Боліхов О. Л., Красильников С. В., Степанов В. А. Електротехнічний університет (ЛЕТІ) Санкт-Петербург – 197376, Росія Тел.: (812) 3464516; e-mail: Iab16@vilan.spb.ru

Taking into account that in accordance with (6b) and (3) respectively,it is

possible to find the connection between the coefficients bw and N0 using also the linear dispersion equation for the MSW. Alternatively the method of direct expansion of the (nonlinear) dispersion equation (analogical to those used in the paper [1]) can be used, but using NLL gives

the more accurate result. Coefficients N, b20 can be

found in the similar way. Numerical computations have been done using eq. (6a) for bell-shaped and rectangular 25 ns pulses propagating in the FF with parameters closed to those used in the paper [4] with exception of FF width. The width of the FF is 10 mm and antenna length is equal to 5mm, FF thickness is L = 1.56jum, wavenumber is

. Input amplitude (the amplitude of the pulse at the input (z = 0) of the FF, Ulnp = U(z = 0)) is characterized by the value

. The scale of amplitude, U0, is determined in such a way, that in accordance with eq. (6b), N = 1. In this case, where


Finally, input amplitude is characterized by the value kthr. The computations mentioned

above are not shown here due to lacking of the place. The following qualitative results are obtained, when calculations for nonzero (6 ф 0) and zero (6 = 0) nonlinear

dispersion and diffraction coefficients are compared (for all other nonzero coefficients in the eq. (6)).

IV. Results of the numerical calculations

(1) In both cases of zero and non-zero nonlinear diffraction and dispersion coefficients, the intense 2D compression (“collapse-like behavior”) of pulses is demonstrated for kthr = 15 at the distance z~2.25 mm from the

input antenna. The only quantitative difference between the compressed pulses is obtained in this case, namely the square of their amplitudes differ by about two times.

(2)For the value kthr = 20 rather qualitative difference is demonstrated for the pulses at the distance z~6 mm under the conditions^ 20 Ф Oand bl20 = 0 . In this case pulses drop into three main pulses along the direction of propagation. If bx 20 Ф 0, the last (in time) of the pulses

has maximal amplitude, while for bl20 = 0 the amplitude

of the second of the pulses is maximal. Also, if bx 20 = 0, all pulses including the last of them are splitted in the transverse direction, in distinction to the case bl20 ^ 0. Therefore behaviors of the pulses in both cases are different for sufficiently large input amplitudes and distances from the input antenna. For the rectangular pulses this difference is more clear than for the bell-shaped pulses.

V.                             References

1. A. K. Zvezdin, A. F. Popkov. Zh. Eksp. Teor. Fiz (Translation: Sov. Physics JETP), 1983, 84, No. 2, 606-615.

2.  A. N. Slavin, I. V. Rojdestvenski. IEEE Trans. On Magnetics

1994,               30, No. 1, 37-45.

3.  R. W. Damon, J. R. Eshbach. J. Phys. Chem. Solids. 1961,

19, Nos. 3/4, 308-320.

4.  H. Hia etal. Phys. Rev. B, 1998, 58, № 5, 2708-2715.

Джерело: Матеріали Міжнародної Кримської конференції «СВЧ-техніка і телекомунікаційні технології»