ε (30) 10 −6 2.5 10 −5 10 −4 6 10 −4 2.5 10 −3 10 −2 2.2 10 −2 0.09

νoε (28) 3 10 −16 10 −12 3 10 −11 3 10 −9 10 −7 3 10 −6 2.5 10 −5 8. 10 −4
3 νo = ψ M
10 −9 10 −7 10 −6 1.6 10 −5 1.2 10 −4 10 −3 3.3. 10 −3 0.03
(29)

Table VI : Non linear effect of the confined plasma

Soliton factorization in Fourier series. The short time between two impulses is equal to 1 / f r . The time return
to the balanced plasma is done following the law : exp(-t/ τ p ), with the τ p relaxation time equal to :

τ p = (1 / e)(ε o m e / 2n e )1 / 2 << 1 / f r (33).

The plasma electronic pulsation is : ω pe = e (n e / ε o m e )1 / 2 (34).

−30 19 3 −12
With me = 0.91.10 Kg, ne = 1.6 10 / m , we obtain : f p = 36 GHz and τp = 3ps = 3.10 s .
∞
We have : τ p << 1 / f r . Let ψ ( t ) be : ψ(t ) = b o + ∑ b N cos (Nω r t ) (35)
N =1

With (25) the b and b N coefficients are found :

b o = 3f r ν o ∫
1 / 4f r
−1. / 4f r
sec h 2 [ (ν o ε / 2)1/ 2 . ωp t] dt
i
(36)

b N = 6f r ν o ∫
1 / 4f r
−1 / 4f r
sec h 2 [ (ν o ε / 2)1/ 2 . ωp t ]. cos (N 2 π f r t) dt
i
(37)

ωpi is the ionic plasma pulsation : ωpi = e ( n i εo mi )1 / 2 (38). With mi = 6.63 10−26 Kg ,
n i = 1.6 1019 / m3 , we find : ωpi =0.84 109 . The numerical calculus shows the independence of the bo and
b N coefficients in view of the frequency modulation f r just like is the soliton : ΨM = ∆n i / n i .
With ν o = 1.4 10−9 , ν oε = 1.6 10−15 and 9
ωpi =0.84. 10 we calculate (36) : bo = 1.65.10−9 and bN
given in table VII :

N 1 2 3 4 5 6 7 8 9
bN (37) 2.10-9 9.10-15 7.10-10 4.10-15 4.10-10 10-15 -3.10-10 1.6.10-12 2.3.10-10

Table VII : Fundamental and harmonics frequencies

7
 G. Dubost and A. Bellossi, J. Bare

The even harmonic levels are weaker than the odd ones as it has been found experimentally. On the other hand
for a higher normalized potential ex : ( φM )1 = 2.5.10−2 instead of 1.2.10−3 all calculated harmonics are of
some idea of the amount.

Expansion to a highly ionized plasma. With (13) and (32) we deduced the radiated electric field which is
independent of f r :
(
E θ = e LVp n i 4 π εo r 3 ) [(r
o (
/ d e ) . exp − d e λ De )] 3 (39)

With (2) and (3) we can write :

n i = α .P [KTo(1 − α )] (40)

where P is the gas partial pressure at To and α the ionisation degree. In (39) we have to note the LVp
homogeity : a length to the power of four.

From (6), (39) and (40) and expressed in units of international system we can write :

3
L.Vp.n i2 6.7.10− 24 ⎡ ⎛ 0.015 n1 / 6 ⎞⎤
Eθ = ⎢ exp ⎜ − i ⎟⎥ (41)
3 3
⎢ ⎜ 1/ 2 ⎟⎥
r Te ⎣ ⎝ Te ⎠⎦
and :
ni = α .P.2.4 1020 (42)
1− α

3
P is expressed in pascals, n i the ion number per m , L in meters, Vp in cubic meters, r in meters and E θ
in Volts per meter.

We have shown [10] that the interaction between solitary waves and dielectric matter can give rise to Zenneck
wave or surface waves, practically propagating without attenuation.

SUMMARY
We brought to the fore the solitary waves radiation emitted by a confined plasma column weakly ionized by a
high frequency discharge. The high frequency is modulated in square signals at low frequencies. Between two
pulses the plasma is balanced again for the modulation frequency is very much lower than the plasma electronic
one. The argon low ion speed creates non dispersive waves called “pseudo sonorous”. Their evaluation appeals
to the non linear effects inside the plasma discharge following the hydrodynamic theory. The theoretical results
are in good agreement with the experimental ones. In particular absorbed progressive waves due to some
ionized mercury atoms immersed in a hot plasma of argon (80 %) and neon (20 %) have been visually detected.
The solitary waves are radiated by the plasma discharge which acts like a doublet antenna functioning in TM
mode. We showed that the useful ion number in the discharge is independent of the modulation frequency and
its harmonics. We have expressed the soliton amplitude in terms of the fundamental caracteristics of the plasma
(32) : the Landau length, the mean distance between two ions and the Debye screen wavelength. The theoretical
near electric field radiated by the “plasma antenna” appears in terms of the volume and the length of the
column, the ion density, and the electronic temperature (41). This emission of low frequency pulsed
electromagnetic fields can be the subject of ionics currents induced along a nervous fiber and reradiation of that
low frequency but high amplitude electric field in the interstitial medium [11], and then the excitation of ions
which can induce an ultraviolet radiation liable to penetrate into the cells [13].

8
 Solitary-waves radiated by a confined plasma column

REFERENCES

[1] A. BELLOSSI, G. DUBOST, J.P. MOULINOUX, M. HIMDI, M. RUELLOUX, C. ROCHER : « Biological

Effects of Millimeter-wave irradiation on Mice. Preliminary Results”. IEEE Transactions on Microwave Theory
and Techniques. Vol. 48, n° 1, Nov. 2000 (p. 2104 to 2110).

[2] A. BELLOSSI, G. DUBOST : “Préventions des effets biologiques des ondes millimétriques aux fréquences
28, 38, 60, 77 et 90 à 100 GHz”, R.S.T.D. : Revue Scientifique et Technique de la Défense, 50 – 2000-4, Déc.
2000, (p. 45 à 55).

[3] G. DUBOST, A. BELLOSSI, M. HIMDI : “Effects of 60 GHz electromagnetic fields on motor activity of
healthy mice”. 6th congrès annuel de la Société Française de Pharmacologie, Avril 2002, Rennes, France.

[4] A. BELLOSSI, G. DUBOST, M. HIMDI : « Effets biologiques du champ électrique et du champ

d’induction magnétique aux très basses fréquences, R.S.T.D. : Revue Scientifique et Technique de la Défense,
1997-4 (p. 83 à 96) .

[5] G. DUBOST : « Propagation libre et guidée des ondes électromagnétiques. Applications aux guides et fibres
optiques ». Masson, 3è Edition, 1995 (291 pages).

[6] G.G. BORG, J.H. HARRIS, N.M. MARTIN, D. THORNCRAFT, R. MILLIKEN, D.G. MILJAK, B.
KWAN, TNG, J. KIRCHER : « An investigation of the properties and applications of plasma antennas ».
Switzerland-Millenium Conf. on Antennas and Propagation Davos, April 2000.

[7] D. QUEMADA : « Ondes dans les plasmas ». Collection de Physique du CNES. 1968.

[8] J.P. DAVID : Cours de Radioélectricité Générale. Tome 4, « Propagation des ondes », ENST – Paris, 1948.

[9] J.L. DELCROIX, A. BERS : « Physique des plasmas ». Tomes I (375 pages), et II (499 pages). Inter
Editions/CNRS, 1994.

[10] G. DUBOST, A. BELLOSSI : “Rayonnement et propagation d’ondes solitaires générées par un plasma
confiné”. R.S.T.D. : Revue Scientifique et Technique de la Défense, Sept. 2002, p. 115 à 126.

[11] G. DUBOST, A. BELLOSSI : “Rerayonnement in situ d’une fibre nerveuse irradiée par un champ
électromagnétique externe de très basse fréquence”. R.S.T.D. : Revue Scientifique et Technique de la Défense,
61, Octobre 2003, p. 127 à 140.

[12] G. DUBOST, A. BELLOSSI, M. HIMDI, J.P. DANIEL : “Safety irradiation of humans by R.F near fields
due to dipoles”. J.I.N.A. 98, Nice, Int. Symposium on Antennas, p. 639 to 641.

[13] G. DUBOST, A. BELLOSSI : “Modèle physique de la membrane cellulaire dans les spectres infrarouge,
visible et ultraviolet ». R.S.T.D. : Revue Scientifique et Technique de la Défense, n° 64, Juin 2004 (15 pages).

View publication stats