Spark plasma sintering of tungsten carbide nanopowders obtained through DC arc plasma synthesis

Accepted Manuscript Spark plasma sintering of tungsten carbide nanopowders obtained through DC arc plasma synthesis V.N. Chuvil'deev, Yu.V. Blagoveshchenskiy, A.V. Nokhrin, M.S. Boldin, N.V. Sakharov, N.V. Isaeva, S.V. Shotin, O.A. Belkin, A.A. Popov, E.S. Smirnova, E.A. Lantsev PII:

S0925-8388(17)30804-6

DOI:

10.1016/j.jallcom.2017.03.035

Reference:

JALCOM 41073

To appear in:

Journal of Alloys and Compounds

Received Date: 27 October 2016 Revised Date:

17 February 2017

Accepted Date: 4 March 2017

Please cite this article as: V.N. Chuvil'deev, Y.V. Blagoveshchenskiy, A.V. Nokhrin, M.S. Boldin, N.V. Sakharov, N.V. Isaeva, S.V. Shotin, O.A. Belkin, A.A. Popov, E.S. Smirnova, E.A. Lantsev, Spark plasma sintering of tungsten carbide nanopowders obtained through DC arc plasma synthesis, Journal of Alloys and Compounds (2017), doi: 10.1016/j.jallcom.2017.03.035. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Spark Plasma Sintering of tungsten carbide nanopowders obtained through DC arc plasma synthesis V.N. Chuvil’deev1, Yu.V. Blagoveshchenskiy2, A.V. Nokhrin1 a, M.S. Boldin1, N.V. Sakharov1, N.V. Isaeva2, S.V. Shotin1, O.A. Belkin1, A.A. Popov1, E.S. Smirnova1, E.A. Lantsev1

Lobachevsky State University of Nizhniy Novgorod (Lobachevsky University, UNN) (Russian

Federation, 603950, Nizhniy Novgorod, Gagarina ave., 23) 2)

RI PT

1)

A.A. Baykov Institute of Metallurgy and Material Science of RAS (Russian Federation, 119991,

Moscow, Leninskii ave., 49)

SC

E-mail: [email protected]

Abstract

M AN U

The paper dwells on the research conducted into high-rate consolidation of pure tungsten carbide (WC) nanopowders using the Spark Plasma Sintering technology. Studies included the effect that the original size of WC nanoparticles and their preparation modes have on density, structure parameters, and mechanical properties of tungsten carbide. Samples of high-density

TE D

nanostructured tungsten carbide characterized by high hardness (up to 31-34 GPa) and improved fracture toughness (4.3-5.2 MPa⋅m1/2) were obtained. It has been found that materials that show abnormal grain growth during sintering have lower values of sintering activation energy as

EP

compared to materials the structure of which is more stable during high-rate heating. A qualitative model is proposed that explains this effect through the dependence of the grain boundary diffusion

AC C

coefficient on the grain boundary migration rate. Keywords: tungsten carbide, nanopowders, spark plasma sintering, DC arc thermal plasma

synthesis, grain growth

1. Introduction Pure tungsten carbide is of interest for a variety of applications (cutting tools, drawing dies, etc.) due to a good combination of physical and mechanical properties (high melting temperature, a

Corresponding Author

ACCEPTED MANUSCRIPT

high hardness, low friction coefficient, and chemical resistance to corrosion and oxidation) [1, 2]. However, high brittleness of tungsten carbide traditionally obtained with the help of a powder sintering technology does not allow to use it in its purest form. Cobalt or any other binding substance added during sintering help to reduce its brittleness, but at the same time the binding

RI PT

phase is considered a weak point in terms of corrosion and strength properties of the material. That is why the issue of achieving high mechanical properties with pure tungsten carbide is crucial [3-5]. Improving the mechanical properties of tungsten carbide is primarily associated with

SC

obtaining ultrafine grains in the sintered material [6-7]. To this end, tungsten carbide nanopowders and ultrafine powders are used as a starting material [6, 8].

M AN U

A series of methods are currently available that help to achieve tungsten carbide nanopowders [9-13]. DC arc thermal plasma synthesis is the most advanced among them [14-22]. However, during traditional sintering of tungsten carbide nanopowders, intense grain growth and irregular grain structure were observed [8]. Samples sintered from nanopowders have relatively high porosity and poor mechanical properties comparable with the properties of the material

TE D

sintered from micron powders [1, 23-24].

To tackle the challenges which arise during sintering of nanopowders, new consolidation

EP

methods have been applied lately, as they should help solve the issue of preserving the ultra-fine grain (UFG) size in the sintered material, but at the same time ensure high density.

AC C

The technology known as Spark Plasma Sintering (SPS) is considered an advanced method applied to obtain materials with high-density nano- and ultrafine grain structure [25-28]. The core of the SPS method lies in the high-rate heating of powder materials in vacuum or inert atmosphere by passing direct pulse large currents through the tooling and samples and simultaneous pressure application. High heating rates become particularly relevant during sintering of nanomaterials as they limit the growth of nanograins and contribute to the formation of a uniform high-density nanostructure at low consolidation temperatures. Metal and ceramic materials obtained using the

ACCEPTED MANUSCRIPT

SPS method are characterized by high density and improved mechanical properties, which opens up new opportunities to produce construction materials for a variety of functional purposes [29-31]. The goal of this paper is to study the structure and properties of tungsten carbide obtained

RI PT

through spark plasma sintering from nanopowders produced using DC arc plasma synthesis.

2. Materials and methods

Starting materials include tungsten carbide nanopowders produced using DC arc plasma

SC

synthesis and subsequent recovery annealing in hydrogen [18, 32]. Nanopowders of the W-C system were synthesized in the flow of thermal plasma from the arc plasma torch using a reactor

M AN U

with restricted jet flow [16, 32]. The plasma jet is generated in an electric discharge generator of thermal plasma with nominal power of 25 kW. Thermal plasma jet causes evaporation and chemical transformation of the feedstock, as well as subsequent formation of nanoparticles of the W-C system (see Fig.1). Plasma jet efflux into the reactor tank with a water cooled shield ensures rapid decline in jet temperature, achieving of condensation temperature, and provides proper conditions

TE D

required for speedy formation of nanoparticles from the gas phase. The second stage of obtaining αWC nanoparticles is recovery annealing in hydrogen. Its goal is to control the volume fraction of

EP

W2C, β-WC (WxC1-x), W, α-WC particles and free carbon Сf in a nanopowder charge based on tungsten monocarbide α-WC (see Fig.2).

AC C

As shown in [18, 33-34], the ratio between volume fractions of the above particles in the

mixture obtained, as well as carbon Сf concentration depend on the modes of DC thermal plasma synthesis, temperature and time of hydrogen annealing. By optimizing the modes of two-stage DC arc thermal plasma synthesis, it is possible to effectively control the composition and size of the particles in the nanopowders produced [18, 33-34]. To study the impact that the particle parameters have on the structure and properties of tungsten carbide-based ceramics, 6 series of powders different in composition were obtained through DC arc plasma synthesis with various average initial particle size (R0). Characteristics of

ACCEPTED MANUSCRIPT

nanopowders obtained are presented in Table 1. The chemical composition of nanopowders under consideration is presented in Table 2. According to X-ray diffraction analysis, these nanopowders, except for α-WC nanopowders, contained a different number of the following phases: W2C, β-WC (WxC1-x), α-W. Carbon Cf did

RI PT

not exceed 0.2 wt% (powders of Series 1), while it reduced to 0-0.02 wt% with the increase in the volume fraction (f) of tungsten monocarbide (powders of Series 5-6). Neither X-ray diffraction analysis nor electron microscopy identified any graphite present.

SC

Microphotographs of α-WC monocarbide nanopowders (Series 6) are shown on Fig.3a. The particle size distribution histogram obtained using the method of scanning electron microscopy is

M AN U

shown on Fig. 3b. Nanopowders form conglomerates with average size of ~0.5-1 µm (see Fig.3c). Fig.4 presents X-ray diffraction patterns for powders of Series 2 (f=91.7%, R0=55 nm), Series 3 (f=93.6%, R0=63 nm) and Series 4 (f=99.5%, R0=113 nm). As seen from these figures, Xray diffraction patterns for powders with reduced tungsten monocarbide volume fraction (f=83.5-

TE D

93.6%) clearly show peaks of the primary α-WC phase, as well as β-WC phase. α-W tungsten particles in the amount of ~0.5 wt% are identified in powders of Series 4 (f=99.5%) at the background level.

EP

For the purposes of comparison, this paper involves research into commercial tungsten monocarbide powder produced by H.C. Starck Company (R0=113 nm).

AC C

The specific surface of powders (SBET, m2/g) was measured using specific surface analyzer

“Micromeritics TriStar 3000”. The accuracy of SBET was ± 0.02 m2/g. The average initial particle size (R0, nm) was calculated according to the formula: R0= 6/(ρthSBET), ρth=15.77 g/cm3 being a theoretical density of tungsten monocarbide [2]. Total carbon concentration was identified with the help of “LECO CS-400” analyzer. In order to determine the oxygen concentration we used “LECO TC-600” analyzer. The accuracy of determining the content of carbon and oxygen was ±0.01 wt %. The chemical composition analysis

ACCEPTED MANUSCRIPT

(for content of metal impurities, see Table 2) was performed using “Ultima 2 ICP” atomic emission spectrometer. Phase composition of samples was studied with “Shimadzu XRD-700” X-ray diffractometer (emission CuKα, scanning rate 2 о/min, in order to identify phases with low intensity of X-ray

RI PT

peaks, the scanning rate was reduced to 0.5 о/min). The accuracy of the volume fraction of α-WC tungsten monocarbide phase was 0.3% at the scanning rate of 0.5 о/min. Phase identification and determination of the volume fraction of each phase were performed using the method of corundum

SC

numbers (Database “DIFFRACplus Evaluation package Release 2009”).

Sintering of nanopowders was performed using a sintering plant «Dr. Sinter model SPS-

M AN U

625» (SPS SYNTEX Inc., Japan) in the temperature range from 1400°C to 1950°C without holding. The heating rate (V) ranged from 25 to 2400°C/min. After attaining the desired temperature, heating was switched off, thus, the sample together with the mold were allowed to gradually cool down. Experiments were carried out at uniaxial compression pressure (P) of 60 and 75 MPa.

TE D

Temperature (T) was measured with «Chino IR-AH» pyrometer focused on the outside of the mold. Sintering took place in 4 Pa vacuum in graphite molds 12 mm in inside diameter. The temperature accuracy was ±10 оС, the shrinkage accuracy was 0.01 mm.

EP

A dilatometer which is part of a unit “Dr. Sinter model SPS-625” was used to measure effective shrinkage (Leff, mm) under continuous heating (see Fig.5a). To take account of thermal

AC C

expansion of the “machine-sample” system into effective shrinkage Leff measured with a dilatometer, additional studies into thermal expansion of the system were carried out without samples (empty mold). The above dependencies L0(Т) were then deduced from Leff(T) dependence, and the dependence of shrinkage rate D (Fig. 5b) on the heating temperature was calculated. Vickers hardness (Hv) was measured using “Struers Duramin-5” microhardness tester with a 2 kg load. Fracture toughness KIC was calculated using the Palmqvist method. The accuracy of Hv and KIC was ±0.1 GPa and ±0.1 MPa·m1/2 respectively.

ACCEPTED MANUSCRIPT

Samples density was measured using the hydrostatic weighing method using “Sartorius CPA” scales. The intercept method was applied to measure grain size using Jeol JSM-6490 scanning electron microscope with a peripheral device for local energy-dispersive microanalysis

RI PT

INCA 350. The average grain size in sintered ceramics was determined using the intercept method.

3. Experimental results

3.1. Impact of the initial particle size on tungsten carbide sintering temperature

SC

Fig.5 presents typical dependencies of the degree of shrinkage L(T) and shrinkage rate D(T) on the heating temperature for the powders under consideration. As seen from Fig.5, D(T)

M AN U

dependence for all the nanopowders under consideration obviously has a three-stage character. Intensity of powder shrinkage is little at an early heating stage. At temperatures below representative temperature Т1, shrinkage rate is growing; with temperature increasing (Т1<Т
TE D

heating rate V=25 оС/min and pressure P=75 MPa. Data analysis presented in Table 1 proves that representative temperatures Т1 and Т2 depend

EP

on the initial particle size: with compositions characterized by low α-WC monocarbide volume fraction (f=83.5-93.6%) increase in the initial size of particles R0 from 46 nm to 63 nm results in Т1

AC C

growing from 1145 оС to 1460 оС and Т2 growing from 1420 оС to 1520 оС. For powders with initial particle size R0=113 nm and f=99.5%, temperature values are

Т1=1330оС and Т2=1380-1390оС (see Table 1).

1

The first maximum value (ТН2) on D(T) curve is associated with degassing or release of residual gases from the pre-

compacted powders. Gases result from high adsorption capacity of DC arc thermal plasma nanopowders [18, 33]. Intensity and level of the maximum value depend on the heating mode and the pressure applied, as well as recovery annealing conditions.

ACCEPTED MANUSCRIPT

Thus, it may be claimed that should SPS nanopowders be added up with second-phase particles, nonmonotonic dependence of representative sintering temperatures Т1 and Т2 on the initial particle size R0 is observed (see Table 1, Fig.6). During sintering of nanopowders of Series 5-7 (volume fraction for tungsten monocarbide

RI PT

close to 100%), increase in the initial particle size R0 from 72 nm to 112 nm leads to a monotonic increase in Т1 from 1365-1375оС to 1430оС and insignificant rise in Т2 from 1490оС to 1515оС.

SC

3.2. Impact of nanopowder composition on parameters of the structure formed

Electron microscopic studies prove that abnormal grain growth is observed in the structure

M AN U

of the material sintered from nanopowders added up with second-phase particles (volume fraction of tungsten monocarbide f=83.5-99.5%). To illustrate the above, Fig.7 presents microphotographs showing the structure of samples of Series 1 (f=83.6%) and Series 3 (f=93.5%). These figures prove that after heating to Т=1460-1550оС, an inhomogeneous consertal structure is formed with average size of abnormal grains da (from 3 to 20 µm) being an order of magnitude bigger than the average

TE D

size of grains in a matrix dm (~0.1-0.3 µm). It is notable that in case of abnormal growth, pores are mainly found along the boundaries of abnormally coarse tungsten carbide grains. With sintering

EP

temperature rising to 1700оС, a uniform coarse-grained structure is formed. In this case pores are found both along the boundaries and in the grains of the sintered material. Fig. 8 shows grain size

AC C

distribution histograms in case of normal grain growth (see Fig. 8a, b) and abnormal grain growth (Fig. 8c, d) during sintering of nanopowders of Series 1 (Fig. 8c, d) and Series 6 (Fig. 8a, b). Analysis of the data presented in Table 1 shows that with tungsten monocarbide volume

fraction rising from 83.5 to 99.5%, the average size of abnormally coarse grains da is growing from 1-3 to 17-20 µm. X-ray diffraction analysis illustrates that abnormally coarse grains are W2C particles. The volume fraction of W2C particles in sintered ceramics ranges from 2.4 to 10.7% depending on the initial volume fraction of tungsten monocarbide and sintering temperature. To illustrate the above,

ACCEPTED MANUSCRIPT

Fig.9 presents diffraction patterns of ceramic samples sintered from powders with tungsten monocarbide volume fraction f = 93.6% (Series 3) and f = 99.5% (Series 4). The volume fraction of W2C particles (fW2C) in ceramic samples of Series 3 and Series 4 approximates ~24.4% and ~5.2%, respectively (Tsint= 1550 оС, V=25 оС/min).

RI PT

For ceramics of Series 5 obtained through sintering of nanopowders, the volume fraction of tungsten monocarbide in which according to X-ray diffraction analysis constitutes 100%, W2C particles are identified in the amount ~9.9%.

SC

This result brings about the conclusion that the initial volume fraction of the second-phase particles acting as sort of “nuclei’s” (“points of grain growth”) for abnormally coarse grains is small

M AN U

and lies within the experimental error range permissible for X-ray diffraction analysis (~1% for αW and W2C phases in α-WC ceramics subject to the analysis being performed using Shimadzu XRD-700 X-ray diffractometer).

Note should be taken of no correlation between the volume fraction of second-phase

TE D

particles (1-f) and volume fraction of newly formed W2C particles: in ceramic samples of Series 3 (f=93.6%), Series 4 (f=99.5%), and Series 5 (f~100%), the volume fraction of W2C particles is fW2C = 24.4%, 5.2%, and 9.9% respectively. Similar results were achieved while studying ceramics

EP

sintered at other temperatures and heating rates. In our opinion the above proves that not all secondphase particles (W2C, β-WC (WxC1-x), α-W) are “nuclei’s” for abnormally coarse W2C grains while

AC C

sintering α-WC ceramics. (The nature of “nuclei’s” for abnormally coarse grains will be discussed below.)

Research into the structure of samples sintered from nanopowders of Series 6-7 (volume

fracture of tungsten monocarbide is 100%) shows that these materials are characterized by normal grain growth and monomodal grain size distribution, although the distribution bell is rather wide. The average grain size for samples sintered at 1400°C and 1800°C is 90 nm and 350 nm respectively. All the materials under consideration are characterized by active grain growth at

ACCEPTED MANUSCRIPT

temperatures above 1800°C which led to the formation of an UFG structure (average grain size is 300-400 nm).

3.3. Impact of the initial particle size on density

RI PT

Reduction in the initial particle size R0 leads to increased density (ρ) of sintered samples: to illustrate this, Fig.10 shows ρ(R0) dependences for samples sintered at Т=1500 оС and 1550 оС (V=25 оС/min).

SC

It should be highlighted that the density of samples sintered at Т=1550оС from powders, the volume fraction of α-WC tungsten monocarbide in which is close to 100% (Series 5-7) but the

M AN U

initial particle size is bigger (line 3, Fig.10), is higher than the density of samples obtained from powders with reduced monocarbide volume fraction and smaller initial particle size (see line 2, Fig.10).

At 1500оС no significant difference is observed in the densities of samples with various α-

TE D

WC monocarbide volume fraction.

3.4. Impact of the sintering temperature on density and mechanical properties

EP

Fig.11 shows that the dependence of density on sintering temperature has a typical two-stage character with no visible change in density in case of SPS temperature rising above 1700°С (P=60

AC C

MPa, V=2400 oC/min).

At sintering temperature of 1800°С (Р=60 MPa), density close to theoretical (series 6,

R0=72 nm) was reached with all samples heated at the rate V≤500 °С/min. In case of higher heating rates, maximum density goes down. When density of sintered samples is above 97%, their hardness is much higher than standard values. At sintering temperature of 1700°С, hardness reaches Hv~31.1 GPa.

ACCEPTED MANUSCRIPT

Similar results were obtained during SPS sintering of H.C. Starck nanopowders. Table 3 shows that the increase in sintering temperature from 1460оС to 1700оС is accompanied by decrease in hardness and fracture toughness (KIC) of sintered samples.

RI PT

3.5. Impact of the heating rate on mechanical properties Materials obtained in this paper through SPS technology at high heating rates are characterized by exceptionally high hardness and at the same time demonstrate an excellent

SC

combination of properties “hardness / fracture toughness” (see Table 3 for comparison with the results in papers [3, 6, 35-38]). At low heating rates (25°С/min), samples are characterized by better

M AN U

fracture toughness but lower hardness values (Hv=24.2 GPa, KIС=6.7 MPa·m1/2). At high heating rates (2400°С/min), samples are characterized by higher hardness but slight reduction in fracture toughness (Hv=31.1 GPa, KIС=5.2 MPa·m1/2).

Table 3 presents data achieved from studying samples obtained at the heating rate of 25оС/min and pressure of 75 MPa. In this case samples with increased hardness are characterized by

TE D

lower fracture toughness. Sintering of tungsten monocarbide nanopowders (Series 6) at Т=1550оС, heating rate of 25оС/min, and pressure 75 MPa allows for samples characterized by high hardness

EP

Hv=34 GPa and satisfactory fracture toughness KIC=4.3 MPa·m1/2.

AC C

4. Discussion

4.1 Generalizing the results While studying the specific features that characterize evolution of the structure of

nanopowder tungsten carbide-based compositions under high rate heating, there is a number of striking effects that require special consideration and analysis. Firstly, an unusual nature of the impact that the initial particle size R0 has on optimal sintering temperature T2 for tungsten carbide samples is discovered (see Fig.6). Reduction in the initial particle size R0 from 112 nm to 72 nm in α-WC tungsten monocarbide expectedly leads to

ACCEPTED MANUSCRIPT

optimal sintering temperature Т2 falling from 1515оС to 1490оС. However, in case of sintering of compositions nanopowders with the volume fraction of tungsten monocarbide f=83.5-99.5%, the dependence of temperature Т2(R0) is nonmonotonic: increase in the initial particle size from 46 nm to 63 nm leads to Т2 growing from 1420 to 1520оС, but in case of further increase in the initial

RI PT

particle size R0 to 113 nm, the optimal sintering temperature Т2 goes down to 1380-1390оС. At the same time the density of the sintered material varies very little and exceeds 98%.

This result is inconsistent with the results presented in papers [35, 37-39], which show that

SC

sintering in the same modes of powders with bigger initial particle size leads to smaller density. Secondly, an unusual dependence of the abnormal grain size on the volume fraction of the

M AN U

second-phase particles is detected. It is usually assumed that decrease in the volume fraction of the second-phase particles in the original composition promotes homogeneity of the sintered material structure. However, in the present case reduction in the volume fraction of the second-phase particles in the original nanopowder composition from 16.5% to 0.5% leads in particular to the

TE D

increase in the average size of abnormally coarse grains from 1-3 to 17-20 µm. It should be also noted that the density of samples sintered from nanopowders with smaller initial particle size but with reduced α-WC monocarbide volume fraction appears to be smaller than

Fig.9).

EP

the density of samples sintered from coarser (72-112 nm) tungsten monocarbide powders (see

AC C

In conclusion it should be again noted that in powders the volume fraction of second-phase particles fr acting as “nuclei’s” for abnormally coarse grains in sintered ceramics appears to be much less than the volume fraction of second-phase particles. It means that not all second-phase particles (W2C, β-WC, α-W) are “nuclei’s” for abnormally coarse grains in sintered ceramics.

4.2 Calculation of energy activation As noted above, dependencies of the degree of shrinkage (L, mm) and shrinkage rate (D, mm/s) on temperature and heating time are considered primary data during Spark Plasma Sintering

ACCEPTED MANUSCRIPT

of tungsten carbide. The procedure of converting L(T) dependence into compression schedules ρ/ρth-T is specified in [40, 41] (ρth=15.77 g/cm3 being a theoretical density of tungsten monocarbide [2]). ρ/ρth(T) dependencies have a three-stage nature typical for compression curves during

RI PT

sintering of powder materials. Our further analysis of ρ/ρth(T) dependencies shall be limited to considering the stage of intensive compression (temperature range Т1<Т<Т2 shown on Fig.5b). Activation energy of the compression process may be determined by the slope angle of

SC

ρ/ρth(T) dependence (presented in double logarithmic coordinates) on reciprocal temperature Tm/T: ln(ln[(ρ/ρth)/(1-ρ/ρth)]) - Tm/T (for more detail see [40, 41]). (Tm=3143 K is the melting temperature

M AN U

for tungsten monocarbide [2]).

Table 1 shows the values of the sintering activation energy calculated for nanopowder compositions with various volume fracture of α-WC monocarbide. In samples of Series 1-4, the volume fraction of tungsten monocarbide in which is under

TE D

100%, the values of the sintering activation energy within the sintering temperature range Т= 11501550 оС is Q=107-135 kJ/mol (~4.1-5.2 kTm), which agrees well with the data obtained in [42] within the same temperature range of 1050-1200 oC during SPS of tungsten carbide nanopowders

EP

amid intensive grain boundary migration: Q =103.5 kJ/mol (~4 kTm). Note shall be taken that the obtained values of Q are much lower than the values of the

AC C

activation energy for grain boundary diffusion Qb of carbon 14C in coarse-grained tungsten carbide obtained through hot pressing (Qb=367 kJ/mol ~14 kTm [43]), as well as the values of the activation energy for 14C diffusion provided in [44] (Qb=300 kJ/mol ~11.4 kTm) и в [41] (Qb=240 kJ/mol ~9.1 kTm). In samples of Series 6-7 with 100% α-WC monocarbide volume fraction, average values of the activation energy Q appear to be considerably higher (7.7-10.8 kTm~ 201-282 kJ/mol) and better comply with the above reported value [43-45].

ACCEPTED MANUSCRIPT

Thus, data provided in Table 1 show that two types of behavior are observed during tungsten carbide sintering: sintering under abnormally low activation energy values and sintering under activation energy corresponding to the equilibrium values of the activation energy for grain boundary diffusion Qb.

RI PT

It should be emphasized that after sintering materials of all the above types significantly differ in structure: samples of type materials (sample of Series 1-4) are characterized by an inhomogeneous structure containing abnormally coarse grains, while samples of type 2 materials

M AN U

Let us analyze the results obtained.

SC

(samples of Series 6-7) form a predominantly homogeneous UFG structure2.

4.3 Impact of the average initial powder particle size and volume fraction of tungsten monocarbide on the grain growth and WC ceramics density

Let us consider the impact that second-phase particles of W2C, β-WC carbides and pure

TE D

tungsten (α-W) have on sintering of tungsten carbide. In line with [6, 8, 15, 38], some of these second-phase particles may act as sort of “nuclei’s” for abnormally coarse grains. (Further, for the sake of brevity, we shall call these particles «active phase particles»).

EP

Let us assume that the typical size to which on average an abnormally growing «active phase particle» may grow is determined by the distance to another such particle λ*. Based on

AC C

microstructural data, it is possible to roughly estimate the volume fraction of such 'active phase particles'. Let us take a simple formula: λ* = R 3 f r , where fr and R stand for volume fraction and size of “active phase particles” respectively. Assuming that λ* value corresponds to half the size of abnormal grains dm (λ*=dm/2) and by substituting values R0=46-113 nm and dm=3-17 µm (see Table 1) into the equation for λ*, we shall

2

Exception to the above are ceramic samples sintered from nanopowders of Series 5 in which W2C particles were

discovered and consequently slight abnormal grain growth was detected: the average size of abnormal grains is da=1.21.3 µm, while the average size of matrix grains is dm=0.35 µm.

ACCEPTED MANUSCRIPT

find that the volume faction of “active phase particles” is fr~3⋅10-5-3⋅10-4%. This value is much smaller than the values of the second-phase particles volume fraction in samples of Series 1-4 (from 0.5 to 16.5%) measured with the help of X-ray diffraction analysis. The obtained results also count in favor of the above made assumption that not all second-phase particles (W2C, β-WC, α-W) are

RI PT

“active”. It is crucial to note that the volume fraction of W2С particles in sintered ceramics is rather high (10-25%), while the volume fraction of W2С particles in the original powder is below the

SC

resolving power of the X-ray diffraction analysis (less than 1%). This means that the sintering process is characterized by «transformation» of the original “active phase particles” into W2C

M AN U

particles which grow rapidly on further heating. Since the experiment shows simultaneous growth in W2C volume fraction and size (size of abnormally coarse grains), it becomes obvious that this process is not a conventional coalescence process.

In our opinion, the most likely candidate to act as active phase particles may be particles of

TE D

α-W metal tungsten in which carbon diffusion at specified sintering temperatures (1300-1500оС) is rather intense (parameters of volume carbon diffusion in tungsten monocrystals: Dv0=3.0 cm2/s, Qv=247 kJ/mol ~ 8 kTm [44, 46]), whereas tungsten at elevated temperatures has a high tendency to

EP

W2C carbide formation [47].

We shall note that W2C phase cannot act as an “active phase”. This is due to the fact that the

AC C

volume fraction of W2C particles in sintered ceramics is much greater than the volume fraction of W2C particles of the original powder. This means that the process of splitting the main phase α-WC into W2C and α-W tungsten (2WC → W2C + C)3 shall take place during the sintering process. Such

3

Chances of α-WC particles disintegration during SPS are beyond the scope of this research. Slight W2C particle

formation (under 5%) in ceramics sintered from nanopowders that entirely consist of α-WC tungsten monocarbide was detected only under such simultaneous conditions as extremely high heating rates (1500 оС/min and more) and high sintering temperatures (1720 оС and more). Since such sintering modes are beyond the considered temperature range and sintering rate range, we shall neglect the impact of this factor of the processes discussed.

ACCEPTED MANUSCRIPT

methods as X-ray diffraction analysis and electron microscopy have not detected tungsten in the structure of sintered ceramics. The hypothesis about the role of tungsten is supported by the results of research [38] into sintering of tungsten carbide with additions of pure tungsten and carbon (pure WC, WC+0.5%W,

RI PT

WC+0.4%C, WC+0.5%W+0.4%C). This paper shows that with 0.5% α-W particles, the fine structure of a sintered sample accommodates abnormally coarse grains of elongated shape, while a coarse equiaxed structure is formed by sintering other systems. Similar results were obtained in

added W2C carbide particles and α-W tungsten particles.

SC

paper [48] devoted to studies of abnormal grain growth in a hard alloy WC-10%Co-0.25%VC with

M AN U

Summarizing the analysis results it may be concluded that big distance between “active phase particles” λ* (the role of which is presumably played by tungsten particles) leads to the fact that with the reduction in the total volume fraction of second-phase particles the size of abnormally

TE D

coarse grains dm is growing.

4.4. Effect of accelerated sintering during abnormal grain growth Let us consider the issue related to the reasons for lowering the optimal sintering

EP

temperature and the reasons why low sintering activation energy values occur in materials with abnormal grain growth.

AC C

In our opinion, the said effects may be related to some changes in the parameters of grain

boundary diffusion due to intense grain growth in such materials (see Fig.12). In accordance with the concepts of the non-equilibrium grain boundary theory [49-51], the

activation energy of grain boundary diffusion Q *b depends on the degree of grain boundary nonequilibrium, which in turn is determined by the density of defects introduced into boundaries: orientational misfit dislocations (OMD) with Burgers vector ∆b and density ρb, as well as products of their delocalization such as a tangential component of delocalized OMD with Burgers vector density wt. As shown in [50, 52], in case of abnormal grain growth when grain boundaries

ACCEPTED MANUSCRIPT

migrating at high speed absorb lattice dislocations distributed in grains (see Fig.13), the value of grain boundary diffusion D *b is exponentially dependent on the density of lattice dislocations and grain boundaries migration rate: D *b = D b exp( g1 (ρ v Vm ) n ) , where Db stands for the diffusion coefficient in equilibrium grain boundaries, ρv is the density of lattice dislocations, Vm is the grain

RI PT

boundaries migration rate, g1 is a numerical parameter that depends on the thermodynamic properties of the material, n=0.25, 0.5 is a numerical coefficient that depends on what types of defects dominate within grain boundaries: OMD (n=0.25) or tangential components of delocalized

SC

dislocations (n=0.5).

The results show an important practical conclusion regarding the impact of the grain

M AN U

boundaries migration on the process of sintering of UFG hard alloys. To ensure tungsten carbide sintering at a reduced temperature, it is necessary to optimize its phase composition and sintering modes so as to provide sufficiently 'fast' migration of grain boundaries while heating. However, it should be borne in mind that in case of sintering second-phase particles powders of tungsten

samples.

EP

Conclusions

TE D

carbide, the abnormal grain growth leads to a decrease in the mechanical properties of sintered

AC C

1. Studies were conducted into the structure and properties of tungsten carbide samples obtained using the method of Spark Plasma Sintering of nanopowders obtained through DC arc plasma synthesis. It was shown that samples sintered from nanopowders with reduced content of tungsten monocarbide have a heterogeneous structure which proves that during Spark Plasma Sintering there was an intensive abnormal grain growth. Samples sintered from nanopowders of tungsten monocarbide have a homogeneous ultrafine structure and high mechanical properties: high hardness and enhanced fracture toughness. It was found that the heating rate has a significant impact on mechanical properties of tungsten carbide samples sintered from α-WC nanopowders obtained through DC arc plasma

ACCEPTED MANUSCRIPT

synthesis. At low heating rates (25°С/min), samples are characterized by better fracture toughness but lower hardness values (Hv=24.2 GPa, KIС=6.7 MPa·m1/2). At high heating rates (2400°С/min), samples are characterized by higher hardness but slight reduction in fracture toughness (Hv=31.1 GPa, K1С=5.2 MPa·m1/2).

RI PT

2. One of the most probable causes of abnormal grain growth during Spark Plasma Sintering of tungsten carbide is the presence of pure tungsten particles in nanopowders obtained through DC arc plasma synthesis. At elevated temperature tungsten particles react with carbon (2W + C →

SC

W2C), which results in W2C particles that are characterized by high migration mobility of boundaries.

M AN U

3. Rapid migration of boundaries leads to 'sweeping' by boundaries of lattice dislocations and reduced activation energy of grain boundaries diffusion. This reduces the optimal sintering temperature for tungsten carbide powders. However, materials obtained at low sintering temperatures have low hardness and fracture toughness because of heterogeneity of the grain

TE D

structure.

4. The effect of reducing the optimal sintering temperature for tungsten carbide powders during Spark Plasma Sintering is related to the effect of grain boundary diffusion acceleration

EP

caused by abnormal grain growth.

AC C

Acknowledgements

This research is carried out with the support of the Russian Foundation for Basic Research

(Grant No. 15-33-21007) and the Ministry of Education and Science of the Russian Federation.

References [1]. Panov V.S., Chuvilin A.M., Fal’kovskii V.A. Technology and Properties of Sintering Hard Alloys and Units on their Base. Moscow: National University of Science and Technology “MISiS”. 2004. 468 p. (in Russian).

ACCEPTED MANUSCRIPT

[2]. Andrievskii R.A., Spivak I.I. Strength of Refractory Compounds and Related Materials. Chelyabinsk: Metallurgiya. 1989. 368 p. (in Russian). [3]. Shon I.-J., Kim B.-R., Doh J.-M., Yoon J.-K., Woo K.-D. Properties and rapid consolidation of ultra-hard tungsten carbide // Journal of Alloys and Compounds. 2009. V. 489. Iss. 1. P. L4-L8.

RI PT

(http://dx.doi.org/10.1016/j.jallcom.2009.09.040). [4]. Zhang J., Zhang G., Zhao Sh., Song X. Binder-free WC bulk synthesized by spark plasma sintering // Journal of Alloys and Compounds. 2009. V. 479. Iss. 1-2. P. 427-431.

SC

(http://dx.doi.org/10.1016/j.jallcom.2008.12.151).

[5]. Tsai K.-M. The effect of consolidation parameters on the mechanical properties of binderless

M AN U

tungsten carbide // International Journal of Refractory Metals and Hard Materials. 2011. V. 29. P. 188-201. (http://dx.doi.org/10.1016/j.ijrmhm.2010.10.006).

[6]. Zhao J., Holland T., Unuvar C., Munir Z.A. Sparking plasma sintering of nanometric tungsten carbide // International Journal of Refractory Metals and Hard Materials. 2009. V. 27. P. 130-139. (http://dx.doi.org/10.1016/j.ijrmhm.2008.06.004).

TE D

[7]. Sivaprahasam D., Chandrasecar S.B., Sundaresan R. Microstructure and mechanical properties of nanocrystalline WC-12Co consolidated by spark plasma sintering // International Journal of Metals

and

Hard

Materials.

2007.

V.

25.

P.

144-152.

EP

Refractory

(http://dx.doi.org/10.1016/j.ijrmhm.2006.03.008).

AC C

[8]. Sommer M., Schubert W.D., Zobetz E., Warbichler P. On the formation of very large WC crystals during sintering of ultrafine WC-Co alloys // International Journal of Refractory Metals and Hard Materials. 2002. V. 20. P. 41-50. (http://dx.doi.org/10.1016/S0263-4368(01)00069-5). [9]. Panov V.S. Nanostructured sintered WC-Co hard metals (Review) // Powder Metallurgy and Metal Ceramics. 2015. V.53. Iss.11-12. P. 643-654. (http://dx.doi.org/10.1007/s11106-015-9670-2). [10]. Kear B.H., McCandlish L.E. Chemical processing and properties of nanostructured WC-Co materials // Nanostructured Materials. 1993. V.3. Iss.1-6. P.19-30. (http://dx.doi.org/10.1016/09659773(93)90059-K).

ACCEPTED MANUSCRIPT

[11]. McCandlish L.E., Kear B.H., Bhatia S.J. Spray conversion process for the production of Nanophase composite powders. Pat. 5352269 USA. Oct. 4 1994. [12]. Panov V.S., Zaitsev A.A. Developmental tendencies of technology of ultradispersed and nanosized WC-Co hard alloys alloyed with tantalum carbide: Review // Russian Journal of Non-

RI PT

Ferrous Metals. 2015. V. 56. Iss. 4. P. 477-485. (http://dx.doi.org/10.3103/S106782121504015X). [13]. McCandlish L.E., Kear B.H. Kim B.K. Processing and properties of nanostructured WC-Co // Nanostructured Materials. 1992. V. 1. Iss. 2. P. 119-124. (http://dx.doi.org/10.1016/0965-

SC

9773(92)90063-4).

[14]. Tsvetkov Y.V., Panfilov S.A. Low-Temperature Plasma in Restoration Processes. Moscow:

M AN U

Nauka. 1980. 359 p. (in Russian).

[15]. Blagoveshchenskiy Y.V., Samokhin A.V., Tsvetkov Y.V., Alexeev N.V., Isaeva N.V., Kornev C.A., Melnik Y.I. Nanopowders of WC-Co system with different inhibitor additions manufacturing by plasmochemical process. Proc. 17 Plansee Seminar. Reutte, Austria. May 25-29, 2009. V. 3. P. GT23/1-6.

TE D

[16]. Patent RU №2349424 C1 «Method of powder receiving on basis of tungsten carbide». Blagoveshchenskij J.V., Alekseev N.V., Samokhin A.V., Mel’nik J.I., Tsvetkov J.V., Kornev S.А.

EP

(in Russian). Application No. 2007138445/02 at 18.10.2007. Published 20.03.2009. [17]. Samokhin A.V., Alekseev N.V., Kornev S.A., Sinaiskii M.A., Blagoveshchenskiy Y.V.,

AC C

Kolesnikov A.V. Tungsten carbide and vanadium carbide nanopowders synthesis in DC plasma reactor // Plasma Chemistry and Plasma Processing. 2013. V. 33. Iss. 3. P. 605-616. (http://dx.doi/org/10.1007/s11090-013-9445-9). [18]. Krasovskii P.V., Malinovskaya O.S., Samokhin A.V., Blagoveshenskiy Y.V., Kazakov V.A., Ashmarin A.A. XPS study of surface chemistry of tungsten carbides nanopowders produced through DC thermal plasma/hydrogen annealing process // Applied Surface Science. 2015. V. 339. P. 46-54. (http://dx.doi.org/10.1016/j.apsusc.2015.02.152).

ACCEPTED MANUSCRIPT

[19]. Krasovskii P.V., Samokhin A.V., Fadeev A.A., Alexeev N.V. Thermal evolution study of nonmetallic impurities and surface passivation of Cu nanopowders produced via a DC thermal plasma synthesis // Advanced Powder Technology. 2016. V. 27. Iss. 4. P. 1669-1676. (http://dx.doi.org/10.1016/j.apt.2016.05.031).

RI PT

[20]. Lessard J.D., Shekhter L.N., Gribbin D.G., Blagoveshchensky Y., McHugh L.F. A new technology platform for the production of electronic grade tantalum nanopowders from tantalum scrap sources // International Journal of Refractory Metals and Hard Materials. 2015. V. 48. P. 408-

SC

413. (http://dx.doi.org/10.1016/j.ijrmhm.2014.09.027).

[21]. Samal S. Thermal plasma technology: The prospective future in material processing // Journal Cleaner

Production.

2017.

V.

142.

Part

M AN U

of

4.

P.

3131-3150.

(http://dx.doi.org/10.1016/j.jclepro.2016.10.154).

[22]. Krasovskii P.V., Samokhin A.V., Malinovskaya O.S. Characterization of surface oxide films and oxygen distribution in α-W nanopowders produced in a DC plasma reactor from an oxide //

Powder

Technology.

2015.

V.

286.

P.

144-150.

TE D

feedstock

(http://dx.doi.org/10.1016/j.powtec.2015.07.025). [23]. Seung I. Cha, Soon H. Hong. Microstructures of binderless tungsten carbides sintered by spark

EP

plasma sintering processes // Materials Science and Engineering А. 2003. V. 356. Iss. 1-2. P. 381389. (http://dx.doi.org/10.1016/S0921-5093(03)00151-5).

AC C

[24]. Mannesson K., Borgh I., Borgenstam A., Agren J. Abnormal grain growth in cemented carbides – Experiments and simulations // International Journal of Refractory Metals and Hard Materials. 2011. V. 29. P. 488-494. (http://dx.doi.org/10.1016/j.ijrmhm.2011.02.008). [25]. Tokita M. Spark Plasma Sintering (SPS) Method, Systems, and Applications (Chapter 11.2.3). In Handbook of Advanced Ceramics (Second Ed.). Academic Press. 2013. P. 1149-1177. (http://dx.doi.org/10.1016/B978-0-12-385469-8.00060-5). [26]. Tokita M. Industrial applications of advanced spark plasma sintering // American Ceramic Society Bulletin. 2006. V. 85. № 2. P. 32-34.

ACCEPTED MANUSCRIPT

[27]. Munir Z.A., Anselmi-Tamburini U., Ohyanagi M. The effect of electric field and pressure on the synthesis and consolidation materials: A review of the spark plasma sintering method // Journal of Materials Science. 2006. V. 41. Iss.3. P.763-777. (http://dx/doi/org/10.1007/s10853-006-6555-2). [28]. Tokita M. Development of advanced spark plasma sintering (SPS) systems and its industrial

RI PT

applications // Ceramic Transactions. 2006. V. 194. P. 51-59. [29]. Chuvil’deev V.N., Moskvicheva A.V., Nokhrin A.V., Baranov G.V., Blagoveshenskii Yu.V., Kotkov D.N., Lopatin Yu.G., Belov V.Yu. Ultrastrong nanodispersed tungsten pseudoalloys

SC

produced by High-Energy Milling and Spark Plasma Sintering // Doklady Physics. 2011. V. 56. No. 2. P. 109-113. (http://dx/doi/org/10.1134/S1028335811020017).

M AN U

[30]. Chuvil’deev V.N., Nokhrin A.V., Baranov G.V., Moskvicheva A.V., Boldin M.S., Kotkov D.N., Sakharov N.V., Blagoveshensky Yu.V., Shotin S.V., Melekhin N.V., Belov V.Yu. Study of the structure and mechanical properties of nano- and ultradispersed mechanically activated heavy tungsten

alloys

//

Nanotechnologies

in

Russia.

2013.

V.

8.

No.1-2.

P.

108-121.

(http://dx/doi/org/10.1134/S1995078013010047).

TE D

[31]. Orlova A.I., Volgutov V.Yu., Mikhailov D.A., Bykov D.M., Skuratov V.A., Chuvil’deev V.N., Nokhrin A.V., Boldin M.S., Sakharov N.V. Phosphate Ca1/4Sr1/4Zr2(PO4)3 of the NaZr2(PO4)3

EP

structure type: synthesis of a dense ceramic material and its radiation testing // Journal of Nuclear Materials. 2014. V. 446. Iss. 1-3. P. 232-239. (http://dx.doi.org/10.1016/j.jnucmat.2013.11.025).

AC C

[32]. Patent RU №2311225 C1 «Plasma Device for Producing Nano-powders». Application No. 2006110838/15 at 05.04.2006. Published 27.11.2007. [33]. Krasovskii P.V., Blagoveshchenskii Yu.V., Grogorovich K.V. Determination of oxygen in WC-Co

nanopowders

//

Inorganic

Materials.

2008.

V.

44.

Iss.

9.

P.

954-959.

(http://dx/doi.org/10.1134/S0020168508090100). [34]. Voldman G.M., Levinsky Yu.V., Balgoveshchensky Yu.V., Isaeva N.V., Melnik Yu.I., Blagoveshchenskaya N.V. Kinetics of the processes in W-C plasmochemical nanopowders under

ACCEPTED MANUSCRIPT

annealing in hydrogen atmosphere // Physics and Chemistry of Materials Treatment. 2012. No. 4. P. 23-27 (in Russian). [35]. Kim H.C., Yoon J.K., Doh J.M., Ko I.Y., Shon I.J. Rapid sintering process and mechanical properties of binderless ultra fine tungsten carbide // Materials Science and Engineering A. 2006. V.

RI PT

435-436. P. 717-724. (http://dx.doi.org/10.1016/j.msea.2006.07.127). [36]. Kim H.C., Shon I.J., Yoon J.K., Lee S.K., Munir Z.A. One step synthesis and densification of ultra-fine WC by high-frequency induction combustion // International Journal of Refractory Metals

SC

and Hard Materials. 2006. V. 24. P. 202-209. (http://dx.doi.org/10.1016/j.ijrmhm.2005.04.004). [37]. Kim H.C., Shon I.J., Garay J.E., Munir Z.A. Consolidation and properties of sub-micron

M AN U

tungsten carbide by field-activated sintering // International Journal of Refractory Metals and Hard Materials. 2004. V. 22. P. 257-263. (http://dx.doi.org/10.1016/j.ijrmhm.2004.08.003). [38]. Gubernat A., Rutkowski P., Grabowski G., Zientara D. Hot pressing of tungsten carbide with and without sintering additives // International Journal of Refractory Metals and Hard Materials, 2014. V. 43. P. 193-199. (http://dx.doi.org/10.1016/j.ijrmhm.2013.12.002).

TE D

[39]. Srivatsan T.S., Woods R., Petraroli M., Sudarshan T.S. An investigation of the influence of powder particle size on microstrucutre and hardness of bulk samples of tungsten carbide // Powder

EP

Technology. 2002. V. 122. P. 54-60. (http://dx.doi.org/10.1016/S0032-5910(01)00391-6). [40]. Potanina E., Golovkina L., Orlova A., Nokhrin A., Boldin M., Sakharov N. Lanthanide (Nd,

AC C

Gd) compounds with garnet and monazite structures. Powders synthesis by "wet" chemistry to sintering ceramics by Spark Plasma Sintering // Journal of Nuclear Materials. 2016. V. 473. P. 9398. (http://dx.doi.org/10.1016/j.jnucmat.2016.02.014). [41]. Chuvil’deev V.N., Boldin M.S., Dyatlova Ya.G., Rumyantsev V.I., Ordan’yan S.S. A comparative study of hot pressing and Spark Plasma Sintering of Al2O3-ZrO2-Ti(C,N) powders // Inorganic

Materials.

2015.

V.

(http://dx/doi/org/10.1134/S0020168515090034).

51.

Iss.

10.

P.

1047-1053.

ACCEPTED MANUSCRIPT

[42]. Nanda A.K., Watabe M., Kurokawa K. The sintering kinetics of ultrafine tungsten carbide powders

//

Ceramics

International.

2011.

V.

37.

Iss.

7.

P.

2643-2654.

(http://dx.doi.org/10.1016/j.ceramint.2011.04.011). [43]. Buhsmer C.P., Crayton P.H. Carbon self-diffusion in tungsten carbide // Journal of Materials

RI PT

Science. 1971. V. 6. Iss. 7. P. 981-988. (http://dx/doi/org/10.1007/BF00549949). [44]. Properties, Production and Application of Refractory Compounds. Handbook / Ed. Kosolapova T.Y. Moscow: Metallurgiya. 1986. 928 p. (in Russian)

SC

[45]. Erdelyi G., Beke D.L. Chapter 11 Dislocation and grain boundary diffusion in non-metallic system. In book “Diffusion in Non-Metallic Solids (Part 1)”. London-Bornstein – Group III

M AN U

Condensed Mater. 1999. V. 33B1. P.1-48.

[46]. Samsonov G.V., Vinitskii I.M. The Refractory Compounds. Moscow: Metallurgiya. 1976. 560 p. (in Russian)

[47]. Kharatyan S.L., Chatilyan H.A., Arakelyan L.H. Kinetics of tungsten carbidization under nonisothermal conditions // Materials Research Bulletin. 2008. V. 43. Iss. 4. P. 897-906.

TE D

(http://dx.doi.org/10.1016/j.materresbull.2007.05.003). [48]. Adorian A., Schubert W.D., Schön A., Bock A., Zeiler B. WC grain growth during the early

EP

stages of sintering // International Journal of Refractory Metals and Hard Materials. 2006. V. 24. P. 365-373. (http://dx.doi.org/10.1016/j.ijrmhm.2005.11.009).

AC C

[49]. Chuvil’deev V.N. Non-equilibrium grain boundaries in metals. Theory and applications. Moscow: Fizmatlit. 2004. 304 p. (in Russian) [50]. Segal V.M., Beyerlein I.J., Tome C.N., Chuvil’deev V.N., Kopylov V.I. Fundamentals and Engineering of Severe Plastic Deformation – New York: Nova Science Publishers, 2010, 542 p. [51]. Chuvil'deev V.N., Kopylov V.I., Zeiger W. Non-equilibrium grain boundaries. Theory and its applications for describing nano- and micro-crystalline materials processed by ECAP // Annales de Chimie: Science des Materiaux. 2002. V. 27. Iss. 3. P. 55-64.

ACCEPTED MANUSCRIPT

[52]. Nokhrin A.V. Effect of grain-boundary diffusion acceleration during recrystallization in submicrocrystalline metals and alloys prepared by severe plastic deformation // Technical Physics

AC C

EP

TE D

M AN U

SC

RI PT

Letters. 2012. V. 38. Iss. 7. P. 630-633. (http://dx.doi.org/10.1134/S1063785012070073).

ACCEPTED MANUSCRIPT

Table 1. Impact of the initial particle size and volume fraction of tungsten monocarbide on standard SPS temperatures and structural parameters of sintered tungsten carbide samples (V=25 оС/min, pressure 75 MPa)

Standard SPS R0 ,

m2/g

nm

1

8.34

46

2

6.87

3

Series

Average gain size(3)

temperatures

f, %

о

Activation energy,

о

dm, µm

(5)

Т1, С

Т2, С

83.5

1145

1420

0.1

55

91.7

1370-1405

1495

0.1

6.01

63

93.6

1460

1520

0.3

4

3.36

113

99.5(2)

1330

1380-1390

2.5-10

5

5.03

80

100

1425

1505

6

5.27

72

100

1365-1375

1490

7(1)

3.40

112

100

1430

da, µm

Qb/kTm

Hardness Hv, GPa(3)

5.2

28.0

8

5.0

26.0

20

4.7

20.0

17-20

4.1

16.5

0.35

1.2-1.3

10.4

34.0

0.1-0.15

(4)

10.8

29.0

(4)

7.7

31.0

SC

1-3

M AN U 1515

(5)

RI PT

SBET,

0.2-0.3

(1)

– H.C. Starck powder.

(2)

– volume fraction was calculated at X-ray scanning rate of 0.5 о/min (in all the rest cases the

scanning rate was 2оC/min). – after sintering at 1550оС.

(4)

– no abnormal grain growth

(5)

– accuracy of dm depended on the grain size distribution histogram and did not exceed 10% of the

TE D

(3)

AC C

EP

average value for dm and did not exceed 20% of the average da value.

ACCEPTED MANUSCRIPT

Table 2 Chemical composition of studied nanopowders (wt. %) Series

W

C

O

Fe

Cu

Co

Cr

Ca

Mg

Al

Mo

Ni

Si

1

Balance 6.32 0.28 0.017 0.085 0.0017 0.008

0.72 0.026 0.003 0.031 0.003 0.96

2

Balance 6.24 0.19 0.025 0.063 0.0025 0.032

0.98 0.023 0.004 0.019 0.01

3

Balance 6.12 0.22 0.035 0.165 0.0038 0.037

2.40 0.041 0.004 0.001 0.025 0.93

4

Balance 6.13 0.09 0.066

0.29 0.0066 0.008 3.74

5

Balance 6.12 0.05 0.052

0.17 0.0003 0.02

6

Balance 6.13 0.05 0.058 0.063 0.0025 0.083

0.004

1.77 0.032 0.006 0.003 0.005 0.37 1.87 0.067 0.02 -

0.064 0.001

EP

TE D

M AN U

SC

– H.C. Starck powder

AC C

(1)

-

0.21 0.025 0.027 0.028 1.38

RI PT

7(1) Balance 6.12 0.02 0.009 0.095

1.20

0.02 0.018 0.88 -

-

0.15

ACCEPTED MANUSCRIPT

Table 3 Density and mechanical properties of tungsten carbide samples obtained through SPS and characteristics of samples obtained by other authors Material

ρ, %

Samples of tungsten carbide obtained using SPS method from α-WC

98.3 98.6 99.1 99.3 99.5 99.7 99.7 99.2 99.4 99.8

tungsten carbide nanopowder synthesized through DC arc plasma

(f=100%, R0=72-112 nm)

Samples of tungsten carbide obtained using SPS method from α-WC

(R0=40-70 nm, α-WC + 2.0 at. %C)

M AN U

Samples obtained using SPS method [6]

Samples obtained using SPS method [6] (R0=40-70 nm)

TE D

Samples obtained using high frequency

induction heating sintering method (HFIHS) [3] (f=100%)

(f=100%)

EP

Samples obtained using HFIHS method [34]

Samples obtained using HFIHS method [35]

AC C

(W2C phase is present)

Tungsten monocarbide samples obtained using SPS method [36]

Samples obtained using hot pressure method [37] (R0=500-700 nm)

94.7 95.5 97.8 99.2

23.5 23.7 25.4 26.2

-

100 100 ~100 100 100 99.0

26.8 26.6 27.0 26.3 25.6 25.8 29.6

8.3 7.1

96.9 98.5 97.5 98.5

26.7 28.0 20.1 26.5 25.0

6.3 7.1 7.0 4.4 4.8

95.0 97.6 93.0 84.0 80.0 97.5 98.4 97.9 96.5

22.8 24.3 21.4 10.1 7.8 ~26.0 ~24.8 ~17.0 ~15.5

7.2 6.6 5.8 ~6.1 ~5.8 ~5.4 ~5.0

SC

H.C. Starck nanopowder (present paper) (f=100%, R0=113 nm)

K1c, MPa·m1/2 6.0 5.7 6.7 5.4 5.2 5.2 4.3 5.1 4.0 4.5

RI PT

method (present paper)

Hv, GPa 16.5 19.5 24.2 26.5 27.1 31.1 34.0 26.6 31.0 26.0

α-WC α-WC + 1.61 at.%C α-WC + 2.36 at.%C α-WC + 3.17 at.%C R0 = 106 nm R0 = 87 nm

R0 = 400 nm R0 = 1300 nm R0=45 nm R0=73 nm R0=400 nm R0=1300 nm R0=2400 nm R0=4300 nm α-WC α-WC + 0.5 wt.%W α-WC + 0.4 wt.%C α-WC + 0.4 wt.%C + 0.5 wt.%W

ACCEPTED MANUSCRIPT

List of figures

Figure 1. X-ray diffraction analysis (a) и microphotography (b) of W-C system after DC air thermal plasma synthesis.

annealing in hydrogen according to X-ray diffraction analysis.

RI PT

Figure 2. Changing phase composition of powders in tungsten-carbide system during recovery

SC

Figure 3. Microphotographs (a), particle size distribution histogram (b) and conglomerate size distribution histogram (c) of tungsten carbide nanopowder of Series 6. Scanning electron

scaled particles (conglomerates).

M AN U

microscopy. Fig. 3b, c: N – number of particles (conglomerates) of R size, N∑ - total number of

Figure 4. X-ray diffraction patterns for powders of Series 2 (R0=55 nm, f=91.7%) (a), Series 3

TE D

(R0=63 nm, f=93.6%) (b) and Series 4 (R0=113 nm, f=99.5%) (c).

Figure 5. Dependence of shrinkage amount (а) and shrinkage rate (b) on heating temperature of

о

EP

tungsten carbide nanopowders during SPS. Series 4 (f= 99.5%, R0=113 nm). Heating rate V=25

AC C

С/min, pressure 75 MPa.

Figure 6. Dependence of SPS optimal temperature on the initial size of tungsten carbide particles (heating rate V=25 оС/min, pressure 75 MPa, vacuum).

Figure 7. Abnormal grain growth in tungsten carbide during SPS (heating rate 25 оС/min, pressure 75 MPa): a – Series 1 (volume fraction of monocarbide f=83.5%, sintering temperature Т=1550оС, average size of matrix grains 0.1 µm, average size of abnormal grains ~1-3 µm); b – Series 3 (volume fraction of monocarbide f=93.6%, sintering temperature Т=1460оС, average of matrix

ACCEPTED MANUSCRIPT

grains 0.1 µm, average size of abnormal grains ~20 µm). Scanning electron microscopy of fractures.

Figure 8. Grain size distribution histograms in case of normal grain growth (Fig. 8a, c, e) and

RI PT

abnormal grain growth (Fig. 8b, d, f) while sintering nanopowders of Series 1 (Fig. 8b, d, f) and Series 6 (Fig. 8a, c, e). Sintering temperature: 1500 оС (a, b), 1550 oC (c, d), 1720 oC (e, f)

SC

Figure 9. Diffraction patterns of ceramics sintered from nanopowders with volume fraction of tungsten monocarbide f=93.6% (a) and 99.5% (b). Heating rate V=25 оС/min, sintering temperature

M AN U

1550оС, pressure 75 MPa

Figure 10. Dependence of relative density on the initial particle size of tungsten monocarbide obtained through DC arc plasma synthesis. SPS at temperatures 1500оС (line 1) and 1550оС (lines 2

TE D

and 3). Line 2 – sintering of nanopowders with volume fraction of monocarbide under 93%. Line 3 – sintering of nanopowders with volume fraction of monocarbide 99.5-100%.

EP

Figure 11. Dependence of density, hardness and grain size in tungsten carbide sintered samples on

AC C

sintering temperature at heating rate 2400 °C/min. Series 6.

Figure 12. Dependence of sintering activation energy on grain boundaries migration rate

Figure 13. Accumulation of dislocations along migrating grain boundaries during annealing of UFG materials. OMD are markedred. Scheme.

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

20 nm

RI PT

(b)

Intensity

(a)

Figure 1

ACCEPTED MANUSCRIPT 1 – α-WC 2 – W 2C 3 – WxC1-x 4 – α-W

1

1

o

DC synthesis 1030 C, 5 h. 2 SBET=6.9 m /g 1 1

1

1

1

1

RI PT

1 1

o

1

DC synthesis 970 C, 5 h. 2 SBET=8.89 m /g

1

M AN U

1

1

1

1 1

SC

4

o

DC synthesis 930 C, 5 h. 2 SBET=13.5 m /g

2

3

4

3

1

11

TE D

1

o

DC synthesis 870 C, 5 h. 2 SBET=21.4 m /g

2

4

AC C

EP

2

4

4

3

2

3 Initial powders. 2 SBET=27.4 m /g 3

3

3

4 3

Figure 2

3

3 3

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

(a)

AC C

EP

TE D

(b)

Figure 3

(с)

Figure 4

α-WC

α-WC α-WC

α-WC

β-WC α-WC

α-W

α-W

α-WC

AC C α-WC

α-WC

EP

α-WC

α-WC

α-WC

α-WC α-WC

β-WC

α-WC

α-WC

M AN U

β-WC

TE D

α-WC

SC

β-WC

β-WC

α-WC

α-WC

α-WC

α-WC α-WC

RI PT β-WC

β-WC

β-WC

α-WC

α-WC

α-WC

ACCEPTED MANUSCRIPT (a)

(b)

(c)

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

(a)

Figure 5

(b)

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 6

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

(a)

Figure 7

(b)

ACCEPTED MANUSCRIPT (a)

(d)

AC C

EP

TE D

M AN U

(c)

SC

RI PT

(b)

Figure 8

(e)

(f)

Figure 9

α-WC

W 2C

W 2C

W 2C

W 2C

α-WC

AC C EP

α-WC

TE D α-WC

W 2C α-WC

SC W 2C

W 2C

W 2C

α-WC

M AN U

α-WC

W 2C

RI PT

α-WC

α-WC

α-WC

ACCEPTED MANUSCRIPT

(a)

(b)

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 10

R0, nm

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 11

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 12

ACCEPTED MANUSCRIPT

Db

RI PT

Vm

SC

δ=2b Db* > Db

TE D

M AN U

Vm

Db* >> Db Vm

AC C

EP

δ=2b

δ=2b

Figure 13

ACCEPTED MANUSCRIPT

Highlights

Samples of nanopowders tungsten carbide were made using DC arc plasma synthesis.

•

Samples of ceramic WC with high mechanical properties were made using the SPS.

•

Abnormal grain growth while SPS tungsten carbide stems from α-W and W2C particles.

•

Under abnormal grain growth, the values of the SPS activation energy are low.

AC C

EP

TE D

M AN U

SC

RI PT

•