Authors Copy - Semantic Scholar
Authors Copy Geochjrnjca et Cosmochimica Acta Vol. 46, pp. 246 —2469 © Pergamon Press Ltd. 982. Printed in U.S.A.
0016-7037/82/122461-09503.oOfO
Analytical electron microscope study of eight ataxites P. M. NOv0TNY, J. I. GOLDsTEIN, and D. B. WILLIAMS Department of Metallurgy and Materials Engineering, Lehigh University, Bethlehem, PA 18015 (Received July 27, 1981; accepted in revised form August 30, 1982) Abstract—Optical and electron optical (SEM, TEM, AEM) techniques were employed to investigate the fine structure of eight ataxite-iron meteorites. Structural studies indicated that the ataxites can be divided into two groups; a Widmanstàtten decomposition group and a martensite decomposition group. The Widmanstätten decomposition group has a Type I plessite microstructure and the central taenite regions contain highly dislocated lath martensite. The steep M shaped Ni gradients in the taenite are consistent with the fast cooling rates, 500”C/my, observed for this group. The martensite decomposition group has a Type III ptessite microstructure and contains all the chemical group IVB ataxites. The maximum taenite Ni contents vary from 47.5 to 52.7 wt% and are consistent with slow cooling to low temperatures 350°C at cooling rates 25nC/my. Ordered FeNi and the cloudy border structure were not observed in any of the ataxites. Modest reheating to 350nC may have been responsible for the lack of these structures.
INTRODUCTION THE ATAXITES are a class of iron meteorites, which have a structure too fine to be discerned by the un aided eye (Buchwald, 1975). Meteorites of chemical group IVB as well as several anamolous irons have an ataxite structure. The general ataxite structure consists of occasional kamacite (cv-bcc) plates or rods in a Widmanstätten orientation within a matrix of fine plessite, a fine-grained intergrowth of kamacite and taenite. The submicron size of the kamacite and taenite (7-fcc) phases within the plessite has generally discouraged investigation of this class of iron mete orites since optical or electron microprobe techniques lack the necessary resolution. Recently, Jago (1981) has investigated several atax ites and plessitic octahedrites by transmission elec tron microscope (TEM) techniques. Of the three ataxites that were studied, one, Arltunga, contains a plessite matrix consisting of a micro-Widmanstàtten pattern of kamacite plates and residual taenite. Mas salski et a!. (1966) has termed this structure Type I plessite; the structure forms by the reaction 7 — a where untransformed taenite transforms to ka + macite and taenite. Two other ataxites, Warburton Range and Tawallah Valley, contain a plessite matrix consisting of fine rods of taenite in a kamacite matrix. Massalsld et al. (1966) have termed this structure as
Type III plessite. Buchwald (1966) observed the same structure in Kokomo, a typical IVB ataxite. The structure forms by the reaction 7 — a — a + 7 where untransformed taenite first transforms to mar tensite, a2, below the martensite start, M5, tempera ture. The a2 transformation structure is called —
Type II plessite. At lower temperatures the martensite may decompose to a + 7 through a diffusion-con trolled process. For Type III plessite a2 decomposes
to form rods in an a matrix. This plessite structure was first observed by Buchwald (1975) for Hoba, Kokomo, Chinga and other IVBs using optical tech-
niques and by Lin et al. (1977) using TIM tech niques. A cloudy zone structure, consisting of a ho neycomb-like mixture of kamacite and taenite, was identified only in the Warburton Range ataxite. The cloudy zone structure was initially observed by Scott (1973) using electron microscope techniques. To understand further the fine structure of the ataxites it is necessary to obtain chemical and struc tural data on the submicron scale. The development of the analytical electron microscope (AEM) has en abled investigators to analyze submicron regions in various materials. For example, Goldstein and Wil liams (1977), and L in eta!. (1977, 1979) have applied
the AIM in their studies of the IIICD iron meteorites allowing much more detailed explanations of the microstructure. The purpose of this investigation is to use the AIM as well as other electron optical tech niques to study the microstructure of a representative number of ataxites. The thermal histories of these meteorites will be discussed in terms of the microstructural evidence obtained. EXPERIMENTAL PROCEDURE Samples of eight ataxites were supplied by the Smithson ian Institution through the courtesy of Dr. R. S. Clarke. The samples are listed in Table 1 along with their bulk chemistry and cooling rates. Five ataxites (Cape of Good Hope, Hoba, Chinga, Tawallah Valley and Weaver Mountains) are from chemical group IVB while the remaining three ataxites (Arltunga, Guffey and Nordheim) are not members of a chemical group. Both Tawallah Valley and Arltunga were studied by Jago (1981). All of the ataxites have been studied optically by Buchwald (1975). Sections of these meteorites were mounted in lucite and polished using standard metallographic polishing tech niques. After polishing, the samples were etched with 2% nital for —35 sees. Light optical examination of the samples was performed using a Zeiss Axiomat microscope. Scanning Electron Microscope (SEM) examination was performed using an ETEC Autoscan. After light optical and SEM examination, samples of four ataxites, Tawallah Valley, Hoba, Nordheim and Arltunga, were sectioned and sliced into wafers —300 .em thick on
2461
Authors Copy P. M. Novotny, J. I. Goldstein and D. B. Williams
2462
TABLE 1 Ataxites Investigated in this Stody
Chemical Groop
+
÷
+
Cooling Rate fOC/My)*
Meteorite
USNW9
Azltonga Coffey Nordheim
2467 4832 3190
Anon. Anom. Anon.
9.91 10.3 11.67
0.63 0.55 0.51
0.24 0.02 0.04
>500
Cope of Good Mope Hoba Chinga Tawallah Valley Weaver Mto.
2706 3390 t585 1458 1624
IVB tVB LVB WE 158
16.32 16.4 16.58 17.6 17.72
0.84 0.76 0.55 0.69 0.82
0.12 0.07 0.05 0.1 0.10
7 —9 25 —20 20
+
Ni(wtS)
Co(wt2.)
P(wti)
—
—
-
Chemical analyses listed by Bachwaid (1975) Goldstein cod Short (1967)
a slow-speed diamond saw. Electrical discharge machining was used to trepan the wafers into discs 3 mm in diameter. The 3 mm discs were thinned by fine grinding on silicon carbide paper to a thickness of —50 tim. The discs were jet polished in a Fischione unit using a perchloric acid/ethanol solution. The electron transparent discs were then cleaned in an ion beam thinner for approximately 10 to 20 minutes to remove any chemical residue from the jet polishing. The samples were examined in a Philips EM400T equipped with an EDAX X-ray detector and 9100 computer based ana lyzer. X-ray microanalysis data were obtained using an accel crating voltage of 100 kV, electron probe sizes of 200, 100, 50 and 20 A and a counting time of 100 secs. X-ray intensity data were converted to composition data using the ratio method (Cliff and Lorimer 1975; Goldstein, 1979): CF. —
i—Ni
(‘Fe kFeNil
(1)
\19j
where CF. and C9 are the compositions in wt% for Fe and 100) and ‘Fe and I9 are the Ni respectively (Cpe ÷ C9 measured x-ray intensities corrected for continuum back ground at a given point. The sensitivity factor kFeNi was measured with the Lehigh Philips EM400T using two well
characterized Fe-Ni alloys. The kFeN, value employed in the calculation was 0.85 ± 0.00$ at the 95% confidence level. RESULTS
Light and SEM microstructure examination Typical micrographs of the structure of the eight ataxites are shown in Figs. 1 to 8. These micrographs indicate that, on the basis of the plessite structure, the ataxites can be divided into two groups, namely, a Widmanstatten decomposition group and a mar tensite decomposition group. The Widmanstätten decomposition group consisting of the meteorites Aritunga, Nordheim and Guffey, has a Type I plessite microstructure where Widmanstatten kamacite forms in a taenite matrix. Kamacite grains are separated by ribbon-like rims of taenite (Figs. 1, 2 and 3). In some of the wider taenite rims some internal structure can be observed (Fig. I b). The kamacite is multigranular
and several kamacite grains are found between the
FIG. Ia. Light photomicrograph of Arltunga. 2% nital etch. Marker = 25 tim. lb. SEM photomicrograph of Arltunga. Internal structure within the taenite is very evident. 2% nital etch. Marker = 2.5 jim.
I
(j
)
Authors Copy Eight ataxites
2463
FIG. 2a. Light photomicrograph of Nordheim. Large Widmanstätten a platelets can be seen in center of picture. 2% nital etch. Marker = 25 m. 2b. SEM photograph of Nordheim. No internal structure is seen within the taenite. 2% nital etch. Marker = 2.5 #m. taenite ribbons. No cloudy zone was observed in these meteorites. The martensite decomposition group consists of the meteorites Tawallah Valley, Hoba, Weaver Mountains, Cape of Good Hope, and Chinga. The structure of these meteorites contains some optically resolvable Widmanstätten a precipitates (Fig. 4a) but consists mainly of Type III plessite formed from martensite by the decomposition reaction y a2 a + -y (figs. 4—8). TypicaL Type III plessite consists of oriented -y rods in an a matrix. The prior a2-mar—
D
tensite plates are outlined by the -y rims that surround regions of -y rods in the a matrix (Figs. 4b and 5c). Some Type II plessite, that is retained armartensite, can be observed in a few regions at the edge of the taenite rims of Widmanstatten a precipitates (Figs. 6 and 8). Within the martensite decomposition group, Tawallah Valley and Weaver Mountains have a much higher number of Widmanstätten a precip itates than the other three members of the group. No cloudy zone was observed by optical or SEM ex amination.
FIG. 3a. Light photomicrograph of Guffey showing Widmanstätten a — a + -y type plessite. 2% nital etch. Marker = 10 m. 3b. SEM photomicrograph of Guffey. Fine structure within taenite can be seen. 2% nital etch. Marker = 5 m.
Authors Copy 2464
P. M. Novotny, J. I. Goldstein and D. B. Williams
0
FIG. 4a. Light photomicrograph of Tawallah Valley. Widmanstàtten a platelets are shown in a matrix of fine plessite. 2% nital etch. Marker = 20 m. 4b. SEM photomicrograph of Tawallah Valley. Fine
plessite from decomposed martensite is present between a platelets. Arrow indicates prior armartensite plate. 2% nital etch. Marker = 3 tim.
TEM microstructure examination Since the microstructures of the meteorites within each structural group were very similar, only two meteorites from each group were chosen for exami nation by TEM. Arltunga and Nordheim represented the Widmanstätten decomposition group, while Ta wallah Valley and Hoba represented the martensite decomposition group. TEM examination of taenite in the Widmanstàtten decomposition group was consistent with SEM evi dence and showed that the structure of the central regions of the taenite was martensite. Figures 9 and 10 illustrate the highly dislocated lath martensite in the rim-like taenite from Aritunga and Nordheim respectively. The kamacite and taenite phases of Nordheim are more highly dislocated than those of Arltunga. No orientation relationships between a, y and a2 were obtained due to the complex mixture of the three phases. In addition no cloudy zone was observed. TEM examination of plessite in the martensite decomposition group was consistent with the SEM evidence and showed that the decomposed martensite in the plessite consists of fine y rods in an a matrix surrounded by a taenite rim. Figures 11 and 12 illustrate the plessite structure in Tawallah Valley and Hoba respectively. In some cases one ptessite region containing fine y rods is adjacent to another plessite region containing large y rods. The Type III plessite structure observed is similar to that shown by the IIICD irons (Lin et at. 1979). Type III plessite structure was also observed by Jago (1981) during TEM examination of the Tawallah Valley and Warburton Range ataxites. Diffraction results from the Tawallah Valley meteorite indicate that the orien
‘
rim surrounding
tation relationship between the Widmanstätten a precipitates and the clear taenite rim and between rods and the a matrix of the plessite regions the is Nishyama-Wasserman, namely { 111 }//{0l 1 },. and KOl l)//1000 A wide taenite rims in Tawallah Valley ranged from 47.5 to 52.7 wt% Ni, no ordered FeNi phase (tetrataenite) was observed. The compositions of very fine rods, 500CC! 106 yrs, while the martensite decom position group has cooling rates 25 CC! 106 yrs (Ta ble 1). Arltunga is the only one of the three Wid manstàtten decomposition group meteorites to have a published cooling rate (Goldstein and Short, 1967). However, the similar microstructures of Guffey and Nordheim indicate that they too have cooling rates 500°C/106 yrs.
The microstructures of the meteorites in each de composition group are consistent with the cooling rates of the meteorites. The martensite in the center of the taenite and the very steep “M” composition profiles of the Widmansttten decomposition group meteorites are indicative of fast cooling. The taenite regions in these meteorites are the remnants of the original taenite that was the host y phase in the — a + ‘y reaction. As the a-kamacite nucleated and grew, excess Ni was rejected into the taenite. However, because of the very fast cooling rate, dif fusion did not occur at a rate which allowed for the redistribution of Ni within the taenite. As a result the very steep “M” shaped composition profile was re tained. The low Ni content regions of the “M” com position profile in taenite decomposed to martensite at low temperatures. Further decomposition to Type III plessite probably did not occur due to the fast cooling rates. The five martensite decomposition group mete orites have microstructures indicative of slow cool ing. The relatively flat Ni gradient observed in the taenite (Fig. 1 4a) is indicative of a slow cooled mi crostructure such as those observed in the IIICD irons (Lin et a!., 1979). Slow cooling probably allowed Type III plessite to form after a2 was formed in un transformed taenite. No ordered FeNi (tetrataenite) or cloudy zone was observed in any of the meteorites. In the Widman stätten decomposition group the maximum Ni con tent of the taenite did not exceed 40% Ni. Therefore ordered FeNi would not form although the cloudy zone which contains 30 to 40 wt% Ni would be ex pected to develop. For the martensite decomposition group, AIM composition profiles indicate that the Ni content of the taenite was high enough for the cloudy zone (30—40 wt% Ni) or even ordered FeNi
I
FIG. 9. Lath-like nature of martensite is again evident in 0.2 m. this TEM bright field from Arltunga. Marker
FIG. 10. Two thin taenite regions in highly dislocated kamacite are seen in this bright field of Nordheim. Marker = l.Oflm.
0
C)
Authors Copy Eight ataxites
2467
FiG. 11 a. Detail of TIM bright field image of plessite structure in Tawallah Valley. TIM bright field image clearly shows various size taenite rods in o matrix. Marker = 0.5 zm. 1 lb. TIM bright field image of another plessite region in Tawallah Valley. Marker = 0.2 tim.
(
(48—52 wt% Ni) to form. The lack of ordered taenite and the cloudy zone indicates that these five mete orites may have been reheated. Alternately the low temperature cooling history was too rapid for ordered taenite or the cloudy zone to form. In agreement with this study, Jago (1981) did not observe cloudy zone or ordered FeNi in Aritunga or Tawallah Valley. However a cloudy zone was ob served by Jago in Warburton Range which is a mem ber of the martensite decomposition group. In con trast to this result, all five ataxites of the martensite decomposition group considered in this study did not contain a cloudy zone structure. Jago (1981) argues that a significant reheating episode took place which removed the cloudy zone structure in Tawallah Val ley. Apparently such a reheating episode did not oc cur in Warburton Range. None of the theories for the formation of the cloudy zone have been con firmed and the temperature at which the structure develops is unknown. However if reheating does obliterate the cloudy zone, it must not affect the del icate structure of Type III plessite. Therefore the cloudy zone must decompose at relatively low tem peratures, probably
0016-7037/82/122461-09503.oOfO
Analytical electron microscope study of eight ataxites P. M. NOv0TNY, J. I. GOLDsTEIN, and D. B. WILLIAMS Department of Metallurgy and Materials Engineering, Lehigh University, Bethlehem, PA 18015 (Received July 27, 1981; accepted in revised form August 30, 1982) Abstract—Optical and electron optical (SEM, TEM, AEM) techniques were employed to investigate the fine structure of eight ataxite-iron meteorites. Structural studies indicated that the ataxites can be divided into two groups; a Widmanstàtten decomposition group and a martensite decomposition group. The Widmanstätten decomposition group has a Type I plessite microstructure and the central taenite regions contain highly dislocated lath martensite. The steep M shaped Ni gradients in the taenite are consistent with the fast cooling rates, 500”C/my, observed for this group. The martensite decomposition group has a Type III ptessite microstructure and contains all the chemical group IVB ataxites. The maximum taenite Ni contents vary from 47.5 to 52.7 wt% and are consistent with slow cooling to low temperatures 350°C at cooling rates 25nC/my. Ordered FeNi and the cloudy border structure were not observed in any of the ataxites. Modest reheating to 350nC may have been responsible for the lack of these structures.
INTRODUCTION THE ATAXITES are a class of iron meteorites, which have a structure too fine to be discerned by the un aided eye (Buchwald, 1975). Meteorites of chemical group IVB as well as several anamolous irons have an ataxite structure. The general ataxite structure consists of occasional kamacite (cv-bcc) plates or rods in a Widmanstätten orientation within a matrix of fine plessite, a fine-grained intergrowth of kamacite and taenite. The submicron size of the kamacite and taenite (7-fcc) phases within the plessite has generally discouraged investigation of this class of iron mete orites since optical or electron microprobe techniques lack the necessary resolution. Recently, Jago (1981) has investigated several atax ites and plessitic octahedrites by transmission elec tron microscope (TEM) techniques. Of the three ataxites that were studied, one, Arltunga, contains a plessite matrix consisting of a micro-Widmanstàtten pattern of kamacite plates and residual taenite. Mas salski et a!. (1966) has termed this structure Type I plessite; the structure forms by the reaction 7 — a where untransformed taenite transforms to ka + macite and taenite. Two other ataxites, Warburton Range and Tawallah Valley, contain a plessite matrix consisting of fine rods of taenite in a kamacite matrix. Massalsld et al. (1966) have termed this structure as
Type III plessite. Buchwald (1966) observed the same structure in Kokomo, a typical IVB ataxite. The structure forms by the reaction 7 — a — a + 7 where untransformed taenite first transforms to mar tensite, a2, below the martensite start, M5, tempera ture. The a2 transformation structure is called —
Type II plessite. At lower temperatures the martensite may decompose to a + 7 through a diffusion-con trolled process. For Type III plessite a2 decomposes
to form rods in an a matrix. This plessite structure was first observed by Buchwald (1975) for Hoba, Kokomo, Chinga and other IVBs using optical tech-
niques and by Lin et al. (1977) using TIM tech niques. A cloudy zone structure, consisting of a ho neycomb-like mixture of kamacite and taenite, was identified only in the Warburton Range ataxite. The cloudy zone structure was initially observed by Scott (1973) using electron microscope techniques. To understand further the fine structure of the ataxites it is necessary to obtain chemical and struc tural data on the submicron scale. The development of the analytical electron microscope (AEM) has en abled investigators to analyze submicron regions in various materials. For example, Goldstein and Wil liams (1977), and L in eta!. (1977, 1979) have applied
the AIM in their studies of the IIICD iron meteorites allowing much more detailed explanations of the microstructure. The purpose of this investigation is to use the AIM as well as other electron optical tech niques to study the microstructure of a representative number of ataxites. The thermal histories of these meteorites will be discussed in terms of the microstructural evidence obtained. EXPERIMENTAL PROCEDURE Samples of eight ataxites were supplied by the Smithson ian Institution through the courtesy of Dr. R. S. Clarke. The samples are listed in Table 1 along with their bulk chemistry and cooling rates. Five ataxites (Cape of Good Hope, Hoba, Chinga, Tawallah Valley and Weaver Mountains) are from chemical group IVB while the remaining three ataxites (Arltunga, Guffey and Nordheim) are not members of a chemical group. Both Tawallah Valley and Arltunga were studied by Jago (1981). All of the ataxites have been studied optically by Buchwald (1975). Sections of these meteorites were mounted in lucite and polished using standard metallographic polishing tech niques. After polishing, the samples were etched with 2% nital for —35 sees. Light optical examination of the samples was performed using a Zeiss Axiomat microscope. Scanning Electron Microscope (SEM) examination was performed using an ETEC Autoscan. After light optical and SEM examination, samples of four ataxites, Tawallah Valley, Hoba, Nordheim and Arltunga, were sectioned and sliced into wafers —300 .em thick on
2461
Authors Copy P. M. Novotny, J. I. Goldstein and D. B. Williams
2462
TABLE 1 Ataxites Investigated in this Stody
Chemical Groop
+
÷
+
Cooling Rate fOC/My)*
Meteorite
USNW9
Azltonga Coffey Nordheim
2467 4832 3190
Anon. Anom. Anon.
9.91 10.3 11.67
0.63 0.55 0.51
0.24 0.02 0.04
>500
Cope of Good Mope Hoba Chinga Tawallah Valley Weaver Mto.
2706 3390 t585 1458 1624
IVB tVB LVB WE 158
16.32 16.4 16.58 17.6 17.72
0.84 0.76 0.55 0.69 0.82
0.12 0.07 0.05 0.1 0.10
7 —9 25 —20 20
+
Ni(wtS)
Co(wt2.)
P(wti)
—
—
-
Chemical analyses listed by Bachwaid (1975) Goldstein cod Short (1967)
a slow-speed diamond saw. Electrical discharge machining was used to trepan the wafers into discs 3 mm in diameter. The 3 mm discs were thinned by fine grinding on silicon carbide paper to a thickness of —50 tim. The discs were jet polished in a Fischione unit using a perchloric acid/ethanol solution. The electron transparent discs were then cleaned in an ion beam thinner for approximately 10 to 20 minutes to remove any chemical residue from the jet polishing. The samples were examined in a Philips EM400T equipped with an EDAX X-ray detector and 9100 computer based ana lyzer. X-ray microanalysis data were obtained using an accel crating voltage of 100 kV, electron probe sizes of 200, 100, 50 and 20 A and a counting time of 100 secs. X-ray intensity data were converted to composition data using the ratio method (Cliff and Lorimer 1975; Goldstein, 1979): CF. —
i—Ni
(‘Fe kFeNil
(1)
\19j
where CF. and C9 are the compositions in wt% for Fe and 100) and ‘Fe and I9 are the Ni respectively (Cpe ÷ C9 measured x-ray intensities corrected for continuum back ground at a given point. The sensitivity factor kFeNi was measured with the Lehigh Philips EM400T using two well
characterized Fe-Ni alloys. The kFeN, value employed in the calculation was 0.85 ± 0.00$ at the 95% confidence level. RESULTS
Light and SEM microstructure examination Typical micrographs of the structure of the eight ataxites are shown in Figs. 1 to 8. These micrographs indicate that, on the basis of the plessite structure, the ataxites can be divided into two groups, namely, a Widmanstatten decomposition group and a mar tensite decomposition group. The Widmanstätten decomposition group consisting of the meteorites Aritunga, Nordheim and Guffey, has a Type I plessite microstructure where Widmanstatten kamacite forms in a taenite matrix. Kamacite grains are separated by ribbon-like rims of taenite (Figs. 1, 2 and 3). In some of the wider taenite rims some internal structure can be observed (Fig. I b). The kamacite is multigranular
and several kamacite grains are found between the
FIG. Ia. Light photomicrograph of Arltunga. 2% nital etch. Marker = 25 tim. lb. SEM photomicrograph of Arltunga. Internal structure within the taenite is very evident. 2% nital etch. Marker = 2.5 jim.
I
(j
)
Authors Copy Eight ataxites
2463
FIG. 2a. Light photomicrograph of Nordheim. Large Widmanstätten a platelets can be seen in center of picture. 2% nital etch. Marker = 25 m. 2b. SEM photograph of Nordheim. No internal structure is seen within the taenite. 2% nital etch. Marker = 2.5 #m. taenite ribbons. No cloudy zone was observed in these meteorites. The martensite decomposition group consists of the meteorites Tawallah Valley, Hoba, Weaver Mountains, Cape of Good Hope, and Chinga. The structure of these meteorites contains some optically resolvable Widmanstätten a precipitates (Fig. 4a) but consists mainly of Type III plessite formed from martensite by the decomposition reaction y a2 a + -y (figs. 4—8). TypicaL Type III plessite consists of oriented -y rods in an a matrix. The prior a2-mar—
D
tensite plates are outlined by the -y rims that surround regions of -y rods in the a matrix (Figs. 4b and 5c). Some Type II plessite, that is retained armartensite, can be observed in a few regions at the edge of the taenite rims of Widmanstatten a precipitates (Figs. 6 and 8). Within the martensite decomposition group, Tawallah Valley and Weaver Mountains have a much higher number of Widmanstätten a precip itates than the other three members of the group. No cloudy zone was observed by optical or SEM ex amination.
FIG. 3a. Light photomicrograph of Guffey showing Widmanstätten a — a + -y type plessite. 2% nital etch. Marker = 10 m. 3b. SEM photomicrograph of Guffey. Fine structure within taenite can be seen. 2% nital etch. Marker = 5 m.
Authors Copy 2464
P. M. Novotny, J. I. Goldstein and D. B. Williams
0
FIG. 4a. Light photomicrograph of Tawallah Valley. Widmanstàtten a platelets are shown in a matrix of fine plessite. 2% nital etch. Marker = 20 m. 4b. SEM photomicrograph of Tawallah Valley. Fine
plessite from decomposed martensite is present between a platelets. Arrow indicates prior armartensite plate. 2% nital etch. Marker = 3 tim.
TEM microstructure examination Since the microstructures of the meteorites within each structural group were very similar, only two meteorites from each group were chosen for exami nation by TEM. Arltunga and Nordheim represented the Widmanstätten decomposition group, while Ta wallah Valley and Hoba represented the martensite decomposition group. TEM examination of taenite in the Widmanstàtten decomposition group was consistent with SEM evi dence and showed that the structure of the central regions of the taenite was martensite. Figures 9 and 10 illustrate the highly dislocated lath martensite in the rim-like taenite from Aritunga and Nordheim respectively. The kamacite and taenite phases of Nordheim are more highly dislocated than those of Arltunga. No orientation relationships between a, y and a2 were obtained due to the complex mixture of the three phases. In addition no cloudy zone was observed. TEM examination of plessite in the martensite decomposition group was consistent with the SEM evidence and showed that the decomposed martensite in the plessite consists of fine y rods in an a matrix surrounded by a taenite rim. Figures 11 and 12 illustrate the plessite structure in Tawallah Valley and Hoba respectively. In some cases one ptessite region containing fine y rods is adjacent to another plessite region containing large y rods. The Type III plessite structure observed is similar to that shown by the IIICD irons (Lin et at. 1979). Type III plessite structure was also observed by Jago (1981) during TEM examination of the Tawallah Valley and Warburton Range ataxites. Diffraction results from the Tawallah Valley meteorite indicate that the orien
‘
rim surrounding
tation relationship between the Widmanstätten a precipitates and the clear taenite rim and between rods and the a matrix of the plessite regions the is Nishyama-Wasserman, namely { 111 }//{0l 1 },. and KOl l)//1000 A wide taenite rims in Tawallah Valley ranged from 47.5 to 52.7 wt% Ni, no ordered FeNi phase (tetrataenite) was observed. The compositions of very fine rods, 500CC! 106 yrs, while the martensite decom position group has cooling rates 25 CC! 106 yrs (Ta ble 1). Arltunga is the only one of the three Wid manstàtten decomposition group meteorites to have a published cooling rate (Goldstein and Short, 1967). However, the similar microstructures of Guffey and Nordheim indicate that they too have cooling rates 500°C/106 yrs.
The microstructures of the meteorites in each de composition group are consistent with the cooling rates of the meteorites. The martensite in the center of the taenite and the very steep “M” composition profiles of the Widmansttten decomposition group meteorites are indicative of fast cooling. The taenite regions in these meteorites are the remnants of the original taenite that was the host y phase in the — a + ‘y reaction. As the a-kamacite nucleated and grew, excess Ni was rejected into the taenite. However, because of the very fast cooling rate, dif fusion did not occur at a rate which allowed for the redistribution of Ni within the taenite. As a result the very steep “M” shaped composition profile was re tained. The low Ni content regions of the “M” com position profile in taenite decomposed to martensite at low temperatures. Further decomposition to Type III plessite probably did not occur due to the fast cooling rates. The five martensite decomposition group mete orites have microstructures indicative of slow cool ing. The relatively flat Ni gradient observed in the taenite (Fig. 1 4a) is indicative of a slow cooled mi crostructure such as those observed in the IIICD irons (Lin et a!., 1979). Slow cooling probably allowed Type III plessite to form after a2 was formed in un transformed taenite. No ordered FeNi (tetrataenite) or cloudy zone was observed in any of the meteorites. In the Widman stätten decomposition group the maximum Ni con tent of the taenite did not exceed 40% Ni. Therefore ordered FeNi would not form although the cloudy zone which contains 30 to 40 wt% Ni would be ex pected to develop. For the martensite decomposition group, AIM composition profiles indicate that the Ni content of the taenite was high enough for the cloudy zone (30—40 wt% Ni) or even ordered FeNi
I
FIG. 9. Lath-like nature of martensite is again evident in 0.2 m. this TEM bright field from Arltunga. Marker
FIG. 10. Two thin taenite regions in highly dislocated kamacite are seen in this bright field of Nordheim. Marker = l.Oflm.
0
C)
Authors Copy Eight ataxites
2467
FiG. 11 a. Detail of TIM bright field image of plessite structure in Tawallah Valley. TIM bright field image clearly shows various size taenite rods in o matrix. Marker = 0.5 zm. 1 lb. TIM bright field image of another plessite region in Tawallah Valley. Marker = 0.2 tim.
(
(48—52 wt% Ni) to form. The lack of ordered taenite and the cloudy zone indicates that these five mete orites may have been reheated. Alternately the low temperature cooling history was too rapid for ordered taenite or the cloudy zone to form. In agreement with this study, Jago (1981) did not observe cloudy zone or ordered FeNi in Aritunga or Tawallah Valley. However a cloudy zone was ob served by Jago in Warburton Range which is a mem ber of the martensite decomposition group. In con trast to this result, all five ataxites of the martensite decomposition group considered in this study did not contain a cloudy zone structure. Jago (1981) argues that a significant reheating episode took place which removed the cloudy zone structure in Tawallah Val ley. Apparently such a reheating episode did not oc cur in Warburton Range. None of the theories for the formation of the cloudy zone have been con firmed and the temperature at which the structure develops is unknown. However if reheating does obliterate the cloudy zone, it must not affect the del icate structure of Type III plessite. Therefore the cloudy zone must decompose at relatively low tem peratures, probably
