Extreme-mass ratio inspirals in Schwarzschild - de Sitter spacetime I: Weak-field orbits John Adrian N. Villanueva, Ian Vega Abstract The inspiral of a compact object into a black hole is a key source of low-frequency gravitational waves for future space-based detectors like LISA. While models of this process have advanced, they typically focus on asymptotically flat spacetimes. In this paper, we explore how the absence of asymptotic flatness affects the slow, adiabatic orbital evolution due to radiation reaction. This lack of asymptotic flatness can arise from external environments or an expanding universe. Using the Schwarzschild-de Sitter (SdS) spacetime, where the deviation from flatness is governed by the cosmological constant, we study bound orbits characterized by their semi-latus rectum p p and eccentricity e e . We calculate how the cosmological constant shifts the separatrix between bound and plunging orbits and alters the relationship between the binary’s binding energy, angular momentum, and orbital parameters. Assuming the orbital timescale is much shorter than the inspiral timescale, we apply a modified quadrupole formula to examine the impact of a small positive cosmological constant on the orbital evolution in the weak-field limit. We find that the cosmological constant accelerates the decrease in eccentricity, reducing inspiral plunge times, which could influence event rate estimates for space-based detectors. 1 Introduction Gravitational waves have emerged as powerful probes of the universe, opening a fundamentally new window onto its most energetic and previously inaccessible phenomena. The recent direct detections of gravitational-wave signals [ 1 ] have firmly established their role as essential tools in modern astrophysics. Together with solar and supernova neutrinos, diffuse gamma-ray backgrounds, and electromagnetic counterparts in binary neutron star mergers [ 23 , 60 ] —gravitational waves form an additional pillar of multimessenger astronomy. By enabling us to “hear” the gravitational dynamics of compact objects, they reveal aspects of the cosmos that were once beyond observational reach, transforming our understanding of an otherwise silent universe. Gravitational wave astronomy could also help cosmology on the problem of the expansion of the universe [ 63 ] . The general rate of expansion, known as the Hubble parameter [ 26 , 64 ] is closely related to the cosmological constant, \Lambda . While estimates of this parameter have been obtained, most notably by the Planck collaboration [ 3 ] , this will still be tested more thoroughly using gravitational wave signals [ 73 ] . Borrowing the idea of standard candles from stars to determine the gravitational redshift distance between celestial bodies, standard sirens have also been put forward for gravitational wave sources [ 44 , 37 , 81 ] . Relevant sources include black hole binary mergers [ 71 ] and neutron star binaries [ 75 ] , both of which have already been observed by the ground-based triple gravitational wave interferometer network consisting of LIGO (Hanford and Livingston) and Virgo (Italy). The types and number of sources available for analysis will increase significantly with the upcoming launch of the space-based LISA (Laser Interferometer Space Antenna) gravitational wave detector [ 82 , 35 , 9 , 28 ] . Both ground-based and future space-based detectors will be pivotal in analyzing gravitational wave signals to estimate the Hubble parameter and the size of the universe. Low-frequency gravitational wave sources, particularly in the millihertz range, will be key targets for space-based detectors like LISA. Extreme-mass ratio inspirals (EMRIs) [ 5 , 50 ] in particular, where compact objects spiral into massive black holes with a mass ratio of approximately 10^{-6} , are expected to have large number of gravitational wave cycles within the millihertz band (typically on the order of the inverse of the mass ratio) which should enable us to precisely probe strong-field gravitational effects [ 15 , 4 ] . One could envision an EMRI as a binary system containing a stellar-mass compact object, and a supermassive black hole. However, the typical isolated two-body problem in classical physics provides an idealized model of a celestial binary system, neglecting various external physical factors in a real astrophysical context. To accurately model the motion of such a binary, one must account for physically realistic external agents that contribute gravitationally. These external influences are often referred to as environmental effects, and have been extensively discussed in the literature [ 16 , 84 , 87 , 24 ] . In addition to the effects of matter [ 18 , 55 , 17 , 25 ] and external fields [ 36 , 39 ] , we argue that the expansion of spacetime itself can also be considered an environmental effect. While there have been numerous studies on the radiation reaction in binary systems, fewer have focused on the influence of the expanding universe on binary dynamics [ 22 , 47 ] . The inclusion of the cosmological constant not only means that the spacetime has a constant acceleration of expansion, it adds new features in the spacetime geometry [ 77 ] such as the emergence of a cosmological horizon and new properties to the geodesics of the spacetime, as studied throroughly in [ 78 ] and [ 40 ] . The de Sitter solution, which provides the simplest description of an expanding universe, generates a nontrivial gravitational flux that can be detected at infinity due to spacetime curvature. This phenomenon has been rigorously examined by [ 10 , 13 , 20 , 22 , 14 , 11 , 12 , 32 , 33 , 49 , 21 ] . When a black hole binary is placed within an expanding universe, the resulting spacetime is modeled as Schwarzschild-de Sitter, also known as Kottler spacetime [ 52 ] . This solution represents the simplest black hole solution to Einstein’s field equations that includes a cosmological constant [ 77 ] . The objective of this work is to investigate the impact of spacetime expansion due to the cosmological constant on an inspiraling binary system, building on the energy and angular momentum flux formulas provided by [ 47 ] . We focus on the orbital evolution of binaries primarily driven by gravitational wave radiation and its associated radiation reaction. Specifically, we consider a binary consisting of a stellar-mass compact object gravitationally bound to a black hole, such as an EMRI. We assume that the compact object survives tidal disruption by the Schwarzschild black hole, allowing it to successfully inspiral and eventually plunge. The binary is placed within a de Sitter universe, which expands at a constant rate. As a result, outside the black hole, the spacetime is effectively described by the Schwarzschild-de Sitter metric. In this framework, we analyze both conservative and dissipative effects of the cosmological constant on the binary’s dynamics. To provide a comprehensive description, we consider generic orbits with eccentricities ranging from e=0 (circular orbits) to e=1 (unbound orbits), treating eccentricities approaching unity as representing unbound orbits. Although [ 47 ] have already examined the effect of radiation on orbital evolution in de Sitter space, they did not modify the full Keplerian dynamical quantities to include the de Sitter terms nor fully explore how the cosmological constant affects orbital dynamics. In this work, we extend their analysis by providing a complete treatment of the weak-field, adiabatic dissipative dynamics driven by radiation reaction for a compact binary. We also present preliminary results on strong-field energy fluxes, offering a check on the validity of the weak-field flux approximations. Additionally, we explore other conservative effects of the cosmological constant on the binary, such as the separatrix and the parameters of the last stable orbit, as well as the influence of radiation reaction on the periapsis advance, which is also modulated by the cosmological constant. Moreover, we calculate adiabatic waveforms for a continuously evolving orbit due to radiation reaction in Schwarzschild-de Sitter spacetime, highlighting shifts in the waveform’s amplitude and phase. We estimate the impact of a realistic cosmological constant, placing particular emphasis on its role in the dissipative dynamics driven by gravitational radiation reaction. While the contribution from \Lambda is generally subdominant relative to leading post-Newtonian effects, its relative importance increases near the in certain regions of the parameter space (which will be discussed below), where it may approach the magnitude of post-Newtonian corrections. More broadly, our results may be interpreted in terms of an environmental contribution to the gravitational potential of the form \delta V\sim\Lambda r^{2} . In this sense, \Lambda serves as a proxy for an r^{2} perturbation to the effective potential. An example arises in studies of accretion or tidal interactions in a binary system analyzed in the co-rotating frame. In this setting, the Roche potential contains an additional contribution associated with the binary’s angular velocity \omega , including terms that scale as \sim(\omega\times r)^{2} [ 38 ] . Our results may therefore be mapped onto such systems by interpreting a large effective cosmological constant \Lambda as corresponding to a large orbital angular velocity induced from a Coriolis force. Similarly, the framework developed here could be extended to describe an EMRI embedded in an external uniform magnetic field, where the magnetic-field contribution to the metric functions scales as \sim Br^{2} [ 69 , 51 ] . In this way, the present analysis captures a broader class of physical scenarios characterized by quadratic modifications to the effective gravitational potential. The paper is divided into five parts. Section 2 contains an elementary review of bound timelike geodesics in Schwarzschild-de Sitter spacetime, which represents the conservative trajectories followed by particles. We parametrize bound orbits using either their energy and angular momentum (E,L) or their semilatus rectum and eccentricity (p,e) . An important result of this section is the effect of the cosmological constant on the separatrix surface, the boundary in parameter space separating unstable and stable bound orbits. In Section 3 , we build on the results of Hoque and Aggarwal to describe the influence of the cosmological constant on orbital evolution due to radiation reaction in the weak-field, adiabatic limit. We check the accuracy of these results with a preliminary calculation of the fluxes derived from black hole perturbation theory in Schwarzschild-de Sitter spacetime. Additionally, we provide sample orbital shapes to illustrate the impact of the cosmological constant on the dynamics of the binary system. Section 4 focuses on the calculation of adiabatic waveforms, demonstrating how the cosmological constant modifies the amplitude and phase of gravitational waveforms for both circular and eccentric binaries. We also explore the constraints that the adiabatic approximation places on the mass ratios and the cosmological constant, determining the conditions under which these approximations remain valid. Lastly, in Section 5 we investigate the effect of the cosmological constant on the plunge times of generic orbits and examine how this relates to the timescales for detecting gravitational wave sources. We adopt the usual conventions by Misner, Thorne and Wheeler, which uses the mostly plus metric signature and the geometrized units of naturalizing the constants G=c=1 . For astrophysical applications, these naturalized constants applied to conventional units translate to 1\,M_{\odot}\simeq 5\times 10^{-6} s \simeq 1.5 km and 1 pc \simeq 10^{8} s. 2 Orbital mechanics in Schwarzschild-de Sitter spacetime The first exact solution to the Einstein field equation representing a black hole immersed in an expanding universe is known as the Schwarzschild-de Sitter (SdS) solution [ 34 ] : g_{\mu\nu}=\text{diag}(-f(r),\,f(r)^{-1},\,r^{2},\,r^{2}\sin^{2}\theta), (2.1) where we have the metric function, f(r)=1-2M/r-\Lambda\,r^{2}/3 . This metric is a special case of the McVittie solution [ 59 ] , for positive spatial curvature and a constant Hubble parameter, H(t)=H_{0} . The SdS metric satisfies the vacuum Einstein field equation with a positive cosmological constant, \Lambda : G_{\mu\nu}-\Lambda\,g_{\mu\nu}=T_{\mu\nu}=0. (2.2) When 0<\Lambda<\Lambda_{\textrm{max}}:=(3M)^{-2} (or 1/\sqrt{\Lambda}>3M ), the spacetime will feature two horizons (i.e. the roots of equation ( 2.1 )): the event horizon, which marks the boundary of a black hole, and the cosmological horizon, which sets the scale of the observable universe. When \Lambda>(3M)^{-2} , the spacetime is not static for any r (i.e. there no longer exists a static patch), and the SdS spacetime is no longer a good representation of a black hole immersed in an expanding universe. We shall only be interested in the former case, and we will focus on orbits between the event horizon and cosmological horizon. It is convenient to use dimensionless variables \tilde{r}:=r/M and \lambda:=M^{2}\Lambda/3 . For both horizons to exist, as we required in the preceding paragraph, we thus require \lambda<\lambda_{\textrm{max}}:=1/27 . A value of \lambda approaching this critical value would imply an unrealistic scenario in which a black hole is comparable in size to the entire universe. Based on our best estimates for the cosmological constant and the Hubble parameter [ 72 ] , we instead find that \lambda\ll 1 . For a supermassive black hole (SMBH) of mass M=10^{6}M_{\odot} and \Lambda=10^{-50} km -2 , our dimensionless cosmological constant \lambda is of order \lambda=10^{-38} . This is an extraordinarily small number, leading us to expect that any resulting effects from \lambda will be practically negligible. Nonetheless, we find it worthwhile to explore the potential effects when \lambda is not exceedingly small, as this may serve as a proxy for other environmental factors with similar r^{2} -scaling in the effective potential [ 31 ] . Sticking to a cosmological constant scenario, we note that a non-negligible \lambda=10^{-5} , for \Lambda\sim 10^{-50} km -2 , would imply a black hole mass of M\approx 10^{22}M_{\odot} . We are thus fully justified in exploiting \lambda\ll 1 as a small parameter in all of our calculations. As an example, consider the locations of the horizons, which are obtained by solving for the roots of the metric function: \displaystyle f(\lambda;r)=1-\frac{2}{r}-\lambda r^{2}=0. (2.3) Up to \mathcal{O}(\lambda^{2}) , the roots are \displaystyle r_{h} \displaystyle= \displaystyle 2+8\lambda+\mathcal{O}(\lambda^{2}) (2.4) \displaystyle r_{c} \displaystyle= \displaystyle-1+\lambda^{-1/2}-\frac{3}{2}\lambda^{1/2}-4\lambda+\mathcal{O}(\lambda^{2}) (2.5) \displaystyle r_{o} \displaystyle= \displaystyle-1-\lambda^{-1/2}+\frac{3}{2}\lambda^{1/2}-4\lambda+\mathcal{O}(\lambda^{2}) (2.6) with r_{h} as the horizon of the SdS black hole, r_{c} as the cosmological horizon and r_{o} as an unphysical negative root. We further see that requiring that the black hole is inside the universe, r_{\textrm{h}}0 spacetime require more energy input to separate. Looking at the angular momentum in equation ( 2.16 ), and requiring it to be always positive along with p , we could also limit the possible values of the eccentricity of an orbit. A particular SdS orbit can therefore only have eccentricity values, 00 makes most eccentric orbits unbound. We see in the red and violet curve that a sufficiently high cosmological constant would also make orbits with high separation (larger orbits) also unbound. In (b) we show a continuous variation of \lambda onto the phase space (p,e) wherein the marginally bound orbits are the points lying on the surface. (a) (b) Figure 3 : Similar contour plot and 3D plot to Figure 2 but with the scattering separatrix equation S(p,e,\lambda) (equation ( 2.30 )). We can solve for the real roots in p of equation ( 2.3.1 ) perturbatively using the method of dominant balance [ 19 ] and see that there are two relevant curves, \displaystyle p_{\mathrm{inner,pl}}=6+2e+\lambda\frac{8(3-e)^{2}(3+e)^{3}}{\left(1-e^{2}\right)^{2}}+O(\lambda^{2}) (2.27) and \displaystyle p_{\mathrm{outer,pl}} \displaystyle= \displaystyle-\frac{1}{12}(e-3)^{2}(e+1)+\left(\frac{(e-1)^{2}(e+1)^{3}}{4\lambda}\right)^{1/3}+\mathcal{O}(\lambda^{1/3}). (2.28) The two curves (equations ( 2.3.1 ) and ( 2.28 )) form approximately the boundary of the region of bound orbital parameters, one as an “inner” separatrix, and one as an “outer” separatrix. The main observation here is the coupling of the cosmological constant with eccentricity, producing a tighter constraint for allowable orbits of higher eccentricity. Both separatrices, equations ( 2.27 ) and ( 2.28 ), are also placed in Figure 1(a) and 1(b) to see how they limit the constant curves of (E,L) and (p,e) . This boundary extends to very large p for small enough \lambda but for a nontrivial \lambda , we see here a distinct feature of the SdS spacetime. The separatrices may imply that the presence of the cosmological constant limits the existence of very large orbits, making them unstable since the dynamics due to the cosmological repulsion dominates the opposing central gravitational attraction of the black hole. We further check this claim in the next part by examining the nature of marginally bound orbits associated with an outer unstable extrema. 2.3.2 Scattering separatrix The earlier analysis only applies on marginally bound orbits in relation to the inner unstable extrema, wherein we have analyzed the region of bound orbits merging with the plunge region. This amounts to solving for the extremas with the radial distance evaluated at the periapsis. We can now examine the locus of parameters that merges the apoapsis to the outer unstable extrema by exmaining: \displaystyle\dfrac{d\,V_{\text{eff}}}{dr}\Big|_{r=r_{a}}=0 (2.29) which produces a distinct separatrix polynomial obtained earlier: \displaystyle S(p,e,\lambda)=( \displaystyle e-1)^{3}(e+1)^{2}(2e+p-6) \displaystyle-\lambda\,p^{3}\left(e^{3}+5e^{2}-5e-4p+15\right)+O\left(\lambda^{2}\right)=0. (2.30) Figure 4 : Potential shapes, V(r) vs r , (black curves in the right with dashed line as the orbital energy, E ) of selected orbital parameters (black dots) in (p,e) space. The intersection of orange and blue region contains the space of all bound orbits. The orange boundary (orange dashed line) marks the point at which orbits are marginally bound and may scatter away to the cosmological horizon while the blue boundary (blue dashed line) marks the points at which orbits are also marginally bound but may plunge to the black hole. The complement of the colored regions are unbound and may either plunge or scatter away. The contours of \lambda satisfying S(p,e,\lambda) is distinct from the separatrix polynomial, P(p,e,\lambda) , (equation ( 2.3.1 )), and can be seen in Figure 3 . We call P(p,e,\lambda) as the plunge separatrix since orbits outside the region it bounds will plunge to the black hole while we call S(p,e,\lambda) as the scattering separatrix since orbits outside the region it bounds will scatter to the cosmological horizon. The scattering separatrix also has maximum parameters \lambda=4/16875\sim 2.37\times 10^{-4} at (p=7.5,e=0) equivalent to the parameters derived from the limits of the plunging separatrix. When solved for p using a dominant balance method, we also obtain two real physical solutions in p distinct from equation ( 2.27 ) and ( 2.28 ), \displaystyle p_{\mathrm{inner,sc}} \displaystyle= \displaystyle 2(3-e)-\frac{8(e+3)^{2}(e-3)^{3}\lambda}{\left(e^{2}-1\right)^{2}}+\mathcal{O}(\lambda^{2}), (2.31) \displaystyle p_{\mathrm{outer,sc}} \displaystyle= \displaystyle\frac{1}{12}(3+e)^{2}(1+e)+\left(\frac{(1-e)^{3}(1+e)^{2}}{4\lambda}\right)^{1/3}+\mathcal{O}(\lambda^{1/3}). (2.32) This solution marks the boundary at which orbits will stay bounded and not scatter away as opposed to equation ( 2.28 ) which is the curve that marks the boundary at which orbits will stay bounded and not plunge to the black hole. We can now establish a hierarchy of lengthscales where our perturbative solutions will be valid, by examining the colored regions in Figure 4 . We impose the hierarchy of the following appropriate lengthscales as: p_{h}\dfrac{8}{9}-24\,\lambda and L^{2}>12-2592\,\lambda for circular orbits. For eccentric orbits, neglecting high eccentricities of order \mathcal{O}((1-e)^{-1}) , we have E^{2}<16-768\,\lambda and L^{2}<4-48\,\lambda . As expected from earlier results, the energy and angular momentum at the separatrix is decreased by a factor due to the cosmological constant. This would mean that lower energy and angular momentum would be required for orbits to plunge the black hole due to the expansion of spacetime, taking it longer to inspiral if we naively assume equal rates of energy and angular momentum loss for particles in a Schwarzschild and SdS spacetime. 3 Adiabatic orbital evolution As the compact object inspirals toward the black hole, it emits gravitational radiation, which carries away energy and angular momentum from its motion. In the leading post-Newtonian (pN) approximation, the quadrupolar formula for the secularly averaged energy and angular momentum flux measured at the cosmological horizon, has been studied extensively by Bonga among others [ 22 , 47 , 49 ] . For sources inside asymptotically flat spacetimes, the gravitational radiation is defined at the “far-away wave zone” using a 1/r expansion and then performing a temporal average over the orbital period. For asymptotically - de Sitter ( \Lambda>0 ) spacetimes, the technique used by Bonga and Ashtekar [ 11 , 14 ] was to define the radiative modes in the future null infinity ( \mathcal{I}^{+} ) using a late-time expansion replacing the expansion in 1/r . In particular, the radiation field was defined to be the following: \mathcal{Q}_{ij}:=\left(\mathcal{L}_{T}^{3}Q_{ij}^{(\rho)}+3H\mathcal{L}_{T}^{2}Q_{ij}^{(\rho)}+2H^{2}\mathcal{L}_{T}Q_{ij}^{(\rho)}+H\mathcal{L}_{T}^{2}Q_{ij}^{(p)}+3H^{2}\mathcal{L}_{T}Q_{ij}^{(p)}+2H^{3}Q_{ij}^{(p)}\right)\Bigg|_{t\rightarrow t_{\mathrm{ret}}} (3.1) with H=\left(\frac{\Lambda}{3}\right)^{1/2} , \mathcal{L}_{T} is the Lie derivative along the de Sitter time translational Killing vector and is related to the coordinate time t by \mathcal{L}_{T}=\partial_{t}-2\left(\frac{\Lambda}{3}\right)^{1/2} . The energy of the gravitational radiation at \mathcal{I}^{+} is defined to be the conserved current associated with a modified time-translation vector field. The averaged fluxes of energy and angular momentum radiated through \mathcal{I}^{+} of a compact source was derived [ 46 ] to be the integral of the mass ( \rho ) and pressure ( p ) quadrupole moments \mathcal{Q}_{ij}=\partial_{t}^{3}Q_{ij}^{(\rho)}-(3\Lambda)^{1/2}\partial_{t}^{2}Q_{ij}^{(\rho)}+\frac{2}{3}\Lambda\partial_{t}Q_{ij}^{(\rho)}+\left(\frac{\Lambda}{3}\right)^{1/2}\partial_{t}^{2}Q_{ij}^{(p)}-\frac{\Lambda}{3}\partial_{t}Q_{ij}^{(p)} , \displaystyle\langle\mathcal{P}\rangle= \displaystyle\frac{G}{5}\langle\mathcal{Q}_{ij}\mathcal{Q}_{ij}-\frac{1}{3}\mathcal{Q}^{2}\rangle (3.2) \displaystyle\langle\dot{L}_{k}\rangle= \displaystyle\frac{2G}{5}\epsilon_{ikm}\left\langle\mathcal{Q}_{ij}\left(\mathcal{L}_{T}^{2}Q_{ij}^{(\rho)}+\left(\frac{\Lambda}{3}\right)^{1/2}\left(\mathcal{L}_{T}Q_{ij}^{(\rho)}+\mathcal{L}_{T}Q_{ij}^{(p)}\right)+\frac{\Lambda}{3}Q_{ij}^{(p)}\right)\right\rangle evaluated at the retarded time t_{ret}=-\left(\frac{\Lambda}{3}\right)^{-1/2}\ln\left(\left(\frac{\Lambda}{3}\right)^{1/2}(r-\eta)\right) with \eta as the de Sitter conformal time. We reexpress the fluxes in terms of the orbital parameters (p,e) noting that we also neglect the pressure quadrupole moment in the leading order, \displaystyle\dfrac{dE}{dt} \displaystyle= \displaystyle-\frac{32}{5}\frac{1}{p^{5}}\left(1+\frac{73}{24}e^{2}+\frac{37}{96}e^{4}\right)(1-e^{2})^{3/2} \displaystyle-\lambda\frac{8}{3p^{2}}\left(4-(1-e^{2})^{1/2}\right)(1-e^{2})^{3/2}+\mathcal{O}(\lambda^{2}) \displaystyle\dfrac{dL}{dt} \displaystyle= \displaystyle-\frac{32}{5}\frac{1}{p^{7/2}}\left(1+\frac{7}{8}e^{2}\right)(1-e^{2})^{3/2} (3.4) \displaystyle-\lambda\frac{8}{p^{1/2}}(1-e^{2})+\mathcal{O}(\lambda^{2}). It is important to note that these expressions are valid only within the the range specified by equation ( 2.33 ). Under the conditions described by equation ( 2.33 ), for p=p_{\text{stable}} , the flux contribution from the cosmological constant remains subdominant compared to the Peters-Mathews fluxes. However, this contribution becomes comparable in magnitude only for p>p_{\text{stable}} , corresponding to unbound orbits. The expressions above (equations ( 3 ) and ( 3.4 )) indicate that a nonzero energy and angular momentum flux implies the energy and angular momentum of a bound orbit must change over time, although implicitly through the orbital parameters, (L,E)\rightarrow(L(p(t),e(t)),E(p(t),e(t))). (3.5) The flux of energy and angular momentum will cause the orbital trajectory to evolve continuously, resulting in a dynamic orbit characterized by time-dependent orbital parameters. Given that we have defined the bound orbits as ellipses (equation ( 2.10 )), long-term secular changes [ 56 ] in the orbital parameters will manifest as an evolving ellipse, or an osculating orbit. We now outline the derivation for the orbital evolution equations. To begin, we observe that the time derivatives of energy and angular momentum can be directly related to time derivatives of the orbital elements, \dfrac{dE}{dt}=\dfrac{\partial E}{\partial p}\dfrac{dp}{dt}+\dfrac{\partial E}{\partial e}\dfrac{de}{dt}\,,\quad\dfrac{dL}{dt}=\dfrac{\partial L}{\partial p}\dfrac{dp}{dt}+\dfrac{\partial L}{\partial e}\dfrac{de}{dt}. (3.6) We utilize the Jacobian J_{ij}=\partial_{i}\rho_{j} where i corresponds to the elements of (p,e) and \rho_{j} represents the elements of (L,E) . The evolution of the orbital parameters can then be expressed as: \displaystyle\left(\begin{array}[]{c}\dfrac{dp}{dt}\\ \\ \dfrac{de}{dt}\end{array}\right)=(J_{ij})^{-1}\left(\begin{array}[]{c}\dfrac{dL}{dt}\\ \\ \dfrac{dE}{dt}\end{array}\right). In this derivation, circular orbits, e=0 are excluded, as they render J_{ij}^{-1} singular as discussed in Section 2.2 . The evolution of circular orbits will instead be addressed separately in the next subsection, Section 3.1 . Using the fluxes (equations ( 3 ) and ( 3.4 )) and the derivatives of the corrected energy and angular momentum (equations ( 2.16 ) and ( 2.17 )), the orbital evolution equations given by: \displaystyle\dot{p}(p,e) \displaystyle= \displaystyle 2\sqrt{p}\;\dot{E}_{N} \displaystyle-\lambda\Bigg[2\sqrt{p}\;\dot{E}_{\lambda}+\epsilon\frac{3p^{7/2}}{\left(e^{2}-1\right)^{2}}\dot{L}_{N}-\epsilon\frac{4p^{5}}{\left(e^{2}-1\right)^{3}}\dot{E}_{N}\Bigg]+\mathcal{O}(\lambda^{2}) \displaystyle\dot{e}(p,e) \displaystyle= \displaystyle\frac{p}{e}\;\dot{L}_{N}+\frac{\left(e^{2}-1\right)}{\sqrt{p}e}\;\dot{E}_{N} \displaystyle-\frac{\lambda}{e}\Bigg[p\;\dot{L}_{\lambda}+\frac{\left(e^{2}-1\right)}{\sqrt{p}}\;\dot{E}_{\lambda}+\epsilon\frac{7p^{5/2}}{2\left(e^{2}-1\right)}\dot{L}_{N}-\epsilon\frac{4\left(e^{2}+1\right)p^{4}}{\left(e^{2}-1\right)^{3}}\dot{E}_{N}\Bigg]+\mathcal{O}(\lambda^{2}). The orbital evolution equations involve de Sitter terms of order \epsilon(=1) , a fiducial parameter, to highlight the effect of the corrected conserved energy, E=E_{N}+\epsilon\lambda E_{\lambda} , and angular momentum, L=L_{N}+\epsilon\lambda L_{\lambda} (equations ( 2.2 ) and ( 2.14 )). These orbital evolution equations represent the correct form of the orbital dynamics up to leading \mathcal{O}(\lambda) and leading p - dependence. Notably, the earlier work by Hoque and Aggarwal [ 47 ] did not account these linear \lambda terms, which should also contribute to the evolution of the orbital parameters. We further observe that the orbital evolution equations are only valid within the eccentricity range 0p_{0,circ} , the effect of the cosmological constant on reducing the orbital size becomes noticeable before the effect on eccentricity decay. For a fixed orbital size, we can numerically integrate the equation for \dot{e} to estimate the circularization time and its dependence on e and \lambda . Specifically, we compute t_{circ}=t(e,p_{fixed})=\int(de/dt)^{-1}de . Figure 8 shows the results for a fixed orbit size of p_{fixed}=50 . We observe that at a given value of cosmological constant, the circularization time is longest at high eccentricities. However, at larger values of \lambda the circularization time shortens at higher eccentricities. Figure 8 : Numerical curves for t_{circ} for a fixed p=50 (with the assumption that it stays constant) and its dependence on e for varying \lambda . We see here that circularization time are reduced by the cosmological constant. As we increase the value of \lambda , highly eccentric orbits circularizes much faster. (a) (b) Figure 9 : Numerical curves of p(t) and e(t) with initial parameters (p_{0}=20,\;e_{0}=0.9) for different cosmological constant values. We show here the general effect of the cosmological constant which is to reduce plunge and circularization times as the cosmological constant increases. To analyze the time evolution of the orbital parameters (p,e) , we numerically solve the system of ODE (equations ( 3.16 ) and ( 3.17 )) for given initial configurations. The resulting solutions, illustrated in Figure 9 , demonstrate that the presence of the cosmological constant is to increase the rate of orbital decay. As a consequence, the cosmological constant accelerates the inspiral of the CO, leading to an earlier plunge into the central BH. We reconstruct hybrid orbital trajectories in the following manner. The method of osculating orbits [ 68 , 66 ] supposes the true physical orbit evolves in time through its orbital parameters, such that at every instant, the orbit is a geodesic. The quasi-Keplerian equation of the orbit is of the form: r(\psi)=\frac{pM}{1+e\cos{\psi}}\rightarrow r(t)=\frac{p(t)M}{1+e(t)\cos{\psi(t)}}. (3.30) where the orbital parameters \{p,e,\psi\} are now functions of time. The angular frequency, \omega , of the CO orbiting the central BH is derived directly from the geodesics of the SdS spacetime. Using the definition \omega=(d\phi/d\tau)(dt/d\tau)^{-1} , and considering the orbital plane at \theta=\pi/2 , with the angular momentum L and energy E expressed in terms of (p,e) and r(\psi) for a bound orbit, the angular frequency in the large- p limit is given by: \omega=\left(\frac{L}{r^{2}}\right)\left(\frac{E}{f(r)}\right)^{-1}=n(1+e\cos(\psi))^{2} (3.31) where n is the corrected mean motion or the angular frequency of a circular orbit modified by the cosmological constant, n=\frac{1}{p^{3/2}}-\lambda\left(\frac{p^{3/2}}{2}+\mathcal{O}(\lambda p^{1/2})\right). (3.32) To eliminate oscillations in the angular frequency due to dependence on the true anomaly \psi , we perform an averaging procedure over a full orbital cycle. \displaystyle\frac{1}{P}\int_{t_{0}}^{t_{0}+P}\omega dt \displaystyle= \displaystyle\frac{1}{P}\int_{t_{0}}^{t_{0}+P}n(1+e\cos(\psi))^{2}dt (3.33) Assuming that the argument \delta of the periapsis remains constant over one complete orbit, \omega=\frac{d\phi}{dt}=\frac{d\psi}{dt}+\frac{d\delta}{dt}\sim\frac{d\psi}{dt}, (3.34) and we set dt=\omega^{-1}d\psi . Then we integrate: \int_{0}^{2\pi}(1+e\cos(\psi))^{-2}d\psi=\int_{t_{0}}^{t_{0}+P}n\,dt=\mathcal{M}=2\pi (3.35) over \psi=(0,2\pi) corresponding to the angle swept after a period P telling us that over a complete radial cycle, the orbit with angular frequency equal to the mean motion n , completes a full cycle in \psi . Since the right hand side corresponds to the mean anomaly \mathcal{M} , which is equal to 2\pi , the mean motion n serves as the averaged angular frequency over a single cycle from \psi=0 to \psi=2\pi for a generic eccentric orbit. We consider the mean motion as the characteristic angular frequency of the orbit, which will directly govern the phase evolution of the emitted gravitational wave. The osculating equations of the orbital parameters is appended by the equation for \psi , \displaystyle\frac{d\psi}{dt}=n(p)(1+e\cos{\psi})^{2} (3.36) with n(p) (equation ( 3.32 )) as the orbital mean motion as a function of p . Within a single cycle (an orbital period timescale), we impose that \psi only changes significantly, and that the changes in p and e happen only as a secular drift (adiabatic timescale, or the radiation reaction timescale). Within the adiabatic timescale, such that the true anomaly remains the single orbital parameter that changes within an orbital period, our hybrid orbital trajectory will be valid. Considering the radial motion along an orbital x-y plane, we have the positions: x=r(t)\cos{\phi(t)},\quad y=r(t)\sin{\phi(t)} (3.37) with the coordinate azimuthal angle \phi=\psi+\delta , related to the angle at the periapsis, which also becomes an evolving parameter \delta\rightarrow\delta(t) . We only include the leading order (1pN) rate of change in \delta , \Delta\delta_{1pN}=\frac{6\pi}{p}. (3.38) Since \delta depends on the evolving orbital size, we can estimate how much it changes due to the contribution of the cosmological constant to the gravitational flux. Using the dimensionalized inspiral rate \dot{p} from ( 3.28 ), we find that after one complete orbit, the periapsis shifts by a coordinate angle: \phi=2\pi\left(1+\frac{3}{p}\frac{GM}{c^{2}}\right)\sim 2\pi\left(1+\left(\frac{1\;\mathrm{km}}{p}\right)\left(\frac{M}{M_{\odot}}\right)\right). (3.39) For an EMRI orbit over a 1-year evolution, the change in the orbital size is approximately p\sim(p_{0}-10^{-30}) km for a realistic \Lambda value. The expected change in periapsis phase shift due of to the SdS gravitational flux is of the order \Delta\delta=\delta_{Sch}-\delta_{SdS}\sim 10^{-30} . Averaged over an orbital period (single radial cycle) P=2\pi/n(p) , the rate of periapsis precession becomes the following expression: \frac{d\delta}{dt}=\frac{\Delta\delta_{1pN}}{P}=\frac{3n(p)}{p} (3.40) which becomes implicitly an osculating orbital element through p and e . Along with the radiation reaction equations, ( 3.16 ) and ( 3.17 ), which could be interpreted as the (secularly averaged) p and e change after a single radial cycle, the complete adiabatic system are the system of equations: \displaystyle\frac{d\psi}{dt} \displaystyle= \displaystyle n(p)(1+e\cos{\psi})^{2} (3.41) \displaystyle\frac{d\delta}{dt} \displaystyle= \displaystyle\frac{3n(p)}{p} (3.42) \displaystyle\frac{dp}{dt} \displaystyle= \displaystyle-\frac{64}{5}p^{-3}(1-e^{2})^{3/2}\left(1+\frac{7}{8}e^{2}\right) (3.43) \displaystyle-\lambda\Bigg[16(1-e^{2})-\frac{4\left(26e^{4}-283e^{2}-168\right)}{15\left(1-e^{2}\right)^{3/2}}+\mathcal{O}(\lambda p)\Bigg] \displaystyle\frac{de}{dt} \displaystyle= \displaystyle-\frac{304}{15}p^{-4}e(1-e^{2})^{3/2}\left(1+\frac{121}{304}e^{2}\right) (3.44) \displaystyle-\lambda\Bigg[\frac{32}{3}p^{-1}e^{-1}\left(1-e^{2}\right)^{3/2}\left(1-(1-e^{2})^{1/2}\right) \displaystyle-p^{-1}e^{-1}\frac{2\left(73e^{6}-784e^{4}-965e^{2}-24\right)}{15\left(1-e^{2}\right)^{3/2}}+\mathcal{O}(\lambda p^{-2})\Bigg]. Upon solving for the evolution of the osculating elements, we can reconstruct a sample orbit, with the fast evolution of \psi , as well as the secular drift of \{p,e,\delta\} . We note here that the equations for \psi and \delta brings about conservative changes to the orbit while the equations for p and e comes from the dissipation of energy and angular momentum of the orbit. We present orbital solutions for a selected set of initial conditions, (p_{0}=70,\lambda=10^{-8}) and (p_{0}=50,\lambda=10^{-7}) with eccentricities e_{0}=0.7 , and e_{0}=0.4 in Figure 10 . Additionally, Figure 11 illustrates the evolution of circular orbits for p_{0}=50 and \lambda=10^{-7} . These initial conditions and timestamps were chosen to highlight noticeable orbital deviations induced by the cosmological constant. Beyond accelerating the decrease in binary separation r(\psi) , the cosmological constant also introduces a phase shift, contributing to enhanced precession of the periapsis. Lastly, we also estimate the conservative corrections to the orbital period P due to the cosmological constant. \displaystyle P=\frac{2\pi}{n}=2\pi\sqrt{\frac{p^{3}}{GM}}\left(1+\frac{\Lambda}{6}p^{3}\frac{c^{2}}{GM}\right) (3.45) where we have brought back the physical constants for increased readability. Considering an EMRI of mass \mu=10M_{\odot} orbiting a black hole of mass M=10^{6}M_{\odot} in a de Sitter universe with a cosmological constant \Lambda\sim 10^{-50} , we obtain the following estimate for the orbital period: \displaystyle P \displaystyle\sim \displaystyle 2\pi\times 10^{4}\mathrm{yrs}\left(\frac{p}{1\;\mathrm{pc}}\right)^{3/2}\left(\frac{10^{6}M_{\odot}}{M}\right)^{1/2}\left(1+10^{-18}\left(\frac{p}{1\;\mathrm{pc}}\right)^{3}\left(\frac{10^{6}M_{\odot}}{M}\right)\left(\frac{\Lambda}{10^{-50}\mathrm{km}^{-2}}\right)\right). At an orbital separation of 1 pc, the correction from the cosmological constant is negligible. However at much larger separations ( ~\mathcal{O}(10^{6}) pc), its effect on the orbital period may become noticeable. Nonetheless, when accounting for gravitational radiation reaction, this correction is further reduced, since the orbital period depends on the evolving orbital size. (a.1) p_{0}=50 , e_{0}=0.7 at t=3.1\times 10^{4} (a.2) p_{0}=50 , e_{0}=0.4 at t=2.6\times 10^{4} (b.1) p_{0}=70 , e_{0}=0.7 at t=7.3\times 10^{4} (b.2) p_{0}=70 , e_{0}=0.4 at t=6.3\times 10^{4} Figure 10 : Sample eccentric orbit shapes for various initial p for Schwarzschild orbits (blue, dashed) and for SdS orbits (red, solid) with (a) \lambda=10^{-7} and (b) \lambda=10^{-8} . The corresponding colored dots are the position at the given time. We show here sample orbital trajectories achievable by initial (p_{0},\,e_{0}) indicated below each figure. We also show the time stamps below each figure for the initial parameters to achieve the corresponding orbital shape. We see here that the deviation caused by the cosmological constant is more relevant for orbits of higher eccentricity. We also notice a lag in the phase to complete a radial cycle caused by the cosmological constant. Figure 11 : Sample circular orbit in Schwarzschild (blue-dashed) and SdS (red-solid) at \lambda=10^{-7} spacetime with initial separation p_{0}=50 at times t=(0,1.18\times 10^{5}) . At short timescales, the orbital deviations are negligible, but at long timescales, SdS orbits deviate noticeably from the Schwarzschild orbit, with the cosmological constant facilitating faster evolution, reaching the separatrix faster (black) or plunging the black hole earlier. 4 Gravitational waveforms In this section, we examine the gravitational waveforms generated by an orbit under adiabatic evolution, where the dominant governing equation is the balance of energy and angular momentum flux. Adiabatic evolution happens when changes in the background metric due to variations in the geodesics are negligible, allowing us to ignore perturbations to the spacetime. The energy and angular momentum flux influences only the orbital parameters, while the trajectory remains constrained to the background geodesics. Adiabatic approximation Adiabatic evolution is defined by the requirement that any relevant orbital parameter q=(p,e) evolves much more slowly than the radial orbital period P=2\pi/\Omega_{r} , such that \Delta q=|\dot{q}|P\ll q [ 30 ] and \Omega_{r} as the radial angular frequency. By analyzing the adiabatic evolution condition, we can effectively divide the full orbital parameter space into two distinct regions: • Regions of slow evolution , where the weak-field gravitational fluxes will be sufficient to describe the evolution of the orbit. • Regions of fast evolution , where a strong-field formalism for energy fluxes (such as black hole perturbation theory) is required, along with corrections to the full metric to account for changes in the background geodesics (self-consistent evolution with self-force effects). Adiabatic evolution may still hold in both regions, but as we will see later, additional conditions must be met, imposing further constraints on the orbital parameters. 4.1 Orbits near the separatrix Near the separatrix, orbits are already expected to be affected by the black hole so strongly that the orbital evolution is described fully by the nonlinearities in the Einstein field equation. This happens when the CO further closes its distance with the black hole continuously, increasing the rate of gravitational energy release. A full computation of the gravitational fluxes in the strong-field regime is beyond the scope of this paper and will be presented in a companion work, where black hole perturbation theory is applied to Schwarzschild - de Sitter spacetime. However, the adiabatic approximation can still be examined in this region to determine the additional conditions necessary for its validity. We derive the radial period P near the separatrix, p\sim p_{\text{sep}} (equation ( 2.27 )) to examine the time scale at which adiabatic evolution holds. Assuming that changes in angular frequency and orbital period remain within the adiabatic limit, we approximate the orbital period at the separatrix up to linear in \lambda and \Delta p\sim p-p_{\text{sep}} : \displaystyle P|_{\Delta p\sim p-p_{\text{sep}}\ll 1} \displaystyle\sim \displaystyle 12\sqrt{6}\pi+3\sqrt{6}\pi\,\Delta p+\lambda\left(1296\sqrt{6}\pi+972\sqrt{6}\pi\,\Delta p\right). (4.1) We assume here that orbits near the separatrix have already circularized, and we introduce \Delta p to describe deviations from the separatrix. The validity of the adiabatic approximation is then tested by examining the evolution of p under the prescribed weak-field fluxes: \displaystyle\frac{|\dot{p}|P}{p} \displaystyle\sim \displaystyle\frac{16}{45}\sqrt{\frac{2}{3}}\pi q-\frac{4}{27}\sqrt{\frac{2}{3}}\pi\Delta p\;q+\lambda\;q\left(\frac{56\sqrt{6}\pi}{5}\Delta p\;+\frac{672\sqrt{6}\pi}{5}\right) (4.2) where we brought back the mass ratio scaling q:=\mu/M from equations ( 3.16 ) and ( 3.17 ). This leads to a constraint on the mass ratio: \displaystyle\frac{\mu}{M} \displaystyle\sim \displaystyle\frac{45\sqrt{\frac{3}{2}}}{16\pi}+\frac{75\sqrt{\frac{3}{2}}\;}{64\pi}\Delta p-\lambda\left(\frac{25515\sqrt{\frac{3}{2}}}{8\pi}+\frac{93555\sqrt{\frac{3}{2}}}{32\pi}\Delta p\right). (4.3) We now assess whether the mass ratio constraint can remain within the physical range q\in(0,1) . In the absence of a cosmological constant (\lambda=0) , the contraint leads to unphysical scenarios, producing values outside the range of q for \Delta p , i.e., for p>p_{sep} . This invalidates the use of weak-field fluxes for adiabatic evolution in the strong-field region. Furthermore, inclusion of a cosmological constant does not help loosen the validity of this approximation; for the constraint to yield physically meaningful mass ratios, an unrealistically large cosmological constant (\lambda\sim 10^{-3}) well beyond the bound for stable orbits (\lambda\sim 10^{-4}) . Although the breakdown of our initial assumptions (such as the applicability of pN fluxes near the separatrix) suggests that this analysis is not strictly valid in the strong-field regime, this calculation serves as a preliminary attempt to explore the influence of the cosmological constant on the limits of adiabatic evolution. We now present preliminary results comparing the weak-field energy flux formula with the strong-field energy flux at infinity, obtained from the perturbation of spherically symmetric spacetimes due to an orbiting point mass. This comparison provides insight into the validity range of the weak-field formula. The strong-field energy flux is obtained using methods of Schwarzschild perturbation theory [ 58 , 57 , 45 ] , generalized to spherically symmetric spacetimes to accommodate the Schwarzschild-de Sitter spacetime. We briefly outline the method, with the complete procedure detailed in a companion paper. The goal is to numerically calculate the energy flux valid near the separatrix of the Schwarzschild-de Sitter spacetime: \langle\dot{E}^{\infty}_{\ell m}\rangle=\frac{1}{64}\frac{(l+2)!}{(l-2)!}\sum_{n}\omega^{2}_{mn}|C^{+}_{\ell mn}|^{2}. (4.4) Here \omega_{mn} is the angular frequency of the orbiting particle in terms of the (m,n) modes of its radial and azimuthal oscillations. The normalization coefficient C^{+}_{\ell mn} is obtained by numerically integrating the inhomogeneous Regge-Wheeler-Zerilli equation in the frequency domain with the tortoise coordinate defined by \frac{dr}{dr_{*}}=f^{-1} , \left(\frac{d^{2}}{dr_{*}^{2}}-V_{\ell}+\omega_{mn}^{2}\right)R_{\ell mn}(r)=Z_{\ell mn}(r) (4.5) where R_{\ell mn} satisfies the general solution of the form: R_{\ell mn}(r)=c_{\ell mn}^{+}(r)\hat{R}_{\ell mn}^{+}(r)+c_{\ell mn}^{-}(r)\hat{R}_{\ell mn}^{-}(r) (4.6) with the coefficients defined as: \displaystyle\begin{aligned} &c_{\ell mn}^{+}(r)\equiv\frac{1}{W_{\ell mn}}\int_{r_{\min}}^{r}dr^{\prime}\frac{\hat{R}_{\ell mn}^{-}\left(r^{\prime}\right)Z_{\ell mn}\left(r^{\prime}\right)}{f\left(r^{\prime}\right)},&C_{\ell mn}^{+}\equiv c_{\ell mn}^{+}\left(r_{\max}\right)\\ &c_{\ell mn}^{-}(r)\equiv\frac{1}{W_{\ell mn}}\int_{r}^{r_{\max}}dr^{\prime}\frac{\hat{R}_{\ell mn}^{+}\left(r^{\prime}\right)Z_{\ell mn}\left(r^{\prime}\right)}{f\left(r^{\prime}\right)},&C_{\ell mn}^{-}\equiv c_{\ell mn}^{-}\left(r_{\min}\right).\end{aligned} (4.7) The source term Z_{\ell mn}(r) is derived from the stress-energy tensor of a point particle, of mass \mu , in a generic orbital trajectory, x_{p}(\tau) : T^{\mu\nu}\left(x^{\alpha}\right)=\mu\int\frac{d\tau}{\sqrt{-g}}u^{\mu}(\tau)u^{\nu}(\tau)\delta^{4}\left[x-x_{p}(\tau)\right]. (4.8) The homogeneous solutions \hat{R}_{\ell mn}^{\pm}(r) are obtained by solving the homogeneous Regge-Wheeler-Zerilli equation with the following asymptotic ingoing and outgoing wave boundary conditions: \hat{R}_{\ell mn}^{\pm}(r_{*}\rightarrow\pm\infty)=\exp(\pm i\omega_{mn}r_{*}). (4.9) Although this method allows for a generic orbital configuration, we restrict the preliminary analysis to the circular limit and consider only the dominant (l,m) energy flux mode. This simplification is sufficient to provide a rough verification of the weak-field formula. Specifically, we solve for E^{\infty}_{22} using the Zerilli potential, modified to incorporate the cosmological constant (with the full derivation included in the companion paper): \displaystyle V_{\text{even}}= \displaystyle\frac{f}{r^{2}\tilde{\Lambda}^{2}}\left[2\tilde{\lambda}^{2}\left(\tilde{\lambda}+1+\frac{3M}{r}\right)+\frac{18M^{2}}{r^{2}}\left(\tilde{\lambda}+\frac{M}{r}\right)\right]-\frac{f}{r^{2}\tilde{\Lambda}^{2}}(6M^{2}\Lambda) (4.10) with constants (not related to \lambda or \Lambda ) \tilde{\Lambda}=\tilde{\lambda}+3M/r and \tilde{\lambda}=(l+2)(l-1)/2 . We note here however, that ( 4.4 ) is the expected radiation rate if the spacetime is asymptotically flat. This is a problem since we know that SdS is asymptotically de - Sitter, and the energy flux in terms of the fields should be modified by the cosmological constant. Luckily, many prescriptions on computing the energy flux have been put forward by analyzing de Sitter in Bondi frame, and computing the mass loss in terms of the news tensor, notably by [ 11 , 49 , 80 ] and more recently, de Sitter quadrupolar linearized waves by [ 48 , 21 , 29 ] . It was shown that the even parity, \ell=2 fields (at the Bondi frame of the asymptotically de Sitter null infinity) carry energy far from the source, r\rightarrow\infty , (a la’ Teukolsky waves [ 83 ] ) as \dot{E}^{\ell=2,m=\pm 2}=-\frac{3}{8\pi G}\Big|\partial_{u}^{3}A_{m}-\frac{4}{3}\Lambda\;\partial_{u}A_{m}\Big|^{2}-\Lambda\;|\partial_{u}^{2}A_{m}|^{2}+\mathcal{O}(\Lambda^{2}) (4.11) where A_{m}(u) ’s are even-parity solutions to the linearized Einstein field equations that reduces to the asymptotically flat limit at \Lambda\rightarrow 0 . Throughout our energy flux comparisons, we only use the asymptotically flat limit of the energy flux formula, since all succeeding terms from the \Lambda correction have subdominant contributions compared to the leading term. The energy flux computed from the solutions of the Regge-Wheeler-Zerilli equations is expected to be independent to a far-field (large- p ) approximation, as these equations are derived from perturbations of the SdS spacetime due to an orbiting test mass. Hence, the strong-field energy flux formula should remain valid even at large orbital separations. This allows us to assess the accuracy of the weak-field energy flux formula (equation ( 3 )) by evaluating its deviation from the strong-field energy flux (equation ( 4.4 )). In Figure 12 , we illustrate the deviation of the energy flux obtained from perturbation theory (equation ( 4.4 )) and that from a weak-field expansion (equation ( 3 )), 1-|\dot{E}_{22}^{\infty}/\dot{E}_{pN}| , for increasing binary separation, r_{p} , and cosmological constant, \lambda . Figure 12 : Deviation of the weak-field energy flux formula (equation ( 3 )) and the strong field energy flux ( l=2,m=2 mode) from SdS perturbation theory (equation ( 4.4 )) for a circular orbit of radius r_{p} . The validity of the weak-field energy flux formula is stronger for larger r_{p} and lower \lambda . For smaller r_{p} , the weak-field formula may still be valid if \lambda is allowed to be of lower value. Figure 13 : Deviation of the weak-field energy flux formula (equation ( 3 )) and the strong field energy flux ( l=2,m=2 mode, n_{max}=30 ) from SdS perturbation theory (equation ( 4.4 )) for generic eccentric orbits in terms of the cosmological constant, \lambda , and cosmological horizon, r_{c} . We also compare the energy flux of circular orbits to eccentric orbits, showing the deviations in Figure 13 . Our results indicate that for higher values of \lambda , significant deviations emerge at sufficiently large r_{p} , confirming that the weak-field quadrupolar energy flux (equation ( 3 )) remains valid only within the particle position satisfying the hierarchy (equation ( 2.33 )). Additionally, for smaller \lambda , the OSCO shifts farther from the black hole, leading to an increasingly larger region of validity for the weak-field fluxes. Overall, we find that higher \lambda values lead to a scaling error in eccentric orbits relative to their circular counterparts. However, as expected, for smaller \lambda , both circular and eccentric orbits exhibit similar deviations in the fluxes. 4.2 Orbits far away from the separatrix It has been shown that in regions far from the separatrix, and thus the black hole, applying adiabatic approximation on orbital evolution results to no strict conditions for the mass ratio of the binary [ 30 ] . We repeat the analysis here to determine whether the presence of a cosmological constant introduces new constraints. Similar to the derivation from previous subsection, the change in the semi-latus rectum in the large- p limit yields the following constraint on the mass ratio: \frac{\mu}{M}\ll\alpha(e)p^{5/2}-\lambda(\beta(e)p^{11/2}+\mathcal{O}(\lambda p^{7/2})) (4.12) where \alpha(e) and \beta(e) are positive functions of eccentricity for all values of e . This result confirms that the constraint is automatically satisfied for large p , implying that the validity of the weak-field fluxes for adiabatic evolution remains independent of the mass ratio. Hence, at large distances, the evolution of the orbital parameters is necessarily adiabatic. We note however that the presence of a cosmological constant \lambda strengthens the bound at which adiabaticity may be applied to eccentric orbits. This suggests that in the case of eccentric orbits, the strong-field region is effectively shifted towards much lower p . Thus, the weak-field fluxes will be sufficient for the balance formula governing the orbital trajectory and gravitational waveforms. We present below the gravitational waveforms, deriving the phase from the orbital mean motion and the strain magnitude from the binary mass quadrupole moment. The leading order relationship of the time derivatives of the mass quadrupole moment to the waveform strain will be sufficient to show the effect of the cosmological constant up to order \Lambda^{1/2} since the linearized radiation in asymptotically-de Sitter spacetime was shown to be [ 32 , 43 ] : \chi_{ij}\sim\partial_{t}^{2}Q_{ij}+\mathcal{O}\Bigg(\frac{\sqrt{\Lambda}}{D}\Bigg) (4.13) with D as the wave zone distance. The wave zone radiation was derived from the perturbation of de Sitter spacetime (with the cosmological constant related to H by H^{2}=\Lambda/3 ), g_{\mu\nu}=\gamma_{\mu\nu}+\epsilon h_{\mu\nu} (4.14) with the background line element, \gamma_{\mu\nu}dx^{\mu}dx^{\nu}=-dt^{2}+e^{2Ht}\delta_{ij}dx^{i}dx^{j} (4.15) which should be valid far from the source. The resulting linearization of the field equations has the following form, \eta^{\alpha\beta}\partial_{\alpha}\partial_{\beta}\chi_{\mu\nu}+\frac{2}{\eta^{2}}\partial_{0}\chi_{\mu\nu}-\frac{2}{\eta^{2}}\left(\delta_{\mu}^{0}\delta_{\nu}^{0}\eta^{\alpha\beta}\chi_{\alpha\beta}+\delta_{\mu}^{0}\chi_{0\nu}+\delta_{\nu}^{0}\chi_{0\mu}\right)=-16\pi T_{\mu\nu} (4.16) with \eta_{\mu\nu} as the components of the Minkowski metric, and with the radiation field \chi_{\mu\nu}=\exp(-2Ht)\bar{h}_{\mu\nu} (4.17) with \bar{h}_{\mu\nu} as the usual trace-reversed perturbation metric. The radiation field can then be expressed in terms of the source multipole moments, \displaystyle\chi_{ij}= \displaystyle\frac{2}{R}\left(e^{-Ht}+HR\right)\left[\ddot{Q}_{ij}^{(\rho)}-2H\dot{Q}_{ij}^{(\rho)}+H\dot{Q}_{ij}^{(p)}\right] (4.18) \displaystyle-2H\left[\ddot{Q}_{ij}^{(\rho)}-3H\dot{Q}_{ij}^{(\rho)}+H\dot{Q}_{ij}^{(p)}+2H^{2}Q_{ij}^{(\rho)}-H^{2}Q_{ij}^{(p)}\right]. with R as the radial distance of the source, R^{2}=x^{2}+y^{2}+z^{2} . Higher order terms in \Lambda would represent tail terms in the waveform due to backscattering [ 12 ] , which we will ignore here since their contribution to the amplitude and phase is small enough with respect to the leading order. At spatial infinity, the polarization modes of the gravitational wave signal h (in the transverse-traceless gauge), at a distance D from the source, will then be related to the mass quadrupole moment (we suppress the label \rho , Q^{(\rho)}_{ij}=Q_{ij} ), as it is in an asymptotically flat spacetime (up to leading order in \Lambda ) [ 56 , 43 ] by the following expression: \displaystyle h_{+} \displaystyle= \displaystyle\frac{1}{D}\left(\ddot{Q}_{11}-\ddot{Q}_{22}+2\sqrt{\frac{\Lambda}{3}}\left(\dot{Q}_{11}-\dot{Q}_{22}\right)\right)+\frac{\Lambda}{3}\left(\dot{Q}_{11}-\dot{Q}_{22}+\frac{2\sqrt{\Lambda}}{3}\left(Q_{11}-Q_{22}\right)\right), (4.19) \displaystyle h_{\times} \displaystyle= \displaystyle\frac{2}{D}\left(\ddot{Q}_{12}-2\sqrt{\frac{\Lambda}{3}}\dot{Q}_{12}\right)+\frac{\Lambda}{3}\left(\dot{Q}_{12}-2\sqrt{\frac{\Lambda}{3}}Q_{12}\right). (4.20) We assume that the binary components lie in a single plane, say the x-y plane, and the observer is aligned along the z- axis. This orientation corresponds to a face-on view of the binary. We further set the orbital plane such that its origin coincides with one of the foci of the elliptic orbit, which would be the position of the central black hole. The mass quadrupole moment of the binary, with reduced mass \mu , is then expressed into: Q_{ij}=\mu\,r^{2}(\psi)\left(\begin{array}[]{ccc}\cos^{2}(\phi)&\quad\sin(\phi)\cos(\phi)&\quad 0\\ \\ \sin(\phi)\cos(\phi)&\quad\sin^{2}(\phi)&\quad 0\\ \\ 0&\quad 0&\quad 0\end{array}\right). (4.21) Substituting in the solutions of the adiabatic osculating trajectory equations into the strain formulas ( 4.20 ), allows us to generate adiabatic waveforms plots valid in the weak-field region of the parameter space. It would also be sufficient to plot the polarization strains ( 4.20 ) up to the leading order in \Lambda (ignoring the second term) since we are already using the leading \Lambda -corrected solutions in the orbital parameters. We note here that under secular evolution [ 67 ] , the waveform phase would gain additional contribution from the conservative gravitational self-force, which we do not include in this paper. In Fig. 14 , we present a circular waveform under the assumption that the orbit begins with a small initial eccentricity and circularizes almost instantly, making \dot{e} negligible. We observe an phase advance, which can be attributed to the cosmological constant term in the mean motion. This early departure arises from modified angular frequency, with the energy and angular momentum being affected by the cosmological constant the most for nontrivial eccentricities. (a) (b) (c) Figure 14 : Adiabatic waveforms of (p_{0}=50) for Schwarzschild with \lambda=0 (dashed, blue) and with SdS with \lambda=10^{-7} (red). We show here the (a) plus polarization as well as the (b) cross polarization of the waveform in the adiabatic limit of an initially eccentric orbit with eccentricity e_{0}=0.7 . We show here the plus polarization (c) of the waveform in the adiabatic limit of an initially circular orbit. The main effect of the cosmological constant is the increase in amplitude and an initial delay in the wave phase. Lastly, we estimate the change both in the amplitude and the phase of the waveform due to the SdS gravitational flux in an EMRI orbit. We begin by considering the analytical expression for complex time-domain waveform [ 56 ] : \displaystyle h(t)=\mathcal{A}(t)\;\exp(i\Psi(t))=h_{+}-ih_{\times} (4.22) where the SdS change in amplitude arises from an \mathcal{O}(\lambda) deviation, \delta\mathcal{A}_{SdS} , in the amplitude \mathcal{A}=\mathcal{A}_{pN}(1+\delta\mathcal{A}_{SdS}). (4.23) Similarly, the phase correction, \delta\Psi_{SdS} , appears in \Psi=\Psi_{pN}(1+\delta\Psi_{SdS}). (4.24) The post-Newtonian amplitude and phase are related to the Keplerian frequency \omega_{N} as \mathcal{A}_{pN}=4\omega_{N}^{2/3}/D and \Psi_{pN}=2\omega_{N}t . Expressing the semi-latus rectum p in terms of \omega_{N} via \omega_{N}=p^{-3/2} and using the evolution equations (equations ( 3.16 ) and ( 3.17 )) we obtain first-order SdS corrections: \displaystyle\delta\mathcal{A}_{SdS} \displaystyle= \displaystyle-\lambda\;\omega_{N}^{-2}+\mathcal{O}(\lambda^{2}) (4.25) \displaystyle\delta\Psi_{SdS} \displaystyle= \displaystyle-\lambda\;\frac{\omega_{N}^{-2}}{2}+\mathcal{O}(\lambda^{2}). (4.26) These expressions indicate that the amplitude and phase shift are initially small but grow as the frequency increases. This suggests that both the amplitude and phase of the SdS waveform will eventually deviate significantly from their pN counterparts at an earlier stage in the inspiral. The Keplerian mean motion, \omega_{N} may be estimated as: \omega_{N}\sim 1.22\times 10^{-4}\mathrm{yr}^{-1}\left(\frac{1\;\mathrm{pc}}{p}\right)^{3/2}\left(\frac{M}{10^{6}M_{\odot}}\right)^{1/2}. (4.27) Since both amplitude and phase shifts scale similarly with \omega_{N} , we then obtain an estimate for \delta\mathcal{A}_{SdS} and \delta\Psi_{SdS} in a circular EMRI: \displaystyle\delta\mathcal{A}_{SdS}\sim\delta\Psi_{SdS}\approx 10^{-18}\left(\frac{p}{1\;\mathrm{pc}}\right)^{3}\left(\frac{M}{10^{6}M_{\odot}}\right)\left(\frac{\Lambda}{10^{-50}\mathrm{km}^{-2}}\right). (4.28) Using the previous inspiral rate estimate (equation ( 3.28 )), a circular inspiral over one year causes a decrease of p\approx 10^{-25} km, leading to an extremely small reduction in the amplitude of order \mathcal{O}(10^{-93}) . However, as the frequency increases, the amplitude and phase shift eventually surpass their pN counterparts at critical frequency, \omega_{crit}\sim\lambda^{-1/2} . Before reaching \omega_{crit} , the cosmological constant primarily reduces the amplitude of the waveform and delays the phase evolution. For initially eccentric orbits, the phase shift remains similar to the circular case, since we consider the waveform frequency to be related to the mean motion averaged over a full orbit. However, the amplitude shift is modified by the eccentricity: \mathcal{A}(e)D=\frac{2(e+2)\omega^{2/3}}{(e+1)^{3}}-\lambda\frac{4}{\omega^{4/3}(e+1)^{2}}+\mathcal{O}(\lambda^{2})+\mathcal{O}(\omega^{4/3}). (4.29) We see in this expression that more eccentric orbits produce a lower overall amplitude shift due to the cosmological constant compared to circular orbits. The eccentricity affects \omega_{crit} at most by a factor of order e , which increases for higher eccentricities. As a consequence, waveforms from highly eccentric SdS orbits take longer to surpass the amplitude and phase of their circular waveform counterparts. 5 Timescales for gravitational wave detection In this section, we apply the radiation reaction equations (equations ( 3.16 ) and ( 3.17 )) to estimate the inspiral timescales for gravitational wave detection. Detectors will be operating for only a limited amount of time and therefore can only detect transient gravitational wave signals from sources for a given range of lengthscale or timescale. For the space-based detector LISA, the operation time is only up to 4.5 years before closing down for improvements [ 6 , 28 ] . Signals can only enter the detector frequency range and can be detected at the “lifetime” of the detector, which can translate to sources up to a range of length or timescale. Sources that plunge less than the detector lifetime as well as those that cannot reach the frequency of the detector bandwidth will be irrelevant sources for detection. It will then be important to predict sources in a specific range of orbital parameters that can decay in a specific time. In this section, we obtain an analytical inspiral timescale formula in terms of orbital parameters that can describe the decay of a binary, similar to what was derived by [ 65 ] and extract what effect the cosmological constant may bring. For circular orbits, an exact analytic decay timescale with the effect of the cosmological constant can be derived similarly in [ 65 ] . Peters’ argument of minimal dependence of e(p) can also be used to extend the suitability of the timescale to eccentric orbits. We start by deriving the inspiral timescale as the integral of the inverse of the \dot{p} equation. Integrating up to near-coalescence, the timescale, \tau , becomes a function of the initial orbital parameters. Up to \mathcal{O}(\lambda) , the timescale with a cosmological constant is derived as a factor multiplied to Peters’ timescale \tau_{P} , \tau\sim\tau_{P}\left(1+\lambda\;\tau_{\lambda}\right) (5.1) with Peters timescale in (p,e) space is equal to \displaystyle\tau_{P} \displaystyle= \displaystyle\frac{5c^{5}p_{0}^{4}}{256G^{3}M^{3}qf(e_{0})} (5.2) \displaystyle= \displaystyle 1.84\times 10^{24}\text{yr}\left(\frac{p_{0}}{1\;\text{pc}}\right)^{4}\left(\frac{10^{6}\,M_{\odot}}{M}\right)^{3}\left(\frac{10^{-5}}{q}\right)\frac{1}{\;f(e_{0})} wherein we brought back the physical units and scalings for increased readability. The semi-latus rectum p here has units of length unlike the previous usage of the nondimensional p . The function f(e_{0}) is the enhancement function for including initially eccentric orbits, and is equal to: f(e)=\left(1-e^{2}\right)^{3/2}\left(1+7e^{2}/8\right). (5.3) The timescale function linear to \lambda is given by: \tau_{\lambda}=-\frac{19c^{2}p_{0}^{3}}{4GM}\frac{F(e_{0})}{f(e_{0})} (5.4) which is related to a new eccentricity enhancement function: \displaystyle F(e) \displaystyle= \displaystyle\frac{1}{228\left(1-e^{2}\right)^{3/4}}\Big(60\left(1-e^{2}\right)^{5/4}-26\left(1-e^{2}\right)-231\sqrt{1-e^{2}}+425\Big) (5.5) that evaluates to 1 for e\rightarrow 0 . For circular orbits, we have the timescale ratio \sigma(p,e=0,\lambda)=\tau_{P}/\tau has the form: \displaystyle\sigma(p_{0},e=0,\lambda) \displaystyle= \displaystyle 1-\lambda\frac{19c^{2}p_{0}^{3}}{4GM}+\mathcal{O}(\lambda^{2}) (5.6) \displaystyle= \displaystyle 1-\lambda\frac{19}{4}\left(\frac{p_{0}}{1.5\text{ km}}\right)^{3}\left(\frac{1M_{\odot}}{M}\right)^{3}. The main effect of the factor is to reduce the timescale due to the negative sign. Larger initial orbits will have an increased deviation to the Peters’ timescale due to the p_{0}^{3} scaling. As the orbiter starts initially near the BH, it will approximately follow the Peters’ timescale but if it starts initially farther, the timescale correction goes larger as shown in Figure 15 . Figure 15 : Ratio, \sigma(p_{0},\lambda) , of the Peters-Mathews inspiral timescale and the full SdS inspiral timescale for various cosmological constant values with its p dependence and (b) its e dependence. We see here that larger orbits gain an increased effect by the cosmological constant as the orbit will decay faster compared to the Peters’ decay time ( \lambda=0 ). For a stellar mass BH, with mass M=\mathcal{O}(10)M_{\odot} , circular orbits will decay in a timescale reduced by a factor of \sigma\approx 1-10^{-3}\lambda\;p_{0}^{3} , which will only become considerably large if the starting orbit has a size p_{0}\approx 10\lambda^{-1/3}\text{ km} . Near the maximum cosmological constant, say \lambda=10^{-6} , the starting orbit must be around p_{0}\sim\mathcal{O}(10^{3})\text{ km} to have maximum deviation. Beyond this estimate, the ratio will be negative and is thus unphysical, which means that orbits are unbound at these regime. For a SMBH, the correction is very small with an initial orbit having p_{0}\sim\mathcal{O}(10^{8})\text{ km} before being considerably large. This may be of astrophysical significance for scenarios of large-distance capture of inspiralling objects like EMRIs by SMBH [ 6 , 8 ] with gravitational wave induced evolution estimated to be the dominant dynamics starting at a relatively large distance. We note here that a \lambda=10^{-6} means either a relatively small \Lambda ( =10^{-18} ) compared to the measured \Lambda(\sim 10^{-50}) [ 2 ] or hypothetically large black hole (of order M=10^{22}\,M_{\odot} ) inside a small universe (of order \mathcal{O}(10^{9}) km). For the effect of the eccentricity, let us include the enhancement functions from equations ( 5.3 ) and ( 5.5 ). To obtain this, we only need to multiply the earlier factors to the \Lambda dependent term, \sigma(p_{0},e_{0},\lambda)\sim 1-\lambda\frac{19c^{2}p_{0}^{3}}{4GM}\frac{F(e)}{f(e)}. (5.7) The deviation, \tau(e)-\tau(e=0) , brought by the eccentricity can also be visualized by noticing that the correction factor to the timescale is just the eccentricity functions: \displaystyle\frac{\tau_{P}(e)}{\tau_{P}(e=0)} \displaystyle= \displaystyle f(e_{0})^{-1}, (5.8) \displaystyle\frac{\tau_{\lambda}(e)}{\tau_{\lambda}(e=0)} \displaystyle= \displaystyle F(e_{0})/f(e_{0}). (5.9) We show in Figure 16 the comparison of f(e_{0})^{-1} and F(e_{0})/f(e_{0}) . The function F(e_{0})/f(e_{0}) increases faster than f(e_{0}) for more eccentric orbits which means that the effect of eccentricity in modifying the timescale is strengthened by the the presence of the cosmological constant. We see in Figure 16 that an initially eccentric orbit, with e_{0}\sim 0.7 , the deviation to the inspiral timescale is strengthened to a factor of \mathcal{O}(100) . This can also mean that as the eccentricity continues to decrease, the value of eccentricity affects the SdS inspirals faster as compared to pN inspirals. Figure 16 : Eccentricity factors for the circular inspiral time. We see here that initially eccentric orbits are more affected by the cosmological constant which can result to a larger deviation in the decay time with Peters formula. Up to a factor of \sim 100 is to be included for e_{0}\sim 0.7 orbit decay time. However, this will still scale with the value of \lambda linearly. We show in Figure 17 the cosmological constant contours and isochrones in the phase space of orbital parameters (p,e) . If we relate the area in the phase space as the number of potential sources, Figure 17 (a) shows that higher values of cosmological constant increases the population of detectable sources at a fixed inspiral time. In Figure 17 (b) we see that the SdS inspiral time shows a different curve from the \lambda=0 inspiral time implying a shift in the orbital parameters of potential sources. This means that for a given time, we can detect much larger orbits due to the cosmological constant. The general observation here is that the population of sources will be limited to their initial eccentricity and size at capture and will increase due to higher eccentricity with larger orbits decaying faster from the effect of the cosmological constant. (a) Contours of equal \lambda for a fixed inspiral time. (b) Isochrones for a fixed \lambda Figure 17 : Analytical contours for a fixed \tau and fixed \lambda . In (a), we have an increasing \lambda for a fixed \tau=10^{5} with the filled region as the source population with higher decay timescale ( \tau>10^{5} ). The fixed timescale contours show that by increasing the cosmological constant, the total area covered by a particular timescale contour decreases which could mean that the source population will decrease most notably in e for a given cosmological constant and time. However in (b), we compare decreasing \tau for a fixed \lambda , and by comparing timescale contours for a specific cosmological constant, the source population increases in p as compared to the \lambda=0 case. We can now numerically integrate the orbital evolution equations to obtain a more realistic inspiral time measurement across the parameter space. We start by solving the plunge times for a range of initial parameters, then we deduce if those parameters are detectable within a given time. We numerically integrate the radiation reaction equations (equations ( 3.16 ) and ( 3.17 )), considering all possible initial parameters at a given range. Then we stop the integration if the parameters near the separatrix and we then collect the time per initial parameter for the integration to end. We consider the range p_{0}=(10,50),\;e_{0}=(0,1) with the cosmological constant near maximum, \lambda=10^{-7} . The isochrones with the evolving eccentricity is obtained numerically in Figure 18 . We show in Figure 18 that for a given range of time, the range of initial parameters is expanded due to the presence of a cosmological constant. Orbits of the same size will indeed decay faster with the help of the cosmological constant, but the higher the eccentricity, the factor of reduction in the timescale is decreased. Figure 18 : Numerical contour plots of the plunge times in a range of initial orbital parameters for Schwarzschild (\lambda=0) (left) and SdS (\lambda=10^{-7}) (right) spacetime. We note here that incorporating a self-consistent evolution in (p,e) decreases plunge times as the CO starts to inspiral further away from the BH. Given a fixed timescale, say at \tau=10^{5} , a larger region is covered in the initial parameter which can be expected as an increase in the population of potential gravitational waves sources. There is also notable decrease in population for highly eccentric orbits as also seen from the analytical contours previously, which means that highly eccentric orbits will be expected to decay slower as compared to higher values of cosmological constant. 6 Conclusions The cosmological constant introduces both conservative and dissipative effects on the dynamics of a binary system. Key findings of this paper include an accelerated rate of orbital decay, with shorter plunge and circularization times due to the cosmological constant. Additionally, we demonstrate that eccentric orbits are more strongly influenced by the cosmological constant, with this effect being independent of the binary’s separation. For the conservative effects, we analyze changes in the effective potential, where the cosmological constant introduces additional turning points for a given orbital parameter. Depending on its value, the cosmological constant alters the nature of bound and unbound orbits, with a critical threshold causing the complete absence of bound orbits. This leads to an analysis of the cosmological constant’s influence on the orbital parameters governing the existence of bound orbits. We derive the separatrix, represented as a 2D surface, a polynomial function of (p,e,\lambda) , and identify the critical value of the cosmological constant that allows for a single stable bound orbit. Although small in magnitude, the cosmological constant significantly affects the parameter space. We also find the presence of an outer separatrix, which further constrains the region of bound orbits; outside this boundary, the cosmological term dominates, rendering orbits unstable. For the dissipative effects, we use the prescription by Hoque and Aggarwal to compute the post-Newtonian gravitational fluxes for a binary system influenced by the cosmological constant. We show that the cosmological constant increases the rate of decay of the orbital parameters when translated into radiation reaction equations. Due to the increase in Keplerian binding energy and angular momentum, the cosmological constant introduces new dynamical effects in the radiation reaction. While the cosmological constant generally reduces the rate of orbital decay compared to standard post-Newtonian radiation reaction in flat spacetime, we find a nontrivial increase in eccentricity evolution at initially low eccentricities. We attribute this to the non-trivial coupling of the cosmological constant with eccentricity, where binaries with nonzero eccentricity are more sensitive to its effects. We have investigated the effect of the cosmological constant on physical quantities such as circularization, plunge times, and orbital trajectories. Deviations from post-Newtonian orbital shapes are observed for eccentric binaries. We also generate waveforms produced by binaries under radiation reaction, showing an increase in amplitude and phase advance due to the cosmological constant. The phase advance is linked to a decrease in orbital frequency, also caused by the cosmological constant. Finally, we calculate the detection timescales for these waveforms, demonstrating that the cosmological constant increases the number of detectable sources within a given time. Although the contribution from \Lambda is typically subleading compared to dominant post-Newtonian terms, we see a relative importance growing in certain regions of the parameter space (where the cosmological repulsion might be dominant over the radiation reaction dynamics) where it could become comparable in magnitude. Again, our results may be interpreted more generally as describing a quadratic term to the effective gravitational potential, \delta V\sim\Lambda r^{2} , with \Lambda serving as an effective parameter for environmental perturbations exhibiting the same scaling. The formalism developed here therefore extends beyond the purely cosmological context, providing a unified framework for analyzing systems subject to r^{2} corrections to the effective potential and clarifying their combined conservative and radiation-reaction–driven evolution. Lastly, we are currently extending our investigation of the cosmological constant’s effects on radiation reaction to the strong-field regime. Additionally, much work is needed to incorporate other environmental effects from realistic astrophysical scenarios. Future studies may explore cases with non-constant cosmic acceleration, such as in the McVittie black hole spacetime [ 62 ] , which we leave for subsequent research.