Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Nonlocal composite metamaterial in calculation of near-field thermal rectification

Open Access Open Access

Abstract

We consider the near-field radiative heat transfer (NFRHT) between a semi-infinite polar dielectric plate and a composite metamaterial containing semiconductor nanoparticles with spatial dispersion or nonlocality. We show that the enhancement of near-field heat radiation is found owing to the strong coupling of surface phonon polaritons (SPhPs) excited between polar dielectric plate and air, and localized surface plasmon polaritons (LSPPs) in the composite metamaterials. The introduction of nonlocality is helpful to enhance the NFRHT and even the rectification efficiency significantly when the temperature difference is small. For a large volume fraction of semiconductor nanoparticles, the thermal rectification efficiency can be larger than 90% at a small temperature difference within 150 K. Our findings may pave a way for thermal devices based on composite metamaterial containing semiconductor nanoparticles.

© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement

1. Introduction

The study of both experimental and theoretical behavior of near-field radiative heat transfer (NFRHT) at a distance smaller than the thermal wavelength of radiation was investigated in the past decades [17]. The enhancement mechanism of NFRHT is due to the strong coupling between surface phonon polaritons (SPhPs) or surface plasmon polaritons (SPPs) [8,9], or hyperbolic surface plasmon polaritons supported by graphene meta-surface [10]. Owing to its numerous applications for heat flux modulation [1012], near-field thermophotovoltaics [13], energy conversion and management [14], NFRHT was widely investigated between two plates [8,15], sphere-plate [16], two-sphere [17], gratings [18], etc.

Near-field thermal rectification, also known as thermal diode, is a two-terminal device with the thermal conductance dependent on the direction of heat flux. It has been predicted in the systems including the asymmetric interface with different materials [19], dissimilar materials having temperature-dependent properties [8,9,15] or asymmetric geometric structures [20]. For practical applications, highly efficient thermal rectification and/or highly NFRHT is of much interest. For instance, Otey et al. [15] proposed the thermal diode made by SiC-3C and SiC-6H plates, relying on the coupling and decoupling of their SPhPs at different temperatures and frequencies. Joulain et al. investigated high heat flux between two bodies when the exchanging temperatures of the two bodies are from ambient to high temperatures or in a reverse manner, and large rectification was interpreted due to the weakening of the surface polariton with increasing temperature [8].

On the other hand, composite metamaterials provide us more adjustable parameters to enhance NFRHT and/or the thermal rectification. In general, these composite metamaterials are designed or fabricated using nanoparticles embedded in a host (or dielectric) medium or a planar substrate. Zhang et al. [21] studied the enhancement of NFRHT between composite nanostructures, and they demonstrated that the significant enhancement of NFRHT results from thermally excited SPhPs, and SPPs, and even hyperbolic phonon polaritons. In addition, Zhou et al. [22] proposed a thermal diode using a single material (i.e. 3C-SiC nanoporous and 3C-SiC plates) by adopting the effective medium theory (EMT) and obtained rectification coefficient above 0.6 with a volume fraction of 0.25. Ogundare et al. [23] proposed a thermal rectifier using composite metamaterials and a rectification coefficient above 0.8 was achieved.

With the development of science and technology, the tip-sample distances can be down to a few nanometers [24]. In order to fill the gap between experimental and theoretical results of near-field heat transfer, the nonlocal description of the permittivity was introduced [24]. The spatial dispersion or nonlocality means that the permittivity (or the permeability) of the material is dependent both on the incident frequency $\omega $ and the wave vector k. Later, some scientists included the nonlocal effect of the material in the study of NFRHT between homogeneous structures. Chapuis et al. [25] investigated the effects of spatial dispersion/nonlocality in NFRHT between two parallel metallic surfaces. They showed that the radiative heat flux between two metallic plates saturates when the gap size reaches a distance equivalent to skin depth. Singer et al. [26] studied NFRHT between two n-doped semiconductors using Lindhard-Mermin nonlocal permittivity model. They observed that the evanescent heat transfer coefficient saturates at a shorter distance, and the nonlocal effect is significant when the distance is shorter than the Thomas-Fermi length for high doped concentration. Venkataram et al. [27] investigated the thermal radiation between molecules and macroscopic bodies at the mesoscopic scales. They uncovered phonon polaritons can be strongly influenced by the nonlocal response to electromagnetic fluctuations. For composite metamaterials, the size of particle inclusions is also about a few nanometers, and it is natural to expect the existence of the nonlocal properties of nanoparticle inclusions. However, the NFRHT between composite metamaterials containing nonlocal nanoparticles has not been considered. Due to the interaction between the nonlocality and the inhomogeneity, NFRHT from such kind of system may be of much interest.

In this paper, we study the NFRHT between a semi-infinite SiO2 (bottom) plate and a semi-infinite plate with composite metamaterials containing semiconductor nanoparticles randomly embedded in the host medium. In this connection, we introduce the nonlocal permittivity $\varepsilon ({\omega ,k} )$ of nanoparticle inclusions, and adopt nonlocal full-wave electromagnetic scattering theory to derive the effective permittivity of the composite metamaterials [28,29]. With the aid of fluctuational electrodynamics theory, we take one step forward to investigate the performance of NFRHT. We shall observe the strong coupling between the localized surface plasmon polaritons (LSPPs) in the composite metamaterials and surface phonon polaritons (SPhPs) (excited between polar dielectric plate and air). In addition, we analyze the contributions of the nonlocality and the volume fraction of semiconductor nanoparticles to the near-field strong coupling effects. Our work provides us an original coupling mechanism to control heat flux and enhance the thermal rectification, which shows potential applications in NFRHT.

2. Theoretical model

Let us consider the near-field thermal radiation between a semi-infinite (top) plate with composite metamaterial and a semi-infinite SiO2 (bottom) plate of permittivity ${\varepsilon _2}\; $ separated by a vacuum gap d. We aim at the study of near-field thermal radiation for such a system. In the forward temperature biased scenario, the composite metamaterial is maintained at a high temperature ${T_\textrm{H}}$ and the SiO2 plate is at a low temperature ${T_\textrm{L}}$, and one yields the forward radiative heat flux ${Q_f}$. In the reverse case, the temperature of the two semi-infinite plates is swapped, and the reverse radiative heat flux ${Q_r}$ can be achieved. The composite metamaterial consists of semiconductor (InSb) nanoparticles with radius a and the volume fraction f randomly dispersed in the host medium of permittivity ${\varepsilon _s}$. For such a composite metamaterial, we take into account the effect of nonlocality or spatial dispersion due to the small size of the InSb nanosphere. In this connection, the permittivity of the nonlocal InSb nanoparticles consists of the transverse (${\varepsilon _T}$) and the longitudinal (${\varepsilon _L}$) parts, given by [2830],

$$ {{\varepsilon _T}(\omega )= {\varepsilon _\infty } - \frac{{\omega _p^2(T )}}{{\omega [{\omega + i\gamma (T )} ]}}}$$
$$ {{\varepsilon _L}({\omega ,k} )= {\varepsilon _\infty } - \frac{{\omega _p^2(T )}}{{\omega [{\omega + i\gamma (T )} ]- {\beta ^2}(T ){k^2}}}}$$
where ${\varepsilon _\infty }$, ${\omega _p}$, k, and $\gamma $ are the relative permittivity at high frequency, plasma frequency, wave vector, and damping constant. The parameters used are obtained from Ref. [31,32]. In addition, for semiconductors, $\beta (T )$ is given by,
$$ {{\beta ^2}(T )= \frac{{3{\hbar ^2}}}{{5{m^\ast }^2}}{{({3{\pi^2}n(T )} )}^{\frac{2}{3}}}}$$
where $\hbar $, $n(T )$, and ${m^\ast }$ are the reduced Planck constant, carrier concentration, and electron’s effective mass, respectively.

In what follows, we concentrate on the radiative heat flux and the rectification efficiency $\eta $ to characterize the thermal diode performance.

2.1 Near-field thermal rectification efficiency

For the structure shown in Fig. 1, since the radiative heat flux in the forward-biased scenario is ${Q_f}$ with the top plate at a high temperature ${T_\textrm{H}}$, and the reverse scenario is ${Q_r},$ the characteristic of thermal diode performance is measured by its rectification efficiency [8,23,32],

$$ {\eta = \; \frac{{|{{Q_f} - {Q_r}} |}}{{\max ({{Q_f},\; {Q_r}} )}}.}$$

According to the fluctuational electrodynamics theory for a two-body system, the net heat flux of the system ${Q_f}$ or ${Q_r}$ is defined as [32],

$$ {{Q_{f,r}} = \; \frac{1}{{4{\pi ^2}}}\mathop \int \limits_0^\infty \textrm{d}\omega \mathop \int \limits_0^\infty d{k_\rho }\mathop \sum \limits_{\alpha = s,p} {\tau _\alpha }({\omega ,{k_\rho },d} ){k_\rho }[{\mathrm{\Theta }({\omega ,\; {T_\textrm{H}}} )- \mathrm{\Theta }({\omega ,\; {T_\textrm{L}}} )} ],}$$
where $\mathrm{\Theta }({\omega ,\; T} )= \; \hbar \omega /[{\textrm{exp}({\hbar \omega /{k_B}T} )- 1} ]$ is the average energy of the Planck oscillator at the frequency $\omega $, ${k_\rho }$ is the parallel wave vector. Moreover, the transmission coefficient ${\tau _\alpha }$ ($\alpha \; $ denotes s or p polarization) for the propagating (${k_\rho } < {k_0}$) and evanescent waves (${k_\rho } > {k_0}$) is written as [32],
$$ {\; {\tau _\alpha }({\omega ,{k_\rho },d} )= \left\{ \begin{array}{ll} {\frac{{({1 - \textrm{|}{r_{1\alpha }}{|^2}} )({1 - \textrm{|}{r_{2\alpha }}{|^2}} )}}{{|1 - {r_{1\alpha }}{r_{2\alpha }}{e^{2i\kappa d}}{|^2}}},}&{{k_\rho } < {k_0}}\\ {\frac{{4\textrm{Im}({{r_{1\alpha }}} )\textrm{Im}({r_{2\alpha )}}{e^{ - 2\textrm{Im}(\kappa )d}}}}{{|1 - {r_{1\alpha }}{r_{2\alpha }}{e^{ - 2\textrm{Im}(\kappa )d}}{|^2}}},}&{{k_\rho } > {k_0}} \end{array} \right..}$$

In Eq. (6), ${r_{1\alpha }}\; \textrm{and}\; {r_{2\alpha }}$ represent Fresnel reflection coefficients, which have the form,

$$ {{r_{1s}} = \; \frac{{\kappa - {\kappa _1}}}{{\kappa + {\kappa _1}}};\; {r_{1p}} = \; \frac{{\kappa {\varepsilon _{\textrm{eff}}} - {\kappa _1}}}{{\kappa {\varepsilon _{\textrm{eff}}} + {\kappa _1}}}\; }$$
and
$$ {{r_{2s}} = \; \frac{{\kappa - {\kappa _2}}}{{\kappa + {\kappa _2}}};\; {r_{2p}} = \; \frac{{\kappa {\varepsilon _2} - {\kappa _2}}}{{\kappa {\varepsilon _2} + {\kappa _2}}}\; }$$
with ${k_0} = \omega /c$, $\kappa = \sqrt {{k_0}^2 - {k_\rho }^2} $ and ${\kappa _{1,2}} = \sqrt {{k_0}^2{\varepsilon _{\textrm{eff},2}} - {k_\rho }^2} $. For the derivation of net heat flux together with rectification efficiency, we have assumed the effective permittivity of the composite metamaterials to be ${\varepsilon _{\textrm{eff}}}$. In the next subsection, we shall aim at the study of the effective permittivity of the composite metamaterials by taking into account the nonlocality of semiconductor nanoparticles.

 figure: Fig. 1.

Fig. 1. Schematic of two semi-infinite plates at a vacuum distance d maintain at low and high temperatures ${T_\textrm{L}}$ and ${T_\textrm{H}}$, respectively.

Download Full Size | PDF

2.2 Effective permittivity of composite metamaterials

The composite metamaterial is composed of nonlocal semiconductor spherical nanoparticles with volume fraction f, randomly embedded in the host medium. To study the effective permittivity of the composite, we adopt nonlocal effective medium theory [28,29,33], which deals with the scattering problem of coated spheres with nonlocal semiconductor core of radius a and the host shell of radius b embedded in an effective medium with permittivity ${\varepsilon _{\textrm{eff}}}$. Note that when the nonlocality is taken into account, both longitudinal and transverse waves exist in the nonlocal semiconductor nanoparticles. In detail, the wave vector of the longitudinal electromagnetic wave ${k_L}$ propagating in the nonlocal semiconductor nanoparticles is determined by ${\varepsilon _L}({\omega ,{k_L}} )= 0$, while the wave vectors of the transverse mode in the core, the shell, and the host effective medium obey the dispersion relation $k_{s,\; \; T}^2 = \; {k_0}^2{\varepsilon _{s,\; T}}(\mathrm{\omega } )\mu (\mathrm{\omega } )$, and $k_{\textrm{eff}}^2 = \; {k_0}^2{\varepsilon _{\textrm{eff}}}\; {\mu _{\textrm{eff}}}$ with $\mu (\omega )= 1$ and ${\mu _{\textrm{eff}}} = 1$.

Here, the governing equations for the fields which will be expanded in spherical unit vectors $({e_r}$, ${e_\theta }$, and ${e_\phi })$, are

$${{E^I} = \mathop \sum \limits_{l = 1}^\infty Z\left\{ {\nabla \times \left[ {\vec{r} \cdot {j_l}\left( {{k_{\textrm{eff}}}r} \right) \cdot } \pi\right] - i \cdot \frac{1}{{{k_{\textrm{eff}}}}}\nabla \times \nabla \times \left[ {\vec{r} \cdot {j_l}\left( {{k_{\textrm{eff}}}r} \right) \cdot {\prod }} \right]} \right\},}$$
$${{E^R} = \mathop \sum \limits_{l = 1}^\infty Z\left\{ {\nabla \times \left[ {\vec{r} \cdot {a^R}{h_l}\left( {{k_{\textrm{eff}}}r} \right) \cdot }\pi \right] - i \cdot \frac{1}{{{k_{\textrm{eff}}}}}\nabla \times \nabla \times \left[ {\vec{r} \cdot {b^R}{h_l}\left( {{k_{\textrm{eff}}}r} \right) \cdot {\prod }} \right]} \right\},}$$
$${{E^s} = \mathop \sum \limits_{l = 1}^\infty Z\left\{ {\begin{array}{c} {\nabla \times \left[ {\vec{r} \cdot \left[ {a_s^{TM}{j_l}\left( {{k_s}r} \right) + b_s^{TM}{y_l}\left( {{k_s}r} \right)} \right] \cdot }\pi \right] - }\\ {i \cdot \frac{1}{{{k_s}}}\nabla \times \nabla \times \left[ {\vec{r} \cdot \left[ {a_s^{TE}{j_l}\left( {{k_s}r} \right) + b_s^{TE}{y_l}\left( {{k_s}r} \right)} \right] \cdot {\prod }} \right]} \end{array}} \right\},}$$
where $\textrm{Z} = \; {E_0}{e^{ - i\omega t}}{i^l}\frac{{2l + 1}}{{l({l + 1} )}}$, $\pi = P_l^{\left( 1 \right)}\left( {{\text{cos}}\theta } \right){\text{sin}}\phi$, ${\prod } = P_l^{(1 )}({\textrm{cos}\theta } )\textrm{cos}\phi $, ${a_s}$ and ${b_s}$ are the scattering coefficients of magnetic and electric shells, respectively. Equations (911), respectively represent the incident and scattering field in the effective host medium, and local electric field in the shell, respectively. The functions ${j_n}$ (or ${h_n}$) and ${y_n}$ are the spherical (or Hankel) Bessel functions of the first and second kinds.

On the other hand, the electric fields related with the longitudinal and transverse waves excited in the nanosphere are written as,

$$ {{E^L} = \mathop \sum \limits_{l = 1}^\infty Z \cdot \frac{1}{{{k_L}}}\nabla [{a_c^L{j_l}({{k_L}r} )\cdot {\prod }} ],}$$
$$E^T = \mathop \sum \limits_{l = 1}^\infty Z\left\{ {\nabla \times \left[ {\vec{r} \cdot a_{c}^{T} j_{l}\left( {k_{Tr}} \right)\cdot {\pi}} \right]-i\cdot \displaystyle{1 \over {k_0}}\nabla \times \nabla \times \left[ {\vec{r}\cdot b_{c}^{T} j_{l}\left( {k_{Tr}} \right)\cdot {\prod}} \right]} \right\}.$$

By applying the boundary conditions for the outer radius (i.e. $r = b$) and inner radius (i.e. $r = a$) of the core-shell nanosphere, one gets the scattering coefficient of the electric field (${b_n}$),

$${b_n} = \frac{{\left[ {\begin{array}{ccccc} {\frac{{ - {{[j{j_l}({k_{\textrm{eff}}}b)]}^{\prime}}}}{{{k_{\textrm{eff}}}b}}}&{\frac{{ - {{[j{j_l}({k_s}b)]}^{\prime}}}}{{{k_s}b}}}&{\frac{{ - {{[y{y_l}({k_\textrm{s}}b)]}^{\prime}}}}{{{k_\textrm{s}}b}}}&0&0\\ { - \frac{{{k_{\textrm{eff}}}}}{{{\mu_{\textrm{eff}}}}}{j_l}({k_{\textrm{eff}}}b)}&{ - \frac{{{k_s}}}{{{\mu_s}}}{j_l}({k_s}b)}&{ - \frac{{{k_s}}}{{{\mu_s}}}{y_l}({k_s}b)}&0&0\\ 0&{\frac{{{{[j{j_l}({k_s}a)]}^{\prime}}}}{{{k_s}b}}}&{\frac{{{{[y{y_l}({k_\textrm{s}}a)]}^{\prime}}}}{{{k_\textrm{s}}a}}}&{\frac{{ - {{[j{j}({k_T}b)]}^{\prime}}}}{{{k_T}b}}}&{\frac{{{j_l}({k_L}b)}}{{{k_L}b}}}\\ 0&{\frac{{{k_s}}}{{{\mu_s}}}{j_l}({k_s}a)}&{\frac{{{k_s}}}{{{\mu_s}}}{y_l}({k_s}a)}&{ - \frac{{{k_T}}}{{{\mu_T}}}{j_l}({k_T}a)}&0\\ 0&0&0&{l(l + 1)\frac{{{j_l}({k_T}a)}}{{{K_T}a}} \times ({\varepsilon_\infty } - {\varepsilon_T})}&{[{j_l}({k_L}a)]^{\prime} \times ( - {\varepsilon_\infty })} \end{array}} \right]}}{{\left[ {\begin{array}{ccccc} {\frac{{{{[h{h_l}({k_{\textrm{eff}}}b)]}^{\prime}}}}{{{k_{\textrm{eff}}}b}}}&{\frac{{ - {{[j{j_l}({k_s}b)]}^{\prime}}}}{{{k_s}b}}}&{\frac{{ - {{[y{y_l}({k_\textrm{s}}b)]}^{\prime}}}}{{{k_\textrm{s}}b}}}&0&0\\ {\frac{{{k_{\textrm{eff}}}}}{{{\mu_{\textrm{eff}}}}}{h_l}({k_{\textrm{eff}}}b)}&{ - \frac{{{k_s}}}{{{\mu_s}}}{j_l}({k_s}b)}&{ - \frac{{{k_s}}}{{{\mu_s}}}{y_l}({k_s}b)}&0&0\\ 0&{\frac{{{{[j{j_l}({k_s}a)]}^{\prime}}}}{{{k_s}b}}}&{\frac{{{{[y{y_l}({k_\textrm{s}}a)]}^{\prime}}}}{{{k_\textrm{s}}a}}}&{\frac{{ - {{[j{j_l}({k_T}a)]}^{\prime}}}}{{{k_T}a}}}&{\frac{{{j_l}({k_L}a)}}{{{k_L}a}}}\\ 0&{\frac{{{k_s}}}{{{\mu_s}}}{j_l}({k_s}a)}&{\frac{{{k_s}}}{{{\mu_s}}}{y_l}({k_s}a)}&{ - \frac{{{k_T}}}{{{\mu_T}}}{j_l}({k_T}a)}&0\\ 0&0&0&{l(l + 1)\frac{{{j_l}({k_T}a)}}{{{K_T}a}} \times ({\varepsilon_\infty } - {\varepsilon_T})}&{[{j_l}({k_L}a)]^{\prime} \times ( - {\varepsilon_\infty })} \end{array}} \right]}}.$$
Here $j{j_l}(\theta )= \theta \cdot {j_l}(\theta )$, $y{y_l}(\theta )= \theta \cdot {y_l}(\theta )$, $h{h_l}(\theta )= \theta \cdot {h_l}(\theta )$, and $b = \; a/\sqrt[3]{f}$. Equation (14) allows us to deduce the effective permittivity of the composite media. By considering the limit condition ${k_{\textrm{eff}}} \cdot a \ll 1$ with an intermediate-term $n = 1$ in Eq. (14), the effective permittivity for the nonlocal composite metamaterials can be determined when ${b_1} = 0$,
$$ {{\varepsilon _{\textrm{eff}}} = \; \frac{{2{\varepsilon _s}{j_1}({{k_s}b} )Y - \; 2{\varepsilon _s}{y_1}({{k_s}b} )J}}{{{{[{{k_s}b.{j_1}({{k_s}b} )} ]}^{\prime}}Y - {{[{{k_s}b.{y_1}({{k_s}b} )} ]}^{\prime}}J}},}$$
where
$$ {Y = P \cdot {{[{{k_s}a.{y_1}({{k_s}a} )} ]}^{\prime}}{\varepsilon _T} - {y_1}({{k_s}a} ){\varepsilon _s},}$$
$$ {J = P \cdot {{[{{k_s}a.{j_1}({{k_s}a} )} ]}^{\prime}}{\varepsilon _T} - {j_1}({{k_s}a} ){\varepsilon _s},}$$
$$ {P = \; \frac{{{\varepsilon _\infty }{j_1}({{k_T}a} )j_1^{\prime}({{k_L}a} )}}{{{\varepsilon _\infty }{{[{{k_T}a.{j_1}({{k_T}a} )} ]}^{\prime}}j_1^{\prime}({{k_L}a} )- \frac{{2({{\varepsilon_\infty } - {\varepsilon_T}} ){j_1}({{k_T}a} ){j_1}({{k_L}a} )}}{{{k_L}a}}}}.}$$

For the local case, Eq. (15) is naturally reduced to the classical Maxwell-Garnett theory with the form,

$$ {{\varepsilon _{\textrm{eff}}} = \frac{{{\varepsilon _s}{\varepsilon _T}({2f + 1} )+ \varepsilon _s^2({2 - 2f} )}}{{{\varepsilon _T}({1 - f} )+ {\varepsilon _s}({2 + f} )}}.}$$

Here, we would like to mention that Eq. (15) and Eq. (16) are valid for the long-wavelength limit ${k_{\textrm{eff}}} \cdot a \ll 1$ and ${k_0} \cdot a \ll 1$. In general, the particle size a needs to be much smaller than the operating wavelength. On the other hand, $a < d \ll \lambda $ must be satisfied when investigating the near-field thermal radiation between composite metamaterials [21,34,35]. Indeed, Eq. (16) gives the size-independent effective permittivity. Beyond the long-wavelength limit, the effective permittivity with Eq. (16) shall be dependent on the radius of nanoparticles explicitly even though the nonlocality is not taken into account.

3. Results and discussion

We will start with the analysis of the effective permittivity ${\varepsilon _{\textrm{eff}}}$ against the angular frequency $\omega $ by considering nonlocal effects, as shown in Fig. 2. For the nonlocal case, it is observed that as the radius of semiconductor nanoparticles is increased, the real part of the effective permittivity Re(${\varepsilon _{\textrm{eff}}}$) always exhibits a ripple-like line shape, while the resonant peak is red-shifted accompanied with the enhanced magnitudes for the imaginary part $\textrm{Im}({\varepsilon _{\textrm{eff}}})$. It is evident that for further increasing the size, the nonlocal effect can be neglected, and hence the results should tend to be the local case, as expected (see Fig. 2(a) and (b)). For the local case, the effective permittivity is independent of the radius indeed under the long-wavelength approximation limit. As far as the temperature effect is concerned, the decrease in temperature leads to both the decreased magnitude and the red-shift of the resonant peak (see Figs. 2(c) and (d)).

 figure: Fig. 2.

Fig. 2. The effective permittivity ${\varepsilon _{\textrm{eff}}}$ against angular frequency $\omega $ for real part (a) and imaginary part (b) at temperature $T = 450\textrm{K}$, $f = 0.1$ for local and nonlocal cases with different a. The effective permittivity ${\varepsilon _{\textrm{eff}}}$ against angular frequency $\omega $ for real part (c) and imaginary part (d) at different temperatures with a = 25 nm. The inset figure in (a) is the permittivity of SiO2 with several phonon modes [8].

Download Full Size | PDF

Next, we show the net spectral heat flux, denoted by ${q_\omega }$, and contour plots $q({\omega ,{k_p}} )$ in Fig. 3. Figures 3(a) and (b) depict ${q_\omega }$ for the forward and reverse biased scenarios, respectively. To elucidate on the physical mechanism, the parameters $d = 30\textrm{nm}$, $f = 0.1$, $a = 25\textrm{nm}$, and ${T_\textrm{L}} = 300\textrm{K}$ are maintained unless otherwise stated. For the forward scenario as shown in Fig. 3(a), we use different vertical lines to represent the frequency positions of the resonant modes at different temperatures. For local composite plate (blue dotted line) at ${T_\textrm{H}} = 600\textrm{K}$, its LSPP takes place at $\textrm{Re}({{\varepsilon_{T}}} )={-} 2$, marked with a black dashed line. For nonlocal composite metamaterials (lines red-triangle and -square symbols), the LSPPs at ${T_\textrm{H}} = 450\textrm{K}$ and $600\textrm{K}$ are determined by $\textrm{Re}\{{P{{[{{k_s}a \cdot {y_1}({{k_s}a} )} ]}^\mathrm{^{\prime}}}{\varepsilon_{T}}} \}= \textrm{Re}\{{{y_1}({{k_s}a} ){\varepsilon_{h}}} \}$ (i.e. by equating Eq. (15b) to be zero), are marked with black-triangle and -square symbols, respectively. The peak positions of ${q_\omega }$ coincide with these black lines, suggesting that they are caused by LSPPs of the semiconductor nanoparticles. Conversely, in the reverse scenario, the local and nonlocal composite metamaterials are at the receiving temperature ${T_\textrm{L}}$. As shown in Fig. 3(b), since ${\varepsilon _2}$ does not change with ${T_\textrm{H}}$, the reverse heat fluxes ${q_\omega }$ are almost the same for different ${T_\textrm{H}}$ and their peaks are all at $\omega = 0.5 \times {10^{14}}\; \textrm{rad}\,{\textrm{s}^{ - 1}}$, which is just the LSPP frequency position for the nonlocal semiconductor nanoparticles at ${T_\textrm{L}}$. Note that the reverse heat fluxes are much smaller than those of the forward bias, resulting in thermal rectification.

 figure: Fig. 3.

Fig. 3. The net spectral heat fluxes for local, and nonlocal at different ${T_\textrm{H}}$ for forward-biased (a) and reverse-biased (b) scenarios. The contour plots $q({{k_\rho },\omega } )$ for forward-biased scenarios (c) ${T_\textrm{H}} = 450\textrm{K}$ and (e) ${T_\textrm{H}} = 600\textrm{K}$ and their reverse-biased counterpart (d) and (f) for nonlocal conditions, respectively.

Download Full Size | PDF

To better understand the physical mechanism of the thermal rectification effect in our system, we plot the $q({{k_\rho },\omega } )$ with angular frequency $\omega $ and normalized wave vector ${k_\rho }$, as shown in Figs. 3(c)-(f). In these figures, the resonant frequency of the LSPPs supported by nonlocal semiconductor nanoparticles is marked with solid black lines. Meanwhile, it can be seen from the inset figure of Fig. 2(a) that two SPhPs modes, which are ascertained by $\textrm{Re}({{\varepsilon_2}} )={-} 1$, are supported in SiO2 with angular frequencies ${\omega _1} = 0.9 \times {10^{14}}\; \textrm{rad}\,{\textrm{s}^{ - 1}}$ and ${\omega _2} = 2.2 \times {10^{14}}\; \textrm{rad}\,{\textrm{s}^{ - 1}}$ and their frequencies are marked by white solid lines in Figs. 3(c)-(f). In Fig. 3(c) when the composite metamaterials are at ${T_\textrm{H}} = 450\textrm{K}$, it is found that the solid white line and solid black line are very close to each other, and the energy flow is concentrated on near them, indicating that the forward heat flux is caused by the coupling between LSPP and SPhP at ${\omega _1}$. In Fig. 3(e), the LSPP mode supported by semiconductor nanoparticles is located at $\omega = 1.5 \times {10^{14}}\; \textrm{rad}\,{\textrm{s}^{ - 1}}$ at ${T_\textrm{H}} = 600\textrm{K}$ while the SiO2 plate does not support SPhPs near this frequency. However, there is a weak phonon mode of SiO2 at this frequency, which leads to the occurrence of the forward heat flux peak in this case. Following the above principle, due to the existence of coupling modes, energy can propagate through the evanescent wave. In Figs. 3(d) and (f), the heat flux almost exists in the range ${k_\rho }c/\omega < 1$ (propagating wave). The nonlocal composite plates at 300 K support LSPPs at $\omega = 0.5 \times {10^{14}}\; \textrm{rad}\,{\textrm{s}^{ - 1}}$ in the reversed scenario, which are quite far from their corresponding coupling positions ${\omega _1}$ and ${\omega _2}$. As a consequence, one yields the decoupling between these modes, resulting in less energy transmission by propagation wave as shown in Figs. 3(b), (d) and (f).

Then, we concentrate on the analysis of the thermal rectification efficiency in Fig. 4. Figure 4(a) shows the variation between rectification efficiency as the function of the high temperature ${T_\textrm{H}}$ at volume fraction $f = 0.01$. We observe that the rectification efficiency for the nonlocal case is higher than the one for the local case when ${T_\textrm{H}}$ range from 300K to 500K. The high rectification efficiency at these temperatures is quite interesting due to the fact that small temperature difference (since ${T_\textrm{L}} = 300\textrm{K}$) is enough to get high rectification effect. This behavior can be interpreted as follows. As seen in the inset figure in Fig. 4(a), LSPPs frequencies excited from the nonlocal case are closer to the resonant coupling frequency ${\omega _1}$ than the local case in the low temperature region, and hence the coupling between the LSPPs and SPhPs is stronger to enhance the rectification efficiency. In Fig. 4(b), we plot the rectification efficiency as a function of separation distance d. The rectification falls off due to the exponential decay of the surface polaritons. Even for this, the rectification efficiency for the nonlocal case is quite larger than the one for the local case for d less than 100nm.

 figure: Fig. 4.

Fig. 4. (a) Rectification efficiency for nonlocal and local against high temperature ${T_\textrm{H}}$. (b) Rectification efficiency for nonlocal and local at ${T_\textrm{H}} = 450\textrm{K}$ against the vacuum distance d.

Download Full Size | PDF

For the present system, the rectification efficiency can be further enhanced with increasing the volume fraction of nonlocal semiconductor nanoparticles. This is shown in Fig. 5. Figure 5 depicts the influence of volume fraction f on the rectification efficiency. The increase in the volume fraction f leads to the increase in the rectification efficiency. To one’s interest, the maximal rectification efficiency $\eta = 0.92$ occurs when ${T_\textrm{H}} = 420K$ with the volume fraction $f = 0.1$. A higher rectification efficiency can also be seen at higher temperature but a small temperature difference may be quite important for practical applications. Figure 5(b) shows the variation of $\eta $ with the separation d. For large volume fraction, high efficiency can be well kept as d is reduced. For instance, for $f = 0.1$, 50% efficiency can still be found even $d = 100\textrm{nm}$. This is quite similar as the one reported in Ref. [26]. Although at larger separations, the rectification efficiency here is lower than those between VO2 phase transition materials [36], the use of composite metamaterials containing InSb nanoparticles allows us more degrees of freedom to tune the plasmonic properties of semiconductor dynamically, the volume fraction, and so on.

 figure: Fig. 5.

Fig. 5. (a) The rectification efficiency with different volume fraction f for the nonlocal case against high temperature ${T_\textrm{H}}$. (b) The rectification efficiency with different volume fraction f at ${T_\textrm{H}} = 450\textrm{K}$ against the vacuum distance d.

Download Full Size | PDF

 figure: Fig. 6.

Fig. 6. The variation of rectification efficiency against volume fraction f for both local and nonlocal conditions at ${T_\textrm{H}} = 450\; \textrm{K}$.

Download Full Size | PDF

In the end, we plot the variation of the rectification efficiency against the volume fraction in Fig. 6. We show that the rectification efficiencies for both local and nonlocal cases increase monotonically with increasing the volume fraction, and the rectification efficiency for the nonlocal case (denoted with red dashed line) is always larger than the one for the local case. Although the localized surface plasmon resonant frequency is weakly dependent on the volume fractions, large absorption can be found for the composite metamaterials with large volume fractions, which may lead to large rectification efficiency for large volume fractions.

4. Conclusion

In conclusion, we demonstrate the modulation of thermal rectification between nonlocal composite metamaterial containing semiconductor nanoparticles and polar dielectric plates. The enhancement of near-field thermal transfer results from the coupling of localized surface plasmon polariton (LSPP) in composite metamaterials and surface phonon polariton (SPhP) from the surface of SiO2 plates. The introduction of the nonlocality is quite helpful to yield high thermal rectification efficiency at low high-temperature region. Moreover, the thermal rectification efficiency can be further enhanced with increasing the volume fraction of nonlocal semiconductor nanoparticles. We believe that beyond their potential applications in thermal management, our study may provide alternative insights into the design of thermal devices with composite metamaterials.

Funding

National Natural Science Foundation of China (92050104); Suzhou Prospective Application Research Project (SYG202039).

Disclosures

The authors declare no conflicts of interest.

Data availability

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

References

1. H. Wang, S. Hu, K. Takahashi, X. Zhang, H. Takamatsu, and J. Chen, “Experimental study of thermal rectification in suspended monolayer graphene,” Nat. Commun. 8(1), 15843 (2017). [CrossRef]  

2. E. G. Cravalho, C. L. Tien, and R. P. Caren, “Effect of small spacings on radiative transfer between 2 dielectrics,” J. Heat Transfer. 89(4), 351–358 (1967). [CrossRef]  

3. D. Polder and M. Van Hove, “Theory of radiative heat transfer between closely spaced bodies,” Phys. Rev. B 4(10), 3303–3314 (1971). [CrossRef]  

4. A V. Shchegrov, K. Joulain, R. Carminati, and J. J. Greffet, “Near-field spectral effects due to electromagnetic surface excitations,” Phys. Rev. Lett. 85(7), 1548–1551 (2000). [CrossRef]  

5. A. I. Volokitin and B. N. J. Persson, “Resonant photon tunneling enhancement of the radiative heat transfer,” Phys. Rev. B 69(4), 045417 (2004). [CrossRef]  

6. K. Joulain, J. P. Mulet, F. Marquier, R. Carminati, and J. J. Greffet, “Surface electromagnetic waves thermally excited: Radiative heat transfer, coherence properties and Casimir forces revisited in the near field,” Surf. Sci. Rep. 57(3-4), 59–112 (2005). [CrossRef]  

7. C. J. Fu and Z. M. Zhang, “Nanoscale radiation heat transfer for silicon at different doping levels,” Int. J. Heat Mass Transfer 49(9-10), 1703–1718 (2006). [CrossRef]  

8. K. Joulain, Y. Ezzahri, J. Drevillon, B. Rousseau, and D. D. S. Meneses, “Radiative thermal rectification between SiC and SiO2,” Opt. Express 23(24), A1388–A1397 (2015). [CrossRef]  

9. L. Tang, J. DeSutter, and M. Francoeur, “Near-field radiative heat transfer between dissimilar materials mediated by coupled surface phonon- and plasmon-polaritons,” ACS Photonics 7(5), 1304–1311 (2020). [CrossRef]  

10. M. He, H. Qi, Y. Ren, Y. Zhao, and M. Antezza, “Active control of near-field radiative heat transfer by a graphene-gratings coating-twisting method,” Opt. Lett. 45(10), 2914 (2020). [CrossRef]  

11. M. J. He, H. Qi, Y. T. Ren, Y. J. Zhao, and M. Antezza, “Magnetoplasmonic manipulation of nanoscale thermal radiation using twisted graphene gratings,” Int. J. Heat Mass Transfer 150, 119305 (2020). [CrossRef]  

12. J. E. Perez-Rodriguez, G. Pirruccio, and R. Esquivel-Sirvent, “Spectral gaps in the near-field heat flux,” Phys. Rev. Mater. 3(1), 015201 (2019). [CrossRef]  

13. T. Inoue, T. Koyama, D. D. Kang, K. Ikeda, T. Asano, and S. Noda, “One-Chip Near-Field Thermophotovoltaic Device Integrating a Thin-Film Thermal Emitter and Photovoltaic Cell,” Nano Lett. 19(6), 3948–3952 (2019). [CrossRef]  

14. A. Fiorino, L. Zhu, D. Thompson, R. Mittapally, P. Reddy, and E. Meyhofer, “Nanogap Near-Field Thermophotovoltaics,” Nat. Nanotechnol. 13(9), 806–811 (2018). [CrossRef]  

15. C. R. Otey, W. T. Lau, and S. Fan, “Thermal Rectification through Vacuum,” Phys. Rev. Lett. 104(15), 154301 (2010). [CrossRef]  

16. S. Shen, A. Mavrokefalos, P. Sambegoro, and G. Chen, “Nanoscale thermal radiation between two gold surfaces,” Appl. Phys. Lett. 100(23), 233114 (2012). [CrossRef]  

17. R. M. Abraham Ekeroth, P. Ben-Abdallah, J. C. Cuevas, and A. García-Martín, “Anisotropic Thermal Magnetoresistance for an Active Control of Radiative Heat Transfer,” ACS Photonics 5(3), 705–710 (2018). [CrossRef]  

18. X. L. Liu, B. Zhao, and Z. M. Zhang, “Enhanced near-field thermal radiation and reduced Casimir stiction between doped-Si gratings,” Phys. Rev. A 91(6), 062510 (2015). [CrossRef]  

19. R. Messina, P. Ben-Abdallah, B. Guizal, and M. Antezza, “Graphene-based amplification and tuning of near-field radiative heat transfer between dissimilar polar materials,” Phys. Rev. B 96(4), 045402 (2017). [CrossRef]  

20. J. N. Hu, X. L. Ruan, and Y. P. Chen, “Thermal Conductivity and Thermal Rectification in Graphene Nanoribbons: A Molecular Dynamics Study,” Nano Lett. 9(7), 2730–2735 (2009). [CrossRef]  

21. W. B. Zhang, C. Y. Zhao, and B. X. Wang, “Enhancing near-field heat transfer between composite structures through strongly coupled surface modes,” Phys. Rev. B 100(7), 075425 (2019). [CrossRef]  

22. C. L. Zhou, Y. Zhang, H. L. Yi, and L. Qu, “Radiation-based Near-field Thermal Rectification via Asymmetric Nanostructures of the Single Material,” in PhotonIcs & Electromagnetics Research Symposium, (PIERS-Spring, Rome, Italy, 2019), pp. 2652–2658.

23. R. T. Ogundare, W. Ge, and L. Gao, “Photonic thermal rectification with composite metamaterials,” Chin. Phys. Lett. 38(1), 016801 (2021). [CrossRef]  

24. A. Kittel, W. Muller-Hirsch, J. Parisi, S. A. Biehs, D. Reddig, and M. Holthaus, “Near-field heat transfer in a scanning thermal microscope,” Phys. Rev. Lett. 95(22), 224301 (2005). [CrossRef]  

25. P. O. Chapuis, S. Volz, C. Henkel, K. Joulain, and J. J. Greffet, “Effects of spatial dispersion in near-field radiative heat transfer between two parallel metallic surfaces,” Phys. Rev. B 77(3), 035431 (2008). [CrossRef]  

26. F. Singer, Y. Ezzahri, and K. Joulain, “Nonlocal study of the near field radiative heat transfer between two n-doped semiconductors,” Int. J. Heat Mass Transfer 90, 34–39 (2015). [CrossRef]  

27. P. S. Venkataram, J Hermann, A. Tkatchenko, and A. W. Rodriguez, “Phonon-polariton mediated thermal radiation and heat transfer among molecules and macroscopic bodies: Nonlocal electromagnetic response at mesoscopic scales,” Phys. Rev. Lett. 121(4), 045901 (2018). [CrossRef]  

28. J. Sun, Y. Huang, and L. Gao, “Nonlocal composite media in calculations of the Casimir force,” Phys. Rev. A 89(1), 012508 (2014). [CrossRef]  

29. X. Bian, D. L. Gao, and L. Gao, “Tailoring optical pulling force on gain coated nanoparticles with nonlocal effective medium theory,” Opt. Express 25(20), 24566–24578 (2017). [CrossRef]  

30. J. R. Maack, N. A. Mortensen, and M. Wubs, “Two-fluid hydrodynamic model for semiconductors,” Phys. Rev. B 97(11), 115415 (2018). [CrossRef]  

31. D. Feng, S. K. Yee, and Z. M. Zhang, “Near-field photonic thermal diode based on hBN and InSb films,” Appl. Phys. Lett. 119(18), 181111 (2021). [CrossRef]  

32. G. Xu, J. Sun, H. Mao, and T. Pan, “Highly efficient near-field thermal rectification between InSb and graphene-coated SiO2,” J. Quant. Spectrosc. Radiat. Transfer 220, 140–147 (2018). [CrossRef]  

33. Y. Huang and L. Gao, “Equivalent Permittivity and Permeability and Multiple Fano Resonances for Nonlocal Metallic Nanowires,” J. Phys. Chem. C 117(37), 19203–19211 (2013). [CrossRef]  

34. R. Esquivel-Sirvent, “Ultrathin metallic coatings to control near field radiative heat transfer,” AIP Adv. 6(9), 095214 (2016). [CrossRef]  

35. J. E. Pérez-Rodríguez, G. Pirruccio, and R. Esquivel-Sirvent, “Near-Field Radiative Heat Transfer around the Percolation Threshold in Al Oxide Layers,” J. Phys. Chem. C 123(16), 10598–10603 (2019). [CrossRef]  

36. Z. Zheng, X. Liu, A. Wang, and Y. Xuan, “Graphene-assisted near-field radiative thermal rectifier based on phase transition of vanadium dioxide (VO2),” Int. J. Heat Mass Transfer 109, 63–72 (2017). [CrossRef]  

Data availability

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (6)

Fig. 1.
Fig. 1. Schematic of two semi-infinite plates at a vacuum distance d maintain at low and high temperatures ${T_\textrm{L}}$ and ${T_\textrm{H}}$, respectively.
Fig. 2.
Fig. 2. The effective permittivity ${\varepsilon _{\textrm{eff}}}$ against angular frequency $\omega $ for real part (a) and imaginary part (b) at temperature $T = 450\textrm{K}$, $f = 0.1$ for local and nonlocal cases with different a. The effective permittivity ${\varepsilon _{\textrm{eff}}}$ against angular frequency $\omega $ for real part (c) and imaginary part (d) at different temperatures with a = 25 nm. The inset figure in (a) is the permittivity of SiO2 with several phonon modes [8].
Fig. 3.
Fig. 3. The net spectral heat fluxes for local, and nonlocal at different ${T_\textrm{H}}$ for forward-biased (a) and reverse-biased (b) scenarios. The contour plots $q({{k_\rho },\omega } )$ for forward-biased scenarios (c) ${T_\textrm{H}} = 450\textrm{K}$ and (e) ${T_\textrm{H}} = 600\textrm{K}$ and their reverse-biased counterpart (d) and (f) for nonlocal conditions, respectively.
Fig. 4.
Fig. 4. (a) Rectification efficiency for nonlocal and local against high temperature ${T_\textrm{H}}$. (b) Rectification efficiency for nonlocal and local at ${T_\textrm{H}} = 450\textrm{K}$ against the vacuum distance d.
Fig. 5.
Fig. 5. (a) The rectification efficiency with different volume fraction f for the nonlocal case against high temperature ${T_\textrm{H}}$. (b) The rectification efficiency with different volume fraction f at ${T_\textrm{H}} = 450\textrm{K}$ against the vacuum distance d.
Fig. 6.
Fig. 6. The variation of rectification efficiency against volume fraction f for both local and nonlocal conditions at ${T_\textrm{H}} = 450\; \textrm{K}$.

Equations (19)

Equations on this page are rendered with MathJax. Learn more.

ε T ( ω ) = ε ω p 2 ( T ) ω [ ω + i γ ( T ) ]
ε L ( ω , k ) = ε ω p 2 ( T ) ω [ ω + i γ ( T ) ] β 2 ( T ) k 2
β 2 ( T ) = 3 2 5 m 2 ( 3 π 2 n ( T ) ) 2 3
η = | Q f Q r | max ( Q f , Q r ) .
Q f , r = 1 4 π 2 0 d ω 0 d k ρ α = s , p τ α ( ω , k ρ , d ) k ρ [ Θ ( ω , T H ) Θ ( ω , T L ) ] ,
τ α ( ω , k ρ , d ) = { ( 1 | r 1 α | 2 ) ( 1 | r 2 α | 2 ) | 1 r 1 α r 2 α e 2 i κ d | 2 , k ρ < k 0 4 Im ( r 1 α ) Im ( r 2 α ) e 2 Im ( κ ) d | 1 r 1 α r 2 α e 2 Im ( κ ) d | 2 , k ρ > k 0 .
r 1 s = κ κ 1 κ + κ 1 ; r 1 p = κ ε eff κ 1 κ ε eff + κ 1
r 2 s = κ κ 2 κ + κ 2 ; r 2 p = κ ε 2 κ 2 κ ε 2 + κ 2
E I = l = 1 Z { × [ r j l ( k eff r ) π ] i 1 k eff × × [ r j l ( k eff r ) ] } ,
E R = l = 1 Z { × [ r a R h l ( k eff r ) π ] i 1 k eff × × [ r b R h l ( k eff r ) ] } ,
E s = l = 1 Z { × [ r [ a s T M j l ( k s r ) + b s T M y l ( k s r ) ] π ] i 1 k s × × [ r [ a s T E j l ( k s r ) + b s T E y l ( k s r ) ] ] } ,
E L = l = 1 Z 1 k L [ a c L j l ( k L r ) ] ,
E T = l = 1 Z { × [ r a c T j l ( k T r ) π ] i 1 k 0 × × [ r b c T j l ( k T r ) ] } .
b n = [ [ j j l ( k eff b ) ] k eff b [ j j l ( k s b ) ] k s b [ y y l ( k s b ) ] k s b 0 0 k eff μ eff j l ( k eff b ) k s μ s j l ( k s b ) k s μ s y l ( k s b ) 0 0 0 [ j j l ( k s a ) ] k s b [ y y l ( k s a ) ] k s a [ j j ( k T b ) ] k T b j l ( k L b ) k L b 0 k s μ s j l ( k s a ) k s μ s y l ( k s a ) k T μ T j l ( k T a ) 0 0 0 0 l ( l + 1 ) j l ( k T a ) K T a × ( ε ε T ) [ j l ( k L a ) ] × ( ε ) ] [ [ h h l ( k eff b ) ] k eff b [ j j l ( k s b ) ] k s b [ y y l ( k s b ) ] k s b 0 0 k eff μ eff h l ( k eff b ) k s μ s j l ( k s b ) k s μ s y l ( k s b ) 0 0 0 [ j j l ( k s a ) ] k s b [ y y l ( k s a ) ] k s a [ j j l ( k T a ) ] k T a j l ( k L a ) k L a 0 k s μ s j l ( k s a ) k s μ s y l ( k s a ) k T μ T j l ( k T a ) 0 0 0 0 l ( l + 1 ) j l ( k T a ) K T a × ( ε ε T ) [ j l ( k L a ) ] × ( ε ) ] .
ε eff = 2 ε s j 1 ( k s b ) Y 2 ε s y 1 ( k s b ) J [ k s b . j 1 ( k s b ) ] Y [ k s b . y 1 ( k s b ) ] J ,
Y = P [ k s a . y 1 ( k s a ) ] ε T y 1 ( k s a ) ε s ,
J = P [ k s a . j 1 ( k s a ) ] ε T j 1 ( k s a ) ε s ,
P = ε j 1 ( k T a ) j 1 ( k L a ) ε [ k T a . j 1 ( k T a ) ] j 1 ( k L a ) 2 ( ε ε T ) j 1 ( k T a ) j 1 ( k L a ) k L a .
ε eff = ε s ε T ( 2 f + 1 ) + ε s 2 ( 2 2 f ) ε T ( 1 f ) + ε s ( 2 + f ) .
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.