-
The action can be expressed as [70, 86])
$ S = \int d^4x\sqrt{-g}\Big[R-2\nabla_\alpha\phi \nabla^\alpha\phi-K(\phi) F_{\alpha\beta}F^{\alpha\beta} -V(\phi) \Big]\, , $
where new quantities in the action are ϕ and
$ K(\phi) $ , which represent the massless scalar field and scalar field function, respectively. It should be noted that$ K(\phi) $ is also referred to as the coupling function describing the relation between dilaton fields and the electromagnetic$ F_{\alpha\beta} $ . The last term$ V(\phi) $ in the action denotes the potential pertaining to the cosmological constant Λ, related to the de-Sitter black hole solution with the dilaton field within the EMS theory; i.e.,$ V(\phi) = \dfrac{\Lambda}{3} (e^{2\phi}+4+ e^{-2\phi}) $ [85]. Then, the metric describing a spherically symmetric charged black hole within the EMS theory expressed in Schwarzschild coordinates (i.e.,$ V(\phi) = 0 $ ) is given by [86]$ {\rm d}s^2 = -U(r){\rm d}t^2+\frac{{\rm d}r^2}{U(r)}+f(r)\left({\rm d}\theta^2+\sin^2\theta {\rm d}\varphi^2\right)\, , $
(1) with radial functions
$ U(r) $ and$ f(r) $ for$K(\phi) = \dfrac{2{\rm e}^{2\phi}}{\beta {\rm e}^{4\phi}+\beta-2\gamma}$ expressed as$ \begin{aligned}[b] f(r) =\;& r^2\left(1+\frac{\gamma Q^2}{Mr}\right)\, , \\ U(r) =\;& 1-\frac{2M}{r}+\frac{\beta Q^2}{f(r)}\, . \end{aligned} $
(2) Note that M and Q are respectively referred to as the mass and electric charge of the black hole, while β and γ are dimensionless constants within the EMS theory. It should also be noted that
$ f(r) $ and$ U(r) $ can recover the Schwarzschild and Reissner-Nordström black hole solutions in the case of various combinations of parameter β and γ [70, 71]. Moreover,$ f(r) $ and$ U(r) $ recover the Schwarzschild solution in case the β and γ parameters are switched off. Similarly, it reduces to the Reissner-Nordström black hole in the case of$ \gamma = 0 $ and$\, \beta = 1 $ . However, the above solution turns into the dilation solution when$\, \beta = 0 $ and$ \gamma = -1 $ (see for example [70, 71]). The black hole horizon$ r_{h} $ can be easily determined by setting$ U(r) = 0 $ , which is given by$ \frac{r_h}{M} = 1-\frac{\gamma Q^2}{2 M^2}+\sqrt{1+\frac{Q^2 (\gamma -\beta )}{M^2}+\frac{\gamma ^2 Q^4}{4 M^4}}\, . $
(3) Note from the above equation that the black hole horizon no longer exists in the case of a larger parameter β, thus resulting in the space–time being a naked singularity. We demonstrate it in Fig. 1, which plots the parameter space between the charge parameter Q and dimensionless parameter β of the black hole for various combinations of the parameter γ. As can be observed from Fig. 1, a black hole sustains its existence in the region which is separated from naked singularity regions by the curves. We can also approach this issue from a different perspective, i.e., black hole extremes can be determined by imposing the condition
$ U(r) = U'(r) = 0 $ , thereby obtaining the limiting values of black hole parameters asFigure 1. (color online) Parameter space plot between the charge parameter
$Q/M$ and dimensionless parameter β of the black hole within the EMS theory for various combinations of parameter γ.$ \frac{(r_h)_{\rm min}}{M} = 2-\frac{\beta }{\gamma }+ \frac{\sqrt{\beta ^2-2 \beta \gamma }}{\gamma }\ , $
(4) $ \frac{Q_{\rm extr}^2}{M^2} = \frac{2 \left(\beta- \sqrt{\beta^2 -2 \beta \gamma }-\gamma \right)}{\gamma ^2}\, . $
(5) Figure 2 represents possible extreme values of Q as a function of γ while keeping β fixed. As can be observed from Fig. 2, extreme values of black hole charge can reach large values as a consequence of an increase in the value of γ, while the opposite occurs for β. The extreme conditions given by Eqs. (4) and (5) implicitly imply that
$ \,\beta_{\rm{max}} = 2 \gamma $ and$ {Q_{\rm{extr}}}/{M} = \sqrt{{2}/{\gamma}} $ [89], addressing the limiting values of black hole parameters.Figure 2. (color online) Extreme values of
$Q_{\rm extr}/M$ plotted as a function of the parameter γ for various combinations of β.To study shadow formation, it is necessary to consider the motion of test particles around a static and spherically symmetric black hole solution in the EMS theory metric given by Eq. (1). The Lagrangian corresponding to this metric is
$ {\cal{L}} = \frac{1}{2}\left[-U(r)\dot{t}^{2} +\frac{1}{U(r)}\dot{r}^{2}+f(r)\left(\dot{\theta}^{2}+\sin^{2}\theta\dot{\varphi}^{2}\right)\right], $
(6) To obtain the geodesic equations, we use the following Hamilton-Jacobi equation:
$ \frac{\partial {\cal{S}}}{\partial \sigma } = -\frac{1}{2}g^{\mu \nu }\frac{ \partial {\cal{S}}}{\partial x^{\mu }}\frac{\partial {\cal{S}}}{\partial x^{\nu }}, $
(7) where
$ {\cal{S}} $ is the Jacobi action. The Jacobi action separable solution reads$ {\cal{S}} = -Et+\ell \varphi +{\cal{S}}_{r}\left( r\right) +{\cal{S}} _{\theta }\left( \theta \right) , $
(8) where E and
$ \ell $ are the two Killing vectors of the metric expressed by Eq. (1), given by$ E = \frac{{\rm d}{\cal{L}}}{{\rm d}\overset{\cdot }{t}} = -U\left( r\right) \dot{t} $
(9) $ \ell = \frac{{\rm d}{\cal{L}}}{{\rm d}\overset{\cdot }{\varphi }} = f(r)\sin ^{2}\theta \overset{\cdot }{\varphi }\text{.} $
(10) Thus, the geodesic equations are
$ \frac{{\rm d}t}{{\rm d}\sigma } = \frac{E}{U\left( r\right) },\qquad \frac{{\rm d}\varphi }{{\rm d}\sigma } = -\frac{\ell }{f(r)\sin ^{2}\theta }, $
(11) $ r^{2}\frac{{\rm d}r}{{\rm d}\sigma } = \pm \sqrt{{\cal{R}}\left( r\right) },\qquad r^{2}\frac{{\rm d}\theta }{{\rm d}\sigma } = \pm \sqrt{\Theta \left( \theta \right) }, $
(12) where
$ {\cal{K}} $ is the Carter separation constant and$ {\cal{R}}\left( r\right) = r^{4}E^{2}-\left( {\cal{K}}+\ell ^{2}\right) r^{2}U\left( r\right) , $
(13) $ \Theta \left( \theta \right) = {\cal{K}}-\ell ^{2}\cot \theta . $
(14) Dimensionless quantities called impact parameters are introduced as
$ \eta = \frac{{\cal{K}}}{E^{2}},\qquad \text{ }\zeta = \frac{\ell }{E}. $
(15) It depends on the values of critical parameters whether the photon is captured, scattered to infinity, or bound to orbits. Our interest is in spherical light geodesics constrained on a sphere of constant coordinate radius r with
$ {\dot r} = 0 $ and$ {\ddot r} = 0 $ , also known as spherical photon orbits. Without any loss of generality, we set the equatorial plane as$ \theta = \pi/2 $ . Circular orbits correspond to the maximum effective potential, and unstable photons must satisfy the following conditions:$ V_{\rm eff}\left( r\right) \left\vert _{r = r_{ps}}\right. = 0, \;\;\;\;V_{\rm eff}^{\prime }\left( r\right) \left\vert _{r = r_{ps}}\right. = 0, $
(16) or,
$ {\cal{R}}\left( r\right) \left\vert _{r = r_{ps}}\right. = 0,\;\;\;\;{\cal{R}} _{\rm eff}^{\prime }\left( r\right) \left\vert _{r = r_{ps}}\right. = 0. $
(17) where
$ r_{ps} $ is the photon sphere and determines the location of the apparent image of the photon rings; the Carter separation constant disappears on recasting Eqs. (13) and (14). If we consider the metric expressed by Eq. (1) then we can express the radius of the photon sphere as the solution of the equation$ f^{\prime }\left( r_{ps}\right) U\left( r_{ps}\right) -f\left( r_{ps}\right) U^{\prime }\left( r_{ps}\right) = 0, $
(18) or explicitly
$ \begin{aligned}[b]& 6M^{4}r^{2}-2M^{3}\left( r^{3}+Q^{2}r\left( 2\beta -5\gamma \right) \right) \\&\quad +M^{2}Q^{2}\left( -3r^{2}-2Q^{2}\left( \beta -2\gamma \right) \gamma -Q^{4}r\gamma ^{2}\right) = 0. \end{aligned} $
(19) The above equation is a cubic equation, that is, it has three analytical roots. Using the Mathematica 11 software, the only real root is given by
$ \begin{aligned}[b] r_{ps} =\;& \frac{2M^{2}-Q^{2}\gamma }{2M}+\frac{A}{2^{2/3}3M^{2}\left( B+\sqrt[3] {4A^{3}+B^{2}}\right) ^{1/3}}\\&- \frac{\left( B+\sqrt[3]{4A^{3}+B^{2}}\right) ^{1/3}}{2^{1/3}6M^{2}}, \end{aligned} $
(20) where
$ A = 3M^{2}\left( -12M^{4}+6M^{2}Q^{2}\left( \beta -\gamma \right) -Q^{4}\gamma ^{2}\right) $
and
$ B = 108M^{5}\left( -4M^{4}+4M^{2}Q^{2}\left( \beta -\gamma \right) -Q^{4}\gamma ^{2}\right) . $
To illustrate this, Fig. 3 represents the radius of photon sphere with respect to the parameters β and γ. As can be seen from Fig. 3, the photon sphere increases as the magnitude of the γ parameter increases. However, as the parameter β increases, the photon sphere decreases first and then remains constant, regardless of how much it increases. Figure 4 depicts a three-dimensional plot of the radius of the photon sphere with respect to the parameters β and γ, revealing their effect on
$ r_{ps} $ .Figure 3. (color online) Plot of Eq. (20) illustrating the dependence of
$r_{ps}$ on the black hole parameters β and γ; here,$Q/M=0.66$ .Figure 4. (color online) Three-dimensional plot of Eq. (20) illustrating the dependence of
$r_{ps}$ on the black hole parameters β and γ; here,$Q/M=0.66$ .Accordingly, the radius of the shadow
$ R_{sh} $ is defined by the lensed image of the photon sphere as$ \begin{aligned}[b] R_{sh} =\;& \left. \sqrt{\frac{f\left( r_{ps}\right) }{U\left( r_{ps}\right) }} \right\vert _{r = r_{ps}} \\ =\;& \sqrt{-\frac{r_{ps}^2(Mr_{ps}+Q^2\gamma)^2 }{M(2M^2r_{ps}-M(r_{ps}^2+Q^2(\beta-2\gamma))-Q^2r_{ps}\gamma)}}. \end{aligned} $
(21) This coincides with the value of the impact parameter itself. Figures 5 and 6 depict the variation of the shadow observable
$ R_{sh} $ and contours plot, respectively, for the charged black hole solution within the EMS theory in the (β, γ) space. They show how the shadow size varies. Note that an increase in the magnitude of the γ parameter leads to an increase in the size of the black hole shadow. In contrast, the β parameter has the opposite effect: it decreases the shadow of the black hole.Figure 5. (color online) Variation of the shadow observable
$R_{sh}$ according to Eq. (21) for the charged black hole within the EMS theory; here,$Q/M=0.66$ .Figure 6. (color online) Variation of the contour plot for the charged black hole within the EMS theory; here,
$Q/M$ =0.66.Celestial coordinates are used to describe the shadow of the black hole seen on an observer's frame [91]. Thus, we define the celestial coordinates X and Y by
$ X = \lim\limits_{r_{0}\rightarrow \infty }\left( -r_{0}\sin \theta _{0}\left. \frac{ {\rm d}\varphi }{{\rm d}r}\right\vert _{r_{0},\theta _{0}}\right) , $
(22) $ Y = \lim\limits_{r_{0}\rightarrow \infty }\left( r_{0}\left. \frac{{\rm d}\theta }{{\rm d}r} \right\vert _{r_{0},\theta _{0}}\right) , $
(23) where
$ (r_{0},\theta_{0}) $ are the position coordinates of the observer. Assuming that the observer is on the equatorial hyperplane, Eqs. (22) and (23) follow$ X^{2}+Y^{2} = R_{sh}^{2}. $
(24) Table 1 lists the numerical values of
$ r_{ps} $ and$ R_{sh} $ for a specific set of parameters. The profile of the shadows cast by the charged black hole within the EMS theory is shown in Fig. 7 under the influence of the parameters β and γ. Figure 7 clearly shows that the shadow radii decrease in black holes as β increases and that the decrements of the shadow radii also increase with different intervals.$\gamma =-1/2$ $\gamma =-3/4$ $\gamma =-1$ $r_{ps}/M$ $R_{sh}/M$ $r_{ps}/M$ $R_{sh}/M$ $r_{ps}/M$ $R_{sh}/M$ $\beta =1$ 0.5377 0.0678 0.8268 0.1433 1.124 0.24206 $\beta =3/2$ 0.52669 0.05842 0.80688 0.12522 1.0949 0.21377 $\beta =2$ 0.52024 0.05210 0.79469 0.11269 1.0762 0.19371 Table 1. Numerical results for the values of
$r_{ps}$ and$R_{sh}$ for a black hole within the EMS theory; here,$Q/M=0.66$ . -
In this section, we examine the weak gravitational lensing around the black hole within the EMS theory. Interestingly, the deviation of a ray of light from its original path occurs when it passes through the close vicinity of massive objects. For a weak-field approximation, the following relation can be used as a metric tensor:
$ g_{\alpha \beta} = \eta_{\alpha \beta}+h_{\alpha \beta}\, , $
(25) where
$ \eta_{\alpha \beta} $ and$ h_{\alpha \beta} $ refer to expressions for the Minkowski spacetime and perturbation gravity field describing the EMS theory, respectively. For the weak gravitational field to be satisfied, the following expressions for$ \eta_{\alpha \beta} $ and$ h_{\alpha \beta} $ must be satisfied:$ \begin{array}{l} \eta_{\alpha \beta} = {\rm diag}(-1,1,1,1)\ , \\ h_{\alpha \beta} \ll 1, h_{\alpha \beta} \rightarrow 0 \;\; \;\;\text{under}\;\; x^{\alpha}\rightarrow \infty \ ,\\ g^{\alpha \beta} = \eta^{\alpha \beta}-h^{\alpha \beta},\;\;\;\;\;\; h^{\alpha \beta} = h_{\alpha \beta}\, , \end{array} $
(26) where
$ x^{\alpha} $ refers to the spacetime coordinate.Using the fundamental equation, we can obtain the expression for the deflection angle around a compact object within the EMS theory as follows:
$ \hat{\alpha }_{\text{b}} = \frac{1}{2}\int_{-\infty}^{\infty}\frac{b}{r}\left(\frac{{\rm d}h_{33}}{{\rm d}r}+\frac{{\rm d}h_{00}}{{\rm d}r}\right){\rm d}z\ , $
(27) We can express the line element in Eq. (1) as follows:
$ {\rm d}s^2 \approx {\rm d}s_0^2+ \Big(\frac{2M}{r}-\frac{\beta Q^2}{f(r)}\Big){\rm d}t^2+ \Big(\frac{2M}{r}-\frac{\beta Q^2}{f(r)}\Big){\rm d}r^2, $
(28) where
${\rm d}s^2_0 = -{\rm d}t^2+{\rm d}r^2+r^2({\rm d}\theta^2+\sin^2\theta {\rm d}\phi^2)$ . We can easily find components$ h_{\alpha \beta} $ of the metric element in Cartesian coordinates as follows:$ h_{00} = \frac{2M}{r}-\frac{\beta Q^2}{f(r)}, $
(29) $ h_{ik} = \Bigg(\frac{2M}{r}-\frac{\beta Q^2}{f(r)}\Bigg)n_i n_k ,$
(30) $ h_{33} = \Bigg(\frac{2M}{r}-\frac{\beta Q^2}{f(r)}\Bigg) \cos^2\chi \ , $
(31) where
$ \cos^2\chi = z^2/(b^2+z^2) $ and$ r^2 = b^2+z^2 $ .Now, we can define the derivatives of
$ h_{00} $ and$ h_{33} $ using radial coordinates. Subsequently, we can calculate the deflection angle$ \hat{\alpha}_b $ .Accordingly, we determine an explicit form of the deflection angle analytically. To this end, we restrict the location of the observer to the equatorial plane, i.e.,
$ \theta = \pi/2 $ . To analyze the deflection angle, we use the Hamilton formalism to evaluate the geodesic equations. The standard Hamiltonian is expressed as$ H(x,p) = \frac{1}{2}g^{\alpha \beta}(x)p_{\alpha}p_{\beta}\, , $
(32) with
$ \dot{p_{\alpha}} = -\frac{\partial H}{\partial x^{\alpha}} \quad \text{and} \quad \dot{x^{\alpha}} = \frac{\partial H}{\partial p_{\alpha}}\, . $
(33) From the Hamilton-Jacobi equation, we can also derive
$ \begin{align} \dot{\varphi} = \frac{\partial H}{\partial p_{\varphi}} = g^{\varphi \varphi}p_{\varphi} \quad \text{and} \quad \dot{r} = \frac{\partial H}{\partial p_r} = g^{rr}p_{r}\, . \end{align} $
(34) From the above equations, a simplified form is given by
$ \left(\frac{\dot{r}}{\dot{\varphi}}\right)^2 = \left(\frac{g^{rr}p_r}{g^{\varphi \varphi}p_{\varphi}}\right)^2\, . $
(35) It should be noted that the Hamiltonian can be considered as
$ H = 0 $ for the null particle. Thus, Eq. (32) can be rewritten on the basis of$ p_t = -E $ and$ p_{\varphi} = l $ , i.e.,$ g^{rr}p^2_r = -(g^{tt}E^2+g^{\varphi \varphi}l^2)\, , $
(36) leading to
$ \left(\frac{\dot{r}}{\dot{\varphi}}\right)^2 = -\frac{g^{rr}}{(g^{\varphi \varphi}l)^2}(g^{tt}E^2+g^{\varphi \varphi}l^2)\, . $
(37) Taking
$ b = {E/}{l} $ into consideration, which is referred to as the impact parameter, we can rewrite Eq. (37) as follow:$ \left(\frac{\dot{r}}{\dot{\varphi}}\right)^2 = -\frac{g^{rr}}{(g^{\varphi\varphi})^2}(g^{tt}b^2+g^{\varphi\varphi})\, . $
(38) As a matter of fact, the deviation of the light ray leads to the deflection angle by which the ray is bent from its original path when passing through a massive object. Hence, the deflection angle can be evaluated when the light is bent from its original path at the closest distance from the massive object (i.e.,
$ r = r_0 $ ). Furthermore, to determine the impact parameter at$ r = r_0 $ , one can set the following condition:$ \left(\frac{\dot{r}}{\dot{\varphi}}\right)\Big|_{r = r_0} = 0\, , $
(39) with
$ g^{tt}|_{r = r_0} = G^{tt},\quad g^{\varphi\varphi}|_{r = r_0} = G^{\varphi\varphi}, \quad g^{rr}|_{r = r_0} = G^{rr}\, . $
(40) Considering Eqs. (39) and (40) together, the impact parameter can be expressed as follows:
$ b^2 = -\frac{G^{\varphi\varphi}}{G^{tt}}\, . $
(41) According to Eqs. (38) and (41), the integral form of the deflection angle by which the light is deviated from its original path can be defined as
$ \int_0^{\bar{\alpha}} {\rm d} \varphi = \pm 2 \int_{-\infty}^{\infty} \Bigg[\frac{-g^{rr}}{(g^{\varphi \varphi})^2}\left(g^{tt}b^2+g^{\varphi \varphi}\right)\Bigg]^{-1/2}{\rm d}r\, . $
(42) The important point to be noted here is that one can take into account π when evaluating the deflection angle of the light ray deviating from its original trajectory if and only if the coordinate's center refers to the compact object. Accordingly, the real deflection angle by which the light is bent from its original path can be defined as
$ \hat{\alpha_b} = \bar{\alpha}-\pi $ . However, the analytical integration of Eq. (42) for the deflection angle is complicated. Therefore, we resort to the numerical evaluation of the deflection angle$ \hat{\alpha_b} $ . To gain a deeper understanding on the deflection angle of the light ray, we analyzed its behaviour. Its dependence on the impact parameter for various values of black hole parameters is shown in Fig. 8. Note from Fig. 8 that the deflection angle of the light ray decreases with the increase in the impact parameter$ b/M $ , whereas the curves shift downwards to smaller values with the increase in the black hole charge and parameter β.Figure 8. (color online) Deflection angle
$\hat{\alpha}$ plotted as a function of the impact parameter b for different combinations of parameter β (left panel) and black hole charge (right panel) for a fixed value of γ.Let us now examine the brightness of the image using the light's deflection angle around the black hole within the EMS theory. To this end, let us consider the following expression, given in terms of the light angles, such as
$ \hat{\alpha_b} $ , θ and β [36, 41, 92]):$ \theta D_\mathrm{s} = \beta D_\mathrm{s}+\hat{\alpha_b}D_\mathrm{ds}\, . $
(43) Note that in Eq. (43) we represent the distances between the source and observer,
$ D_\mathrm{s} $ , the lens and the observer,$ D_\mathrm{d} $ , and the source and the lens,$ D_\mathrm{ds} $ , whereas θ and β denote the angular position of the image and source, respectively. Based on Eq. (43), the angular position β of the source is expressed as$ \beta = \theta -\frac{D_\mathrm{ds}}{D_\mathrm{s}}\frac{\xi(\theta)}{D_\mathrm{d}}\frac{1}{\theta}\, . $
(44) Note that we have used
$ \xi(\theta) = |\hat{\alpha}_b|\,b $ , with$ b = D_\mathrm{d}\theta $ [92]. Accordingly, one can determine the shape of the image as Einstein ring using the radius$ R_s = D_\mathrm{d}\,\theta_E $ , provided that its shape behaves like a ring. In Eq. (44), the angular part$ \theta_E $ , which appears owing to spacetime geometry between the source images, can be expressed as [36]$ \theta_E = \sqrt{2R_s\frac{D_{ds}}{D_dD_s}}\, . $
(45) Next, we examine the magnification of brightness defined as [41, 93−96]
$ \mu_{\Sigma} = \frac{I_\mathrm{tot}}{I_*} = \mathop \sum \limits_k \bigg|\bigg(\frac{\theta_k}{\beta}\bigg)\bigg(\frac{{\rm d}\theta_k}{{\rm d}\beta}\bigg)\bigg|, \quad k = 1,2, \cdot \cdot \cdot , j\, ,\\ $
(46) where
$ I_\mathrm{tot} $ denotes the total brightness whereas$ I_* $ denotes the unlensed brightness of the source. The total magnification can thus be expressed as$ \mu_\mathrm{tot} = \frac{x^2+2}{x\sqrt{x^2+4}}\, . $
(47) Here,
$ x = {\beta}/{\theta_E} $ refers to a dimensionless quantity. Let us now explore the magnification of the source numerically. In particular, we analyzed the dependence of the total magnification on the black hole electric charge for different values of the parameter β, as shown in Fig. 9. Note from Fig. 9 that the total magnification decreases with the increase in the black hole charge, whereas its curves shift downwards to smaller values as a consequence of an increase in the value of the parameter β.
Shadows and weak gravitational lensing for black holes within Einstein-Maxwell-scalar theory
- Received Date: 2024-05-14
- Available Online: 2024-09-15
Abstract: In this study, we investigated the optical properties of charged black holes within the Einstein-Maxwell-scalar (EMS) theory. We evaluated the shadow cast by these black holes and obtained analytical solutions for both the radius of the photon sphere and that of the shadow. We observed that black hole parameters γ and β both influence the shadow of black holes. We also found that the photon sphere and shadow radius increase as a consequence of the presence of the parameter γ. Interestingly, the shadow radius decreases first and then remains unchanged owing to the impact of the parameter β. Finally, we analyzed the weak gravitational lensing and total magnification of lensed images around black holes. We found that the charge of the black holes and the parameter β both have a significant impact, reducing the deflection angle. Similarly, the same behavior for the total magnification was observed, also as a result of the effect of the charge of the black holes and the parameter β.