The influence of porogen dispersion on the structure and permeability of highly porous material from nickel nanopowder

Introduction

Porous materials are used in many industries. The porous structure, typically regarded as a defect in structural materials, provides unique properties that can be utilized for specific purposes [1]. Highly porous materials can be effectively used as electrodes [2], filters that separate impurity particles larger than the pore size [3–5], and are often employed as thermal insulators [6; 7]. Another application of porous materials is in biocompatible implants [8]. The relatively high internal surface area makes highly porous materials excellent catalysts [9; 10].

Depending on the specific application and the required porous structure, various fabrication methods can be employed to produce such materials, including partial sintering, the use of temporary porogens, direct foaming, and others. In the first case, the powder material is sintered in such a way that pores remain between the particles [11; 12]. This is due either to too low temperature and duration of sintering or to the low density of the initial green sample. In the second method, the added porogens decompose into volatile components or are washed out of the material during its production. The porous structure is controlled by appropriately selecting the porogenic substances. For porous materials produced using dispersed porogens, the shape and size of the pores depend on the shape and size of the porogen particles, while porosity is controlled by the quantitative content of the porogen [13]. This methodology allows for higher porosity values compared to the partial sintering technique. The approaches applied to the fabrication of highly porous materials from powders of various natures with the addition of temporary porogens are seen by the authors as promising, as they enable wide-ranging control over porosity and pore size in the resulting material.

It is important to note that simply having a highly porous structure is insufficient for certain applications. For filters and catalysts, it is necessary to create a porous material with a high proportion of open interconnected pores. This ensures good permeability, an important property for ensuring the reliable operation of the products in which they are used. Permeability is defined as the coefficient that relates the pressure gradient to the flow rate of the medium passing through the sample. It depends on the porous structure and can vary sharply with changes in the pore size distribution or the spatial arrangement and shape of the pore channels [10; 14–17]. It should be noted that a high porosity value does not always indicate good permeability.

In many studies where the authors create a porous material and investigate its structure, insufficient attention is paid to this parameter. However, some researchers provide permeability data without a thorough analysis of their relationship with the morphology of the porous space. Most studies focusing on permeability examine the flow of media through porous structures governed by Darcy’s law or Forchheimer’s law [18–25]. These laws are phenomenological and do not contain any condition describing the influence of the material’s microstructure. Therefore, researchers face the pressing task of finding ways to accurately assess permeability based on models developed considering the material’s microstructure parameters and allowing for predictions of permeability levels [26; 27]. Existing models do not fully account for all the features of the porous structures of modern materials, and creating new models requires a significant amount of experimental data on the relationship between permeability and various structural characteristics. Consequently, to better understand the processes, experimental studies are necessary to identify the connection between the structural features of the porous material and its permeability.

The aim of this work was to establish the influence of the porogen particle size and its volume fraction on the porosity and permeability of the nickel nanopowder material produced through pressing and sintering.

Research methodology

Nickel nanopowder with an average particle size of less than 100 nm, produced by wire explosion technology, was used as the starting material for the production of porous samples. This powder contained a small amount of larger spherical particles, up to 3 µm in size, which is a characteristic and drawback of this method for obtaining nanopowders. Ammonium carbonate (NH₄)₂CO₃ powders were used as the porogen. To study the influence of the volume fraction and dispersion of the porogen on the structure and permeability, ammonium carbonate powders with particle sizes of d = 40÷63, 100÷160, 200÷250, and 315÷400 µm were selected.

To determine the effect of the particle size distribution of the porogen on the structure and permeability, additional mixtures of the porogen were prepared, using particles from two size ranges: 40–50 and 315–400 µm, as well as 100–125 and 315–400 µm. For simplicity, these porogen powders will hereafter be referred to as “bidisperse”. In each mixture, the amounts of both fractions of the powders were varied in steps of 25 % – from 100 % content of the powder with d = 40÷50 µm (or 100÷125 µm) to 100 % content of the powder with d = 315÷400 µm.

The production of porous nickel materials consisted of three technological operations: preparing the initial mixtures of nickel nanopowder with the porogen, compacting the green samples, and sintering them. The volume fraction of the porogen in the initial mixtures was 60, 80, 85, and 88 %, with a compaction pressure of 300 MPa. The samples were produced by uniaxial compaction on a hydraulic press (Knuth, Germany) in a split mold with a diameter of 13.6 mm. The height of the compacts before sintering was 10 mm. To remove ammonium carbonate, the compacts were heated in an argon flow to a temperature of 100 °C at a rate not exceeding 1 °C/min. The optimal sintering temperature for the nickel nanopowder was determined to be 550 °C, with a sintering time of 120 min. The heating rate to the sintering temperature did not exceed 2 °C/min, which was necessary for the slow removal of the decomposition products of the porogen. Higher heating rates resulted in structural defects in the samples, such as microcracks. The thermal treatment of the samples was conducted in a tube furnace (MTI GSL1500X, USA).

Porosity was measured using the hydrostatic weighing method, with a relative error not exceeding 0.6 %. The permeability of the obtained porous nickel samples was determined using a method based on Darcy’s law. For this, while a liquid flowed through the sample, the pressure drop across its ends and the flow rate of the liquid – determined by the known volume of liquid passing through the sample over a fixed period – were recorded. The study was conducted by passing distilled water under pressure at room temperature. The pressure drop across the tested samples varied from 0 to 0.02 MPa, with the value recorded using a digital manometer (DM5002M, Manotomy JSC, Russia) with an allowable measurement error of 2·10^–5 MPa. The relative error in the permeability measurement did not exceed 10 %.

Results and discussion

Fig. 1 presents the microstructure of the fracture surfaces of samples with porosities of 79.3 and 88.5 %, sintered in a hydrogen atmosphere at a temperature of 550 °C. The volume fraction of the porogen in the mixture from which these samples were pressed was 80 and 88 %, respectively. Scanning electron microscopy revealed that, due to the thermal decomposition of the porogen, a pore structure was formed, which can be considered a replica of the removed porogen, with some modification in their shape and size as a result of compaction and sintering. Due to the high activity of the nanopowders, sintering was conducted at a relatively low temperature, resulting in samples with sufficient strength necessary for further investigation of their permeability.

In samples with an initial porogen content of 88 %, a significant number of thin walls, with a thickness of no more than 1–3 µm, were observed, featuring “windows” formed at the points of contact between porogen particles as well as from the escape of decomposition products. The small amount of nickel powder present in the framework of the highly porous material apparently defined a “lace-like” structure in these walls, characterized by numerous smaller holes or voids compared to the windows. Moreover, the smaller the porogen used, the more pronounced this structure became.

The influence of the volume fraction of the porogen and its dispersion on the porosity (P) and permeability (K) of the sintered material was investigated. Fig. 2 presents the dependencies of the porosity of the sintered material on the particle size (d) of the porogen for samples in which the volume fraction of the porogen was 65, 80, 85, and 88 %. It is evident that the value of P increases with increasing values of d. When using a porogen with d > 100 µm, the porosity of the sintered material equals or exceeds the expected value, which is equal to the volume fraction of the porogen in the initial mixture. For d = 40÷63 µm, the value of P was lower than expected, except for samples with a porogen volume fraction of 65 %. The closed porosity of all materials did not exceed 1 %. The porogen (NH₄)₂CO₃ decomposes during sintering at temperatures below 100 °C. This leads to the release of pore space and the formation of channels that connect the pores to the free surface of the sample. The interconnection of pores in such materials determines the high proportion of open porosity and their permeability.

The dependence of permeability on the volume fraction of the porogen and its dispersion was investigated (Fig. 3). It was found that as the volume fraction of the porogen increases, the permeability also rises. Additionally, this increase is achieved through the enlargement of the porogen particle size while maintaining a constant volume fraction in the initial powder mixture. For example, with a porogen content of 65 vol. %, the permeability (K, 10^–12) increases from 0.1 to 0.4 m², at 80 vol. %, it rises from 1 to 2.9 m², at 85 vol. %, from 1.8 to 4.6 m², and at 88 vol. %, from 3.9 to 8.4 m².

Fig. 4 illustrates the dependence of permeability on the porosity of the material for samples obtained using porogen powder with particles of a specified size range.

Using the compaction and sintering regimes described above, samples were obtained from mixtures of nickel nanopowder and bidisperse porogen, with the latter fixed at 85 vol. %. Data from Fig. 5 show that the samples contain pores corresponding to the sizes of the particles of the porogen used – large pores from particles with diameters of 315–400 µm and small pores from particles with diameters of 40–50 µm or 100–125 µm (Fig. 5, a, b). No inhomogeneity in the pore distribution within the volume of the samples was detected.

It is noteworthy that in samples containing the porogen with smaller particle sizes (40–50 µm), there was a greater number of windows on the surfaces of the larger pores. These windows are comparable in size to the smaller porogen particles and were likely formed due to their contact with the larger ones. It can be hypothesized that a greater number of windows will provide better permeability in this series of samples, along with the high permeability achieved by using porogen with particle sizes of 315–400 µm. However, according to the results of the study (Fig. 6), the samples obtained from mixtures with smaller porogen particles (40–50 µm) exhibited lower permeability compared to those using powders with diameters of 100–125 µm. The content of larger porogen particles (315–400 µm) ranging from 0 to 50 % did not lead to a significant change in the value of K: for samples made with porogen of 40–50 µm, the permeability was (1.3±0.1)·10^–12 m², while with 100–125 µm it was (2.2±0.2)·10^–12 m². Further increasing the proportion of 315–400 µm particles in the porogen resulted in an increase in K to 4.6·10^–12 m².

Conclusions

As a result of the conducted studies on the structure and permeability of the obtained porous materials, the following conclusions were established.

1. The permeability of highly porous materials made from nickel nanopowders increases with both the volume fraction of the porogen, ranging from 60 to 88 %, and the particle size of the porogen. The maximum permeability achieved was 8.4·10^–12 m² in a sample with a porosity of 88.5 %.

2. Using a bidisperse porogen facilitates smoother regulation of the permeability in nickel nanopowder materials. As the proportion of larger particles (315–400 µm) in the porogen powder increases, the samples exhibit enhanced permeability. The lowest permeability was recorded when only fine porogen was used (e.g., with particle sizes of 40–50 µm or 100–125 µm). When up to 50 vol. % of larger particles is added, the permeability varies by approximately 10 %. A notable increase in permeability, reaching 4.6·10^–12 m², is observed at a 100 % volume fraction of porogen particles sized 315–400 µm in the initial mixture.