Journal of Materials Science & Technology  2020 , 37 (0): 114-122 https://doi.org/10.1016/j.jmst.2019.06.017

Research Article

Improvement on compressive properties of lotus-type porous copper by a nickel coating on pore walls

Hao Dua*, Chuanyu Cuiab, Housheng Liuab, Guihong Songc*, Tianying Xionga

a Division of Surface Engineering of Materials, Institute of Metal Research, Chinese Academy of Sciences, Shenyang, 110016, China
b School of Materials Science and Engineering, University of Science and Technology of China, Shenyang, 110016, China
c School of Materials Science and Technology, Shenyang University of Technology, Shenyang, 110870, China

Corresponding authors:   ∗Corresponding authors.E-mail addresses: hdu@imr.ac.cn (H. Du), ghsongsut@126.com (G. Song).∗Corresponding authors.E-mail addresses: hdu@imr.ac.cn (H. Du), ghsongsut@126.com (G. Song).

Received: 2019-01-1

Revised:  2019-06-25

Accepted:  2019-06-26

Online:  2020-01-15

Copyright:  2020 Editorial board of Journal of Materials Science & Technology Copyright reserved, Editorial board of Journal of Materials Science & Technology

More

Abstract

The aim of this work is to understand the effect of a thin coating on the compressive properties of the porous metal. In our work, the uniaxial compressive behavior and the energy absorption properties of the lotus-type porous copper deposited with Ni coatings with thickness from 3.9 to 4.8 μm on pore walls were investigated. It is found that the Ni coating on pore walls shows a clear enhancement effect on compressive properties of the lotus-type porous copper, in which the specific yield strength and the energy absorption per unit mass at densification strain increase from 5.27 to 7.31 MPa cm3 g-1 and from 11.50 to 18.21 J g-1 with the Ni coating, respectively. Furthermore, the enhancement appears to be insensitive to the coating thickness. It is considered that the resistance of the interface between the nickel coating and the pore walls to the dislocation slip plays an important role in the improvement on compressive properties of the lotus-type porous copper.

Keywords: Lotus-type porous copper ; Coating ; Pore walls ; Compressive properties

0

PDF (2769KB) Metadata Metrics Related articles

Cite this article Export EndNote Ris Bibtex

Hao Du, Chuanyu Cui, Housheng Liu, Guihong Song, Tianying Xiong. Improvement on compressive properties of lotus-type porous copper by a nickel coating on pore walls[J]. Journal of Materials Science & Technology, 2020, 37(0): 114-122 https://doi.org/10.1016/j.jmst.2019.06.017

1. Introduction

Porous and foamed metals can be applied as the engineering materials with excellent properties for dual attributes of both structural and functional characteristics [[1],[2]]. In addition to a reduction in density, the main benefit of introducing pores in metals is some new functions provided by the large surface area, such as tunnel, supporter for applications in the fields of separation, filtration, energy absorption, heat exchange, sound/vibration absorption, biological transplant, catalyzing reaction, and electrochemical process [[3],[4]]. However, the pores in porous and foamed metals deteriorate the mechanical properties inevitably for not only a reduced constraint at the pore surface by an absence of metals but also a stress concentration around pores [[5],[6]].

Various studies have been conducted to improve the mechanical properties of porous and foamed metals. Among them, two main approaches have been taken. One is to control their pore structural parameters. Andrews et al. [7] investigated the effect of the ratio of specimen size (L) to cell size (d), L/d, on mechanical properties of foamed aluminum. They revealed that Young’s modulus and compressive yield strength increased with an increase in L/d. In the case of lotus-type porous metals, it has been demonstrated that the mechanical properties depend on porosity, pore diameter, and arrangement of pores [[6],[8]]. The compressive properties in the direction parallel to the pore direction decrease almost linearly with increasing porosity [[9],[10]]. However, the improvement of mechanical properties by controlling pore structure is generally associated with a sacrifice on porosity (increasing density) and specific surface. In fact, in some cases, it is more important to keep or reduce the density than to increase the strength of the porous and foamed metals. Another approach is the addition of second treatment, such as equal-channel angular extrusion (ECAE) process and depositing a film or a coating. Suzuki et al. [11] reported that Vickers hardness of lotus-type porous copper can be increased through the ECAE process because of the work hardening effect. However, the pore morphology of the lotus-type porous copper changed too much by the extrusion. It has been demonstrated that mechanical properties of the foamed metals could potentially be increased by thin films [12] or coatings such as Ni [13], Ni alloy [[14],[15]], Al2O3 [16] in the thickness of several tens to several hundred of micrometers. In the case of thick coatings, the mechanical property improvement can be explained by a composite enhancement effect that the coating is regarded as an ultra-high strength skin encapsulating a foamed metal [[17],[18]]. However, the obvious variation of pore structure including porosity (density) pore size must be considered on the foamed metals by the thicker coatings in some applications. From a practical standpoint, a thin coating for an improvement on the mechanical properties for porous and foamed metals is more promising.

According to the available reports, Ni coating is generally known to provide high strength, good wear and corrosion resistance for foamed metals. So, similar merits can also be expected for lotus-type porous metals. It has been achieved by the authors that a thin Ni coating can be deposited on pore walls of lotus-type porous copper uniformly by electroplating [19]. It is interesting to find that the thin Ni coating can increase the compressive strength and energy absorption about 30%-50% for the porous copper. So far, only limited data are available, further information about the coating effects on the pore structure, the specific strength and the energy absorption efficiency of the lotus-type porous copper are not clear. Furthermore, the mechanism for the improvement was not discussed in detail.

The main purpose of this work is to further understand the enhancement effect of a thin Ni coating on the compressive properties of lotus-type porous copper. To achieve this goal, the effects of the coating thickness on the porosity and the pore size of the lotus-type porous copper were analyzed first. Then, the effects of the Ni coatings on the compressive strength, the energy absorption and the energy absorption efficiency of the lotus-type porous copper were evaluated, the mechanism for the enhancement effect is discussed.

2. Experimental

2.1. Fabrication of the lotus-type porous copper specimens

A lotus-type porous copper was employed as the substrate. The porosity, pore density, average pore size, and the average pore length of the substrate are 47.1%, 180 cm-2, 438.5 μm and 9.2 mm, respectively. Among these parameters, the porosity was measured by both the weighing method and the image analyzing method, and the others were obtained only by the image analyzing method using a SISC Image Analyzing software (KYKY Technology Development Ltd., China). In the analyzing process, the edge detection filter and the intensity threshold in the software were adjusted to form a sharp binary image for determining pore size, pore length and pore density. The porous Cu specimens with the size of 30 mm × 15 mm × 6 mm were cut by the electric discharge machine (DK 7763, Longhao Digital-Controlled Machine Corp., China). The specimens were electroplated with Ni coatings on the pore walls with thickness from 3.9 to 4.8 μm by adjusting the deposition time. The details on the fabrication of the porous copper and deposition of the Ni coating as well as their structural characteristics have been introduced in the other references [[19],[20]].

2.2. Evaluation on compressive properties and microstructure characterization

The specimens, both the coated and the uncoated, were cut into the dimensions of 5 mm × 5 mm × 6 mm by the electric discharge machine for compressive measurement, considering the size effect on the compressive properties [21]. Compressive tests were performed on the universal testing machine (Model 5582, Instron Co. Ltd., USA) at room temperature. The contact surfaces between the specimen and the two supporting plates were pretreated with a lubricant spray to minimize friction. The cross-head speed was set to be 0.1 mm min-1, corresponding to a strain rate of 2.78 × 10-4 s-1 by the specimen thickness. The compressive yield strength (0.2% offset strength) was determined and reported by the testing machine automatically from the stress-strain curve. The mean values and the standard deviations of the compressive yield strength were obtained from the results tested on 5 specimens in each case. All the specimens were compressed to the strain of 80%, which is well beyond the densification strain (generally 50%-70%) for the lotus-type porous copper.

The energy absorption per unit volume (W), the energy absorption per unit mass (WM) and the energy absorption efficiency (E) of each specimen were calculated according to the stress-strain curve by using the following equations [[22],[23]]:

$W=\int_{0}^{1}\sigma(\varepsilon)d\varepsilon$ (1)

$W_{M}=\frac{\int_{0}^{1}\sigma(\varepsilon)d\varepsilon}{\rho}$ (2)

$E=\frac{\int_{0}^{1}\sigma(\varepsilon)d\varepsilon}{\sigma_{max}l}$ (3)

where σ(ε) is the compressive stress, ε is the strain, l is the maximum strain, ρ is the density of the measured specimen, and σmax is the maximum compressive stress corresponding to l. In the cases of W and WM, the values at densification point which is defined as the strain corresponding to the end of the stress plateau are regarded as the energy absorption capacity for each specimen.

In order to investigate the effects of the Ni coatings on the deformation behavior and deformation mechanism of the lotus-type porous copper, the compressive tests were also carried out by interrupting at the strains of 20%, 40% and 80% on both the coated and the uncoated specimens, then the pore walls of the compressed specimens were observed by the scanning electron microscope (SSX-550, Shimadzu Corp., Kyoto, Japan) to clarify the microscopic deformation behavior in both the copper matrix and the Ni coatings.

3. Results and discussion

3.1. Effect of the Ni coating on the structure of the lotus-type porous copper

Since the pore size of the foamed metals is generally bigger than 1 mm or even several millimeters, the variation of pore structure by coatings with several micrometers to several tens’ micrometers can be neglected. However, the pore size of lotus-type porous copper is only several hundred of micrometers, so, after the Ni coatings with thickness from 3.9 to 4.8 μm were deposited on the pore walls, it is necessary to understand the changes on the pore structure of the lotus-type porous copper, such as the porosity, the pore size, which are two important structural parameters influencing its mechanical, thermal, and energy absorption properties. With the pore density of 180 cm-2, the average pore size of 438.5 μm mentioned above, the pore surface of the lotus-type porous copper specimen (5 mm × 5 mm × 6 mm) used as substrate in this work is calculated as 371.95 mm2 by π×D × L×n (D, L, and n are diameter, length, and number of pores, respectively), assuming all pores are open or half open.

Assuming that the pore size and the inter-pore spacing are uniform, the lotus-type porous copper can be regarded as a structure with cylindrical pores distributed in a regular hexahedron structure [24], as shown in Fig. 1. In this schematic diagram, r is the radius of pores, L is half the distance between the neighboring pores, so the thickness of the pore wall is 2L-2r, and the side length and the area of the unit regular hexahedron are $\frac{2L}{\sqrt{3}}$ and $2\sqrt{3}L^{2}$, respectively. Thus, the porosity P can be expressed as $\frac{\pi r^{2}}{2\sqrt{3}L^{2}}$, by which the effect of the coating thickness on the calculated porosity is listed in Table 1. It can be found that the coating thickness is about 2.19% (4.8 μm, thickest) and 1.78% (3.9 μm, thinnest) of the pore radius. After depositing the Ni coatings, the pore radii decrease a little, while distances between two neighbor pores keep the same for the lotus-type porous copper. Furthermore, the calculated porosity of the lotus-type porous copper decreases to 45.06% in the case of the specimen with the thickest coating of 4.8 μm, and 45.44% in the case of the specimen with the thinnest coating of 3.9 μm, from the original value of 47.1%. The measured porosities by the weighing method also prove this point, as shown in Table 1. The difference between the calculated and the measured values on porosity is caused by the pores not being complete cylindrical in the lotus-type porous copper.

Fig. 1.   Image of the coated lotus-type porous copper specimen and its schematic diagram for the cross section of the pore structure.

Table 1   Effect of coating thickness on the average radius of pores, the half distance between neighbor pores and porosity for the lotus-type porous copper.

Coating thickness (μm)Average radius of pores (mm)Half distance between neighbor pores (mm)Calculated porosity (%)Measured porosity (%)
00.21930.304347.10
3.90.21540.304345.4445.65
4.00.21530.304345.4045.79
4.1021520.304345.3645.62
4.30.21500.304345.2745.51
4.40.21490.304345.2345.48
4.60.21470.304345.1545.52
4.80.21450.304345.0645.42
39.00.18030.304331.84

New window

According to the results mentioned above, it is indicated that the pore structure of the lotus-type porous copper does not change with the deposition of a thin coating on the pore walls. However, the variation on pore structure cannot be neglected if the coating thickness reaches several tens micrometers, e.g. as shown in Table 1, when the coating thickness is 39.0 μm, the porosity is only 31.84%.

3.2. Effect of the Ni coating on compressive behavior of the porous copper

Since the difference in the stress-strain curve among the coated specimens with different Ni coating thickness remains in experimental error, the lotus-type porous copper deposited with the Ni coating (thickness = 4.4 μm), which possesses the highest values on compressive property in this work, is chosen as an example, whose a compressive stress-strain curve is shown in Fig. 2 and compared with that of the uncoated sample as a reference. Unlike the compressive stress-strain curves of foamed metals [[25],[26]], no constant plateau stress appears for both the uncoated and the coated lotus-type porous coppers, but the compressive stress continuously increases as the deformation progresses, indicating that stress concentration does not occur around the pores during compression along the pore direction. The deformation of both the coated and the uncoated specimens can be divided into three regions, a linear region, an analogous plateau region characterized by a small slope and a densification region characterized by a relatively steep slope. It is hard to identify the densification strain point (onset of densification) σD precisely due to no abrupt transition between the analogous plateau region and the densification region. Formula εD=0.905-0.432×$\frac{\rho^{*}}{\rho S}$ (r2=0.87) (εD is the corresponding strain to the densification point, ρ* is the density of the porous metal, ρs is the density of the base metal that the porous metal is made of, and  r2 is the coefficient of determination of this equation) by Simone and Gibson [27] is employed, in which the densification point of 67% is chosen for both the uncoated and the coated porous copper, respectively.

Fig. 2.   Compressive stress-strain curve between the coated and the uncoated lotus-type porous copper (a) and the magnified part (b) in the strain range to 0.05.

It should be mentioned that there is about several megapascal difference in stress between the coated specimen and the uncoated specimen when the compressive strain is smaller than 3%, which is supposedly connected mainly with a prestretch from the Ni coating to the copper substrate at the interface. There should be an internal stress between the nickel coating and the pore wall, consisting of two main components: a thermal stress due to the difference in thermal expansion, and a structural stress due to the lattice mismatch (about 2.51% in lattice constant [28]) between the Ni coating and the Cu substrate. The coefficient of thermal expansion (CTE) of nickel (13.2 × 10-6 K-1 at 293 K) is lower than that of copper substrate (17.1 × 10-6 K-1 at 293 K) [29], thus the substrate contracts more than the surface coating, resulting in a tensile stress from the surface layer during the cooling process from the electroplating temperature (55 °C) to room temperature (15 °C). Basrour and Robert [30] also reported that there was residual stress smaller than 40 MPa between the copper substrate and electroplated nickel coating. The tensile stress in the surface layer will increase the magnitude of the compressive stress, especially at the beginning of the compressive process. It is also worth pointing out that several (periodic) stress drops with multiple serrations appear on the coated specimen when the strain is higher than 40%, as indicated by the arrows in Fig. 2. Such stress drop is caused by the appearance of cracking in the Ni coating, not the breakage or decohesion of the Ni coating from the pore walls, which will be shown and discussed in 3.4.

According to the stress-strain curves, the compressive yield strength and the energy absorption capacity are obtained for both the uncoated and the coated lotus-type porous copper, and the data are listed in Table 2. Considering the density being an important property of a porous material, specific compressive yield strength and specific energy absorption capacity are also calculated. It is shown that the yield strength and the energy absorption capacity per volume increase from 22.1 to 29.6 MPa (about 34% on growth rate) and from 48.2 to 73.7 MJ m-3 (about 53% on growth rate) by the Ni coating for the lotus-type porous copper. In the cases of the specific yield strength (yield strength per mass) and the specific energy absorption capacity (energy absorption capacity per mass), the values increase from 4.69 to 6.10 MPa cm3 g-1 and from 10.24 to 15.19 J g-1, respectively.

Table 2   Density, compressive strength and energy absorption capacity of both the uncoated and the coated lotus-type porous coppers.

ParametersDensity
(g cm-3)
Compressive strength (MPa)Specific compressive strength (MPa cm3 g-1)Energy absorption capacity per volume (MJ m-3)Energy absorption capacity per mass (J g-1)
Uncoated4.708122.1 ± 2.04.69 ± 0.4248.2 ± 3.010.24 ± 0.64
Coated-3.9 μm4.837228.2 ± 2.85.83 ± 0.5869.1 ± 3.714.29 ± 0.76
Coated-4.0 μm4.824729.3 ± 2.76.07 ± 0.5671.6 ± 3.914.84 ± 0.81
Coated-4.1 μm4.839828.4 ± 2.55.87 ± 0.5269.5 ± 3.514.36 ± 0.72
Coated-4.3 μm4.849629.4 ± 2.86.06 ± 0.5872.2 ± 4.014.89 ± 0.82
Coated-4.4 μm4.852329.6 ± 3.16.10 ± 0.6473.7 ± 4.115.19 ± 0.84
Coated-4.6 μm4.848728.8 ± 2.65.94 ± 0.5470.1 ± 3.514.46 ± 0.72
Coated-4.8 μm4.857629.2 ± 2.96.01 ± 0.6071.3 ± 3.914.68 ± 0.80

New window

Ni coating is beneficial to the compressive properties of lotus-type porous copper. Although there is a little increase in density of the porous copper, the Ni coating increases not only the compressive strength, energy absorption, but also the specific compressive properties, which are of great interest for lotus-type porous metals in lightweight structural applications. Furthermore, it is indicated that there is no coating thickness effect on the enhancement in this work.

3.3. Effect of the Ni coating on energy absorption with strain of the porous copper

The energy absorption per mass and the energy absorption efficiency of the uncoated and the coated lotus-type porous copper with the strain are calculated up to the strain of 80% and shown in Fig. 3. It is indicated that the energy absorption per mass of both the uncoated and the coated lotus-type porous copper increases with the increasing strain. The energy absorption is poor when the strain is at the linear elastic region for both specimens in Fig. 2, little energy is absorbed in this short region; while it increases significantly when the strain is higher than 10%, where the compression reaches the analogous plateau region, and higher compressive force is necessary to deform the porous copper (and the Ni coating) plastically; the energy absorption increases more significantly when the strain is more than 70%, corresponding to the densification region, in which much higher compressive force has to be applied because of the work-hardening effect. It is interesting to find that the Ni coating significantly improves the energy absorption per mass of the lotus-type porous copper for a given strain during the compression, and the improvement turns more significant with the increasing strain. Since the stress amplitude and the strain range at the plateau region or the analogous plateau region determine the amount of energy absorption of foamed metals [3] and lotus-type porous metals [[8],[10]], it is desirable to enhance stress amplitude and extend the analogous plateau stress region at the same time to increase the energy absorption for the lotus-type porous copper. From the stress-strain curve (Fig. 2), the Ni coating can enhance the stress amplitude during the whole compression process, especially at the analogous plateau stress region and the densification region for the lotus-type porous copper, which results in the improvement on energy absorptions per volume and per mass. Furthermore, the energy absorption per mass of the lotus-type porous copper deposited with a Ni coating reaches 19.8 J g-1 at the strain of 80%, much higher than that of foamed metals [23].

Fig. 3.   Relationship between energy absorption per mass (a), energy absorption efficiency (b) and strain for both the uncoated and the coated lotus-type porous copper.

In the case of the energy absorption efficiency, the Ni coating does not increase the efficiency for the lotus-type porous copper, the corresponding value of the coated porous copper is even lower than that of the uncoated porous copper when the strain is lower than 0.15 or higher than 0.7, which may stem from the steeper stress-strain curve mainly. In the case of foamed metals, there is generally a flat plateau region in the compressive stress-strain curve, in which the stress keeps almost at a constant value with the increasing strain [[25],[31]]. It has been accepted that the flatter and the longer the plateau region is in the stress-strain curve, the higher energy absorption efficiency the foamed metal possesses [32], which follows Eq. (3). In the case of lotus-type porous copper, there is only an analogous plateau region, in which the stress does not keep constant but increases gradually with the increasing strain. From the stress-strain curve of the coated lotus-type porous copper in this work, the analogous plateau region is between 0.15 and 0.7, while the stress-strain curve is steeper for the coated porous copper out of this range, resulting in a poorer energy absorption efficiency. However, the absorption energy efficiency is a little higher in the analogous plateau region although the stress-strain curve of the coated lotus-type porous copper is steeper than the uncoated lotus-type porous copper.

3.4. Mechanism for the enhanced compressive properties and energy absorption capacity of the porous copper

The results mentioned above reveal that a thin Ni coating on pore walls increases not only the compressive strength and the flow stress but also the energy absorption capacity of the lotus-type porous copper. The improvement on the compressive properties is attributed to the increased constraint on the pore walls, from the stress-free cut pore edges to a new interface between the pore walls and the Ni coating.

The first effect by the increased constraint is the composite effect. The porous metals can be regarded as a composite structure with the solid body such as walls and edges, and empty spaces such as voids and pores dispersed within a solid framework. The empty spaces do not contribute to the overall strength [33]. In the case of lotus-type porous metals, the simple rule of mixtures of empty pores and solid body can be applied. Following this consideration, the load-bearing capacity of a porous metal deposited with a coating is shared between the metal body and the coating [26]. In the case of the lotus-type porous copper deposited with a Ni coating, assuming a uniform strain within both the nickel coating and the copper, the mechanical properties can be estimated using a rule of mixtures approach considering the pore characteristics by:

$E_{s}=f_{Ni}E_{Ni}+(1-f_{Ni})E_{Cu}$ (4)

where Es, ENi and ECu are mechanical properties of the whole specimen, the nickel coating and the copper substrate, respectively; fNi is the fraction of nickel and 1- fNi is the fraction of copper in the composite. However, the contribution to the mechanical properties by the Ni coating should be low because the thickness of the Ni coating deposited on pore walls is much lower (3.9-4.8 μm) than that of copper pore walls (0.3 mm). Furthermore, the observed results show the compressive strength is not affected by the coating thickness, indicating there should be some other reasons for the improvement.

The second consideration is the obstacle effect exerted by the Ni coating or the interface between the coating and the pore walls on dislocation glide. The deformation mechanism in pore walls of the lotus-type porous copper is like that in the dense copper. It is generally accepted that dislocations in metal will slide along a proper surface, and the sliding surface in metal (especially with a higher toughness) will rotate from an unfavorable position to a favorable position under deformation. Porous metals possess larger free surface than the corresponding bulk metals, which results in a lower strength or mechanical properties as the dislocations escape off more easily from the free surface on pore walls during the deformation. However, the situation changes when a coating is deposited on the pore walls. The Ni coating influences both creation and motion of the dislocations over a significant portion, in which the coating acts as the barriers to the dislocations sliding out from the internal part of the porous metal body during deformation, the blocked dislocation may form pileups and lead to further resistance to the following dislocation, so higher stress and more energy are necessary to move the dislocations, which result in an enhancement not only in the mechanical properties such as strength but also in the compressive energy absorption. To prove this point, pore walls of both the uncoated and the coated specimens compressed to 20%, 40% and 80% were observed and are compared in Fig. 4, Fig. 5. The entire images of the inner surface of pore walls for both specimens are displayed. According to Fig. 4, a lot of slip lines cover the surface of the pore walls in the uncoated specimen, which propagate from the inner volume and exit the surface of the specimen to form surface. At the low strain (20%), conventional fine slip lines are clearly found. When the strain is up to 40%, slip deformation becomes concentrated on particular slip planes, forming thick slip lines as shown in Fig. 4(d). With the further increased strain, the dislocations become increasingly difficult to move. Thus, deformation occurs only on limited planes, resulting in the formation of large deformation bands as shown in Fig. 4(f). In the case of the coated specimen, no slip lines but a coating surface can be found, as shown in Fig. 5. It is assumed that dislocation will try to escape from the copper matrix through pore walls on the uncoated specimen during the compression. But in the case of the coated specimen, as observed in Fig. 5, although the stress is higher, a new interface (Cu/Ni interface) acts as a barrier or a sink for dislocations, thus prevents the dislocations from escaping and slipping out the pore walls as before. Furthermore, it is suggested the dislocations will form pileups [[34],[35]] beneath the Cu/Ni interface when they reach an obstacle such as a strong interface during plastic deformation, as shown in a schematic of a uniaxial compressive test on a specimen consisting of a copper substrate and a Ni coating (Fig. 6). Thus, dislocations in the coating or interface are subject to two types of stress, the externally applied stress and the internal stress imposed by the pileups. The former provides the driving force for further plastic flow, while the latter acts as resistance [36]. At last, the work hardening occurs in the deformed regions in both copper and nickel parts at the same time.

Fig. 4.   SEM images of pore walls after being compressed to 20% (a), 40% (c) and 80% (e) of the uncoated porous copper in low magnification and their corresponding high magnification images are shown in (b), (d) and (f), respectively.

Fig. 5.   SEM images of pore walls after being compressed to 20% (a), 40% (c) and 80% (e) of the coated porous copper in low magnification and their corresponding high magnification images are shown in (b), (d) and (f, g), respectively.

Fig. 6.   Schematic of a uniaxial compressive test of a specimen consisting of a copper substrate and a Ni coating.

The third consideration is the increased constraint stems from the crack effect, which appears at higher strain during deformation of the pore wall. In this work, some small cracks can be distinguished from the coating surface compressed to the strain of 40%, big and long cracks with the length of several tens micrometer and the width of several micrometers could be seen at the strain of 80%, as shown in Fig. 5(d) and (f). Considering the combination of the brittleness of the Ni coating and the concentration of stress around the coating, microcracks form when the internal stress during the compression exceeds the strength of the nickel coating for stress relief, which was also reported in the case of chromium coating on copper substrate by Pina et al. [37]. After the relief, the stress increases again at the interface with the further compression, resulting in a new relief following a new stress increase, which is proven by the multiple serrations in the stress-strain curve. Therefore, the coating cracks grow and release the stress between the base metal-copper and the Ni coating during the compression, resulting in an improvement on energy absorption capacity for the lotus-type porous copper. From Fig. 5(c) and (e), the crack formation in the Ni coating is also associated with a strut rotation. It should be mentioned that from the images of compressed specimens, the Ni coating has a very good adhesion to the copper walls because Ni and Cu have good wettability at the interface, which means the Ni coating still adheres to the porous copper even after being compressed to 80%.

Since the Ni coating is supposed as a barrier for the dislocation slip during the compression, it improves not only compressive strength but also energy absorption capacity of the lotus-type porous copper. Furthermore, the cracks relieve part of compressive energy on the porous copper when the strain is more than 40%, resulting in a further increase in energy absorption. As it has been indicated that the interface between nickel and copper played an important role in the compressive properties of the lotus-type porous copper, it is still to be studied in future, not only by TEM observation on dislocation accumulation, but also taking into account the modification of the interface between nickel coating and copper wall by annealing. It can be predicted that the adhesion between the coating and the copper substrate will be improved and a new phase may appear by the annealing. On the other hand, the apparent interface between the copper substrate and the Ni coating may disappear for their diffusion, which may deteriorate the obstacle effect.

4. Conclusion

The use of surface coating on the pore wall opens the possibilities for the design of the porous metal. The Ni coating deposited on pore walls with thickness from 3.9 to 4.8 μm improves significantly both the yield strength and the energy absorption in compression for the lotus-type porous copper, which is attributed to the internal stress between the coating and the pore wall, the obstacle to dislocation slip during deformation of the pore wall and the energy release by the coating cracks. The new interface between the Ni coating and the pore walls prevents dislocations from exiting the copper and allows the build-up of significant back stresses, which plays a more important role in the improvement. Furthermore, the specific yield strength and the energy absorption per unit mass also increase with the Ni coating and appear to be insensitive to coating thickness. It is indicated that in the applications where a porous metal is used for a structural role and the weight must be minimized, a coating procedure could be an advantage.

Acknowledgments

This work was supported financially by the National Science and Technology Project (No. 2017ZX02201001) and the National Natural Science Foundation of China (No. 51772193).


/