The impact of varying EDTA dosage on the flower-like morphology and thermoelectric properties of Ce-doped Bi₂Te₃

Abstract

Thermoelectric materials have significant applications in energy utilization and environmental protection. The effect of different EDTA (ethylenediaminetetraacetic acid) dosages on the flower-like morphology and thermoelectric properties of Ce-doped Bi₂Te₃ nanoparticles were investigated. The Ce-doped Bi₂Te₃ nanoparticles were sucessfully prepared via the hydrothermal method, and the influence of EDTA dosage on the morphology, structure, and thermoelectric properties of the materials was analyzed. The experimental results showed that an appropriate amount of EDTA can promote the formation of a flower-like morphology in Ce-doped Bi₂Te₃ nanomaterials and enhance their thermoelectric properties. The ZT values of the x = 0.15 sample made by the better flower-like morphology of nanopowders are all around 1, which reach 1.15 at 398 K. This work demonstrates the synergistic effects of combining nanostructure engineering and chemical doping strategies for thermoelectric performance enhancement.

Introduction

Thermoelectric materials, as functional materials capable of converting heat energy into electrical energy and vice versa, have significant applications in energy utilization and environmental protection^1–3. The properties of the thermoelectric materials is decided by the figure of merit ZT which is defined by . As the equation indicated, a high ZT value requires a large Seebeck coefficient S, low electrical resistivity ρ and low thermal conductivity k^4–6. Bi₂Te, as a typical thermoelectric material, has attracted widespread attention due to its excellent thermoelectric properties. As reported, the morphology of the synthesized nanoparticles by hydrothermal method has significant effect on the ZT values of the thermoelectric materias, and the flower-like morphology is beneficial to improve the ZT values^7–9. Pillai et al. ⁷ utilized a cold-pressed method to prepare the nanostructured rGO/GO-bismuth telluride samples to retain the initial morphology of the host nanomaterials, revealing that larger diameter bismuth telluride nanomaterials exhibited higher Seebeck coefficients and lower thermal conductivity (0.26 W/m–K) compared to smaller diameter and nanowire-based bismuth telluride thermoelectric materials. Kim et al.⁸ reported that nanocrystalline Bi₂Te₃ nanotubes synthesized using the solution phase method offered improved ZT values compared to bulk Bi₂Te₃ near room temperature due to enhanced Seebeck coefficients and suppressed thermal conductivity. Kimura et al.⁹ reported enhanced thermoelectric properties of n-type Bi₂(Se_xTe_1-x)₃ nanoplates synthesized using solvothermal methods by tuning selenium composition. EDTA, a commonly used surfactant, plays a crucial role in regulating the morphology and properties of nanomaterials during the preparation process¹⁰.

Although the thermal conductivity of Bi₂Te₃-based materials has been significantly reduced through powder processing, it remains challenging to compete with newly emerged TE materials that exhibit intrinsically high phonon anharmonicity or liquid-like sublattices. Further reductions in thermal conductivity could potentially enhance the TE performance of Bi₂Te₃-based materials¹¹. It is well-known that Ce can form resonance electron states near the Fermi level, strongly influencing electronic transport properties, particularly thermopower¹². Therefore, studying the impact of varying EDTA dosages on the flower-like morphology and thermoelectric properties of Ce-doped Bi₂Te₃ is of great significance. This work successfully prepared Ce-doped Bi₂Te₃ nanoparticles via the hydrothermal method and analyzed the effect of EDTA dosage on the morphology, structure, and thermoelectric properties of the materials. The results demonstrate the synergistic effects of combining nanostructure engineering and chemical doping strategies for enhancing thermoelectric performance.

Experimental section

Material preparation

The methods of the material preparation and performance testing are same as our previous reports¹². Analytical-grade chemicals including 1.7 mmol BiCl₃, 0.3 mmol Ce(NO₃)₃(6H₂O) and 3 mmol 5 N pure Te powder were used as the precursors for the synthesis of the Ce_0.3Bi_1.7Te₃ nanopowders. The precursors were placed in a Teflon-lined autoclave filled with distilled water. Sodium hydroxide (NaOH) was added as PH-value controller, and a certain amount of EDTA was added as surfactant. Then, the Teflon-lined autoclave was placed on a magnetic stirrer and agitated for 30 min. Subsequently, 0.35 g of NaBH₄ was introduced into the solution as the organic complex reagent. The autoclave was then sealed and maintained at 443 K for 24 h. The resulting powders were filtered and thoroughly washed several times using distilled water, dehydrated ethyl alcohol, and acetone to eliminate impurities. Following this, the powders were dried in a vacuum oven at 373 K for 6 h. Afterwards, they were hot-pressed into bulk pellets with diameters of either 12.5 mm or 15 mm and a thickness of approximately 2 mm at a temperature of 773 K under a pressure of 60 MPa in a vacuum environment.

Performance testing

The phase composition of the powders was analyzed using X-ray diffraction (XRD) with Cu-Kα radiation (Rigaku D/MAX-2550p diffractometer, Japan). The data were collected within a 2θ range of 10–80°, with a step size of 0.02° and a scanning speed of 4°/min. The morphologies of both the powders and the bulk pellets were observed using a field emission scanning electron microscope (SEM) (JSM-6700F, Japan). The densities of the bulk pellets were determined by the Archimedes method. The electrical resistivity and Seebeck coefficient of rectangle bars cut from the ϕ15 mm pellets were measured at several temperature points using an LSR-3/800 Seebeck coefficient/Electric Resistance Measuring System (LINSEIS, Germany) under He atmosphere. The measuring error of the electrical resistivity ρ and Seebeck coefficient S is about 5%. The thermal conductivity of the ϕ12.5 mm pellets was measured using a thermal diffusivity system (FLASHLINETM 3000, ANTER, USA) using Pyroceram as the reference sample (Provided by ANTER) at several temperature points. To avoid the fluctuation in the thermal conductivity measurement several times measurement were perform. The measuring error of the thermal conductivity is about 10%¹⁴. The power factors p were calculated by the equation p = S²/ρ. The figures of merit, ZT, was calculated by the measured data and . So the errors are 10% for power factor and 11% for ZT.

Results and discussion

As in ref¹⁵ when the amounts of EDTA is more than 0.2 g, the synthesized Ce_0.3Bi_1.7Te₃ nanoparticles exists obvious impure phase in current synthesized conditions, so in this work the amounts of EDTA used is less than 0.2 g. Figure 1 presents the x-ray diffraction patterns of Ce-doped Ce_0.3Bi_1.7Te₃ nanoparticles synthesized using varying amounts of EDTA (x = 0.125 g, 0.15 g, 0.175 g, respectively). It can be seen that all the nanopowders show rhombohedral structure as for the Bi₂Te₃ alloy, and there are no obvious impurity phases as compared to the standard JCPDS card of Bi₂Te₃ (15-0863 card). It indicates that in current synthesized conditions pure phase of the Ce_0.3Bi_1.7Te₃ can be formed for the Ce doping element. And the amount of EDTA used does not affect the phase composition of the synthesized nanopowders.

Fig. 1.

XRD patterns of the Ce_0.3Bi_1.7Te₃ nanopowdres prepared with different EDTA amount x, which are 0.125 g, 0.15 g, 0.175 g, respectively.

Figure 2 depicts the morphologies of Ce_0.3Bi_1.7Te₃ nanopowders prepared with different EDTA dosages. It can be seen from the figure, all samples exhibit a well-defined flower-like morphology consisting of larger sheets with thicknesses less than 100 nm. Notably, the nanosheets forming the flower-like structures in the x = 0.125 g and x = 0.15 g samples are significantly smaller compared to those in the x = 0.175 g sample. This suggests that EDTA promotes the growth of larger nanosheets. Bi₂Te₃ crystal has a layered hexagonal structure with Te and Bi atom layers arranged in the order of –Te(1)–Bi–Te(2)–Bi–Te(1)– along the c-axis. In natural growth process Bi₂Te₃ crystal grows much faster along a-axis and b-axis than that along c-axis, thus flake morphology is easy formed. However, this natural growth process can be changed greatly by adding some surfactants¹⁵. During the growth process, EDTA acts as a chelating agent, binding with Bi³⁺ ions to form large molecular complexes, thereby facilitating the growth of Bi₂Te₃ nanosheets. It is indicated that EDTA is helpful to the growth along the a-axis and b-axis of the Bi₂Te₃ crystal which is consistent with our previous reports^10,12. In this work Te powders were used as Te source which makes it possible for neighboring nanosheets to laterally connect each other and consequently form large flower-like clusters as mentioned by our previous reports¹⁵.

Fig. 2.

SEM images of of Ce_0.3Bi_1.7Te₃ nanopowders with different EDTA dosage. (a) x = 0.125 g; (b) x = 0.15 g; (c) x = 0.175 g.

Figure 3 shows the microstructure of Ce_0.3B_i1.7Te₃ bulk samples obtained by hot pressing the flower-like nanopowders with varying EDTA dosages. The densities of the samples are 7.3 g/cm³, 7.2 g/cm³, and 7.5 g/cm³ for x = 0.125 g, 0.15 g, and 0.175 g, respectively, which are close to the theoretical density of Bi₂Te₃ (7.7 g/cm³). Upon hot pressing the flower-like morphologies are no longer visible, but the sheet-shaped crystals still exist with thicknesses less than 100 nm which is beneficial to enhanced thermoelectric (TE) properties. Moreover, the sheet sizes increase significantly along directions parallel to the sheet surfaces due to the high temperature during hot pressing. For the x = 0.125 g and 0.15 g samples, the microstructure of the hot-pressed bulk pellets comprises a mixture of small grains embedded in the larger sheets, which can scatter phonons effectively but not scatter carriers much thus improving the TE properties of Bi₂Te₃-based alloys as reported^10,12. In contrast, the x = 0.175 g sample exhibits a primarily large sheet microstructure, which is caused by the larger nanosheets observed in the corresponding nanopowder as shown in Fig. 2c.

Fig. 3.

SEM images of hot pressed Ce_0.3Bi_1.7Te₃ bulk samples with different EDTA dosage. (a) x = 0.125 g; (b) x = 0.15 g; (c) x = 0.175 g.

Figure 4 displays the electrical resistivity of Ce_0.3Bi_1.7Te₃ bulk samples with different EDTA dosages as a function of temperature. All samples exhibit an increase in resistivity with temperature, indicating a degenerate semiconductor behavior. Notably, the resistivities of the x = 0.15 g and x = 0.175 g samples are lower than that of the x = 0.125 g sample. This may be attributed to the larger sheets in the x = 0.15 g and 0.175 g samples, which facilitate carrier transport and enhance carrier mobility¹³. The electrical resistivity is generally expressed as¹⁶:

1

where μ is the carrier mobility, n is the carrier concentration, and e is the charge of an electron. As indicated by Eq. 1, a higher carrier mobility results in a lower electrical resistivity.

Fig. 4.

Electrical resistivity of the Ce_0.3Bi_1.7Te₃ bulk samples with different EDTA dosage versus temperature.

Figure 5 presents the Seebeck coefficients of the samples versus temperature. All samples exhibit n-type conduction evidenced by negative Seebeck coefficients within the measured temperature range, which indicate that Ce substitution for Bi acts as a donor dopant. The Seebeck coefficients of the samples all show a nonmonotonic variation, which initially increase with temperature, and after reaching a maximum then decrease with further temperature increase. This decrease in Seebeck coefficient at higher temperatures can be attributed to the rapid increase in minority carrier concentration¹⁷. While the electrical resistivity of the x = 0.15 g and x = 0.175 g samples is lower than that of the x = 0.125 g sample (Fig. 4), their Seebeck coefficients (absolute values) are also lower (Fig. 5). This is consistent with the common knowledge that lower electrical resistivity leads to lower Seebeck coefficients. However, the electrical resistivity of the x = 0.15 sample is little smaller than that of the x = 0.175 sample, as depicted in Fig. 4. Fig. 5 reveals that the Seebeck coefficient (in absolute value) of the x = 0.15 samples is much lower than that of the x = 0.175 sample, while the electrial reisitivity of the x = 0.15 samples is similar with that of the x = 0.175 sample. The possible reason can be attributed to the interplay between electron concentrations and scattering mechanisms, which are influenced by the microstructure of the samples. The Seebeck coefficient in degenerate semiconductors can be formulated as¹⁸:

2

Fig. 5.

Seebeck coefficient of Ce_0.3Bi_1.7Te₃ bulk samples with different EDTA dosage versus temperature.

Here, represents the Boltzmann constant, is the temperature in Kelvin, is the reduced Planck constant, n denotes the carrier concentration, and is the scattering factor. As commonly observed from Eq. 2, a larger scattering factor leads to an increased Seebeck coefficient, whereas a higher carrier concentration n, corresponds to a lower Seebeck coefficient S¹⁸. As seen in Fig. 3, the size of the large sheets in the x = 0.15 sample is comparable to that of the x = 0.175 sample. However, the microstructure of the x = 0.15 sample contains tiny grains embedded within the large sheets, leading to more grain boundaries and subsequently more scattering. Nevertheless, for the Seebeck coefficient, the influence of increased carrier concentration outweighs the effect of scattering, which can evident that the microstructure corresponding to the flower-like morphology do not scatter carriers much. Therefore, despite having a slightly smaller electrical resistivithy, the Seebeck coefficient of the x = 0.15 sample is much lower than that of the x = 0.175 sample.

Figure 6 illustrates the power factor (S²/ρ) of the samples as a function of temperature. It is evident from Fig. 6 that, the power factor of the x = 0.175 sample is consistently higher than that of the x = 0.15 and x = 0.125 samples in the tested temperature range. This is attributed to its lower electrical resistivity and relatively higher Seebeck coefficient, as shown in Figs. 4 and 5. Notably, the power factor of the x = 0.15 sample surpasses that of the x = 0.125 sample above 448 K and is comparable to the x = 0.175 sample at 523 K. This indicates that the optimized flower-like morphology achieved through the use of the appropriate EDTA amount is beneficial for enhancing the power factor.

Fig. 6.

Power factor of Ce_0.3Bi_1.7Te₃bulk samples with different EDTA dosage versus temperature.

Figure 7 displays the thermal conductivities of the samples versus temperature. The thermal conductivities of all samples are approximately 1 W/mK and much lower for the x = 0.15 sample, which is lower than that of commercially zone-melted Bi₂Te₃ ingots¹⁹. The possible reason are as follows: firstly, Ce-doping can introduce a number of point defects in the crystals which can effectively enhance phonon scattering and hence reduce the thermal conductivity. In addition, the thickness of the sheets in their microstructures is less than 100 nm which is helpful to decrease thermal conductivity as mentioned above. Additionally, the thermal conductivity of the x = 0.15 sample is lower than the other two samples. This may also be linked to the differences in microstructure of the samples. For the x = 0.15 bulk sample, the microstructure comprises tiny grains embedded in larger sheet which is superior as compared with the other two, as depicted in Fig. 3. The larger sheets can facilitate carrier transport, contributing to the electrical resistivity. And while the tiny grains can effectively scatter phonons, further reducing thermal conductivity as compared with the other two samples. Taking into account the results of the electrical resistivity and thermal conductivity, the microstructure corresponding to flower-like morphology of the x = 0.15 nanopowders is superior as compared with the other two samples, which tcan scatter phonons effectively simultaneously not scatter carriers much. It can also concluded that the amounts of the EDTA is a key factor to form the optimized microstructure.

Fig. 7.

Thermal conductivities of of Ce_0.3Bi_1.7Te₃ bulk samples with different EDTA dosage versus temperature.

Figure 8 presents the ZT values of the bulk samples. The ZT values of the x = 0.15 sample, featuring the improved flower-like morphology of nanopowders, are approximately 1, reaching 1.15 at 398 K. As n-type thermoelectric materials, the ZT value of the x = 0.15 sample is relatively high, given that the ZT value of n-type Bi₂Te₃-based alloys has not been significantly improved beyond 1²⁰. It is noted that the ZT values of the x = 0.15 sample are all around 1 in the testing temperature and reached 1.15 at 398 K which is higher than that of other related reports^21–26, as shown in Table 1. This demonstrates that the flower-like morphology of nanopowders is effective in enhancing the thermoelectric properties of Ce_0.3Bi_1.7Te₃ bulk materials. Importantly, the appropriate amount of EDTA is crucial for forming this optimal flower-like morphology, which can reduce thermal conductivity much and simultaneously keep electrical resistivity at a relatively low extent.

Fig. 8.

The figures of merit (ZT) of of Ce_0.3Bi_1.7Te₃ bulk samples with different EDTA dosage versus temperature.

Table 1.

The thermoelectric fgure-of-merit of the n-type Bi₂Te₃ alloys.

References	Year	Composition	ZTmax (K)
12	2013	Ce0.3 Bi1.7Te3	0.93 at 386
21	2005	Ce-Bi2Se0.3Te3	0.22 at 450
2,2	2014	CeBiTe3	0.29 at 450
2,3	2023	n-type Bi2Te3	0.2 at 500
2,4	2020	Bi1.9Gd0.1Te3	0.75 at 420
25	2020	Bi2Se0.6Te2.4	1.0 at 380–500
26	2021	Bi2Se0.3Te2.7	1.05 at 423
This work		Ce0.3 Bi1.7Te3	1.15 at 398

Conclusion

In this paper, Ce-doped Bi₂Te₃ nanomaterials with a flower-like morphology were successfully synthesized via the hydrothermal method. The influence of varying EDTA dosages on their flower-like morphology were investigated. The phase composition of the powders was analyzed by XRD. And the morphology of the samples were observed by SEM. The results reveal that an optimal amount of EDTA significantly facilitates the development of the flower-like morphology in Ce-doped Bi₂Te₃ nanomaterials. and The effect of varying EDTA dosages on thermoelectric properties was thoroughly investigated. Notably, the appropriate amount of EDTA emerges as a crucial factor in achieving a superior flower-like morphology, which can reduce thermal conductivity much and simultaneously keep electrical resistivity at a relatively low extent. Consequently, the ZT values of the x = 0.15 sample, featuring the optimized flower-like morphology, attained all around 1, peaking at 1.15 at 398 K which is higher than that of the zone melt ingots²⁷. This study demonstrates the synergistic potential of combining nanostructure engineering with chemical doping strategies for enhancing thermoelectric performance.

Author contributions

Fang Wu wrote the main manuscript text and prepared figures. All authors reviewed the manuscript.

Data availability

All data generated or analysed during this study are included in this published article.

Declarations

Competing interests

The authors declare no competing interests.