Signal amplification based on the local nonlinear Mach–Zehnder interferometer Arpita Srivastava, Man Mohan Gupta, S. Medhekar n Department of Applied Physics, Birla Institute of Technology, Mesra, Ranchi 835215, India article info Article history: Received 14 June 2011 Received in revised form 22 August 2011 Accepted 23 August 2011 Available online 13 September 2011 Keywords: Cross phase modulation Mach–Zehnder interferometer Optical amplifier abstract Using the phase modulation of spatial solitons, a new scheme for all-optical signal amplification has been proposed in this paper. The considered structure is composed of the nonlinear Mach–Zehnder interferometer (NMZI) with the straight control waveguide (CWG), the uniform nonlinear medium (NLM) and the linear output waveguide. The local NMZI functions like a phase shifter. The light-induced index changes in the local nonlinear MZI are responsible for the input beam routing in the uniform nonlinear medium. The coupling of the input beam to the output waveguide depends on its propagation direction in the NLM. It is shown that the signal launched at CWG can deflect the beam launched at the NMZI (input beam) and a modulated (amplified) output could be obtained at the output waveguide. Further, signal pulse may be reshaped by appropriately increasing the NLM length. In addition, amplification factor may be enhanced by increasing the NLM length and injecting an appropriate continuous wave beam along with the signal beam at CWG. & 2011 Elsevier Ltd. All rights reserved. 1. Introduction One of the most important components in integrated optical circuits is the Mach–Zehnder interferometer (MZI). MZI has been extensively utilized for optical devices, for example, multiplexing and modulation, low-loss combiner, WDM applications, optical power limiter [1–4], etc. Proposals for various applications using MZI based devices exist in the literature. NMZI (one or both arms are made up of nonlinear materials) has thoroughly been studied for all-optical devices [5–14]. Recently, all-optical switching and all-optical logic gates are proposed using a novel structure consisting of an NMZI along with a CWG and uniform nonlinear medium [15,16]. Using a structure similar to that of Refs. [15,16], we propose signal amplifi- cation and signal reshaping in this paper. 2. The device We consider an NMZI with one arm (NLA) made up of a Kerr nonlinear material as shown in Fig. 1. A nonlinear medium (NLM) is buffered in-between the V-junction of the NMZI and the output waveguide. A control waveguide (CWG) has also been considered as shown. When a beam is launched into P 1 , it splits into two equal parts. One part propagates through NLA and its counterpart through LA. The part propagating through NLA experiences cross phase modula- tion (XPM) if a control beam is present in CWG. The split parts of the beam recombine at the V-junction of the NMZI placed just before the NLM and enter into the NLM in the form of a single beam. The direction of propagation of this single beam in the NLM depends on the relative phase difference of the split parts at the V-junction. As the phase of the part propagating through NLA could be altered by injecting a control beam at P 2 , the beam in the NLM can be deflected in a desired manner by creating appropriate phase difference between the split parts at the V-junction. Moreover, if the combined power of the split parts is equal to the solitonic or nearly solitonic power of the considered NLM, they will form a solitonic/nearly solitonic beam in the NLM (see Fig. 2). It is obvious that the output at P 0 will be maximum if the input beam exactly falls on the straight waveguide (as in Fig. 2b) and it will be zero if the beam falls on either side of the straight waveguide (as in Fig. 2a and c). The Decay medium (DM) is the section of the CWG with a very high loss, where the control beam (signal beam) gets lost after accomplishing its job. The mentioned structure can be thoroughly analyzed using the beam propagation method (BPM), i.e., by solving the below mentioned nonlinear Schrodinger equation (NLSE) using the split step Fourier method [17,18] (for the sake of brevity and saving calculation time, we treat a one-dimensional problem): @E j @z ¼i 1 2kn 0 @ 2 E j @x 2 ik n j ðx, zÞn 0   E j ; j ¼ 1, 2 ð1Þ Here, E j ð¼ ffiffiffi I j p expðx 2 =2x 2 0 ÞÞ is the transverse field envelop of beams, I j is the axial intensity, x the transverse coordinate, x 0 the Contents lists available at SciVerse ScienceDirect journal homepage: www.elsevier.com/locate/optlastec Optics & Laser Technology 0030-3992/$ - see front matter & 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.optlastec.2011.08.020 n Corresponding author. E-mail address: sarangmedekar@rediffmail.com (S. Medhekar). Optics & Laser Technology 44 (2012) 492–496