G. Krauss

The morphology of martensite in iron alloys

Light and electron microscopy have been used to determine the main structural differences between the two major types of martensite in ferrous alloys. In the martensite that forms in dilute alloys of iron, the basic transformation unit takes the shape of a lath, and hence the term lath martensite is appropriate for identifying this morphology. Each lath is the result of a homogeneous shear, and successive shears produce a packet of parallel laths containing a high density of tangled dislocations. The other type, plate martensite, differs in the shape taken by a transformation unit and its transformation sequence is characterized by nonparallel plate formation. Investigation of a large number of binary ferrous systems shows that alloy composition and the transformation temperature influence the transition from lath to plate martensite. These two factors are discussed in terms of their possible effects on the plastic deformation mechanisms which must occur in the parent austenite and product martensite during transformation.

The tempering of Fe-C lath martensite

The changes in matrix structure that occur during tempering of an Fe-0.2C martensite at 400° to 700°C have been investigated. Light and electron metallographic observations show that when tempered, the fine martensitic lath structure coarsens while retaining the elongated packet-lath morphology. The as-quenched hardness 504 Khn and total grain boundary area per unit volume 50,800 cm−1 decrease abruptly at the higher tempering temperatures and in seconds reach relatively stable values that decrease slowly with time. The decrease in low angle boundaries accounts for most of the initial grain boundary area change, while the large angle boundary component of total boundary area decreases gradually with tempering time. Recovery processes are responsible for the initial changes in matrix structure, and carbide boundary pinning suppresses recrystallization until grain growth dominates in the later stages of tempering.

Metastable phases produced by laser melt quenching

A Nd: glass laser has been used to produce and retain metastable phases which are similar to phases obtained by rapid quenching (“splat cooling”) techniques. A single laser pulse of 10 msec duration was focused onto the surface of an alloy specimen to provide a beam intensity of approximately 3.7 × 105 watts per sq cm. A small volume of metal in the relatively large specimen immediately underwent melting. Upon cessation of the laser pulse, the molten metal solidified very quickly. Extremely rapid cooling was achieved due to the presence of nearly ideal conditions for conduction cooling. Laser melting of various alloys of the eutectic Ag-Cu system resulted in a complete series of solid solutions, as verified by X-ray analysis. In addition, the laser melt quenching technique provided a threefold increase in microhardness of the resolidified metal.

Microcracking and fatigue in a carburized steel

A carburized coarse-grained AISI 8620 steel was subjected to three postcarburization heat treatments: a) direct oil quench from the carburizing temperature (1700°F), b) direct oil quench, reheat to 1550°F and oil quench, and c) slow cool, reheat to 1550°F and oil quench, reheat to 1450°F and oil quench. The latter two treatments refined the austenitic grain size over that resulting from the direct quench and caused a reduction in the size of the marten-site plates and of the number and/or size of the microcracks within the plates. The refine-ment of the microstructure and the reduction of the number of microcracks resulted in greatly improved fatigue resistance,i.e., from fatigue limits of 140 to 250 ksi maximum cyclic stress for the direct quench and double reheat conditions, respectively. Subcritical crack growth was transgranular, but the mode of unstable crack propagation was mixed transgranular and intergranular in the direct quench and single reheat specimens. Obser -vations of microcrack coalescence and fracture surface features suggest that microcracks are instrumental in the transgranular mode of failure.

The Development of Martensitic Microstructure and Microcracking in an Fe-1.86C Alloy

The development of the martensitic microstructure in a 1.86 wt pct C steel has been followed by quantitative metallographic measurements over the transformation range of 0.12 to 0.50 fraction transformed (f). The transformation kinetics are described by the equationf = 1 − exp [−0.008 (Ms − Tq)] where Ms and Tq are the martensite start and the quenching temperatures respectively. Fullman’s analysis shows that the average volume per martensite plate decreases by almost an order of magnitude over the transformation range studied, but this decrease is less than that predicted by the Fisher analysis for partitioning of austenite by successive generations of martensite. Microcracking increases with increasingf up to 0.3, but does not increase forf above 0.3 where transformation proceeds by the nucleation of large numbers of small martensite plates. These observations indicate that a critical size of martensite plate is necessary to cause microcracking.

Microcracking sensitivity in Fe-C plate martensite

Metallographic analysis was used to study the effect of carbon content, grain size, quench rate, and retained austenite on microcracking in Fe-C martensites. It was found that microcracking is directly related to an increase in the carbon content of the martensite and that there exists a carbon content which corresponds to both the onset of microcracking and the formation of plate martensite. Retained austenite indirectly affects microcracking in that more complete transformation yields more martensite and consequently more microcracking. Grain size changes from 100 to 1200 μ, introduced by varying the austenitizing temperature from 1800° to 2400°F and varying the time at 2000°F for 15 hr, did not affect microcracking or the amount of retained austenite. Finally, the investigation emphasizes that microcracking is a manifestation of the impingement of martensite plates and is not a function of the stress state introduced by the quenching medium.

The effect of austenite grain size on microcracking in martensite of an Fe-1.22C alloy

Austenitic grain sizes of ASTM No. 9 and coarser were produced in an Fe-1.22 pct C alloy austenitized by immersion in molten lead at 1640†F (893°C), a temperature just above theAcm for this alloy, for periods between 20 s and 1 h. Microcracking sensitivity,Sv, measured as crack area/unit volume martensite, was determined as a function of grain size in brine quenched specimens. Two locations of microcracks were observed in this investigation: 1) intragranular, resulting from the impingement of one martensite plate with another, and 2) grain boundary or intergranular resulting from the impingement of martensite plates at prior austenite grain boundaries. Intragranular microcracking sensitivity, the subject of previous investigations, increased and became the dominant type of cracking with increasing grain size, and reached a constant level for grain sizes of ASTM No. 4.5 and coarser. Total microcracking sensitivity, consisting of both intragranular and grain boundary microcracks, also increased with increasing grain size, then decreased to approach the intragranular value for grain sizes coarser than ASTM No. 3.5. On the other end of the scale, grain boundary microcracking made up a much larger proportion of the total microcracking in the fine grained specimens.

The effects of stabilizing additives on the microstructure and properties of electroless copper deposits

The microstructure and physical properties of copper deposits from an EDTA complexed electroless solution maintained at 50°C and stabilized with either sodium cyanide, 2-mer-captobenzothiazole, or vanadium pentoxide were studied. Copper thicknesses up to 25 microns were plated on pretreated epoxy glass substrates. Scanning electron microscopy showed that the NaCN stabilized baths produced deposits with faceted particles 1 to 5 microns in diameter, the V2O5 baths produced deposits with needlelike particles less than 1 micron in diameter, and the deposit surfaces from the MBT stabilized baths consisted of separated clusters of columnar growths 10 to 50 microns in diameter and 20 to 80 microns high. Sulfur content of 0.2 pct was found in the deposits from the MBT stabilized baths by electron microprobe analysis. The electrical resistivities of all the deposits were measured with a four-point probe and found to be approximately 2 micro-ohm-cm at thicknesses above 25 microns. All the deposits were brittle, but those from the NaCN baths were less brittle than the deposits from the V2O5 and MBT baths. A spiral contractometer was used to measure the residual internal stresses. These stresses were 700 kg/cm2 compressive for the deposits from the NaCN baths and 350 kg/cm2 compressive for the deposits from the V2O5 baths. The deposits from the MBT stabilized baths developed residual tensile stresses that decreased to 150 kg/cm2 with increasing thickness.

Coercive force and structure in a Co-Fe-Au alloy

Strip samples of 82 Co-12 Fe-6 Au alloy were solution annealed and quenched to produce a supersaturated γ (fcc) structure. The samples were then cold worked to 50, 75, and 90 pet R.A. by rolling. Magnetic measurements showed that cold work increasedHc from the 1 oe level in well annealed samples to about 5 oe for all amounts of cold work, and that annealing of the cold worked samples further increasedH1 to maximum values of 9, 11, and 13 oe for the 50, 75, and 90 pet cold worked samples, respectively. The structural changes produced by annealing were followed quantitatively by light metallography. The progress of a discontinuous precipitation reaction, which simultaneously relieved the supersaturation of the solution treated matrix and effectively recrystallized the cold worked matrix was found to correlate well with the measured increases in coercive force. Electron transmission and diffraction verified that fine particles of gold on the order of 1000Å in diam were formed by the discontinuous precipitation.

Coercive force and microstructure in a Zr-permalloy

An alloy containing 80.0 pct Ni, 12.65 pct Fe, 6.74 pct Mo, 0.36 pct Zr, and 0.25 pct Mn by weight was cast, homogenized, and successively cold rolled into thin strips with area reductions of 0, 50, 75, and 90 pct. Annealed samples were studied by optical and electron microscopy, electron diffraction, and magnetic testing to determine the effects of cold work and annealing upon the microstructure and magnetic properties of the alloy. Cold work produced a high initial hardness together with high coercive force. Recrystallization of the cold worked structures occurred upon annealing at 600°C (873 K) and above and caused significant and parallel decreases in hardness and coercive force. The activation energy for recrystallization was found to be 80.5 kcal/g mole (337.0 kJ/g mole) for the 50, 75, and 90 pct cold worked specimens. After annealing at 600°C (873 K), a small number of spherical Ni4Mo particles were observed, but the particles produced little change in magnetic properties apparently because of their relatively coarse size and large spacing. Beginning at 700°C (973 K) ribbon-shaped particles of a Ni5Zr intermetallic compound also precipitated out of solid solution. Both the Ni4Mo and Ni5Zr precipitates were the result of a homogeneous continuous precipitation reaction within the grains. A peak in coercive force at 800°C (1073 K) is attributed to domain wall pinning associated with the fine distribution of rodlike Ni5Zr particles. Cold working 90 pct and aging at 800°C (1073 K) was found to increase coercive force by almost 60 pct from the minimum produced by complete recrystallization. Annealing, however, decreased hysteresis and improved squareness.