Reply to “Comments on “Novel Broadband Artificial Magnetic Conductor With Hexagonal Unit Cell””

The authors are thankful for the useful comments [1] on their letter [2]. The presented artificial magnetic conductor (AMC) is not composed of square but hexagonal unit cells. The referenced first paragraph of [2, Section II], “Planar AMC Design,” makes reference to a general behavior based on a very simple LC model for the unit-cell geometry. Generally, the unit cell can be considered a frequency selective surface (FSS) backed by a dielectric slab with metallic ground plane. To increase the bandwidth of an AMC, it is necessary to reduce the total capacitance C. When the distance between unit cells increases, the gap capacitance decreases. If this is the main capacitive element contributing to the total C, the bandwidth increases. This happens, for example, for the patch unit cell in [3, Table I], a work referenced by [1]. In [3], only one capacitive element is used to model the unit cell, and the objective is to maximize the bandwidth while preserving the resonance frequency. The point is that the total C of the unit-cell model has to be expressed as a series or parallel combination of capacitive elements (gap, parallel plates, fringing effects, etc.), which obviously depends on the geometry of such a unit cell. However, the geometry of the presented unit cell (formed by dipoles) is mainly inductive due to the narrow strips. The behavior of the presented unit cell is described in the third paragraph of [2, p. 616]. It is clearly mentioned that there are two main capacitive elements that contribute to the total C: coplanar C (or gap C) and parallel plates C. Moreover, it is stated that “in sum, the wider AMC operation bandwidth for smaller unit-cell size (W) of this novel design is due to both slightly increased L and more significantly C decreased values of the parallel LC equivalent circuit.” This does not contradict [1] since, according to Fig. 3 and Table I of [2], for smaller unit cell W, the distance between cells 2d (the gap) decreases, and so the gap capacitance.