Demonstrating the Simplicity and In Situ Temperature Monitoring of the Mechanochemical Synthesis of Metal Chalcogenides Suitable for Thermoelectrics

Summary

Here, we present a protocol to synthesize two metal chalcogenides (Cu_1.8S and SnSe) suitable for thermoelectrics via an ultrafast (second-range), solvent-free, and one-step mechanochemical synthesis using elemental precursors. Simultaneously, we demonstrate the monitoring of the temperature in the jar during planetary ball milling in situ by the newly developed device.

Abstract

Mechanochemical synthesis is an extremely useful strategy to reach thermoelectric materials due to its solvent-free one-step character, as the targeted thermoelectricity (TE) materials in a nanocrystalline format can be prepared by mere high-energy milling of elemental precursors. Nevertheless, the subsequent densification method (e.g., spark plasma sintering or hot pressing) is required afterward, similarly to other synthetic methodologies. In this study, the simplicity of mechanochemical synthesis is presented for two selected metal chalcogenides, namely copper sulfide (Cu_1.8S, digenite) and tin selenide (SnSe, svetlanaite), which are known for high ZT values. These compounds can be prepared via a mechanically induced self-propagating reaction (MSR), which is a combustion-like process instantly yielding the products in a very short timeframe (within 1 min). The occurrence of MSR can be well-tracked by in situ temperature monitoring since an abrupt temperature increase occurs at the moment of MSR. We have developed a device which is capable of monitoring the temperature inside the milling jar every 80 ms during planetary ball milling, and it is therefore possible to very precisely track the moment of MSR ignition. The developed device presents an improvement in the monitoring capabilities in comparison with commercially available analogs. This contribution aims to provide a visual insight into all steps, with simple high-energy ball milling of elements to reach TE materials and in situ temperature monitoring being the central points.

Introduction

Statistically, more than 60% of energy in the world is lost, mostly as waste heat. Utilizing the waste heat for thermoelectricity (TE) applications has a great potential. TE offers a suitable method to convert waste heat into electrical energy. Special applications, like electrical energy sources in radioactive thermoelectric generators for space research and/or replacing the old Hg-Zn batteries in cardiac pacemakers, can be mentioned¹.

Among various TE materials, chalcogenides belong among the favorites, especially if they are composed of abundant and non-toxic elements. Chalcogenides with tellurium, lead, and germanium content were reported as perspective TE materials in the past, with Bi₂Te₃ and (Bi,Sb)₂Te₃ being among the most prominent examples. However, both Bi and Te are rare and/or toxic, making the mass production of TE materials with this composition challenging². Looking forward to selection among chalcogenides, the new alternatives that bear in mind non-toxicity, earth-abundancy, and TE efficiency are considered. Two systems that fulfill these criteria are copper sulfides Cu_2-xS and tin selenide SnSe.

Copper sulfides are present frequently in nature as minerals in several compositions, with chalcocite Cu₂S and covellite CuS as border members. In between, several non-stoichiometric compounds exist³. Among them, Cu_1.97S and Cu_1.98S, with interesting properties, were already synthesized by directly melting the elements Cu and S^4,5.  Also, digenite Cu_1.8S is particularly interesting for thermoelectrics.

Tin selenide SnSe represents a high TE figure among chalcogenides. The synthesis at 1223 K for over 9.5 h led to its ultralow thermal conductivity and subsequent high thermoelectric efficiency⁶. Accompanying phenomena were not studied.

Synthesis routes of copper sulfides and tin selenides encompass mostly high-temperature treatment of reaction precursors^4,7,8,9,10. However, there are also alternative, more environmentally sound synthesis routes such as mechanochemical synthesis^3,11,12,13. The mechanochemical synthesis of chalcogenides from elements can, under some circumstances, occur as a mechanically induced self-propagating reaction (MSR), which is a combustion-like process instantly yielding the products in a very short timeframe^14,15,16. For both systems reported in this study, the MSR was reported- for Cu_1.8S, it was done instantly, albeit the Cu:S ratio 1.6 needed to be used due to the volatility of sulfur^16,17, and for SnSe, it occurred in about 15 s¹⁶.

The ignition of an MSR is accompanied by a sudden increase in temperature and pressure. Upon monitoring these characteristics via specifically engineered milling jars, it is possible to determine the MSR onset. However, the commercially available devices for planetary ball milling monitoring offer only the data collection every 2 s, and due to the location of the sensors, MSR can be detected only via pressure monitoring, neither by temperature one^16,18. Moreover, the mentioned system is not transferrable and can only be both purchased and used together with the specifically engineered milling jar, which is both limiting and costly. We have recently developed a transferrable device capable of collecting temperature data every 80 ms¹⁹. This advanced measuring system developed for in situ temperature monitoring during mechanochemical synthesis significantly enhances the capabilities over existing commercial solutions. This system employs an NRBG104F3435B2F NTC thermistor, featuring a resistance tolerance of ±1% at 25 °C and a beta value tolerance of ±1%, ensuring high-precision temperature measurements. With a data capture frequency of every 80 milliseconds, the system provides a high-resolution monitoring crucial for detecting the initiation of MSRs. The thermistor's high sensitivity to temperature changes, indicated by a steep resistance-temperature relationship, ensures accurate detection of rapid temperature spikes. The temperature sensor is strategically placed within an existing screw mechanism used for pressure release and gas addition, located in the hole of a massive cap. This placement protects the sensor from mechanical collisions and signal noise caused by the milling balls, ensuring stable and reliable temperature readings. The limitation is that the ball diameter needs to be larger than the hole diameter. With 10 mm balls, there is no problem. The system's wireless communication capability and robust sealing mechanism prevent material or heat leakage, thereby enhancing the reliability and accuracy of the temperature data collected during the milling process. Designed to be cost-effective and portable, this system represents a significant advancement in the real-time temperature monitoring of chemical reactions during planetary ball milling, offering critical insights for the optimization of materials synthesis.

This study aims to demonstrate the performance of this newly developed device by monitoring temperature during the mechanochemical synthesis of two selected metal chalcogenides that are interesting for TE applications. Another objective is to show the sustainable, simple, and time-saving character of the mechanochemical synthesis, which is boosted when the reaction occurs as an MSR.

Protocol

1. Preparation of CuS mixture with the stoichiometry 1.6:1

1. Tare the weighing paper.
2. Weigh 7.6024 g of elemental copper and 2.3974 g of elemental sulfur powder to achieve the stoichiometry ratio of Cu and S at 1.6:1, with a total mass of 10 g.
3. Before milling, mix the Cu and S powder. After weighing, introduce both Cu and S powders into a plastic weighing dish and mix intensively with a spatula, until a powder of homogeneous color free from large lumps of sulfur is obtained.
   NOTE: The purpose of the mixing is to homogenize the powder and ensure a uniform distribution of powders before the milling experiment.

2. Preparation of SnSe mixture with the stoichiometry 1:1

1. Tare the weighing paper.
2. Weigh 6.0055 g of Sn and 3.9945 g of Se to make the stoichiometry ratio 1:1, with a total mass of 10 g.
3. Before milling, mix the Sn and Se powder by using a spatula to ensure homogeneity (the rules from 1.3 also apply here).

3. Sensor setup

1. Place the sensor board on the top of the jar lid and insert the sensor transistor into the small hole that passes through the lid.
2. Switch on the sensor device and connect it to the software on the laptop via Bluetooth.

4. Performing milling with in situ temperature monitoring

NOTE: The necessary equipment, including the scheme of the temperature monitoring device, is shown in Figure 1.

1. Insert tungsten carbide balls, as specified in Table 1, into the milling jar using the tweezers or just "pour" them inside using gravitational force.
2. Transfer the prepared sample either from section 1 or section 2 to the tungsten carbide milling jar for the synthesis of Cu_1.8S or SnSe.
3. Close the milling jar with the lid that has been set up with the sensor from section 3.
4. Load the jar into the mill by placing the jar and the counterweight into the P7 planetary mill and setting the parameters on the display as specified in Table 1.
5. Type the name of the sample on the active software.
6. Press the Start button on the milling display.
7. After hearing the milling start, click Start in the active software for the sensor to begin recording the temperature during milling.
8. When the MSR occurs, indicated by the sudden increase in temperature, stop the milling and the temperature measurement immediately.
   NOTE: Repeat the experiment with the same system once again for reproducibility

5. Collecting samples

1. Open the jar over a paper sheet in the fume hood, separate the milling balls by sieving the powder through the strainer. The balls remain on the strainer while the fine powder drops on the paper. In the case of large agglomerates, these are removed from ther strainer using tweezers. Collect the sample from the paper.

6. Transferring the powders

1. Transfer the powders from the paper to the glass vials using the gravitational force and spatula, label them, and store them in the desiccator before the measurement.

7. Labeling the glass vials

1. Label the glass vials based on the sample name.

8. Cleaning the jar and sensor

1. Clean the sensor transistor by wiping it with the tissue soaked in etaben.
2. In the case of the jar, pour 75 mL of the etaben solution into the milling jars and perform milling at 300 rpm for 5 min.
3. Use a steel strainer to collect the balls from the jars, and dispose of the toxic aqueous waste into the container.
4. Repeat steps 8.2 and 8.3 until the jar and the balls are free from the solid powder.

9. Processing data from in situ temperature monitoring

1. After finishing the monitoring, the software saves the data automatically as .xlsx files in the download folder on the computer.
2. Process the data in the data treatment software to plot the graph temperature vs time. The raw data obtained from the measurement are directly stored as a .csv file and are already segregated into columns.

10. Powder X-ray diffraction (XRD) measurement

1. Crush the obtained samples by using a mortar and pestle. The samples' appearance is depicted in Figure 2.
2. Transfer each sample with a spatula to the sample holders and label each sample holder.
3. Gently compress the powder with a glass slide, carefully sliding or rotating it to flatten the surface evenly.
4. Transfer the sample holder to the XRD diffractometer.
5. Set up the XRD measurement in the computer using the XRD commander program programming the given measurement conditions.
   NOTE: The instrument used in this study is an X-ray diffractometer using CuKα (40 kV, 40 mA) radiation. The parameter for the powder X-ray diffraction (PXRD) experiment: 2-theta range: 10°-80°, step time: 1 s, step size 0.05 s.
6. Start the PXRD measurement. The XRD data is saved as a ".raw file "on the computer disk.
7. After finishing the measurement, collect the powder back from the sample holder into the glass vial using paper.
8. Convert the .raw file into another file type that is suitable for processing in data treatment software (e.g., Origin)
   NOTE: The PowDLL converter will allow one to convert the file extensions to the desired extension, for example, to .xy format, which is required for Rietveld refinement.

11. Rietveld refinement

1. Do the proper semi-quantitative phase analysis using the XRD software to identify the phases that will be included in the refinement.
2. Download corresponding .CIF files from the internet, e.g., from Crystallography Open Database. Download the one for CuS, rhombohedral Cu_1.8S, cubic Cu_1.8S, SnSe, and SnSe₂.
3. Run JEdit and powder diffraction data analysis software.
   NOTE: The powder diffraction data analysis software used here is Topas Academic software. JEdit was formerly modified to be capable of working with the powder diffraction data analysis software.
4. Create an input file in JEdit. Ensure that the file contains information about the diffractometer and structural parameters of the phases that are included in the refinement.
5. Decide which parameters will be refined.
6. Run the refinement in powder diffraction data analysis software. The software saves the result in the .out file, which automatically becomes the new input ".inp" file if another refinement is run.
7. Modify the input parameters to get the best refinement possible (determined by the Rwp factor) and run the refinement again.
8. Once it is not possible to further improve the refinement, modify the input file in such a way as to save also the results as a .xyd file, which can be read by Origin software.
9. Run the refinement for one last time and export the .xyd file.
10. Find the information about crystallite size and phase composition in the refinement result (there is an option to run the refinement to provide this information) and note it down.
11. Process the data in data analysis software and make the final figures.
    NOTE: This study used Origin software.

Results

The temperature during milling was recorded using Project SAV 1.0 software and plotted accordingly. Figure 3 demonstrates the changes in temperature with milling time. For the Cu_1.8S samples (Figure 3A), the ignition times fall within the range of 0-0.6 s. In the sample Cu_1.8S-1, the MSR occurred before temperature data collection began. Therefore, when performing the two next experiments (Cu_1.8S-2 and 4), data collection was sta...

Discussion

Mechanically induced self-propagating reactions (MSR) are an immediate transformation of precursors into products via an exothermic combustion-like process activated by mechanical action (similar to self-heat sustaining reactions where similar processes are activated by heat). The occurrence of MSR can often be identified by changes in the physical appearance of the product, a distinct smell at the moment of the reaction, or a scratching sound from the milling jar. However, empirical evidence suggests that these sen...

Materials

Name	Company	Catalog Number	Comments
Copper	Pometon, Germany	7440-50-8	Red powder
D8 Advance diffractometer	Bruker, Germany	M88-E03036	X-ray instrument
DiffracPlus Evaluation package release	Bruker, Germany	DOC-M85-EXX002	Diffraction analysis software
Etaben	Mikrochem, Slovakia	64-17-5	solution
Jedit	Open Source software		Programmer's text editor
Project SAV 1.0			Software developed to record data from in situ temeprature monitoring
Pulverisette P7 planetary mill	Fritsch, Germany	07.5000.00	The milling device, utilized in the synthesis of Cu1.8S and SnSe
Selenium	Acros Organic, Germany	7782-49-2	Gray powder
Sulfur	Sigma Aldrich, Germany	7704-34-9	Yellow powder
Tin	Merck, Germany	7440-31-5	Gray powder
Topas Academic	Coelho Software		General non-linear least squares software driven by a scripting language. Its main focus is in crystallography, solid state chemistry and optimization.