Silicon carbide SiC particles present more and more interest in many material processes. In ceramic elaboration, particles or whiskers are incorporated to melts. This component got advantages such as a low pressing temperature; it also improves mechanical properties and confers super elasticity to composite at high temperature (Sakka et al., 1995; Vassen and Stober, 1999; Shinoda et al., 1999). But in solution, SiC particles tend to agglomerate causing then non homogeneity in the ceramic microstructure leading to a low density and bad mechanical properties.
Elaboration of new composite constituted of metallic matrix with dispersed
inert particles is a wide investigation field (Tomaszewski et al., 1969;
Snaith and Groves, 1972). Nickel electroplating process consisting in Ni2+
ions reduction form Watt bath is very well established (George, 1992).
This electrochemical process is used to embed inert particles into the metallic
matrix being electrodeposited for composite elaboration.
The Ni/SiC system has widely been studied (Guglielmi et al., 1972; Lavanant, 1991). Presence of SiC particles in the coats improves mechanical properties and prevents 50 times from corrosion compared to only Ni coat after heating at 400°C (Wang et al., 1999). The Ni/SiC co deposition is also applied in car industry mainly for engines coating.
Investigations undertaken on similar systems led to diverse results. Co deposition
of alumina powder in copper matrix has revealed a better incorporation of α-Al2O3
in contrary to γ-Al2O3 particles (Lee et al.,
1988). This was attributed to the negative zeta potential of γ-Al2O3
particles in CuSO4 acidic solutions. The same approach with a nickel
matrix pulled out γ-Al2O3 particles incorporation
level to depend on Ni reduction courant density and a competition between charge
transfer controlled by Ni2+ ions reduction and mass transfer of γ-
Al2O3 particles for high courant (Webb and Neil, 1994).
Silica co deposition in Ni matrix compared to SiC one showed a very low incorporation
level for SiO2 (Nowak et al., 2000). SiO2 particles
were supposed to enhance specific capacitance of the electrode while this should
decrease in presence of SiC particles in the electrolyte. Thus for a given metal,
incorporation may depend on particles and even on its allotropic forms. A physicochemical
activity between particles surface and the electrolyte may exist and could explain
those changes. The interactions between particles surface, electrolyte and the
metal in reduction may so be responsible of incorporation level. Particles surface
may chemically advance according to different parameters such as electrolyte
composition, pH values (Diakova et al., 2002; Hashiba et al.,
1988; Snaith and Groves, 1977) and presence of refracting agents in the colloidal
phases (Beebhas et al., 1990; Wang and Wei, 1995).
The experimental or process parameters also play an important rule in the co
deposition state (Snaith and Groves, 1978; Feng et al., 2006). The best
incorporation level for SiC particles in nickel matrix may be relied to its
surface state during plating (Kaisheva and Fransaer, 2004). SiC surface may
present particular chemical groups that interact with the electrolyte to favour
incorporation in the metallic matrix. In the present study, we try to establish
SiC surface evolution in Watt bath constituted of NiSO4 and NiCl2
and also, the effect of chloride and per chlorate ions at different pH values
in other to find out an explanation to the good incorporation of these particles
in the nickel matrix.
MATERIALS AND METHODS
Solution of NaCl, NaClO4 and of watt bath salt that are NiCl2 and NiSO4 were elaborated at 102M concentration with 2 g L1 SiC particles. The pH of the solution was fixed in the range of 1.5 to 4 with the counter ion acid. The samples were kept shacked for 10 days and than left to sediment during 24 h at least. Then two pH values are measured with a pH meter PH M210 type METERLAB of TACUSSEL-ELECTRONIC.
Foundation of these experiments is based on that, pH values fixed when elaborating samples represent the liquid phase one, when no interaction did occur with the SiC particles. So, pH1 corresponds to the supernatant liquid after dismissing, while pH2 is determined for the stirred phases. Two variables are then established from experimental data according to the following relations:
ΔpH1 may inform on the reaction that takes place during interaction between solution and particles surface while ΔpH2 gives indication on the kind of chemical species that are fixed on the surface. Zeta potential that represents charges moving with particles through an electric field is also determined with a zetasizer 3000HS coupled to zetasiser software. Zeta potential values in accordance with pH variation should permit to propose a chemical reaction scheme that leads to changes noticed and the surface state. Consciousness of the surface state may thus permit to understand the good incorporation of SiC particles and also agglomeration phenomenon in ceramic melts.
RESULTS AND DISCUSSION
Measurements of pH values show that they change confirming thus a chemical activity between particles surface and solutions. Zeta potential also indicates variation revealing evolution of particles surface state.
Analysis for chloride and per chlorate solutions: Figure
1a and b that, respectively represent changes in chloride
and per chlorate solution indicate that ΔpH1 is negative until
pH 3.15 for chloride and 3.8 for per chlorate. On these scales, interaction
has caused acidification of the supernatant liquid phase. The difference of
pH scale on which acidification occurs for both anions reveals their implication
in this evolution. For higher pH values in both cases, ΔpH1 becomes
positive; the interaction on these pH ranges may cause proton consuming. The
pH of some chloride samples have been readjusted to initial values.
and zeta potential evolution versus pH in chloride solution|
||ΔpH and zeta potential evolution versus pH in per chlorate
||ΔpH change of supernatant liquid after pH adjustment
Figure 1c shows that ΔpH1 still varies with
the same shape on the main pH scale. ΔpH1 variation may so be
the result of an equilibrium state between SiC particles surface and the solutions.
ΔpH2 variations on the same scale show that stirred solution
is lest acidic.
This means that basic species are released into the solution and react with proton making pH increase. These basic species may necessary come from particles surface. Acidification is widely higher for per chlorate solutions then chlorides one. ClO4- may activate the interaction that produce proton and this also confirm one more time that the anions present in solution are implicated in the pH evolution.
Zeta potential values for both solutions show that particles got positive surface charges in high acidic solutions and decrease to attain the isoelectric pH point which values are, respectively around 3 for chloride and 2.5 for per chlorate. Surface charges for both become negative for higher pH values. Positive zeta potential values may be relied to protons activity. Protons may adsorb on a specific chemical group at particles surface producing positive charges.
Charges values thus decrease as proton activity decrease. The iso electric pH value (pHiep) in both solutions are much closed to the silica one as determined elsewhere (Yeh and Van, 1994). In aqueous solution, SiO2 creates silanol group that are protonated and make zeta potential positive according to reaction 3:
Reaction 3 does not free proton; so acidification necessary implies another
reaction. Other chemical groups may so be responsible of the supernatant liquid
acidification. Silicon carbide powder have been analysed by XPS (Binner and
Zhang, 2001). These investigations pulled out presence of bonds such as C-C,
Si-O, O-O-Si. In another work, oxycarbides SiO2C2 and
SiO3C have been found on SiC particles surfaces (Mejean et al.,
1995). The developed formula of these chemical groups can be represented as
on Fig. 2a and b.
Those two species are characterised by all prevailing different bonds. Thereby, they can be considered as being the major components at the particles surface. For both components, reaction with water leads to silanols formations that are protonated in acidic middle and confer positive charges to particles:
During sedimentation, Na+ ions that are very badly solvated are spurned to the SiC particles surface to form soda salts and release protons in the supernatant phase that becomes more acidic:
Acidification level is higher in per chlorate solution then chloride one. This
means that the anions are implied in reaction 4 and 5. By its oxygen atoms,
ClO4- ions easily adsorb on more silicon atoms.
SiC2O2 and 2b) SiO3C|
This favours a better protonation leading to reaction 4 and 5 to form more
silanol groups. Chloride ions that are bigger hardly adsorb in the same condition
limiting then the reactions. Acidification becomes less in both solutions when
pH increases; this shows that protonation step is decisive. As proton activity
fails, reactions 4 and 5 hardly take place reducing by the same way acidification
of the supernatant phase.
Difference between the pHiep for chloride and per chlorate solution is to be relied to their adsorption capacity. The pHiep in per chlorate is more acidic then in chloride one. Per chlorate ions do adsorb more so that charge neutralization occurs in more acidic solution. From pH 3.15 in chloride solution and 3.8 in per chlorate one, supernatant liquids become less acidic. And on the same pH range, particles are negatively charged. A reaction that consumes proton occurs. Hydroxide ions may so be released in the liquid phase. Proton activity becomes so feeble that protonation does no longer occur. The main reaction that can produce hydroxides can be schemed as follow:
For reaction 8, a molecule of water reacts with the basic α position oxygen to free an hydroxides. The positive charge variation +δ on the bond liking silicon and the α oxygen protonated is stabilized by the anion X-through its free electrons doublet on the surface and make zeta potential equal to zero.
For reaction 9, the isomeric form of the carbonyl function prevails; the positive charge appearing on the carbon is stabilized by anion adsorption. Then oxygen anion reacts with a molecule of water releasing the hydroxide. Adsorption of other X-anions and hydroxides give a negative charge to the particles. Reactions 4 and 5 still prevail in per chlorate solution until pH 3.8 because its adsorption mode favours these reactions. For SiO3C, it may render the oxygen at silicon atom α position more basic so that despite a feeble proton activity, protonation still occurs. For SiC2O2, adsorption of chlorate ion may stabilize the isomeric form that guide reaction 4, while these no longer exist in chloride solutions after pH 3.15. The molar quantity of protons free by reactions 6 and 7 or consumed by reaction 8 and 9 is calculated by relation 10:
Decimal logarithm representation of free or consumed protons versus pH of the
solution is taken positive when consumed. Figure 3a and b
representing, respectively chloride and per chlorate solutions show that the
reaction that consumes protons does not depend on the pH solution and this confirms
reaction 8 mechanism.
representation of proton produced and consumed versus pH in chloride solutions|
representation of proton produced and consumed versus pH in per chlorate
The molar quantity of proton consumed in both solutions is almost equal to
105 mol L1. This constancy indicates that the origin
of this phenomenon is something constant. In our cases, this can be relied to
SiC particles quantity that is constant.
On the other hand, proton production decreases with the same slop and this indicates that the same reaction occurs in the solutions on respective pH scales. This shows that sodium ion is lest spurned at the particles surfaces when pH increases. This is in accordance with the fact that as proton activity fails, sodium ions are better solvated in the solution by molecules of water. Reactions 6 and 7 are no longer favored.
Analysis for Nickel chloride solution: Chloride ions effect has already been established. The experiments undertaken here with nickel chloride solutions is to pull out whether nickel ions absorb at the particles surface as written elsewhere (Wang and Cheng, 2001; Grosjean et al., 1998). In case of nickel ions adsorption, zeta potential values should be very different form what found in the case of chloride solutions. Figure 4 shows that zeta potential also decreases with almost the same shape on the same scale with chloride solutions ones. This means that no nickel ion adsorbs at the particles surface. The pH of isoelectric point in nickel chloride solution is somewhat more acidic; 2.42 in other of 2.8 in chloride solutions.
This change is due to the fact that Nickel chloride solution contains twice more chloride ions so that absorption is higher inducing a more acidic pHiep. Nickel ion seems not to adsorb. It simply remains in the acidic solution to form Ni (H2O)6 2+ specie very stable in contrary to Na+ that is spurned at particles surfaces.
Analysis for Nickel sulphate solutions: As in the case of chloride and
per chlorate solutions, Fig. 5 also shows an acidification
of supernatant liquid. Except this similarity, Fig. 5 presents
a particular behavior for zeta potential values which are all negative for any
pH values and the stirred solutions that stays more acidic in contrary to what
was noticed for the other solutions.
and zeta potential in nickel chloride solution|
and zeta potential evolution versus pH in nickel sulphate solution|
representation of proton produced and consumed versus pH in sulphate solutions|
These particularities are necessary relied to sulphate ions in the solutions.
Sulphuric acid used to fix pH values, presents a second acidity for pH 2.99.
First acidity is characterised by dissociation that leads to hydro sulphate
ion; HSO4¯. From pH 2.99 the second acidity free sulphate ion
SO42¯. Zeta potential values in solution of pH values
inferior to second acidity stays constant around-6 mV and strongly decreases
after pH 3. The high decreasing of zeta values indicates that ions adsorbing
are more negative. For pH less than 3, the SO42-ions coming
from nickel sulphate salt dissociation react with water to form HSO4-that
adsorbed for a zeta value equal to-6 mV. The constancy of this value shows that
it only depends on NiSO4 concentration that is fixed. After pH 3,
sulphuric acid second acidity occurs so; more sulphate ions with higher negative
charge are disposal to adsorb leading to zeta potential decreasing (Fig.
6). This particularity shows that the origin of supernatant acidification
in the case of nickel sulphate solutions is different from that takes place
with chloride and per chlorate where spurning of sodium ions to the particles
surface permit. One more time, this indicates that nickel ions do not adsorb
on the pH range of investigation.
For nickel sulphate solutions, sulphate ions adsorb on the nucleophilic silicon sites of the components present at the particles surface. Then, protonation of the oxygen at silicon α position occurs. The silicon positive charge is so stabilized and can react with a molecule of water to release the proton conformal to reaction schemed as follows:
Then during sedimentation, the two-OH functions supported by the silicon form a peroxide link and release two protons that acidify the supernatant phase.
Reaction 12 takes place during sedimentation and some protons may so be retained in the dismiss and are released during stirring making the phase more acidic in contrary to what is noticed in chloride en per chlorate solutions. After pH 3, supernatant phase is also more basic. Reactions 8 and 9 may then take place, with SO42¯ adsorbing. Hydroxides are released in the middle and zeta potential becoming more negative because of SO42 charge. Relation 10 applied to these data; Fig. 5 also presents the same shape. This confirms the hypothesis of reactions that occur at the particles surface.
Experimentations undertaken in this study have revealed that SiC particles surface evolutes in zeta potential values according to solution pH and ions present in the middle. Nickel ion does not seem to adsorb and may stay solvated in Ni (H2O)62+ form in the solution for the pH range considered. In the papers pretending adsorption of nickel ions, the working pH was superior to 4.5. In such condition, Ni+-OH may form and adsorption of this specie can occur by oxygen free electrons. Chloride, per chlorate and sulphate ions adsorb on specific function such as SiO2C2 and SiO3C that exist at the surface. Their adsorption is effective on Si electrophilic site. SO42-when present in the middle adsorbs independently of pH, while Chloride and per chlorate ions adsorb when activity of protons that adsorb on silanol groups fails. The SiC particles embedment in the metallic matrix so depends on the pH of the solution. In acidic middle; pH inferior to 3, particles are positively charged by protons adsorption on the silanol group. So near the reduction pole, they disadsorbed to form hydrogen. Hydrogen forming on the surface prevents particles embedment. For pH around 3, particles are neutralized by adsorption of negatives species on electrophilic site of the components on their surfaces. Near the reduction pole, the negative species are vanished; the electrophilic site charge +δ prevailing so, favours sticking of the particles on the cathode for a good embedment. According to this process, particles are taken to the cathode surface, only by hydrodynamic forces and are not affected by the electric reduction field. During reduction, nickel matrix surface is formed by Ni-OH. An interaction that can occur between Sic particles surface as described and Ni-OH can stand as follows:
With P designing the particle; X-the negative specie adsorbed on its electrophilic site. The partial charges +δ on the particle and -δ on the oxygen stabilize the particle in the metallic matrix.
The worst embedment of SiO2 noticed may so be due to none present of positive charge or nucleophilic site ion Si-OH2+ so that near the reduction pole, protons are reduced in hydrogen and the SiO2 particle rejected into the solution. Concerning the worst embedment for γ-Al2O3 noticed, this could be relied to the oxide steric form that may not allow a good sticking by its aluminium electrophilic site. Watt baths are constituted by a same molar quantity of NiCl2 and NiSO4 for a pH value fixed to 3 with H3BO3. In such conditions, for pH value inferior to 3, protonation of silanol groups leads to positive zeta potential values according to what noticed from chloride solutions. Adsorption of hydro sulphate ions present in the middle is up to neutralise the particles charges. For pH equal to 3 the kinetic of protons reduction into hydrogen is slower than nickel ion one. Thereby, when the particle approaches the reduction pole, the hydro sulphate is vanished, the positive charge so prevails to favour sticking and the particle is so embedded in the nickel matrix. For ceramic elaboration, particles agglomeration is to be avoided. Therefore, they must stay charged to assure repulsive forces between them. The isoelectric pH value has to be defined to permit determination of working pH range.