Improved analytical fit of gold dispersion: Application to the modeling of extinction spectra with a finite-difference time-domain method Alexandre Vial,* Anne-Sophie Grimault, Demetrio Macías, Dominique Barchiesi, and Marc Lamy de la Chapelle Laboratoire de Nanotechnologie et d’Instrumentation Optique - CNRS FRE 2671, Université de Technologie de Troyes, 12 rue Marie Curie, BP-2060 F-10010 Troyes Cedex, France Received 13 July 2004; revised manuscript received 14 October 2004; published 23 February 2005 We propose an accurate description for the dispersion of gold in the range of 1.24–2.48 eV.We implement this improved model in an FDTD algorithm and evaluate its efficiency by comparison with an analytical method. Extinction spectra of gold nanoparticle arrays are then calculated. DOI: 10.1103/PhysRevB.71.085416 PACS numbers: 41.20.Jb, 42.25.Bs, 77.22.Ch, 78.20.Ci I. INTRODUCTION The employment of the finite-difference time-domain FDTDmethod in the study of different electromagnetic phenomena has raised constantly increasing interest over the past 15 years. Since then, an extensive number of references describing the principles of the method have been published; see, e.g., Refs. 1 and 2. Also, a wide variety of software based on this technique has been developed and is commer- cially, and noncommercially, available elsewhere. Due to the fact that accurate results for a full spectrum can be obtained in a single run of the program, the FDTD has proven to be well adapted for different kinds of spectro- scopic studies. 3 Nevertheless, a strong limitation is the re- quirement of an analytical model of dispersion. Typical laws used for FDTD simulations are the Debye, Lorentz, or Drude dispersion models. 3–6 Also, a modified Debye law can be used. 7 At least in principle, any dispersion law could be described in terms of a linear combination of Debye and Lorentz laws. 2 Surprisingly, this property has not been used in studies using the FDTD method. Rather, it has been suc- cessfully applied to the description of optical functions for 11 metals over a wide spectrum, 8 or to a calculation of the re- flectance of single wall nanotubes. 9 In this paper, we will employ a scheme similar to the one that appears in this last reference for the study of the optical response of gold nano- structures. The structure of this work is as follows. In Sec. II, we show the results obtained from the implementation of two classic dispersion models employing the FDTD method. In order to validate the numerical approaches, we apply them to the case of a simple structure and compare the results with those obtained using the analytical method described in Ref. 10. In Sec. III, we employ our FDTD-based implementation of the Drude-Lorentz model to the calculation of extinction spectra. In Sec. IV, we present our main conclusions and final remarks. II. MODELS OF DISPERSION A. The Drude model It is well known that in the near infrared, the relative permittivity of several metals can be described by means of the Drude model, 11 D = - D 2 + i D , 1 where D is the plasma frequency and D is the damping coefficient. Nevertheless, if frequencies within the range of the visible spectrum are required for a specific study, the model in Eq. 1may not be complete enough to provide accurate results. To illustrate this fact, we try to fit the rela- tive permittivity of gold JC , tabulated by Johnson and Christy 12 through the optimization of , D , and D for energies between 1.24 and 2.48 eV wavelengths between 500 and 1000 nm. In order to determine the best set of parameters, we define a fitness function sometimes called an objective functionas = j Re JC j - D j  2 + Im JC j - D j  2 , 2 where j are the discrete values of the frequency =2c / for which the permittivity is calculated. The real and imaginary parts of a complex value z are, respectively, Rezand Imz. The minimization of is performed em- ploying the simulated annealing procedure described in Ref. 13, and results are presented in the first row of Table I. The real and imaginary parts of the permittivity D , calculated with the Drude model, are, respectively, plotted with a dotted line in Figs. 1 and 2. It can be seen that neither Re JC nor Im JC are well described for energies above 2.2 and 1.9 eV, respectively.To emphasize this difference, we also plot in Figs. 1 and 2 the relative errors on Re D and Im D . The existence of a strong discrepancy for energies above 1.9 eV when using the single Drude model is evident. Due to the inability of the Drude model to describe the permittivity of metals over a wide range of frequencies, some of the authors working with the FDTD restrict their studies to a zone of the spectrum where the Drude model is valid Ref. 7 for a gold tip, Refs. 14 and 15 for silver and aluminum structures; it should be noted that for these two metals, the Drude model alone works well for the optical wavelengths. Others try to fit the permittivity in the range of interest by modifying the values of the parameters , D , and D of the model, as shown in Ref. 5 for the case of silver PHYSICAL REVIEW B 71, 085416 2005 1098-0121/2005/718/0854167/$23.00 ©2005 The American Physical Society 085416-1