Spectral properties of limiting solitons in

optical fibers

Sh. Amiranashvili,1∗ U. Bandelow,1 and N. Akhmediev2

1 Weierstrass Institute for Applied Analysis and Stochastics,
Mohrenstraße 39, 10117 Berlin, Germany
2 Optical Sciences Group, Research School of Physics and Engineering,

Institute of Advanced Studies, The Australian National University

Canberra ACT 0200, Australia
∗ Shalva.Amiranashvili@wias-berlin.de

Abstract: It seems to be self-evident that stable optical pulses cannot

be considerably shorter than a single oscillation of the carrier field. From
the mathematical point of view the solitary solutions of pulse propagation
equations should loose stability or demonstrate some kind of singular be-
havior. Typically, an unphysical cusp develops at the soliton top, preventing
the soliton from being too short. Consequently, the power spectrum of the
limiting solution has a special behavior: the standard exponential decay is
replaced by an algebraic one. We derive the shortest soliton and explicitly
calculate its spectrum for the so-called short pulse equation. The latter
applies to ultra-short solitons in transparent materials like fused silica that
are relevant for optical fibers.
© 2014 Optical Society of America
OCIS codes: (060.4370) Nonlinear optics, fibers; (060.5530) Pulse propagation and tempo-
ral solitons; (190.4370) Nonlinear optics, fibers; (190.5530) Pulse propagation and temporal
solitons; (320.5550) Pulses.

References and links

1. R. W. Boyd, Nonlinear Optics (Academic, New York, 2008), 3rd ed.
2. T. Brabec and F. Krausz, “Nonlinear optical pulse propagation in the single-cycle regime,” Phys. Rev. Lett. 78,
3282–3285 (1997).
3. G. Genty, P. Kinsler, B. Kibler, and J. M. Dudley, “Nonlinear envelope equation modeling of sub-cycle dynamics
and harmonic generation in nonlinear waveguides,” Opt. Express 15, 5382–5387 (2007).
4. T. Schäfer and C. E. Wayne, “Propagation of ultra-short optical pulses in cubic nonlinear media,” Physica D 196,
90–105 (2004).
5. K. E. Oughstun and H. Xiao, “Failure of the quasimonochromatic approximation for ultrashort pulse propagation
in a dispersive, attenuative medium,” Phys. Rev. Lett. 78, 642–645 (1997).
6. S. A. Skobelev, D. V. Kartashov, and A. V. Kim, “Few-optical-cycle solitons and pulse self-compression in a
Kerr medium,” Phys. Rev. Lett. 99, 203902 (2007).
7. A. Sakovich and S. Sakovich, “The short pulse equation is integrable,” J. Phys. Soc. Jpn. 74, 239–241 (2005).
8. A. Sakovich and S. Sakovich, “Solitary wave solutions of the short pulse equation,” J. Phys. A 39, L361–L367
(2006).
9. S. Amiranashvili, U. Bandelow, and N. Akhmediev, “Dispersion of nonlinear group velocity determines shortest
envelope solitons,” Phys. Rev. A 84, 043834 (2011).
10. S. Amiranashvili, U. Bandelow, and N. Akhmediev, “Few-cycle optical solitary waves in nonlinear dispersive
media,” Phys. Rev. A 87, 013805 (2013).
11. R. Camassa and D. Holm, “An integrable shallow water equation with peaked solitons,” Phys. Rev. Lett. 71,
1661–1664 (1993).
12. S. Amiranashvili, A. G. Vladimirov, and U. Bandelow, “Solitary-wave solutions for few-cycle optical pulses,”
Phys. Rev. A 77, 063821 (2008).

#224183 - $15.00 USD Received 1 Oct 2014; revised 27 Oct 2014; accepted 27 Oct 2014; published 26 Nov 2014
(C) 2014 OSA 1 December 2014 | Vol. 22, No. 24 | DOI:10.1364/OE.22.030251 | OPTICS EXPRESS 30251
 13. S. A. Kozlov and S. V. Sazonov, “Nonlinear propagation of optical pulses of a few oscillations duration in
dielectric media,” JETP 84, 221–228 (1997).
14. D. Gabor, “Theory of communication,” Journal of the Institute of Electrical Engineers 93, 429–457 (1946).

1. Introduction
Description of short optical pulses is usually based on the generalized nonlinear Schrödinger
equation (GNLSE) [1], which yields dynamics of single-cycle [2] and even sub-cycle [3] pulses.
The validity of the GNLSE is because of two reasons, (i) the standard slowly varying envelope
approximation (SVEA) is relaxed by considering an unidirectional regime of pulse propagation,
and (ii) the complex envelope is defined by omitting the negative-frequency components from
the real field without any reference to the SVEA. On the other hand, exact analytical solutions
of the GNLSE are rare and typically require either a simplified quadratic dispersion law or
artificial relations between equation parameters, like in the case of Sasa-Satsuma equation.
An attractive alternative to the GNLSE is to use a special non-envelope propagation model,
one of the so-called short pulse equations (SPEs). The latter are derived using a simplified
dispersion law. For instance, Schäfer and Wayne [4] noticed that the relative permittivity of
bulk fused silica can be precisely approximated by an extremely simple expression
ε(ω) = n2s (1 − ω p2 /ω 2 ), (1)
for all frequencies that are relevant for optical solitons. As opposed by the Drude disper-
sion law, ns and ω p are just fit parameters. For instance, selecting them as ns = 1.4538 and
ω p /(2π) = 21.78 THz, one gains the relative accuracy of . 0.15% between 0.9 µm and 3.6 µm
(from 83 THz to 333 THz), i.e., for two octaves in fused silica. From the physical side ns is
just a typical value of the refractive index, whereas ω p indicates at which frequency the imagi-
nary part of the refractive index cannot be neglected. The rational approximation (1) naturally
applies to the transparency region with ω  ω p .
In contrast to the Taylor expansion of the dispersion relation in the GNLSE, i.e., a local
approach that is very accurate but only within the convergence radius at the carrier frequency
[5], Eq. (1) pretends to be a global approximation. First, it is bounded for ω → ∞, as opposed by
the unphysical behavior of the GNLSE dispersion. Second, one can apply the Kramers Kronig
relations to demonstrate that Eq. (1) is a kind of universal representation of the anomalous
dispersion domain for any transparent material [6]. Third, simplified propagation equations
yield useful analytic solutions for short optical solitons, in some cases even a full solution of
the SPE is available [7]. These solutions can be used to test more general models.
In what follows we use the dispersion relation (1), to address the following fundamental
question: what happens with an optical soliton when its duration approaches a single oscillation
of the carrier field? The known analytical [6, 8, 9] and numerical solutions [10] suggest the
following feature: a singular cusp develops at the soliton top, preventing the soliton from being
too short. Such peaked solitons have originally been found outside optics, for the shallow water
waves [11].
In the frequency domain the limiting shortest soliton is characterized by the following prop-
erty: an exponential decay of the spectral power is replaced by a rational one. This behavior
is of special interest because pulse spectrum is easier to measure than the actual value of the
electric field inside the pulse. In the following we first revisit the derivation of the SPE and its
limiting soliton [6, 12] and then explicitly calculate corresponding spectra.

2. Complex short pulse equation

In this section we outline the derivation of the real [4] and complex [6, 12] SPEs and demon-
strate that the transition to a more simple complex SPE is dictated by the dispersion law (1).

#224183 - $15.00 USD Received 1 Oct 2014; revised 27 Oct 2014; accepted 27 Oct 2014; published 26 Nov 2014
(C) 2014 OSA 1 December 2014 | Vol. 22, No. 24 | DOI:10.1364/OE.22.030251 | OPTICS EXPRESS 30252
 Propagation of a short pulse in a one-dimensional setting, e.g., in a single-mode polarization-
preserving optical fiber, is approximated by a nonlinear wave equation for the electric field
   
∂t2 ε̂E + χ (3) E 3 − c2 ∂z2 E = 0, where ε̂ ∑ Eω e−iωt = ∑ ε(ω)Eω e−iωt . (2)
ω ω

The nonlinear susceptibility of the third order χ (3) is taken constant. Using Eq. (1) one derives
the following self-consistent equation

n2s (∂t2 E + ω p2 E) − c2 ∂z2 E + χ (3) ∂t2 (E 3 ) = 0, (3)

with the following dispersion law β (ω) = (ns /c)(ω 2 − ω p2 )1/2 for a linear ei(β z−ωt) wave.
Recall, that typical carrier frequencies of optical solitons are considerably larger than the fit
parameter ω p from Eq. (1), as discussed in the Introduction. Therefore β (ω) ≈ (ns /c)[ω −
ω p2 /(2ω)] and one concludes that the third harmonics generation (THG) process is a non-
resonant one for our system. Indeed, the corresponding resonance conditions ω1 + ω2 + ω3 = ω
and ω1−1 + ω2−1 + ω3−1 = ω −1 are incompatible for any four positive frequencies.
The SPE results from the asymptotic expansion of Eq. (3) with respect to µ = ω p /ω0  1,
e.g., µ = 0.1 for a standard carrier wave at 1.4 µm. Actually, we will keep O(µ 2 ) terms and
neglect O(µ 4 ) ones. To derive the SPE we first consider a reference carrier wave (the SVEA is
not assumed) ei(β0 z−ω0 t) and calculate the phase
h n i  n  n s ω0
s s
ω0t − β0 z = ω0 t − z(1 − µ 2 )1/2 = ω0 t − z + µ 2 z + O(µ 4 ). (4)
c c 2c
Motivated by this expression we introduce the normalized field F(ζ , τ) and assume the follow-
ing scaling of the new variables
!1/2
 n 
s 2 ns ω0 3χ (3)
τ = ω0 t − z , ζ =µ z, E(z,t) = µns F(ζ , τ). (5)
c c 8

A straightforward calculation yields that Eq. (3) takes the following form
8
2∂ζ ∂τ F + F + ∂τ2 (F 3 ) = µ 2 ∂ζ2 F. (6)
3
After neglecting the last term one obtains the real SPE as derived in [4]. One should mention
that a more general SPE with an additional term accounting for transition to the normal disper-
sion domain first appeared in [13]. To derive the complex SPE we introduce a complex-valued
electric field F (ζ , τ) such that by construction
1 1
∂ζ ∂τ F + F + ∂τ2 (3|F |2 F + F 3 ) = 0. (7)
2 3
It is easy to see that F = (F + F ∗ )/2 is a valid solution of the real SPE. The clear advantage
of the complex representation is that the self-phase modulation (SPM) term and the THG term
are now separated. Neglecting the non-resonant THG term we obtain the complex SPE
1
∂ζ ∂τ F + F + ∂τ2 (|F |2 F ) = 0. (8)
2
Another physical situation that directly leads to the complex SPE is for a wave with circular
polarization. Two components of the real electric field are just combined into a singe complex
field and one obtains the complex SPE (8) without referring to the real equation [6, 12].

#224183 - $15.00 USD Received 1 Oct 2014; revised 27 Oct 2014; accepted 27 Oct 2014; published 26 Nov 2014
(C) 2014 OSA 1 December 2014 | Vol. 22, No. 24 | DOI:10.1364/OE.22.030251 | OPTICS EXPRESS 30253
 To conclude this section we note that strictly speaking Eq. (8) presupposes that F contains
mainly positive (and F ∗ mainly negative) frequencies. That is, the initial condition F (0, τ)
is chosen to be exactly the positive frequency part of F(0, τ) in accord with the most general
definition of the complex envelope [1,14]. In the course of pulse propagation, the complex field
F (z, τ) accrues some small negative-frequency part generated by the SPM term, however, this
process is non-resonant such that the negative frequencies are not further amplified.

3. Solitary solutions of the complex SPE

For a linear F ∼ exp[i(κζ − ντ)] wave, in which the dimensionless frequency ν corresponds
to the physical frequency νω0 and the reference wave corresponds to ν = 1, Eq. (8) yields
κ = −1/(2ν). Therefore the phase and group velocities read Vph = −2ν 2 and Vgr = 2ν 2 , in
particular, Vph +Vgr = 0. Although both velocities slightly change for a solitary wave, they still
have equal magnitudes. Therefore we introduce a positive parameter γ such that

Vphase nonlinear = γVph = −2γν 2 and Vgroup nonlinear = γVgr = 2γν 2 , (9)

and assume that soliton’s phase and amplitude depend on τ − ζ /(γVph ) and τ − ζ /(γVgr ) re-
spectively. In accord with Eq. (9) we then try the following substitution
    
ζ ζ
F (ζ , τ) = f ντ − exp −i ντ + , (10)
2γν 2γν

where f (ξ ) is the complex shape function that depends also on γ and ν. Substituting F (ζ , τ)
into Eq. (8) we obtain the following ordinary differential equation for f (ξ )

00
f − 2γν 2 | f |2 f − (γ − 1) f + 4iγν 2 (| f |2 f )0 + 2γν 2 | f |2 f = 0, (11)

where prime denotes derivative with respect to ξ . At the soliton tails, where f (ξ ) → 0, we have
f 00 = (γ − 1) f , such that localized solitary solutions require γ > 1 and the soliton duration is
proportional to (γ − 1)−1/2 .
We now split the amplitude and the phase by setting γ 1/2 ν f (ξ ) = a(ξ )eiφ (ξ ) and obtain two
real differential equations for a(ξ ) and φ (ξ ). The phase equation can be integrated and yields

(3 − 4a2 )a2
φ0 = − . (12)
(1 − 2a2 )2

The amplitude equation A00 − (γ − 1)a − φ 02 A − 4φ 0 a3 + 2a3 = 0 with A ≡ a − 2a3 can be mul-
tiplied by A0 and then integrated as well. Restricting ourselves to the localized solutions we
obtain a “mechanical” equation for some effective potential U(a)

(1 − 3a2 )a2 (1 − 7a2 + 12a4 )a4

a02 +U(a) = 0, U(a) = −(γ − 1) 2 2
+ . (13)
(1 − 6a ) (1 − 2a2 )2 (1 − 6a2 )2

An exemplary shape of U(a) for γ slightly smaller than 9/8 is shown in Fig. 1(a). The target
soliton starts from the equilibrium position a = 0 for ξ → −∞ and returns to it for ξ → +∞.
Physically acceptable solutions require U(a) ≤ 0. The solution of the latter inequality that is
“connected” to the a = 0 state is given by
" #1/2
4γ − 3 − (9 − 8γ)1/2
0≤a≤ , (14)
8γ

#224183 - $15.00 USD Received 1 Oct 2014; revised 27 Oct 2014; accepted 27 Oct 2014; published 26 Nov 2014
(C) 2014 OSA 1 December 2014 | Vol. 22, No. 24 | DOI:10.1364/OE.22.030251 | OPTICS EXPRESS 30254
 U (a ) (a) a(ξ) (b)

0.06 0.3
0.04
0.02 0.2

0.1
a ξ
0.0 0 .2 0 .4 0.6 -10 0 10

Fig. 1. (a) An exemplary potential U(a) from Eq. (13) for γ = 9/8 − δ with δ = 10−3 . The
red point labels the upper
√ value of a(ξ ) from the inequality (14). As δ → 0, an infinite wall
is formed at a = 1/ 6, resulting in cusp formation at the top of the soliton. (b) Shape of
the shortest soliton calculated from Eq. (15).

the right limit is labelled by the red point in Fig. 1(a). The target solitary solution exists for 1 ≤
γ ≤ 9/8. Classical solitons appear for γ = 1 + ε 2 with ε  1, then a = O(ε) and in the leading
term Eq. (13) yields that a(ξ ) = ε/ cosh(εξ ). As γ increases the soliton becomes shorter, the
final limiting soliton is obtained for γ → 9/8 and is yielded by the equations

a2 (1 − 3a2 ) 1 |ξ |
− √
a02 = with a(0) = √ ⇒ aeB(a) = Λe 2 2 , (15)
8(1 − 2a2 )2 6
 ia B( √1 )
2 2 1/2
h
2 1/2 e 6
B(a) ≡ (1 − 3a ) − ln 1 + (1 − 3a ) , Λ≡ √ ≈ 0.3935. (16)
3 0 6
The phase dependence of the limiting soliton is calculated from Eq. (12). Without loss of
generality we take φ (0) = 0, then φ (−ξ ) = −φ (ξ ). The shape of φ (ξ ) is shown in Fig. 2(a).
It is of interest that the limiting values φ (±∞) = ±φ∞ can be calculated analytically from (12)
and (15)
√
(3 − 4a2 )a2 dξ
 
4( 2 − 1) 1
Z ∞ Z ∞
0
φ∞ = φ (ξ )dξ = − 2 2
da = − − arcsin , (17)
0 0 (1 − 2a ) da 3 3
the quantity φ∞ ≈ −0.8921 appears later in the equation for the soliton spectrum.

4. The limiting spectrum

In this section we explicitly calculate the spectrum of the shortest soliton. At first glance, the
key Eq. (15) looks too complicated and numerical treatment of the soliton spectrum seems to be
the only choice. Actually, this is not the case.
√ Our analytic approach is based on the observation
that for all relevant amplitudes, 0 ≤ a ≤ 1/ 6, the factor eB(a) from Eq.
√ (15) is a slow function
that is close to 1.√
This happens just because B(0) = 0 and B(1/ 6) ≈ −0.037. Actually
a(ξ ) ≈ Λe−|ξ |/(2 2) is already a very good approximation to the full solution of Eq. (15) that
is shown in Fig. 1(b). As a natural generalization we consider ξ > 0√and expand the amplitude
a(ξ ) and phase φ (ξ ) in a power series with respect to X = Λe−ξ /(2 2) .
With the help of any computer algebra system it is easy to obtain that

a3 a5 49a7 X 3 X 5 61X 7
X = aeB(a) = a − + + +··· ⇔ a=X+ + − +··· (18)
4 8 192 4 16 192

#224183 - $15.00 USD Received 1 Oct 2014; revised 27 Oct 2014; accepted 27 Oct 2014; published 26 Nov 2014
(C) 2014 OSA 1 December 2014 | Vol. 22, No. 24 | DOI:10.1364/OE.22.030251 | OPTICS EXPRESS 30255
Inserting a(X) in Eq. (12) and performing direct integration we get the full solution for ξ > 0
√ 19X 4 457X 6
φ = φ∞ + 3 2X 2 + √ + √ + · · · , (19)
2 2 24 2
√ √ !
iφ iφ∞ 1 + 12i 2 3 −143 + 88i 2 5
ae = e X+ X + X +··· , (20)
4 16
√
where X = Λe−ξ /(2 2) , Λ and φ∞ are yielded by Eq. (16) and Eq. (17). Recall that the solution
for ξ < 0 follows from the fact that a(ξ ) and φ (ξ ) are even and odd functions respectively.
Finally Eq. (20), which can be extended to any power, is used to obtain the spectrum
Z ∞ Z ∞
iφ (ξ ) iΩξ
iφ
a(ξ )eiφ (ξ ) eiΩξ dξ + c.c.,


ae Ω
= a(ξ )e e dξ = (21)
−∞ 0

where Ω refers to the normalized frequency. The result reads

R1 + S1 Ω R3 + S3 Ω R5 + S5 Ω
aeiφ + Λ3 + Λ5


=Λ +··· , (22)
1 + 8Ω2
Ω 9 + 8Ω2 25 + 8Ω2
√ √ √ √
√ S1 = −16S, R3 = 3 2C − 72S, S3 = −48 2C − 4S, R5 = −(715/4) 2C −
where R1 = 4 2C,
220S, S5 = −88 2C + 143S, and we set C = cos φ∞ and S
= sin φ∞ for the sake of brevity. For
Ω  1 Eq. (22) should be replaced by the power law, aeiφ Ω ∼ Ω−2 , which should be observed
for any generic cusp solution [Fig. 2(b)]. Actually, such frequencies are beyond validity of
Eq. (1) and SPE, such that Eq. (22) is the only reasonable result for the soliton spectrum.

φ(ξ) (a) (b)

Power spectrum (a.u.)

10-1
0.5
10-5
ξ
10-9
-15 -5 5 15
10-13
Ω
10 -1
10 0
10 1
10 2
10 3

Fig. 2. (a) Phase of the limiting soliton and (b) its power spectrum. Blue line: numerical
solution, red line: expression (22). The two auxiliary thin lines show the spectrum of the
fundamental soliton (dashed), and the Ω−4 power law (solid).

5. Conclusions
Our main result is Eq. (22) for the spectrum of the shortest soliton. As expected, the spectral
power has an algebraic decay, as opposed by the exponential one for ordinary solitons. The
spectral shape is non-trivial, such a function is difficult to reconstruct from a purely numerical
solution. Figure 2(b) shows predictions of Eq. (22) and compares them with the numerical
results for the soliton spectrum. The agreement is reasonably good for all frequencies of interest
and can be easily improved by calculating further terms in Eq. (22).

Acknowledgments
Sh.A. gratefully acknowledges support by The Einstein Center for Mathematics Berlin.

#224183 - $15.00 USD Received 1 Oct 2014; revised 27 Oct 2014; accepted 27 Oct 2014; published 26 Nov 2014
(C) 2014 OSA 1 December 2014 | Vol. 22, No. 24 | DOI:10.1364/OE.22.030251 | OPTICS EXPRESS 30256