We study defect solitons (DSs) in a parity-time (PT) symmetric superlattice with focusing Kerr nonlinearity. The properties of the DSs with a PT symmetrical potential are obviously different from those in a superlattice with a real refractive index. Unusual features stemming from PT symmetry can be found. Research results show that the solitons with a zero defect or a positive defect can exist and stably propagate in the semi-infinite gap, but they cannot exist in the first gap. For the case of a negative defect, the soliton can stably exist in both the semi-infinite gap and the first gap.
© 2011 OSA
Due to the advanced experimental techniques, it is possible to create a new class of materials, parity-time symmetric media, which admire the combined PT symmetry but do not obey parity and time symmetry separately. In 1998, it had already been reported by Bender and Co-workers  that non-Hermitian Hamiltonians with PT symmetry can still show real spectra. Some unusual features stemming from PT symmetry will be found in such kinds of new media. In recent years, there has became more and more interest in PT symmetrical systems because of these unusual features such as double refraction and PT soliton [2–6]. On the topic of PT solitons, Z.H.Musslimani et al.  studied optical solitons in PT periodic potentials. The necessary condition of a Hamiltonian with PT symmetry has also been pointed out in their research work: when the real part of a PT complex potential is an even function of position and the imaginary part of that is an odd function of position, the Hamiltonian is PT symmetric.
The nonlinear defect modes, defect solitons, have been extensively studied in some systems in the past decades [7–13]. Lots of research results indicate that lattice defects can significantly modify the solitons properties. A variety of interesting phenomena resulting from DSs can be found and can further be applied for fundamental studies and practical applications such as controllable filtering , switching , and steering of optical beams in lattices . Recently, the existence and stability properties of linear and nonlinear modes in a PT periodic potential were studied [17,18]. In this paper, we will study defect solitons in a superlattice with a PT symmetric potential. The existence and stability of DSs will be discussed analytically and numerically.
We consider a light beam propagating in a PT symmetric superlattice in the Kerr nonlinear media described by the following Schörding equation :Eq. (1), V(x) and W(x) are related to the real part and imaginary part of the complex refractive-index distribution [2,7]: 18]. A solution different from that in Ref.  will be obtained for considering the effect of nonlinearity. Although we justly study the conserved solitons and do not considered about the case of the symmetry breaking, we will analyze the stability of solitons using the perturbation growth rate. We search the stationary solitons numerical solution of Eq. (1) in the form of U = f(x)exp(−iμz), where μ is the propagation constant and f(x) is the complex function which satisfies the following equation Equation (4) can be solved numerically .
Figure 1(a) shows Blcoh band structure of PT superlattice. From this figure, we can find that the region of semi-infinite gap is μ ≤ −3.26 and the first, second gap is −2.72 ≤ μ ≤ −1.60 and −0.17 ≤ μ ≤ 1.85, respectively. The intensity distributions of PT superlattice with defect: ε = 0, ε = 0.5, and ε = −0.5 are displayed in Fig. 1(b)–1(d), respectively.
To examine the stability of defect solitons, we search the perturbed solutions of Eq. (1) in the fromEq. (5) into Eq. (1) and then linearing Eq. (1), we arrive at an eigenvalue equation
This equation can be solved numerically to get the perturbation growth rate Re(δ). For Re(δ) = 0, the slitons are linearly stable; otherwise, they are linearly unstable.
3. Numerical results and discussion
Random-noise perturbation whose relative amplitude is set at 10% is added to the initial input light to simulate the soliton propagation. We choose ε = −0.5 as a case of the negative defect in the PT superlattices and the corresponding power P of defect solitons versus the propagation constant μ is showed in Fig. 2(a). It can be found in this figure that defect solitons can exist in both the semi-infinite gap and the first gap. In the semi-infinite gap, the stable region of defect solitons is μ ≤ −4.03 and the unstable region is −4.03 < μ ≤ −3.26. In the first gap, the stable region of defect solitons is −2.72 ≤ μ ≤ −1.75 and the unstable region is −1.75 < μ ≤ −1.61. Two stable examples: μ = −6.0 (point A in Fig. 2(a)) and μ = −2.0 (point B in Fig. 2(a)) have been used to certify soliton stability in the semi-infinite gap and the first gap. Their soliton profiles are showed in Fig. 2(c) and 2(d) and the corresponding soliton propagations are showed in Fig. 2(f) and 2(g), respectively. In the first gap, we select an unstable example: μ = −1.74 (point C in Fig. 2(a)) and plot its soliton profile in Fig. 2(e). In Ref. , the real part of soliton solution in the first gap is odd and the imaginary part is even. In Fig. 2(e), we can see an opposite result that the real part of f(x) is even and the imaginary part is odd. The corresponding soliton propagation is showed in Fig. 2(h). Figure 2(b) shows the change of Re(δ) with propagation constant μ. In the regions: μ ≤ −4.03 and −2.72 ≤ μ ≤ −1.75, the slope of power curve is negative and the Re(δ) is equal to zero; while in the regions: −4.03 < μ ≤ −3.26 and −1.75 < μ ≤ −1.61, the slope of power curve is positive and the Re(δ) is not equal to zero. From these results, we can conclude that the stability or instability of defect soliton in the both semi-infinite gap and the first gap is in accordance with the Vakhitov-Kolokolov (VK) criterion.
Power P versus propagation constant μ for a positive defect (ε = 0.5) in the PT superlattices is presented in Fig. 3(a). In this figure, we can see that the defect solitons can only exist in the semi-infinite gap. In these regions which we have considered, defect solitons can stably transmit. We choose two stable example:μ = −6.5, −5.0 (points A and B in Fig. 3(a)) and show their soliton profiles in Fig. 3(b) and 3(c), respectively. Figure 3(d) and 3(e) present their corresponding soltion propagations. In the regions of the semi-infinite gap which we have considered, the negative slope of power curve and the Re(δ) = 0 indicate that the solitons stability is in accordance with the VK criterion.
Figure 4(a) plots power P versus propagation constant μ for the case of a zero defect (ε = 0). This figure indicates that solitons only exist in the semi-infinite gap. In the regions of propagation constant: −4.07 ≤ μ ≤ −3.50, and −3.36 ≤ μ, the solitons cannot stably propagate. One example: μ = −3.3 (point B in Fig. 4(a)) has been introduced to certify the instability of soliton in these regions. Its solitons profile is showed in Fig. 4(d) and the corresponding soliton propagations is showed in Fig. 4(f). In the regions of propagation constant: μ < −4.07 and −3.50 < μ < −3.36, the solitons can stably transmit. The soliton profile of a stable example: μ = −5.0 (point A in Fig. 4(a)) is shown in Fig. 4(c) and its corresponding propagation is showed in Fig. 4(e). To further verify the instability, we plot the change of the Re(δ) with propagation constant μ in Fig. 4(b). For the negative slope of power curve (dP/dμ < 0) and Re(δ) = 0 in the regions: μ < −4.07 and the positive slope of power (dP/dμ > 0) and Re(δ) > 0 in the regions: −3.36 ≤ μ, we can conclude that the solitons stability or instability in these regions is in accordance with the VK criterion.
Figure 5 shows stable and unstable domains (μ, ε) of the PT solitons according to the relation between defect depth ε and the propagation constant μ. When increasing the defect depth of positive defect from ε = 0.1 to 0.8 or negative defect from ε = −0.1 to −0.7, the stable region of defect solitons in the semi-infinite gap will become narrow (see Fig. 5(a) and 5(b)). For the case of ε = −0.8, the stable region will disappear. In the first gap, the stable region of defect soliton will firstly become large and then become narrow when increasing the defect depth from ε = −0.1 to −0.6. The stable region is vanished for the case of ε = −0.7 (see Fig. 5(c)). The stability of solitons propagation is mainly affected by light diffraction and self-focusing resulting from nonlinearity.Beacuse of the counteraction between light diffraction and nonlinearity, the soliton pulses can stably propagate. When diffraction cannot be suppressed by nonlinearity, soliton pulses will finally decay into linear diffractive waves. In the case of a negative defect, the defect site, which has lower light intensity, can leads to that the repusion from the negative defect will increase the light diffraction and then change the existence and stability of PT solitons. Thus, with the effect of a negative defect mode, the PT gap soliton cannot only exist, but also stably propagate in the first gap. In the case of a positive defect, the defect site, which has a higher intensity, is attractive to the light field. The attraction from the positive defect will decrease the light diffraction. As a result, the existence and stability of PT solitons will also be altered, justly like the case of the negative defect.
To summarize, we have demonstrated that stable defect solitons can be formed in a parity-time symmetrical superlattice with focusing Kerr nonlinearity. These solitons properties are remarkably different from those in a conventional superlattice which has not considered a parity-time potential. The parity-time potential can drastically affect the existence and stability of defect solitons. Numerical results in this paper show that the soliton can stably exist in the semi-infinite gap but cannot exist in the first gap for the cases of a zero defect or a positive defect. For the case of a negative defect, the stable soliton can exist in both the semi-infinite gap and the first gap.
This work was supported by the National Natural Science Foundation of China (Grant No. 60978009) and the National Basic Research Program of China (Grant Nos. 2009CB929604 and 2007CB925204).
References and links
1. C. M. Bender and S. Böttcher, “Real spectra in non-Hermitian Hamiltonians having PT symmetry,” Phys. Rev. Lett. 80, 5243 (1998). [CrossRef]
3. K. G. Makris, R. El-Ganainy, D. N. Christodoulides, and Z. H. Musslimani, “PT-symmetric optical lattices,” Phys. Rev. A 81, 063807 (2010). [CrossRef]
6. C. E. Rüter, K. G. Makris, R. El-Ganainy, D. N. Christodoulides, M. Segev, and D. Kip, “Observation of parity-time symmetry in optics,” Nat. Phys. 6, 192 (2010). [CrossRef]
7. J. Yang and Z. Chen, “Defect solitons in photonic lattices,” Phys. Rev. E 73, 026609 (2006). [CrossRef]
10. X. Zhu, H. Wang, T. W. Wu, and L. X. Zheng, “Defect solitons in triangular optical lattices,” J. Opt. Soc. Am. B 28(3), 521 (2011). [CrossRef]
11. Y. Li, W. Pang, Y. Chen, Z. Yu, J. Zhou, and H. Zhang, “Defect-mediated discrete solitons in optically induced photorefractive lattices,” Phys. Rev. A 80, 043824 (2009). [CrossRef]
13. A. Szameit, Y. V. Kartashov, M. Heinrich, F. Dreisow, T. Pertsch, S. Nolte, A. Tünnermann, F. Lederer, V. A. Vysloukh, and L. Torner, “Observation of two-dimensional defect surface solitons,” Opt. Lett. 34(6), 797–799 (2009). [CrossRef] [PubMed]
15. F. Ye, Y. V. Kartashov, V. A. Vysloukh, and L. Torner, “Nonlinear switching of low-index modes in photonic lattices,” Phys. Rev. A 78, 013847 (2008). [CrossRef]
16. A. Piccardi, G. Assanto, L. Lucchetti, and F. Simoni, “All-optical steering of soliton waveguides in dye-doped liquid crystals,” Appl. Phys. Lett. 93, 171104 (2008). [CrossRef]
19. J. Yang and T. I. Lakoba, “Universally-convergent squared-operator iteration methods for solitary waves in general nonlinear wave equations,” Stud. Appl. Math. 118(2), 153–197 (2007). [CrossRef]
20. J. Yang, “Iteration methods for stability spectra of solitary waves,” J. Comput. Phys. 277(14), 6862–6876 (2008). [CrossRef]