«STRUCTURE AND PROPERTIES OF ELECTRODEPOSITED NANOCRYSTALLINE NI AND NI-FE ALLOY CONTINUOUS FOILS by Jason Derek Giallonardo A thesis submitted in ...»
With the exception of samples no. 1-3 the anisothermal anneal curves experienced a minimal amount of exothermic heat release prior to the main heat release corresponding to grain growth. This initial heat release is considered to be due to subgrain coalescence leading to the onset of the abnormal grain growth stage in the annealing process [Klement et
Anisothermal anneal (DSC) curve for sample no. 4 (Ni, 37 nm).
Anisothermal anneal (DSC) curve for sample no. 9 (Ni-32wt.%Fe, 10 nm).
al. (1995a), Klement et al. (1995b)]. If the initial heat release is combined with the main grain growth heat release, both “nucleation” and abnormal grain growth processes may be taking place followed by normal grain growth – this was the case for samples no. 4-6 and the combined release was integrated to determine the total enthalpy. If the initial heat release may be distinguished clearly from the grain growth heat release, “nucleation” is likely the dominating process followed by normal grain growth – this was the case for samples no. 7-9 and only the main grain growth peak was integrated to determine the total enthalpy. In addition to determining the total enthalpy, the maximum or peak temperature, Tp, of the grain growth peak was reported for each sample. It should be noted that the increase in peak temperature, Tp, with increasing Fe is of particular importance to the use these materials in applications where the temperature is relatively high. By alloying, a stabilizing effect is imparted [Erb et al. (2007)] which allows for the use of these materials in various higher temperature applications, for example, wear resistance at elevated temperatures.
In the case of the nanocrystalline Ni samples, the Curie temperatures are relatively constant and agree well with what is reported in the phase diagram (Fig. 4.1). A decrease in grain size increases the measured total enthalpy. This trend is consistent with Turi (1997) who produced anisothermal anneal curves using modulated differential scanning calorimetry (MDSC) with nanocrystalline nickel samples with grain sizes of 10, 20, and 30 nm. In the analysis carried out by Turi (1997), a distinct relationship was made between grain size or intercrystal volume fraction and total enthalpy based on the assumption that the only defects present in considerable quantities were intercrystal defects. Fig. 4.34 shows a plot of the total enthalpy versus grain size for the nanocrystalline Ni samples. The total enthalpy values of the nanocrystalline Ni samples agree well with those of Turi (1997). A direct relationship between the nanocrystalline Ni-Fe samples and grain size cannot be made mainly due to other possible contributing factors, such as ordering to form the FeNi3 [Turi (1997)]. Instead, the total enthalpy values are plotted as a function of Fe concentration in the deposit (Fig.
The Curie temperature for sample no. 6 (Ni-7.3wt.%Fe) was determined to be about 433oC and is slightly lower than what is reported in the phase diagram (Fig. 4.1). Curie temperatures of the remaining Fe containing samples were not detected since they occur at values greater than 500oC which is outside of the upper limit of the scanned range in the DSC measurements. In general, the total enthalpy values for those samples containing Fe are significantly higher when compared to the nanocrystalline Ni samples. This in part due to the Ni-Fe alloys having smaller grain sizes, i.e., samples no. 7, 8, and 9. In addition to this,
Total enthalpy plotted as a function of grain size for the nanocrystalline Ni samples. The dashed lines are approximate trend lines.
non-equilibrium structures are normally observed in electrodeposited Ni-Fe alloys [Fukumuro et al. (2004)]. As a result, the formation of an equilibrium structure from the non-equilibrium structure may represent a significant portion of the measured total enthalpy [Turi (1997)]. Another possible contributing factor in the nanocrystalline Ni-Fe samples is the reduction of growth faults during the grain growth process. As was seen in the HR-TEM image analysis and the XRD analysis, there was a characteristic increase in the occurrence of growth faults with increasing Fe. As a result, there is also more stored energy in the form of lattice defects. Thus, for the nanocrystalline Ni-Fe alloys, it may be said that there is also an alloying effect associated with the total enthalpy released during anisothermal annealing related to a combination of the formation of equilibrium structures and the reduction of lattice defects such as growth faults.
4.6.2. Excess Interfacial Enthalpy Since the energy released in the nanocrystalline Ni samples can almost be exclusively a result of reducing intercrystal defects [Turi (1997)], a direct relationship between the total enthalpy, H total, and the intercrystal volume fraction can be derived to determine the excess interfacial enthalpy, H ic [Wang et al. (1997)]. The relationship is derived using the model developed by Palumbo et al. (1990) that quantifies the intercrystal volume fraction in polycrystalline materials based on grains having a tetrakaidecahedron shape. The change in intercrystal volume fraction as a result of the anisothermal annealing process is given by,
where, d f and d i are the final and initial grain sizes and fic is the intercrystal volume fraction. Substituting the total intercrystal volume fraction equation (Eq. 2-1) given by Palumbo et al. (1990) in to Eq. 4-4, we arrive at the following expression that allows for the
where, Δ in the expression on the right hand side of the equal sign is the grain boundary thickness and is assumed to be 1 nm. Finally, the relationship used to obtain the excess interfacial enthalpy, H ic, is given as follows,
For the final grain size, d f, a reasonable value of 1000 nm was taken. Table 4.9 provides a summary of the values for excess interfacial enthalpy, H ic. The excess interfacial enthalpy or interfacial energy, H ic, may then be multiplied by the density of Ni (8.9 g/cm3) and the grain boundary thickness (1 nm) to determine the grain boundary energy, gb.
Fig. 4.36 shows a plot of the excess interfacial energy with grain size. The grain boundary energy values are also plotted with grain size in Fig. 4.37. In both cases, the plots show a linear decreasing trend for grain boundary energy as a function of grain size which agrees well with the data of Turi (1997) for nanocrystalline Ni, and is consistent with Lu et al. (1993) who studied a nanocrystalline Ni-P system. At grain sizes above 30 nm, the grain boundary energies are comparable to experimentally extrapolated values at their respective
1.2 1.0 0.8 0.6
peak temperatures determined for conventional polycrystalline Ni [Murr (1975)]. At grain sizes less than 20 nm, there is a clear reduction in the grain boundary energy values. There are several possibilities for the observed reduction in grain boundary energy below 20 nm.
Some of the possibilities were considered by Turi (1997): (1) the influence of solute segregation where it is known that the addition of S to grain boundaries can reduce the energy in Ni [Murr (1975)] – in this particular case, the S concentration in the deposit was found to increase with decreasing grain size, (2) the texture of the material which has been shown to be quite different for these two samples may also have an influence whereby grain boundary character distribution may also vary, and (3) the volume fraction of triple junctions and their characteristics (i.e., I-lines or U-lines) may also have a profound effect on the interfacial energy.
4.7. Summary A series of nanocrystalline Ni and Ni-Fe alloys were synthesized using the electrodeposition technique. The nanocrystalline Ni and Ni-Fe alloys were produced with considerable amounts of impurities namely S and C. The S concentration in the deposit was found to be dependent on the amount of Nanovate™ A24 present in the electrolyte. The C concentration in the deposit varied and no relationships with the organics in the electrolyte were noted. In the case of the nanocrystalline Ni-Fe samples, increasing the Fe in the electrolyte resulted in a systematic increase of Fe concentration in the deposit. The deposit quality for all samples was consistent in that they were all smooth, except for the largest grain size sample (no. 1, ~255 nm). There was also no presence of microcracking. Surface oxides were visible on the sample surface which became more prominent with the addition of Fe.
Microstructural characterization was performed on the series of nanocrystalline Ni and Ni-Fe alloys using a variety of analytical techniques. The grain sizes of the samples were initially determined using TEM image analysis. Additional HR-TEM image analysis was performed on selected samples for the purpose of examining the defects present in these materials. Examples of a triple junction, high angle grain boundary, low angle grain boundary and twin boundary were shown. The grain boundaries analyzed all displayed some level of accommodation strain which was consistent with other HR-TEM studies on nanocrystalline materials synthesized using different means. Also present in the HR-TEM images were features in the grain interiors consistent with stacking faults. Since no deformation was imposed on these samples, the stacking faults are likely to be twin faults of
the growth type, or “growth faults”. These growth faults became more prominent with increasing Fe concentration in the deposit.
The lattice parameters determined for all of the samples agreed well with their polycrystalline counterparts. In the case of the nanocrystalline Ni-Fe alloys, increasing Fe was found to increase the lattice parameter consistent with the predictions of Vegard’s law.
All samples contained various degrees of crystallographic texture. Most samples were found to have a strong (200) fibre texture while others had either a weak (200) fibre texture or a double (111)(200) fibre texture. Grain sizes were determined by XRD pattern analysis and compared with those determined using TEM image analysis. XRD grain sizes that were less than 20 nm were found to agree well with the TEM grain sizes. The XRD method provided significant underestimations at TEM grain sizes greater than 20 nm. The growth fault probability of each sample was quantified by XRD pattern analysis based on peak asymmetries. The growth fault probabilities for the nanocrystalline Ni samples were relatively constant over the grain size range. In the case of the nanocrystalline Ni-Fe samples, the growth fault probabilities increased linearly with the addition of Fe. This trend was found to be consistent with polycrystalline Ni-Fe alloys when comparing to the respective stacking fault energies. The finding also supports the earlier HR-TEM studies which suggested the presence of growth faults and their increased presence with increasing Fe.
Finally, thermal analysis of the nanocrystalline Ni samples provided results that were consistent with previous studies. The total enthalpy (or stored energy) release values
compared well with the increasing trend observed with decreasing grain size and consistent results were also observed for the grain boundary energy estimations. The nanocrystalline Ni-Fe alloys displayed total enthalpy (or stored energy) release values which were consistently higher than those of the nanocrystalline Ni samples when related to grain size, which is also consistent with previous studies. A direct relationship with grain size was not found to be feasible due to other influencing factors which contribute to the heat release event corresponding to grain growth. These factors include ordering of the atomic structure since it is likely that they were deposited in a non-equilibrium state and the reduction of growth faults which are known to be present in increasing quantities with increasing Fe concentration in the deposit.
4.8. References Basinski, Z.S., W. Hume-Rothery and A.L. Sutton, Proc. R. Soc. A229 (1955) 459.
Bradley, A.J., A.H. Jay and A. Taylor, Phil. Mag. 23 (1937) 545.
Carter, C.B. and S.M. Holmes, Phil. Mag. 35 (1977) 1161.
Charnock, W. and J. Nutting, Met. Sci. J. 1 (1967) 123.
Cheung, C.K.S., F. Djuanda, U. Erb and G. Palumbo, Nanostruct. Mat., 5 (1995) 513.
Cohen, J.B. and C.N.J. Wagner, J. Appl. Phys. 33 (1962) 2073.
Cranshaw, T.E., J. Phys. F: Met. Phys, 17 (1987) 967.
Cullity, B.D. and S.R. Stock, “Elements of X-Ray Diffraction, Third Edition”, Prentice Hall, New Jersey, 2001.
Czerwinski, F., J. Electrochem. Soc. 143 (1996) 3327.
Czerwinski, F., H. Li, M. Megret, J.A. Szpunar, D.G. Clark and U. Erb, Scripta Mater. 37 (1997) 1967.
Czerwinski, F. and Z. Kedzierski, J. Mater. Sci. 32 (1997) 2957.
Czerwinski, F., Elecrochim. Acta 44 (1998) 667.
Ebrahimi, F. and H.Q. Li, Rev. Adv. Mater. Sci. 5 (2003) 134.
El-Sherik, A.M. and U. Erb, J. Mater. Sci., 30 (1995) 5743.
Erb, U., K.T. Aust and G. Palumbo, in “Nanostructured Materials: Processing, Properties, and Applications – 2nd Edition”, Ed. C.C. Koch, Noyes | William Andrew, New York, 2007.
Fukumuro, N., M. Imai, M. Chikazawa and T. Watanabe, in “Nano-Plating”, Ed. T.
Watanabe, Elsevier, New York, 2004.
Giallonardo, J.D., U. Erb, G. Palumbo, G.A. Botton and C. Andrei, Mater. Sci. For. 706-709 (2012) 1607.
Grimmett, D.L., M. Schwartz and K. Nobe, J. Electrochem. Soc., 140 (1993) 973.
Haasz, T.R., K.T. Aust, G. Palumbo, A.M. El-Sherik and U. Erb, Scripta Metall. Mater. 32 (1995) 423.
He. J., J. Ye, E.J. Lavernia, D. Matejczyk, C. Bampton and J.M. Schoenung, J. Mater. Sci. 39 (2004) 6957.
Howie, A., Proceedings of the Conference on Electron Microscopy, Delft: de Nederlandse Vereniging Voor Electronmicrocopie, 1961, 383.