Size-Controllable Synthesis of Cu Nanoparticles via Hydrogen Reduction at Low Temperatures

Abstract

The general polyol methods used for synthesizing copper nanoparticles (CuNPs) usually require high temperatures to ignite the reduction reaction, which leads to the difficulty of controlling the particle size. Herein, a new polyol method for the size-controllable synthesis of CuNPs at low temperatures by adopting high-pressure hydrogen as the reductant is verified. It is proven that hydrogen is capable to reductively produce CuNPs at temperatures as low as 120 °C. The size of the CuNPs can be controlled between 19.29 nm and 140.46 nm by adjusting the hydrogen pressure, the reaction temperature, and duration. An empirical relationship between temperature and particle size is proposed. This work verifies a low-temperature strategy to synthesize nanoparticles with good size-controllability.

1. Introduction

The development of advanced power electronics based on third-generation semiconductors (e.g., SiC, GaN) poses great demand for high-performance electronic packaging materials [1]. Pastes based on Cu nanoparticles (CuNPs) are a group of ideal packaging materials owing to their high surface activity, low sintering temperature, and good electrical/thermal conductivity [2,3,4]. Cu-based pastes composed of CuNPs with optimal size distribution tend to show the closer packing of particles and, thus, better performance [5]. Therefore, CuNPs with controllable sizes are desired.

The synthesis of CuNPs has been extensively studied using physical and chemical approaches. Although the physical methods are advantageous for mass production and cost-effectiveness [6,7,8], they face significant challenges in achieving precise control over particle size distribution and morphological uniformity. Chemical synthesis strategies, including liquid-phase reduction [9,10] and polyol-mediated processes [11], are particularly noteworthy for their ability to regulate nucleation and growth kinetics, allowing for the controlled modulation of particle dimensions. Nonetheless, the liquid-phase reduction method commonly uses toxic inorganic reductants, such as hydrazine hydrate and sodium borohydride, which poses substantial environmental concerns. The polyol method offers the advantage of low toxicity, with polyols serving both as solvents and reductants in the reaction system [12]. Therefore, the polyol method has the potential for industrial-scale fabrication.

Polyols, when compared to inorganic reductants, often exhibit weaker reducibility and slower reaction rates. To overcome these limitations, the most efficient strategies include elevating the reaction temperature and increasing the pH of the system [13,14]. However, raising the reaction temperature or pH accelerates the growth kinetics of the Cu nuclei, leading to the rapid growth of the CuNPs, which brings about the difficulty of controlling the size. For this concern, an improved polyol method operating at low-reaction temperatures needs to be proposed to fabricate size-controllable CuNPs.

In this work, the size-controllable synthesis of CuNPs was achieved at low temperatures by adopting high-pressure hydrogen as the reductant in a polyol (ethylene glycol) system. CuNPs can be fabricated at temperatures as low as 120 °C in the reaction system. The reduction effect of hydrogen was investigated. The size dependence of CuNPs on hydrogen pressure, the reaction temperature, and duration was studied using X-ray diffraction (XRD) and transmission electron microscopy (TEM). This work verifies a low-temperature polyol strategy to synthesize CuNPs with good size-controllability.

2. Materials and Methods

2.1. Materials

Copper chloride (CuCl₂, AR) was purchased from Shanghai Aladdin Biochemical Technology Co., Ltd. (Shanghai, China); sodium hydroxide (NaOH, AR) and anhydrous ethanol (AR) were purchased from Sinopharm (Shanghai, China); and ethylene glycol (EG, AR) and polyvinylpyrrolidone (PVP, K29-32) were purchased from Shanghai Macklin Biochemical Co., Ltd. (Shanghai, China).

2.2. Preparation and Characterization of CuNPs

PVP was first dissolved in EG at 70 °C to produce a 20 wt.% solution; 2.9 mmol of CuCl₂ was then dispersed in 50 mL of the solution to obtain solution A. In total, 14.7 mL of EG dissolved with 0.5 mol/L NaOH was added dropwise to solution A, which yielded the blue colloid B. Colloid B was transferred into a high-pressure reactor for producing CuNPs at various conditions, i.e., temperatures from 120 to 220 °C, H₂ pressure from 1.0 to 1.8 MPa, and reaction durations from 0.5 to 1.5 h. The selection of the condition was decided by referring to previous studies. Upon cooling the reactor to room temperature, the products were separated by centrifuging at 10,000 rpm for 15 min. The obtained solids were washed at least three times with anhydrous ethanol and deionized water alternately, followed by drying in a vacuum freeze-dryer for 12 h, and they were finally milled to obtain CuNPs.

The microstructure of the CuNPs was characterized by a Talos F200X field emission transmission electron microscope (TEM, Thermo Scientific, Waltham, MA, USA). The diameters of 100 particles were measured to give the average size and the standard deviation by adopting the normal distribution model. The crystal structure of the particles was characterized by DX2700BH X-ray diffraction (XRD, Hao Yuan, Dandong, China).

3. Results and Discussion

3.1. Verification of H₂ as the Reductant

Blue colloid B was subject to a reaction at 130 °C for 0.5 h in the atmosphere of 1.6 MPa H₂ or N₂. As shown in Figure 1A, an XRD investigation is conducted to verify the phase constitution of the product obtained from reacting in N₂. A series of (111) and (200) crystal planes of Cu₂O are identified at 2θ = 36.25 and 41.97, implying the formation of Cu₂O [15]. TEM was used to further characterize the microstructure. As displayed in Figure 1C, the obtained nanocrystalline shows a plane spacing of 0.243 nm, which corresponds to the (111) crystal plane of Cu₂O [16]. Such results suggest that EG can only reduce Cu²⁺ to Cu⁺, rather than Cu, due to its insufficient reducibility at the relatively low temperature of 130 °C. In contrast, the XRD pattern shows a set of peaks characteristic of face-centered cubic Cu for the product obtained in the H₂ atmosphere, which is further verified by the typical (111) crystal plane of Cu with a plane spacing of 0.208 nm in the TEM micrograph (Figure 1D) [17]. The average particle size is measured to be 21.75 ± 10.68 nm (Figure 2A). Therefore, H₂ plays a key role in reducing the Cu²⁺ precursor to Cu at the condition of 130 °C.

To further verify whether H₂ acted as the reductant to obtain CuNPs at 130 °C, a two-step experiment was conducted by replacing the H₂ atmosphere with N₂ halfway through the reaction. Initially, the reactor was filled with H₂ at 1.6 MPa and proceeded at 130 °C for 0.5 h. In the second step, H₂ was replaced by N₂, and the process continued for an additional hour. The size of the generated particles is statistically measured by referring to the TEM image (Figure 2B). The average size is estimated to be 21.93 ± 8.43 nm, which is almost the same compared to the value of 21.75 ± 10.68 nm obtained without further reaction in N₂. Such phenomena demonstrate that the reduction reaction ceases after expelling H₂, confirming that the formation of CuNPs is attributed to reductive H₂ at the temperature of 130 °C. The reaction mechanism is schematically illustrated in Figure 3. With the addition of alkaline, Cu²⁺ initially reacts with OH⁻ to form the Cu(OH)₂ precipitate and the complex [Cu(OH)₄]²⁻, as described by Steps 1 and 2. Subsequently, [Cu(OH)₄]²⁻ further reacts with EG and forms [Cu(OCH₂CH₂O)₂]²⁻, as shown in Step 3. Upon heating, this complex decomposes into Cu₂O, as depicted in Step 4 [18]. Finally, Cu₂O is reduced to metallic Cu by H₂ at the adopted temperature (Step 5). The chemical equations for each step are shown as Equations (1)–(5):

Cu²⁺ + 2OH⁻ = Cu(OH)₂ ↓

(1)

Cu(OH)₂ + 2OH⁻ = [Cu(OH)₄]²⁻

(2)

[Cu(OH)₄]²⁻ + 2HOCH₂CH₂OH = [Cu(OCH₂CH₂O)₂]²⁻ + 4H₂O

(3)

2[Cu(OCH₂CH₂O)₂]²⁻ = CH₃COCOCH₃ + 2(OCH₂CH₂O)²⁻ + H₂O + Cu₂O

(4)

Cu₂O + H₂ = 2Cu + H₂O

(5)

3.2. Control of the CuNP Size

3.2.1. H₂ Pressure

H₂ is used as the reductant in the reaction, and the Cu₂O species, decomposed from the complex [Cu(OCH₂CH₂O)₂]²⁻, is reduced to metallic Cu. Since the solubility of H₂ in the EG solution is proportional to the pressure of gaseous H₂ according to Henry’s law [19], the concentration of the dissolved H₂ is adjusted by applying H₂ pressure in the range of 1.0–1.8 MPa. The reactions run at 130 °C for 0.5 h. The sizes of the CuNPs synthesized at different pressures were characterized, and the results are shown in Figure 4. As summarized in Figure 4F, the size first decreases from 29.18 ± 13.25 nm for 1.0 MPa to 22.80 ± 9.32 nm for 1.4 MPa and then increases to 28.17 ± 15.60 for 1.8 MPa. The size of the CuNPs shows a V-shaped trend with pressure, probably owing to the switch of nucleation-dominant processes to growth-dominant ones. At low H₂ pressures (1.0~1.2 MPa), the nucleation processes govern the formation of the CuNPs. Few H₂ molecules are dissolved in the EG solution, which leads to a low nuclei rate. Owing to the abundant supply of Cu²⁺, a few Cu nuclei undergo free growth, which results in large CuNPs. As the H₂ pressure rises, the number of Cu nuclei increases, leading to an intensified competition for Cu²⁺ supply. As a result, Cu²⁺ is insufficient to support the free growth of these particles, which causes a decrease in the CuNP size. At higher pressures of above 1.4 MPa, the growth processes become dominant. Higher H₂ pressure increases the concentration of reductant H₂ in the EG solution, thus increasing the reduction potential. Therefore, the growth processes of CuNPs are promoted, which finally leads to a larger size.

3.2.2. Reaction Temperature

Temperature is a key factor affecting particle size. The temperature range of 120–220 °C was chosen to explore the effect of temperature on the CuNP size under the H₂ pressure of 1.4 MPa for 0.5 h. TEM results suggest that the sizes of the CuNPs synthesized vary with temperature, as displayed in Figure 5. As shown in Figure 5A, the low temperature of 120 °C can generate CuNPs with an ultrafine size of 19.29 ± 7.79 nm, which suggests that the nucleation process can be activated at this temperature. With the increase in the reacting temperature, the reductivity of H₂ is stronger, which leads to the speeding up of the conversion from Cu²⁺ to Cu. The growth rate of the CuNPs thus increases. Therefore, the sizes of the CuNPs present an increasing trend with temperature, as summarized in Figure 6. The size of the CuNPs can be empirically fitted by the relationship of d = 0.00902 (T−103.104)² + 17.345 (R² = 0.999), where d is the diameter of the CuNPs and T is the temperature. The squared relationship between d and T indicates that the growth rate of the particles becomes faster at high temperatures, which is particularly obvious at 220 °C. To reveal the underlying mechanism, a supplementary experiment was conducted at 220 °C for 0.5 h in a N₂ atmosphere. The results show that CuNPs are obtained with a size of 197.37 ± 95.57 nm (Figure 7), which indicates that EG is capable of acting as the reductant at 220 °C, consistent with the reported work [20]. Therefore, the squared relationship is probably ascribed to the involvement of EG as the secondary reductant at high temperatures.

3.2.3. Reaction Duration

The size of the CuNPs is also related to the reaction duration, as shown in Figure 8. At a low temperature of 130 °C, the kinetics for both diffusion and reduction are relatively slow, which leads to the gradual growth of the CuNP size within the reaction duration. The average size of the CuNPs increases slightly from 24.10 ± 11.91 nm to 39.48 ± 18.67 nm when the reaction duration extends from 0.5 h to 1 h. Further elongation of the duration to 1.5 h only results in a particle size of 49.43 ± 30.40 nm. The insensitivity of the particle size with duration implies a large operation window to control the size of the CuNPs.

3.3. Outlook of the CuNPs

By adopting H₂ as the reductant in a polyol system, the reaction temperature to obtain CuNPs is reduced to 120 °C, which is significantly lower than the reaction in EG. The size of the CuNPs increases with temperature and duration, which is consistent with reported works [15,21]. As for H₂, the present work showed an unexpected V-shaped trend in CuNP size with H₂ pressure, which is distinct from typical synthesis methods. However, this method still suffers from insufficient production efficiency due to the low concentration of reactants, which should be further optimized in subsequent studies. The present CuNPs are expected to be used as the conductive phase to fabricate packaging pastes for the high-density interconnection of advanced power electronics. Commonly, a high packing density is desired for the conductive phase to optimize the connection. For conductive phases that consist of, for example, two kinds of particles with different sizes, the packing density of the conductive phase relies on the relative size of the two particles. The present size-controllable CuNPs enable optimal size matching between two kinds of Cu particles with different sizes. Therefore, the more densely packed the Cu particles, the higher the performance connection of the power electronics.

4. Conclusions

A new method has been developed for the size-controllable preparation of CuNPs using EG as the solvent and H₂ as the reductant. The reduction effect of H₂ in producing CuNPs is proven by replacing H₂ with N₂ halfway through the reaction. The particle size can be controlled by adjusting the H₂ pressure, reaction temperature, and time. The particle size shows a V-shaped trend in the range of 1.0 MPa to 1.8 MPa, reaching a minimum of 22.80 ± 9.32 nm at 1.4 MPa, probably owing to the switch from nucleation-dominant processes to growth-dominant ones. The size of the CuNPs increases from 19.29 ± 7.79 nm to 140.46 ± 89.74 nm with the temperature increasing from 120 °C to 220 °C and from 24.10 ± 11.91 nm to 49.43 ± 30.40 nm with the reaction time extending from 0.5 h to 1.5 h. An empirical relationship between temperature (T) and particle size (d) is established, expressed as d = 0.00902 (T − 103.104)² + 17.345. This work proposes a low-temperature strategy to synthesize CuNPs with controllable sizes and verifies H₂ as a potential regent to fabricate nanoparticles in a moderate way.