Finite element thermoelastic analysis of functionally graded spherical shells Authors: Anil Kumar, Gopal Kumar Gupta, Ajit Kumar Singh, Achche Lal Published: 2026 ABSTRACT AbstractGraphene, a remarkable material, has numerous efficient applications due to its enhanced chemical, mechanical, electrical, and thermal properties. By gradually spreading graphene nanoplatelets into a matrix, graphene functionally graded materials combine remarkable strength and low weight. Thermoelastic analysis is crucial for forecasting performance and guaranteeing dependability in sophisticated structural applications since it assesses how they react to combined mechanical stresses and temperature changes. In the present work, we consider two functionally graded materials: one made of Graphene and Copper, and the other of Titanium and Zirconium. The problem is formulated for a spherical shell under two generalized thermoelastic theories to examine the thermoelastic behavior, specifically for the functionally graded material of Graphene. We use the Laplace transform technique to eliminate the time dependency from the governing equations and the finite element method to address the non-linearity of the governing equations, which is generated by the dependency of thermal parameters on spatial coordinates. A detailed comparative study of the results obtained is discussed to focus on the functionally graded material and the necessity of the finite element method for generalized thermoelasticity. The numerical values of the field variables are shown through different graphs to provide a more compact analysis of generalized thermoelasticity for functionally graded materials.Keywords: Graphene, Spherical Shell, Generalized thermoelasticity, Finite Element method, Functionally graded materialSubject terms: Engineering, Materials science, Mathematics and computing, Physics Introduction There are two main problems with classical thermoelasticity, often known as the theory of uncoupled thermoelasticity. The first is that, contrary to what we observe in physical tests, it implies that the temperature of an elastic body is independent of its mechanical state. The second problem is that the parabolic heat conduction equation in the theory falsely implies that temperature changes can occur at an unlimited speed, which is not a real-world occurrence. The first of these issues was effectively resolved by Biot’s1 linked thermoelasticity theory, but because it also made use of a parabolic-type heat conduction equation, it continued to have problems with infinite speed. Researchers such as Lord and Shulman2, Green and Lindsay3, and Green Naghdi4–6 developed generalized thermoelasticity theories to produce a more realistic model. These new theories predicted a limiting speed for thermal waves in solids by including one or two relaxation intervals into the thermoelastic process. Either the energy equation and the Neuman–Duhamel connection were changed, or Fourier’s heat conduction equation was modified. The thermoelasticity with dual-phase-lag effects hypothesis, which included two phase-lag factors, was later presented by Tzou7 and Chandrasekharaiah8. Roychoudhuri9 then added three distinct phase-lags to the GN-III model, substantially altering it. One imortant property of GN-II model is that most of the recent developed thermoelasticity theories predict finite speeds, but these models are dissipative, whereas GN-II is conservative. The well-posedness of the solutions to these phase-lag thermoelasticity theories has been the subject of numerous investigations. Hetnarski and Ignaczak10 have shown the theoretical significance of these models, while Quintanilla and Racke11 have examined their stability. Quintanilla’s12,13 most recent reworking of the exact three-phase-lag model resulted in a heat conduction equation with a single delay term. To make the DPL equation easier to solve, most researchers employ a first or second-order Taylor expansion. Quintanilla contends that these approximations may result in unstable solutions, where the temperature increases indefinitely, that are absent from the ” exact” model. He concentrates on the mathematical well-posedness, demonstrating that the behavior of the ”exact” phase-lag model differs from that of its truncated approximations and is frequently more stable. In engineering, common structural shapes like spheres, cylinders, and disks are crucial design components, so it is crucial that they are made to endure the heavy stresses they will experience. In 1984, a spaceplane project in Japan gave rise to the idea of Functionally Graded Materials (FGMs). The goal was to create a new thermal barrier material that could withstand a temperature decrease of 1000 K across a thickness of less than 10 mm and a surface temperature of 2000 K. Since then, a lot of research has been done on FGMs, especially as thin films, and they are almost ready for commercial use. These materials are considered state-of-the-art, heat-resistant choices due to their remarkable thermal characteristics and ability to withstand extremely high thermal loads. Additionally, they have outstanding resistance to corrosion and erosion as well as great fracture toughness. The basic idea behind FGMs is to create a continuously changing composite material by mixing metal and ceramic. The spatial composition is designed to enable a gradual change in material qualities from one component to the next, such as thermal conductivity or Young’s modulus. As a result, the mechanical and thermal coefficients that describe the distribution of stress and temperature in generalized thermoelasticity equations are not constant; rather, they depend continuously on the material’s spatial position. Understanding how FGMs respond to different mechanical and thermal loads has garnered a lot of attention throughout the last ten years. However, because the governing equations for temperature and the related thermoelastic fields are frequently nonlinear, it is difficult to find exact solutions. The idea of laminated composites has been used by numerous researchers as a practical means of approximating solutions for FGM thermoelastic problems in order to get around this. This method has been applied in a number of studies to examine transient thermoelastic problems in FGMs. Other approaches have also been investigated. One-dimensional, steady-state, and transient thermal stresses in functionally graded circular hollow cylinders and spheres were investigated by Obata and Noda14,15 using the perturbation method. Assuming a linear composition of the constituent materials, Lutz and Zimmerman16,17 offered an analytical solution for stresses in thick, radially graded spheres under radial thermal loads. Additionally, they came up with a precise solution for a uniformly heated cylinder in which the thermal expansion coefficient and elastic moduli vary linearly with radius. Shabana and Noda’s18 work, which examined thermoelastoplastic stresses in a fully functionally graded plate under a thermal load, is an example of further research. In the meanwhile, Sankar and Tzeng19 calculated the axial stress distribution in a FGM beam by solving the thermoelastic equilibrium equations. When examining the thermal stress in FGM hollow circular cylinders, Liew et al.20 adopted a novel methodology. They created a revolutionary technique that adapts answers from smaller, homogeneous cylinders in place of the intricate equations of non-homogeneous thermoelasticity. In order to make inferences regarding the overall behavior of thermal stresses in these FGM cylinders, they subsequently examined a number of numerical instances. Darabseh and Bani Salameh21 employed a parabolic heat conduction model to study the transient thermal stresses in a FGM cylinder. They made the assumption that the radial direction of the material qualities varied exponentially. They employed an implicit finite difference numerical technique with various thermal boundary conditions to compute the temperature, radial displacement, and stress distributions. Kothari and Mukhopadhyay22 used a variety of thermoelasticity theories to study a functionally graded hollow disk. They used a volume-fraction rule to represent the material properties of the disk as a power-law function of the radial direction. The coupled equations that resulted from their research were solved using a Galerkin type finite element approach. The Earth’s crust contains a lot of carbon, and people can benefit greatly from its different forms, or allotropes. Graphene, a two-dimensional sheet of carbon with atoms organized in a honeycomb lattice by sp2 hybridization, is one prominent allotrope. Graphene is unique because of its exceptional qualities. At 2630 square meters per gram, it has a remarkably high specific surface area. With a high elastic modulus of 1.02 TPa and an inherent strength of 130 GPa, it is exceptionally strong mechanically. With a high charge carrier mobility of 15,000 /V s at room temperature and a thermal conductivity of 5000 W/mK, it is also a very good conductor of heat and electricity. We must examine graphene’s elastic properties in order to comprehend its special mechanical qualities. These consist of its mass density, specific heat, Poisson’s ratio, Lame’s constants, elastic modulus, and negative coefficient of thermal expansion. Graphene is frequently seen as a two-dimensional structure due to its extreme thinness: just one atom thick. This indicates that its in-plane tensile behavior: that is, how it stretches and pulls within that plane: is its most crucial mechanical characteristic. Its elastic modulus and Poisson’s ratio are the two most basic mechanical properties that characterize this behavior. Lee et al.23,24 determined that the elastic modulus of a single layer of graphene was roughly 1.02 GPa using an atomic force microscope and a nanoindentation technique and estimated Young’s modulus of graphene by Raman spectroscopy. Similarly, Zhang and Pan25 obtained a somewhat lower result of roughly 0.89 GPa for monolayer graphene using nanoindentation as well. Zhang and Pan went one step farther and demonstrated that the elastic modulus drops with the number of layers. For two layers, the value was 0.393 TPa; for three levels, it was 0.051 TPa; and for four or more layers, it was 0.027 TPa. They came to the conclusion that as the number of graphene sheets rises, the elastic modulus decreases. In a different work, Frank et al.26 used an atomic force microscope hanging over trenches to test the elastic modulus of graphene stacks (less than five layers). They stated that the figure was roughly 0.5 TPa. The Young’s modulus of graphene is strong at ambient temperature but decreases as the temperature rises, according to Georgantzinos et al.27 curiously, a linear growth is seen at absolute zero. Additionally, an ongoing experiment was used to validate the study’s conclusions. Under normal circumstances, pristine graphene exhibits a positive Poisson’s ratio, according to early numerical simulations by Zakharchenko et al.28. However, their study also revealed that when the temperature exceeds about 1700 K, this value turns negative. Los et al.29 corroborated this conclusion by stating that the experimental system size is responsible for the Poisson’s ratio, which can vary between 0.15 and 0.3. This range is in line with Gui et al.30’s numerical simulation results, which showed a value of about 0.17. There is a contradiction in the scientific literature, as several other research31–36 have claimed a negative Poisson’s ratio for graphene in spite of these findings. Anisotropic graphene will have important theoretical investigations in the thermo-mechanical and mechanical domains in recent years reported by scientists such as Sfyris et al.37–40, Favata et al.41, and Delfani et al.42. The theoretical and practical analysis of thin graphene sheets on substrates is a particularly challenging subject. Several mathematical models have been presented to better understand the behavior of monolayer graphene. A sixth-order structural tensor can be used to characterize the zigzag and armchair orientations of graphene, which is treated as an anisotropic hexagonal 2-lattice in one such model. Researchers were able to obtain closed-form solutions for the nonlinear theory by using a finite elasticity model. By developing models for freestanding monolayer graphene as a 2D 2-lattice, further work43–45 built upon this. According to these theories, the shift vector and the strain and curvature tensors have a quadratically dependent effect on the material’s characteristics. Graphene and Copper can be fabricated using powder metallurgy, where layers with varying concentrations of Graphene are combined and sintered. This approach allows for a gradient in properties, such as hardness, across the material. FGMs of graphene and copper have potential applications in areas requiring materials with tailored mechanical and thermal properties, such as in aerospace and other high-performance applications. The Finite Element Method (FEM) is more than simply a computational tool in the context of Functionally Graded Materials; it is the link that turns the hypothetical ”super-material” into a manufacturable reality. The material must typically be uniform in order to use traditional analytical solutions. But with FGMs, the density ϱ (r), thermal conductivity K(r), Young’s modulus E(r) and other elastic parameters vary at each point along the space variables. Abbas et. al46. has used the FEM to address the plane waves in generalized thermo-microstretch elastic solid with thermal relaxation. Further, Othman et. al47–49. have studied the effect of rotation on plane waves in generalized thermomicrostretch elastic solid, fundamental solution of generalized thermo-viscoelasticity and generalized magneto- thermo-microstretch analysis in elastic solids, respectively using FEM. Laplace transform method is also applied simultaneously here to tackle the dependency of the problem on time. The combined results are obtained here using Laplace and FEM technique. Othman et. al50. have adopted an analytical technique of the Eigen value expansion method to analyze on plane waves through magneto-thermoelastic microstretch rotating medium with temperature dependent elastic properties. A detailed analysis of the effects of the phase-lags, the angular frequency of thermal vibration and the variability thermal conductivity parameter on the displacement, incremental temperature, and the thermal stresses are presented by Othman and Abouelregal in51. Abouelregal152had examine thermomechanical interactions within a functionally graded unbounded non-homogeneous solid with a spherical hole in a unified way using the Moore–Gibson–Thompson thermoelasticity (MGTE) model.. Thermoelastic analysis of functionally graded materials for fractional heat transfer models and MGT has been carried out by Abouelregal et. al53–56.. Alhashash and Abouelregal57 have presented an analytical framework to investigate the photothermal response of functionally graded semiconductor materials under laser pulse excitation, utilizing a nonlocal thermoelastic model. In the present work, an investigation of thermoelastic behavior on a functionally graded hollow sphere constituted with graphene and Copper under an exact heat conduction model proposed by Quintanilla et. al12,13. together with another generalized thermoelastic models has carried out. The inner boundary of the disk is assumed to be under the thermal shock and stress free. The other boundary is taken insulated with temperature and fixed with some rigid support. We uses the Laplace transform technique to remove the dependency of governing equations on time and finite element method to tackle the non-linearity of governing equations generated due to dependency of thermal parameters on spacial Coordinate. A detailed study on the obtained result is discussed to focus on the functionally graded material and the necessity of the finite element method to the generalized thermoelasticity. The numerical values of the field variables has shown by the different graphs to give the more compact analysis of the generalized thermoelasticity for functionally graded materials. The current study highlights the following points. Unlike classical models where the heat flux is proportional to the temperature gradient, GN-II model makes the heat flux proportional to the thermal displacement gradient. It predicts that thermal waves travel at a finite speed (undamped thermal waves). This is critical for modeling materials that operate at cryogenic temperatures or under extremely fast laser-pulse heating, where the ”lag” in heat response significantly affects structural integrity. Quintanilla’s models account for the micro-structural delay between the application of a heat gradient and the actual movement of heat carriers (electrons or phonons). It introduces two distinct time lags: The time, it takes for heat to start moving and the time, it takes for the material’s lattice to respond. This allows for a more ”memory-dependent” simulation of materials, where the current state of stress depends on the history of thermal loading. Although copper has a high coefficient of thermal expansion and is heavy, it is the gold standard for conductivity. One can make a material that can withstand conditions where conventional copper components would distort or melt by grading it with graphene. Several important real-world implications can be simulated using solution with the Finite Element Method for this particular FGM. Massive heat fluxes are applied to materials in nuclear fusion environments (such as the divertor plates in ITER) while one side is cooled by water. The ”mismatch” in thermal expansion is a common cause of failure for standard copper-to-steel couplings. Construction of the problem We have considered an isotropic medium in the form of FGM spherical shells (see Fig. 1): one () made of Copper and Graphene, and the other () of Titanium and Zirconium.. The effective properties of these FGMs are determined by the properties of their constituent materials. Using a simple rule of mixtures, the effective property of the (Copper and Graphene) can be defined by the following relation: Similarly, the effective property of the (Titanium and Zirconium) can be defined by In this equation: is the volume fraction of Graphene, where . is the volume fraction of Graphene, where . , , and represent the properties of Graphene, Copper, Titanium and Zirconium, respectively. , , and represent the properties of Graphene, Copper, Titanium and Zirconium, respectively. is the effective property of the FGM. is the effective property of the FGM. and are the inner and outer radius of the spherical shell. and are the inner and outer radius of the spherical shell. Design of Spherical Shell and it’s Cross Section. Basic governing equations The basic governing equations in the context of the different generalized thermoelastic theories incorporated form for an isotropic medium in absence of heat source is given by the following relations. Equation of motion The equation of motion for the theory of thermoelasticity by Newton’s second law is given by Stress-strain-temperature relation The stress-strain relation for the theory is given as, Strain displacement relation The relation in strain and displacement is given by From equations (3) and (4) we get Law of heat conduction The unified equation of generalized heat conduction is given as, where the meaning of the the symbols/notations and constants are given in the Table 1. Table 1Symbols/notations and it’s meaning.Symbols/ConstantsMeaningComponents of displacement vectorComponents of elastic strain tensorDilation tensorComponents of stress tensorΘTemperature variationtTimeReference temperatureElastics Lame’s constantsϱDensity of the materialKThermal conductivityConductivity rate of materialCoefficient of linear thermal expansionSpecific heat at constant strainThermal diffusivityRelaxation/delay parameterKronecker delta Symbols/notations and it’s meaning. From the equation (7), the heat conduction equation for Green Naghdi thermoelasticity theory of type-II (GN-II Model) and Thermoelasticity theory given by Quintanilla (Quintanilla Model) can be extracted by assuming the values of parameter constants as follows (i):GN-II Model: z=2,and . GN-II Model: z=2,and . (ii):Quintanilla Model: z=1, and . Quintanilla Model: z=1, and . In the context of current problem, we take an non-homogeneous isotropic elastic functionally graded hollow sphere at rest and reference temperature . Thus, in the present problem, the elastic parameters λ, μ , γ , ϱ, K and are considered as functions of space variable, r. With the cavity’s center serving as the origin, we consider the spherical polar coordinates (r,ϴ,Φ). For the problem addressed in this manuscript is taken as spherically symmetric problem, the only non-zero component of the displacement is the radial displacement component u. The two non-zero strain components are Also, due to axis-symmetric of the hollow sphere, equations (3),(4), and (6)-(7) reduces to the following forms Further, we assume material composition in the functionally graded shell along the radial direction in such a manner that the inner surface and outer surface of the shell are made of ceramic and metal, respectively and the gradual change in composition of the constituent materials in the radial direction is assumed to vary in the form of the law, where r is the radial coordinate and n is the non-homogeneity index that govern the distribution of the constituent materials through the shell (as shown in Fig. 2). Here, the non-homogeneity index, n may be varied to obtain different profiles for the distribution of constituent materials. The present composition implies that the effective properties of shell are followed by Copper (Titanium) in case when n→ 0, and they are followed by Graphene (Zirconium), when . Volume fraction for FGM. Now, to reduce the number of constants and for stability of numerical solution, we introduce the following non-dimensional parameters and notations , , , , , , where is a dimensionless characteristic temperature and subscript m stands for material. In the paper, we have taken Copper and Titanium for () and (), respectively. Using the above parameters and notations and dropping the primes for convenience, equations (10)-(13) take the following forms: where and Boundary conditions We assume that the inner surface of the spherical shell is assumed stress free but exposed to a thermal shock while the outer surface is rigidly fixed and thermally insulated. Thus, the mechanical and thermal boundary conditions at the inner and outer surfaces of the shell are defined as Solution using Galerkin finite element method Applying the Laplace transform to equations (14)-(18), the equations transform to the following forms Now, we employ finite element method to solve the coupled equations (21) and (22). For this, the geometry of the shell is divided into M discrete elements of equal length, h along the radius of the shell. Considering the base element (), l=1,2,3,...M, the displacement and temperature fields over the base element are approximated as where the denotes the shape function that approximates the displacement and temperature fields in the base element and d is the number of nodes in the base element. and are therefore the nodal values of displacement and temperature, respectively. The arrangement of elements are shown in Fig. 3. Graphical representation of elements and nodes of base element for finite element method. Employing the approximated fields for the displacement and temperature in the base element () and applying the Galerkin finite element method over the volume of the base element , Eqns. (21) and (22) yield where i=1,2,...,d. Now, the weak formulation is applied to the terms involving the derivatives of the second order to reduce their orders and to extract mixed boundary conditions with respect to the radial variable. Using the local coordinates , where is the radius of the node of the base element in radial direction and using , Eqs. (25) and (26) reduce to the following equations: Here h is the length of the base element () along the radius. Using Eqn. (24) into Eqns. (27) and (28), we can form the finite element formulation in the Laplace transform domain as where the sub-matrices , , , , of order (d× d) and , , of order (d× 1), defined for the base element () are obtained as follows: where, i,j=1,2,3,...,d. After assembling the element matrices given in the Eqn. (29) for all the elements, we get a system of dM+d number of linear algebraic equations in unknown nodal values of displacement and temperature. While assembling, we interchange the rows and columns of element matrix so that the terms in the right hand sides of the equations (27) and (28) canceled out by each other between any two adjacent elements, except the first node of the first element and last node of the last element. Further in-canceled terms of the right side of Egns. (27) and (28) are known and related to the inner and outer boundary conditions as given by the Egn. (23). After using Eqns. (18) and (23) we get, Further, using Eqn. (23), we obtain Therefore,the sub-matrices of the global force matrix are given by Since in obtained system of algebraic equation, and are known to us. Therefore, two rows and two columns intersecting the nodal values and are deleted from the global stiffness matrix. At the same time in global force matrix , sub-matrices will turn accordingly. Therefore, all the unknown nodal values of displacement and temperature in (r, s) domain can be obtained by solving the linear system of algebraic equations given by Eq.(29). To get the final solution in space-time domain, we employ the numerical method of Laplace inversion proposed by Zakian et. al58 for the better accuracy since this method uses the complex nodes and weights for numerical integration. The outline of the method has been provided in the appendix section. We apply the second order finite difference scheme to get the solution for the stress components. Numerical results and discussion In the previous section, we formulated finite element modeling of the present problem and obtained a system of linear algebraic equations given by Eqn.(29). Now, in this section we carry out our computational work by using MATLAB programming and obtain numerical solution of the present problem.We aim to analyze the effect of non-homogeneity of the material (non-homogeneity index n), time and relaxation parameter, τ on distributions of displacement, temperature and stresses inside the two prescribed spherical shell of two FGMs. The spherical shell of two FGMs are supposed to be initially at reference temperature, . The inner radius () and outer radius () are taken of the shell for both and , whose physical properties are taken in SI units as in the Table 259–61, Table 2Value of Thermoelastic Constants.Thermoelastic constants for each materialMaterial/PropertyGraphene1020−0.33340007002267Copper102.980.334386383.18954Titanium66.20.32118.1808.34410Zirconium117.00.3332.036615.65600 Value of Thermoelastic Constants. The rate of thermal conductivity is taken as, . The value of phase-lag parameters are taken as . The formulation of the present set up is made for the general problem. However, in our computation, we use the linear Lagrangian polynomials as the shape functions in the base element. We break the shape of the spherical shell into M=100 elements along the radius of the shell. Each element, is divided into 2 nodes. The value of u and Θ in Laplace transform domain, (r, s) are obtained by solving the system of Eqns. given by (29). The stresses in the (r, s) domain are computed by employing second order finite difference scheme from Eqns. (19) and (20) using the computed values of displacement and temperature of (r, s) domain. Finally, we employ the Zakian technique58 for the numerical inversion of Laplace transform to obtain the desired field variables in space-time (r, t) domain. The obtained results for displacement (u), temperature (Θ ), radial stress and tangential stress for two FGMs ( and ) under GN-II and Quintanilla model are shown by different 2D plots. The representation of displacement, temperature, radial stress and tangential stress are displayed in the Figs. 4, 5, 6, 7, 8, 9, 10, 11 respectively for the time, t=0.2, 0.3 for n=1,2,3 to show the effect of non-homogeneity index on the distributions of different physical fields predicted by two different models for two FGMs. (a,b) Variation of displacement, u vs. radial distance r at time, t=0.2 and 0.3 for fixed n=1. (a,b) Variation of displacement, u vs. radial distance r at time, 0.2 and 0.3 for GN-II Model (c,d) Variation of displacement, u vs. radial distance r at time, 0.2 and 0.3 for Quintanilla Model. (a,b) Variation of temperature, Θ vs. radial distance r at time, t=0.2 and 0.3 for fixed n=1. (a,b) Variation of temperature, Θ vs. radial distance r at time, 0.2 and 0.3 for GN-II Model (c,d). Variation of temperature, Θ vs. radial distance r at time, 0.2 and 0.3 for Quintanilla Model. (a,b) Variation of radial stress, vs. radial distance r at time, t=0.2 and 0.3 for fixed n=1. (a,b) Variation of radial stress, vs. radial distance r at time, 0.2 and 0.3 for GN-II Model. (c,d) Variation of radial stress, vs. radial distance r at time, 0.2 and 0.3 for Quintanilla Model. (a,b) Variation of tangential stress, vs. radial distance r at time, t=0.2 and 0.3 for fixed n=1. Variation of tangential stress, vs. radial distance r at time, 0.2 and 0.3 for GN-II Model. (c,d) Variation of tangential stress, vs. radial distance r at time, 0.2 and 0.3 for Quintanilla Model. (i).Displacement Displacement Figure 4(a, b) illustrate the changes in deformation u for FGM-1 and FGM-2 under the Quintanilla and GN-II models with a fixed value of homogeneity index, n at time t=0.2 and 0.3. For FGM-1, it is shown that u first declines before beginning to rise and going to zero towards outer boundary of the spherical shell; for FGM-2, this pattern is entirely the opposite. This reaction is explained by the fact that FGM-1 contains Graphene as an ingredient that behaves differently than most materials. Compared to FGM-2, the degree of deformation for FGM-1 is smaller. The greater hardness of the Graphene material found in FGM-1 makes this difference obvious. Additionally, as seen in Figure 4(a, b), it is noted that the displacement’s effective area and magnitude both grow over time. In the models’ context, the Quintanilla model exhibits greater moduli deformation than the GN-II model for FGM-2. On the other hand, the middle region shows a reversal tendency in the case of FGM-1. When the fluctuation of n is taken into account, Figure 5(a, b) illustrate how displacement u varies under the GN-II model for t=0.2 and 0.3, whereas Figure 5(c, d) give the similar description for the Quintanilla model. While there is almost no change for FGM-2 because of the smaller impact of the non-homogeneity index n, Figures 5(a, b) show a noticeable variation in deformation for FGM-1, demonstrating the unique behavior of Graphene in comparison to other materials. The variations in deformation u for both FGMs, however, become increasingly noticeable as time t increases. The Quintanilla model shows a result (shown in Figures 5(c, d)) that is comparable to that shown in Figure 5(a, b) for the GN-II model. A notable effect of n is also shown for FGM-2 as time increases. (ii).Temperature Temperature The temperature variation, Θ, for two FGMs under two prescribed models at t=0.2 and 0.3 for a fixed non-homogeneity index n is displayed in Figures 6(a, b). It is noted that the outcomes at the spherical shell’s boundary satisfy the given boundary data. Under both models, the temperature tends to zero more quickly for FGM-2 than for FGM-1. This suggests that the presence of Graphene in FGM-1 results in improved heat conduction properties. Another noteworthy finding is that, contrary to what is typically observed, the GN-II model does not produce negative temperature readings for functionally graded materials. Additionally, it is clear from the figures that the temperature Θ is higher for FGM-2 than for FGM-1 under both models near the inner boundary of the shell, and this trend reverses toward the outer boundary of the shell as the proportion of the other constituent material in the FGMs gradually increases. The impact of the non-homogeneity index n on temperature at t=0.2 and 0.3 for two FGMs under the GN-II model is shown in Figure 7(a, b). With increasing n, these figures show clear and notable changes in temperature variation for FGM-1 and FGM-2, respectively. Moreover, it is evident how the gradient of FGMs and the fluctuation of n together affect Θ. The temperature against radial distance r for two FGMs under the Quintanilla model at t=0.2 and 0.3 for various values of n is displayed in Figures 7(c, d). With the exception of the temperature change appearing to be exponential, these graphs show a similar pattern of variation for this model also as compared to GN-II. (iii).Radial Stress Radial Stress The radial stress versus r is shown under various thermoelastic models at times t=0.2 and 0.3 for two FGMs with a fixed non-homogeneity index n in Figure 8(a, b). It is clear that the boundary condition for the radial stress at the inner boundary of the shell matches the given data. Additionally, the figures show different patterns for FGM-1 and FGM-2, indicating increasing and decreasing trends, respectively, meaning that the stress at the inner boundary of the shell is tensile for FGM-1 and compressive for FGM-2. The GN-II model has higher values for the radial stress than the Quintanilla model, and the effective region of the stress increases as time increases for both models and materials. The radial stress for the GN-II and Quintanilla models is shown in Figure 9(a, b) and (c, d), respectively, at times t=0.2 and 0.3 for various values of n. It is observed that the stress behaves almost similarly for n=1 and 2 in relation to FGM-2, but the results for n=3 differ significantly. In the case of FGM-1, however, there are notable differences in the results for n=1,2, and 3. The stress approaches zero toward the outer boundary of the spherical shell, and the figures show that the effective region of the GN-II model is smaller than that of the Quintanilla model (Table 3). Table 3Summary of result comparison of for two FGMs.Field Variables/FGMsFGM-1FGM-2Displacementu first declines before beginning to rise. There is much difference in displacement for two different times. The displacement is much affected by homogeneity index n.u first rise before beginning to decline. There is less difference in u with respect to two different times. The displacement here is affected less by n.TemperatureThe temperature tends to zero much slowly for FGM-1. Furthermore, temperature tends quickly zero for GN-II model as compared to Quintanilla model.The temperature tends to zero much quickly for FGM-2. In this case also temperature tends quickly zero for GN-II model.Radial StressThe radial stress at the inner boundary of the shell is tensile. As the time increases, the difference in radial stress decreases. There is significant difference as n varies.The radial stress at the inner boundary of the shell is compressible. As the time increases, the difference in radial stress is still prominent. There is a slight difference as n varies.Circumferential StressThe stress at the inner boundary of the shell is also tensile. Further, variation is much prominent in for each value of non-homogeneity index n.This stress at the inner boundary of the shell is compressible. When there is linear and quadratic (i.e. n=1 and 2) variation in ingredients of FGM-2, it does not affect much the circumferential stress Summary of result comparison of for two FGMs. (iv). Circumferential Stress Circumferential Stress The circumferential stress for two FGM materials under two thermoelastic models at times t=0.2 and 0.3 for a given value of n is shown in Figure 10(a, b). For FGM-1, this stress is clearly tensile, and for FGM-2, it is compressive. The exceptional thermal characteristics of the Graphene material are the cause of this phenomenon. For both FGMs, the GN-II model’s circumferential stress is larger close to the inner edge of the spherical shell, but the Quintanilla model’s circumferential stress increases after a certain radial distance r. In the meantime, in the direction of the shell’s outer edge. Additionally, the figures demonstrate that displays a local minimum close to the shell’s inner boundary. Figure 11(a, b, c, d) shows that the linear and quadratic (i.e. n=1 and 2) variation in ingredients of FGM-2, does not affect much the circumferential stress for both the models while cubic variation (n=3) yields quit different result. Further, variation is much prominent in for each value of non-homogeneity index n. The effective area of for GN-II is much smaller as compared to the Quintanilla model. The geometric domain (such as cylinders, disks, or plates)14–21, and the particular graduation law used (power-law vs. exponential) have a minor impact on the quantitative comparison of contemporary data with previous baseline Functionally Graded Material (FGM) thermoelastic investigations However, comparing contemporary, coupled, or higher-order theories to traditional uncoupled thermoelastic baselines reveals a number of persistent quantitative tendencies46–55,57–60. Conclusions In this study, a thermoelastic analysis has been performed on two types of FGMs: one made of Graphene and Copper, and the other consisting of Titanium and Zirconium. In conventional shells composed of two distinct layers (such as a ceramic coating over metal), the layers frequently delaminate at the joint. Because FGMs change smoothly, there is no ”joint.” This prevents stress accumulation and increases the shell’s strength in extremely hot conditions. In conventional shells composed of two distinct layers (such as a ceramic coating over metal), the layers frequently delaminate at the joint. Because FGMs change smoothly, there is no ”joint.” This prevents stress accumulation and increases the shell’s strength in extremely hot conditions. Most FGM shells are designed with Ceramic on the outside (to handle high heat) and Metal on the inner side (for structural strength). This setup allows the shell to withstand thousands of degrees on the outside while keeping the inside cool enough to protect equipment or passengers. Most FGM shells are designed with Ceramic on the outside (to handle high heat) and Metal on the inner side (for structural strength). This setup allows the shell to withstand thousands of degrees on the outside while keeping the inside cool enough to protect equipment or passengers. A typical shell expands unevenly and may break or lose its shape when heated. Engineers can ”tune” the shell to expand predictably and avoid cracks by altering the Power-Law Index, which is the ratio of metal to ceramic. A typical shell expands unevenly and may break or lose its shape when heated. Engineers can ”tune” the shell to expand predictably and avoid cracks by altering the Power-Law Index, which is the ratio of metal to ceramic. If the FGM shell is supplemented with graphene. Stiffness causes the shell to become significantly lighter yet much stiffer. Before ”imploding” or buckling, it can withstand significantly more external pressure. For shells used in fast-moving missiles or spacecraft, graphene’s ability to reduce vibrations is essential. If the FGM shell is supplemented with graphene. Stiffness causes the shell to become significantly lighter yet much stiffer. Before ”imploding” or buckling, it can withstand significantly more external pressure. For shells used in fast-moving missiles or spacecraft, graphene’s ability to reduce vibrations is essential. The best defense against thermal buckling for a spherical shell is to concentrate the strongest material (such as graphene) close to the inner and outer surfaces (FG-X pattern). The best defense against thermal buckling for a spherical shell is to concentrate the strongest material (such as graphene) close to the inner and outer surfaces (FG-X pattern). Appendix: Zakian method of laplace inversion The Laplace transform, of the function f(t) is defined by where we assume, Re(s)>α and f(t) is integrable and of exponential order α, then it may be inverted by using the economical and efficient formula. iRe()Im()Re()Im()1-3.690208210 E+041.969904257 E+051.283767675 E+011.666063445 E+0026.127702524 E+04-9.540862551 E+0421.222613209 E+0l5.012718792 E+003-2.891656288 E+041.816918531 E+0431.093430308 E+018.409673116 E+0044.655361138 E+03-1.901528642 E+0048.776434715 E+00l.l92185389 E+0l5-l.l87414011 E+02-1.413036911 E+0255.225453361 E+001.572952905 E+01 where is the approximation to the function f(t). The constants and are real or complex numbers. The approximation can be made with arbitrary accuracy since as . For the most applications, sufficient accuracy can be obtained with N=5 and the constants are listed in the Table 1. The arithmetic involved in computing the inversion formula given by A-2 with complex constants and for N=5 can be modified as follows: Thus, the Eqn. A-3 can be used to compute the inverse of Laplace transform of the function f(t). Author contributions Anil Kumar conceptualized and analyze the model. Gopal Kumar Gupta have completed the modelling and analytical part. Ajit Kumar Singh has contributed in numerical and verification of the model. Achche Lal has completed the literature review and drafting of the paper. All authors have contributed equally to this work. Funding Open access funding provided by Symbiosis International (Deemed University). Data availability The datasets used and/or analysed during the current study available from the corresponding author on reasonable request. Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Anil Kumar, Email: anilkumar@allduniv.ac.in. Gopal Kumar Gupta, Email: gopalgupta.iitbhu90@gmail.com. References Associated Data Data Availability Statement The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.