1. Introduction
In recent years, nondiffracting beams have attracted much attention due to their peculiar properties[1,2]. Such beams are able to keep their transverse intensity distribution over a long distance in free space and quickly reestablish their structures when encountering barriers[3]. Since the pioneering work of Durnin et al., Bessel beams, which are the exact solution of the Helmholtz wave equation in circular cylindrical coordinates, were presented[4] and have shown outstanding performance in the fields of optical trapping[5,6], optical microscopy[7,8], material processing[9], and so on. Apart from Bessel beams, plane waves[10], Mathieu beams[11], and Weber beams[12] are the families of propagation invariant beams, which are, respectively, the nonparaxial solutions of the Helmholtz equation under Cartesian coordinates, elliptic cylindrical coordinates, and parabolic cylindrical coordinates. In particular, a family of nondiffracting beams forms a set of complete orthogonal solutions. In this case, any linear superposition of the same type of diffraction-free modes with the same radial wave vector keeps the nondiffracting properties[13]. Recently, Kovalev et al. presented a linear superposition of Bessel modes, whose complex amplitude can be expressed in terms of Lommel functions of two variables. Hence, such a beam is called Lommel beam[14]. Compared to Bessel beam whose transverse intensity profile is radial symmetry, the Lommel beam has a reflective symmetry with respect to both Cartesian coordinate axes[15]. In addition, a Lommel beam has an adjustable symmetrical transverse intensity pattern and continuously variable orbital angular momentum (OAM), which can benefit optical manipulation[16,17] and optical information encoding[18,19].