• PDF / 1,635,706 Bytes
• 8 Pages / 594 x 774 pts Page_size
• 53 Downloads / 310 Views

DOWNLOAD

REPORT

---

Differential Scanning Calorimetry (DSC) was chosen to follow the tempering response during MA. All DSC analyses were performed on a DuPont 9900 Thermal Analysis system. A constant heating rate of 25 K/minute was used. The tempering of iron-carbon hypereutectoid martensites had been studied previously by Won et al. ~]81 The temperature range at which the various stages leading to cementite precipitation during tempering is 375 K to ~625 K. The particular stages used in the martensitic decomposition studied here are shown as the two peaks in Figure 4. The first peak, at about 418 K, corresponds to the decomposition of the primary martensite into e-carbide and a lower carbon martensite. ~191The second peak at about

sample MARTENSITE size 25 30 mg

575 K corresponds to the decomposition of the alloy ultimately into cementite (Fe3C) and ferrite. [2~ Utilizing Won et a I . ' s values for activation energy of these processes, calibration charts were produced relating tempering temperature to remaining convertible cementite in the sample. As can be seen in Figure 5, the higher the temperature, the more cementite is converted during the tempering process, and the less will be available for transformation during DSC analysis (hence the negative slope on the graph). With the tempering vs pct cementite converted relationship developed, we then obtained an estimate of the temperatures seen during MA by milling fresh martensitic powder in the Spex mill and sampling at specified time intervals. Figure 6 shows results for 3 MA runs at different ambient conditions. One run utilized natural conduction

1.0

O 03 c3

.75

O

0.5

41"

I-.rE

.25

o

525

535

542 545 550 555 55B

565

575

Temperature (K) Fig. 5--Amount of martensite converted to cementite v s tempering temperature. Upper and lower curves are the 95 pct confidence intervals on the linear regression line.

E)SC , , Natural convection -,I-----I.--~ N 2 (liquid) v v ~ Forced convection

575 6~

5-

555 4-

Natural convecbon

,-r

o~

2-

I

1

~

555

3-

8.

418

E

o

Forced coonvectlon

545

N20i.~ o

0

535 -1

-2 350

1

2

3

4

5

6

Temperature (K)

MAlime (hrs.) Fig. 4 - - D S C analysis of martensite decomposition. The peaks represent decomposition of the martensite to epsilon carbide (418 K) and cementite (575 K). 2870--VOLUME 19A, DECEMBER 1988

Fig. 6--Powder temperature esumated from tempering response ing time.

vs

mill-

METALLURGICAL TRANSACTIONS A

temperature rises during impact experiments have been reported by Miller e t al. El61While impact heating has been shown to be caused in part by localized shear deformation in metals, t~6'jTlthis explanation may be inadequate for brittle materials where significant plastic deformation may not occur. Using time-resolved infrared radiometry, Miller and associates showed that local temperatures on the order of 675 to 775 K could be attained in impacted nonmetallic, crystalline materials such as NaC1 and ammonium perchlorate. While it was suggested that the observed temperatures were localized in small volume