Graphene-incorporated aluminum with enhanced thermal and mechanical properties for solar heat collectors

Metal matrix composites (MMCs) have a wide range of technological applications owing primarily to their excellent mechanical and thermal properties. Rapidly emerging practical applications such as miniaturized micro-electronic devices, thermoelectric materials, and high-performance mechanical structures can only be made commercially viable with the help of these MMCs.^1–4 For these metal matrix composites, to deliver specific desired outputs, careful observation and tuning of several critical parameters are essential. Factors such as weight percentage, volume fraction, size, shape, and orientation can have a significant impact on the nature of the final composite sample.^5,6 Another important factor is the formation of interfaces between the dispersed phase and the metal matrix, which can impart intriguing characteristics to the final composite structure, yielding unprecedented modifications/enhancements to the physical and chemical properties.^7,8 The arrangement of the dispersed phase inside the bulk metal matrix could either be isotropic or anisotropic, both of which can be controlled to achieve desired results. Of all the possible metal matrix composites, aluminum matrix composites (AMCs) have recently captured significant interest in commercial applications such as aerospace, automobile, electronic/electrical equipment, and a large number of building materials.^9–12 The advantage of aluminum over other prominent metals such as nickel, iron, magnesium, and chromium is that it has the highest abundancy (among all the metals) in the earth’s crust. Besides this, there are several unique features of aluminum, e.g., low density (making it extremely lightweight), corrosion resistance (through the passivation phenomenon), high thermal conductivity (second only to copper), malleability, and ductility. However, the poor tribological property (low wear resistance and average mechanical strength) is one of the prominent drawbacks of aluminum and aluminum based alloys.¹³ Thus, many reinforcement methods have been studied, where the aluminum bulk has been mixed with trace amounts of Al₂O₃, SiC, BN, TiB₂, B₄C, etc., to form AMCs with improved mechanical and thermal properties.^14–18 Only recently, graphene and carbon nanotubes (CNTs) have been discovered as excellent reinforcement materials for AMCs with enhanced electrical, thermal, lubricating, and ductile properties.^19–22 

Graphene is a well-known carbon allotrope with exemplary physicochemical properties, e.g., high carrier mobility, chemical and thermal stability, in-plane electrical and thermal conductivity, etc.^23–25 Graphene, as reinforcement, could impart pronounced improvement to the intrinsic properties of the metal matrix. Significant enhancement of thermal and electrical properties of a metal can be achieved by introducing a low weight fraction of graphene into the metal-matrix. The thermal conductance at the interfaces between metal and graphene is a subject of interest for both fundamental research and industrial applications. To design an efficient electronic- or energy-device, it is very much important to understand the thermal transport between the metal and graphene in the composite/mixture. It is to be noted that the synthesis of research-grade graphene from graphite is a cost-effective method that allows scalability at an industrial scale. The preparation of AMCs using graphene as the reinforcing component can be done with several reported procedures that have been broadly categorized as the dispersion method and interfacial reaction method. Among the two, the dispersion method allows for the even/uniform distribution and better amalgamation of the matrix and dispersed phase, hence preferred over the interfacial reaction method. The two most used dispersion methods through the powder metallurgy technique are high energy ball-milling and flake powder metallurgy processes. While the former can afford a more uniform distribution of the dispersion phase, the latter prevents agglomeration of the dispersion phase inside the metal matrix.

In this report, we have taken aluminum as the metal matrix and graphene (reduced graphene oxide) as the dispersion phase. A cost-effective method involving powder metallurgy and vacuum sintering processes has been followed to obtain Al–Gr (aluminum–graphene) composites with higher thermal conductivity (up to ∼125%), enhanced micro-hardness (up to ∼35%), and lowered values of the co-efficient of thermal expansion (CTE) in contrast to pristine aluminum. Interestingly, all the physical properties of the aluminum content, in the composite, were found to be unaltered. This means that the composite could absorb heat at a high rate and can dissipate the same rapidly, ideal for solar heat collector devices.

Pristine aluminum powder (99.9% purity) and research-grade graphene (of two–five layers with >99.5% carbon content) were purchased from a commercial source. The powder samples were taken as per their respective weight ratios in two different stainless steel (SS) jars along with stainless steel (SS) balls in the presence of toluene as a process control agent. The balls to powder ratio was maintained at 10:1. The milling was carried out for 10 h with 300 rpm in a high energy ball mill. After milling, the Al-based composite powders were synthesized. The green pellets have been prepared using Al–graphene composite materials with the help of a cylindrical die (12.5 mm height and 6 mm diameter), using a hydraulic press machine. During compaction, a load of 1.28 ton with a holding time of 2 min for samples having a diameter of 6 mm and 2.56 ton with a holding time of 5 min for samples having a diameter of 12.5 mm was applied. The green pellets were then subjected to vacuum sintering at 630 °C, for two hours holding time, with a heating rate of 5 °C/min. The thermal fusing of Al-based materials at this temperature was found to be satisfactory with the required strength. The sintered Al–graphene composite pellets were finally fabricated through hydraulic crimping.

In this work, we have inspected pellets made of pure aluminum and also aluminum–graphene composites at different weight percentages of graphene, i.e., 0.25, 0.5, 0.75, and 1 wt. % of graphene, in the aluminum matrix, henceforth labeled as AlGr_0.25, AlGr_0.5, AlGr_0.75, and AlGr_1, respectively. A schematic of the whole process of ball milling and Al–Gr composite formation has been illustrated in Fig. 1. Subsequent morphological and compositional analyses of both pristine aluminum and its graphene composite have been provided in Figs. 2(a)–2(d).

Figures 2(a) and 2(b) show the field-emission scanning electron microscopy (FESEM) images of pristine aluminum and Al–Gr composites (with 1 wt. % of graphene), respectively. The energy-dispersive X-ray spectroscopy data for both the samples have been illustrated in Figs. 2(c) and 2(d), where the former corresponds to pristine Al and the latter is for the Al–Gr composite. As shown in Fig. 2(b), the composite material, AlGr_1, exhibits uniformly distributed spherical particles with individual particle size in the range of 10–15 nm. The distribution also shows the uniform amalgamation of Al and graphene constituents, which is essential for the Al–Gr composites to be used as electrodes for their possible application in solar light devices, etc. Structural analyses for all the samples (i.e., pristine Al and Al–Gr composites at 0.25, 0.5, 0.75, and 1 wt. % of graphene) have been done through the X-ray diffraction technique, and the resultant patterns are illustrated in Fig. 3(a). The Al powder and all the composite materials (AlGr_0.25, AlGr_0.5, AlGr_0.75, and AlGr_1) demonstrate similar phase growths, i.e., planes with Miller indices, (111), (200), (220), and (311), that are characteristic of pristine aluminum. However, the intensities of the samples are quite different from each other and vary from the highest for pristine aluminum to the lowest in the case of the Al–Gr composite with 1 wt. % of graphene content.

The absence of the characteristic graphene peak at ∼27° for the (002) plane in the case of Al–Gr composite samples suggests the high crystalline nature of the aluminum metal (which is the dominant component here) and/or ultra-small thickness of the embedded graphene particles of the order of few hundred nanometers. The vibrational spectra through Raman analysis have been provided in Fig. 3(b), which depicts the characteristic bands of graphene (reduced graphene oxide). The D-band is positioned at a wave number, ∼1300 cm⁻¹, which is a disordered band triggered by the structural defects, edge effects, and dangling sp² carbon bonds. The position of the G-band is close to 1580 cm⁻¹, which is a result of the in-plane oscillations of the sp² hybridized carbon atoms.

The larger intensity of the G-band in comparison to the D-band suggests that the graphene content in the Al–Gr composite readily conserves its pristine properties. It can be observed that both the D- and G-band are quite broad and not sharp, as is usually observed in the case of pristine graphene samples. This might be due to the presence of graphene in trace amounts within the Al–Gr composite.

The Archimedes principle was implemented to determine the bulk density of the sintered Al–Gr pellets. We found that, after sintering, the density values of all the Al–Gr composites remained at ∼97.5% of the original density (for pure aluminum). Furthermore, the micro-hardness test was carried out with a Zwick/Roell machine with the help of the ZH micro-HD software. The hardness of the sintered samples was examined using a Vickers micro-hardness tester with the help of a diamond pyramid indenter. Hardness was inspected from the periphery to the core of the specimen, and the average value was recorded from 10 different test regions. It has been found that sintered AlGr_0.25 has more hardness in terms of the Vickers hardness number (41 VHN) when compared with sintered pure Al (36 VHN). The analysis has been graphically illustrated in Fig. 4(d). Results show that for 0.25% and 0.5% of graphene content, the composites exhibited improved micro-hardness as compared to the rest of the composites. It is interesting to note that the micro-hardness obtains the highest value at the lowest percentage of graphene incorporated into the aluminum bulk. With the increase in the percentage of graphene, the micro-hardness value showed a decreasing trend. The sintered sample, AlGr_0.25, has more hardness as compared to sintered pure Al. The reason behind this may be correlated with the presence of graphene in the metal matrix composite, which can block the emigration of the dislocations and vacancies in the composite. As a result, there is a decrease in the micro-hardness with the increase in the graphene percentage in the Al matrix. In addition, the presence of graphene can obstruct grain growth by grain boundary pinning. The results are well summarized with error bars in Fig. S1. The co-efficient of thermal expansion (CTE) of all the samples has been measured with the help of the L75HX (Linseis, Germany) dilatometer for the purpose of investigating the thermal length change. The sample in pellet form (having a dimension of 6 mm diameter and 6 mm thickness) was placed in a sample holder and was linearly heated and then allowed to cool down. The variation in length is transmitted with the help of a push-rod from the furnace on a linear variable differential transducer (LVDT). The sintered sample temperature was recorded by using a thermocouple, and the measurement was carried out in an argon atmosphere. The co-efficient of thermal expansion is an important factor for the correct estimation of the mechanical stability of the material under thermal stress. The thermal expansion coefficient of the composite depends mainly on its state, constituent materials, and operation environment. In our study, it is observed that the CTE for Al–graphene composites has lower values as compared to pristine Al, at different temperatures [Figs. 4(a) and 4(b)], which indicates better thermal stability in the case of Al–graphene composites. It is to be noted that the coefficient of thermal expansion is inversely proportional to the bond strength of the material and, hence, to the melting point of the material. Therefore, from the CTE experiments, we can conclude that the Al–graphene composite materials would have high melting points than pristine Al. The value of the CTE for pure Al is 26.5 × 10⁻⁶/°C at 200 °C. In our case, for the sintered pure Al, we have obtained a CTE value, 21.36 × 10⁻⁶/°C at 200 °C, which is quite close to the bulk value.

In addition, the change in length (ΔL) of Al–graphene composites and relative expansion (in percentage) at different temperatures are less than those of pristine Al, as can be confirmed from Figs. 4(c) and 4(d). This explains that the Al–graphene composites have high mechanical strength and may be deployed for the application in the transmission line in order to avoid sagging of electrical cables. The thermal conductivity measurement has been carried out from room temperature to 300 °C for the sintered cylindrical pellets (12.7 mm diameter and 3 mm thickness) of the Al–graphene composites with the help of the Linseis, Germany XFA 600 model. It is an indirect method of getting thermal conductivity results, as the measuring instrument gives thermal diffusivity data. The thermal conductivity can be found from thermal diffusivity data as per the following mathematical relation:

where α = thermal diffusivity (m²/s), λ = thermal conductivity (W/mK), C_p = specific heat capacity at constant pressure (J/kg K), and ρ = density (kg/m³). We have afforded a cost-effective and unique process for the enhancement of the thermal conductivity of pristine aluminum by approximately ∼125% by the addition of a trace amount of graphene into the aluminum bulk. Graphene, having carbon purity >99.5%, was incorporated into the bulk aluminum, which was then sintered at a temperature of 630 °C with a soaking time of 120 min under an argon atmosphere. The amount of graphene was varied from 0.25% to 1% by weight into the Al system using the powder metallurgy route. Under the optimum conditions, the enhancement of thermal conductivity up to ∼125% has been observed as compared to pristine aluminum. It has been found that, at room temperature, the increase in the wt. % of graphene (from 0.25% to 1%) in the Al system increased the thermal conductivity of the composites. However, with the increase in the temperatures from room temperature to 300 °C, the thermal conductivity values decrease in the case of both pristine and graphene incorporated aluminum samples, as can be evidenced from Fig. 5(a). Hence, the thermal conductivity values of the samples are higher at room temperature than those obtained at comparatively elevated temperatures. In this report, the thermal conductivity of pristine aluminum has been recorded to be ∼120 W/mK. The decrement in the value of thermal conductivity from the standard reported value (i.e., ∼220 W/mK) can be correlated with the rigorous ball-milling process in which the fine granules of aluminum are readily exposed to the partial surface oxidation process, thereby affecting the overall thermal conductivity. In our case, a higher thermal conductivity value (∼280 W/mK) has been observed for the Al–graphene composite (by incorporating 1% by weight of graphene into the Al bulk), which is more than the thermal conductivity of pristine bulk Al at room temperature, as evidenced in Fig. 5(b). Thus, the technique that has been discussed in this report could be utilized to bring significant improvement to the thermal conductivity of pristine aluminum by introducing only a trace amount of graphene into the aluminum matrix. A detailed comparison of the thermal conductivities of several commercial grade metals and graphene–metal composites has been provided in Table I, for better understanding of the significance of Al–Gr composite as a low-cost alternative for thermal dissipative applications.

The experimental findings have been corroborated by rigorous modeling analyses for all the samples using classical mixture models that are appropriately correlated with the Al–Gr metal matrix composites. The details have been provided in Sec. IV. To check the role of graphene weight percentage in affecting the thermal conductivity in the Al–Gr composites, we have taken 5 wt. % of graphene and prepared the Al–Gr composite, which yielded a thermal conductivity of ∼410 W/mK, which is higher than pristine aluminum and copper. The relevant results have been provided in the supplementary material. The three-dimensional morphological state of AlGr_5 was studied by using a non-destructive technique such as X-ray micro-computed tomography (CT) (model: SkyScan 2211, Bruker, Belgium). The analysis confirms the non-porous nature of the samples with both the internal and external (surface) imaging and micro-structural investigation. The reconstruction of scanned images was performed by using SkyScan’s InstaRecon software. The AlGr_5 sample was scanned with a full 3600 rotations (0.20 increments) under the following conditions: accelerating voltage, 185 kV, and target current, 230 µA. The reconstructed images were transferred into 3D models using CT vox and Avizo 9.0.0 platforms. Figures S2(a) and S2(b) show the external (surface morphology) and internal (structural) images, respectively, recorded for the AlGr_5 sample. The specimen was used in its original form (no processing or preparation such as thin sectioning, surface polishing, etc. was done). Figure S2(c) shows the tomographic image taken through the center of the sample in the X–Z plane of the CT scanner’s coordinate system. From the tomographic study, it is found that our sintered pellet (i.e., AlGr_5) is non-porous in nature and the sample is free from any type of structural defects. The fabricated AlGr_5 sample has been further examined via Electron Probe Micro-Analysis (EPMA) to get precise information regarding the elemental composition through wavelength dispersive X-ray spectroscopy. Figure S3 illustrates the EPMA results for the Al-based composite material, AlGr_5, and its SEM image. It clearly shows the uniform distribution of aluminum and carbon across the given scan region. The micro-structural analysis shows a reasonably uniform distribution of the graphene particles inside the bulk Al matrix.

The estimation and calculation of the composite property are non-trivial, since various structural or geometrical parameters such as volume/weight fraction, geometrical structures, distributions of the filler particles, and the interaction between the filler and matrix control the physical properties of the composite. In other words, the thermal conductivity of the composite depends upon the volume fraction and morphology of the filler particles.^32–34 In order to explore the mechanism of variation of the thermal conductivity of the Al–graphene composite with respect to the graphene content, we further studied various theoretical models to provide a necessary explanation for the results obtained in the experiments. Several models have been proposed for the evaluation of the thermal conductivity of the two-phase mixture/composite: Maxwell’s model, Son-Frey model, Russell’s model, Rayleigh’s model, Lewis–Nielsen model, etc. In this study, graphene particles are considered as the filler particles embedded in the aluminum matrix. All these theoretical models are calculated by considering volume fraction as the base factor to estimate composite conductivity and are valid only when the volume fraction of filler particles in the composite is between 25% and 40% (dilute), because at higher concentration, the interaction between filler particles inside the mixture dominates, which affects the phonon scattering and changes the thermal conductivity. Although in experiment, weight fraction has been taken, all the models described above are based on the volume fraction of the filler and matrix. The conversion of weight fraction to volume fraction in the aluminum–graphene composite is provided in the supplementary material. In summary, it is very much important to understand how experimental conductivity values are correlated with a proposed model. In this paper, the theoretical models used by the researchers to estimate the Al–graphene composite conductivity are described in detail.

At first, two simpler models, i.e., series and parallel models, are used to estimate the upper and lower limit of the composite’s thermal conductivity. In these cases, the two phases are represented by two plates, which are oriented perpendicular (series model) and parallel (parallel model) to the direction of heat flow, as shown in Figs. 6(a) and 6(b). The values obtained in other models lie between these two models. The thermal conductivity of the composite for both the models can be calculated by the following equations:

For the series model,

and for the parallel model,

where V₁ and V₂ are the volume fractions of aluminum and graphene in the composite, respectively, and k₁ and k₂ are the thermal conductivities of the respective phases. However, the estimated values of thermal conductivities obtained from these two models show significantly large deviations from the values obtained experimentally.

On the other hand, the effective thermal conductivities of the two-phase mixture/composite of aluminum and graphene estimated through the Russell model and the Son-Frey model lie in between the values obtained from the series and parallel models and somehow approach experimental values. Here, both the models are based on the dispersion of the discontinuous phase of graphene particles in the continuous phase of the aluminum matrix with a cubic array geometry.^35,36 The Russell’s and Son-Frey models are based on the concept of linear isotherms and linear heat flow, respectively, and numerically obtained by modeling a series–parallel network of the two phases.^37,38 The effective thermal conductivity can be calculated from the following equations:

For the Russell’s model,

and for the Son-Frey model,

where V_d and V_c are the volume fractions of dispersed graphene fillers and the continuous Al matrix, respectively, and k_d and k_c are the corresponding thermal conductivities. Since both the models describe the dispersion of minor phases within a major phase of the matrix, the value obtained is somehow close to the experimental values (Fig. 7). However, through Maxwell’s model,³⁹ we can consider the dilute dispersion of the discontinuous phase of spherically shaped graphene fillers having conductivity, k_d, to be embedded in the continuous aluminum matrix of conductivity k_c, as shown in Fig. 6(c). This model ignores the thermal interactions between filler particles and is valid only in the case of a lower volume fraction of graphene (V_d).^40,41 The effective thermal conductivity of the composite can be evaluated as follows:

On the other hand, the Lewis–Nielsen model can correlate successfully between the experimental and theoretical values of the effective thermal conductivity of the composite having the volume fraction of the discontinuous phase up to 40%, and for higher concentrations, the model fails to predict. The filler particles with various morphologies and distribution patterns inside the continuous matrix can be well described through this model.^41–45 Figures 6(d) and 6(e) represent, schematically, the Lewis–Nielsen model, where filler particles with rod and sphere shapes are arranged in a continuous matrix, respectively. The composite conductivity can be evaluated through this model as follows:

where Φ = V_d = volume fraction of graphene inside the composite and

Here, Φ_m is the maximum volume fraction of the filler particles and A is the shape coefficient (calculated from the aspect ratio) of fillers whose values are provided in the supplementary material for different shapes of fillers.

The experimental thermal conductivity values, along with the calculated values through different models for the aluminum–graphene composite with respect to the different weight fraction of graphene contents, are shown in Fig. 7. The values obtained through the parallel model show marked deviation from the experimental values, whereas the values calculated through the Maxwell’s model, Russell’s model, and Son-Frey model are somehow close to the experimental values. For higher concentrations (>0.8% of weight) of graphene in the composite, Russell’s model is well-matched, indicating the cubical array pattern of graphene inside the cubic Al matrix, where isotherm lines are assumed to be planes.⁴⁶ However, this model results in a slight deviation for a lower volume fraction of graphene. In contrast, Maxwell’s and Son-Frey models closely approach the experimental results for lower weight fractions (<0.6%) and deviate for higher weight fractions. The filler particles develop a continuous conductive network in the case of high graphene contents, whereas the filler particles are dispersed without developing a conductive network in the composite for lower volume fractions. As a result, different models fit with the experimental data for different concentrations.

Furthermore, to get insight regarding the contribution of shape and size and distribution of graphene particles toward heat transport in the Al–graphene composite, we have calculated the effective composite thermal conductivities through the Lewis–Nielsen model for spherical and rod/fiber shapes and compared with the observed values obtained in the experiments (Fig. 8). As shown in Fig. 8(a), the estimated values for spherical graphene fillers are well matching with the experimental results at moderate concentrations, while showing noticeable deviation for higher weight fractions. In this case, the discontinuous phase of graphene particles is assumed to be arranged in a cubic manner (or randomly closed) inside the cubic aluminum matrix. In contrast, considering the graphene fillers as rod or fiber shaped, the model results agree nearly with the experiment at a higher weight fraction of graphene [Fig. 8(b)]. In this case, the fibers are assumed to be distributed randomly in three-dimensional directions inside the matrix with a shape coefficient value of 2.08, implying that lengths are quite larger than the diameters of the fibers. A close observation from Figs. 7 and 8 reveals that the filler morphology plays an integral part in determining the thermal conductivity of the composites. Changing the weight fraction of graphene in the composite sample leads to the modification and distribution of fillers inside the matrix and is responsible for matching the experimental results obtained with different models at different concentrations. Hence, it is clear that the properties of composites are highly dependent on the surface-to-volume fraction, aspect ratio (length/diameter), and distribution pattern of filler graphene inside the matrix.

From the results, we observe that the synthesized Al–graphene composite shows an enhancement of thermal conductivity in comparison to pure aluminum. This excellent heat transfer at the interfaces of aluminum and graphene may be due to the interfacial structures and the interaction between these two phases.⁴⁷ Our sample shows high thermal conductivity with an increase in the concentration of graphene. Due to the atomically thin planar structure of graphene, the interfacial thermal resistance is reduced. As the volume of the discontinuous phase increases inside the aluminum matrix, the filler graphene particles make a continuous network inside the matrix and the phase space states available for phonon scattering start decreasing. As a result, it provides a smooth path for the transport of phonons with less scattering and leads to enhanced conductivity. We observe that the Al–graphene composite sample prepared through the powder metallurgy method has different morphologies of graphene for different concentrations for which different theoretical models fit into the experimental values at different weight fractions.

Since graphene has a much lower density as compared to aluminum, even a small weight fraction of graphene in the composite would contribute a higher volume fraction than the metal matrix (the conversion from weight fraction into volume fraction for this composite is provided in the supplementary material). When the weight fraction of graphene is 1%, the volume fraction reaches almost about 25%. The classical limit, for the volume fraction of filler particles, to be considered for modeling calculations, is up to 25%. Therefore, the modeling calculations, in this report, have been restricted to the composite AlGr_1 and were not considered for the case of AlGr_5, since the value of the volume fraction for the filler particle exceeds the limit of 25%. In this dilute limit, the interaction between the filler particles can be neglected, where the distance between two neighboring filler particles is more than their size. As a result, the heat flux lines created by one filler particle cannot be distorted by the other surrounding filler particles and can be modeled through classical equations to find the effective thermal conductivity of the composite, as has been discussed in the report. However, we have included the results of an Al–Gr composite prepared by incorporating 5 wt. % of graphene in the aluminum matrix, which yielded a higher thermal conductivity value of ∼410 W/mK (which is obvious considering the fact that graphene plays a dominant role here when it comes to thermal conductivity owing primarily to its extraordinarily high thermal conductivity value, ∼3000 W/mK). However, modeling calculation could not be done in this case due to the limitations of the classical models used in this report.

In this work, aluminum–graphene composite materials have been synthesized using a powder metallurgy route and were vacuum sintered prior to morphological and compositional analyses. The composite materials showed uniformly distributed spherical particles with individual particle sizes <10–20 nm. The distribution shows the uniformity of Al and graphene constituents. Elemental composition analyses confirmed the uniform distribution of Al and carbon contents throughout the scanned areas. The recorded thermal and mechanical properties in the case of Al–Gr composites suggest that they can be readily used as efficient solar thermal collectors owing to their excellent thermal conductivities. At present, pristine Cu and/or Al are being used extensively for making solar thermal collectors. The bulk Al and Cu have thermal conductivity values, ∼237 W/mK and ∼40 1W/mK, respectively, at room temperature. In our case, we are getting an improved thermal conductivity (410 W/mK) for Al–graphene composites (incorporation of 5 wt. % of graphene into the Al bulk), which is more than the thermal conductivities of Al and Cu. In addition, we have observed that the co-efficient of thermal expansion (CTE) of Al–graphene composites has lower values as compared to pristine sintered Al, at different temperatures, which indicates that the Al–graphene composites have a higher tolerance toward thermal stress than bare Al.

See the supplementary material for this manuscript, which includes three tables, i.e., Table S1 for the conversion of weight fraction into volume fraction, Table S2 for the maximum packing fraction for various arrangements of filler particles inside the metal matrix, and Table S3 for the aspect ratio and values of the shape coefficient for different shapes of filler particles. Figures include the micro-hardness test result with respect to the weight percentage of graphene at 1 kgf load (Fig. S1), x-Ray micro-CT analysis of Al-based sintered device AlGr_5 (Fig. S2), and the EPMA results for AlGr_5 (Fig. S3).

S.K.P. and M.R.S. contributed equally to this work.

The data that support the findings of this study are available within the article [and its supplementary material].

We gratefully acknowledge the financial support provided by the National Aluminium Company Limited (NALCO), India, via Grant No. RP-115. Part of this work has also been supported by the Ministry of Human Resource Development (MHRD), India, through a Centre of Excellence (Grant No. RP-074), and the Department of Science and Technology (DST-MES), India.