WO2009049201A1

WO2009049201A1 - Copper tin nickel phosphorus alloys with improved strength and formability

Info

Publication number: WO2009049201A1
Application number: PCT/US2008/079573
Authority: WO
Inventors: Carole Lynne Trybus; Peter William Robinson
Original assignee: Gbc Metals, Llc
Priority date: 2007-10-10
Filing date: 2008-10-10
Publication date: 2009-04-16
Also published as: JP5752937B2; TW200934882A; MX2010003995A; JP2011500963A; EP2215278A4; CN101874122A; EP2215278A1; US20090098011A1; CA2702358A1

Abstract

A new copper-based alloy is described along with a processing method to make a strip that can be used for various automotive interconnects. The alloy process combination yields a material with high strength and electrical conductivity with excellent formabiiity. The combination of properties result from a Cu-Sn-Ni-P alloy with optional Mg additions and thermal-mechanical processing to make an alloy with a conductivity of 40%iACS, yield strength of 80 KSI, bend formabiiity of 11/11 minimum, and stress relaxation of 65% at 15O0C after 1000 hours. Processing can be modified to increase formabiiity at the expense of yield strength. Improvements to conductivity come from changes in chemistry as well as processing. The new chemistry-process optimization results in a low cost alloy of Cu-Sn-Ni-P-Mg.

Description

COPPER TIN NICKEL PHOSPHORUS ALLOYS WITH IMPROVED STRENGTH AND FORMABILITY

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to U.S. Provisional Patent Application Serial No. 60/979,064, filed October 10, 2007, the entire disclosure of which is incorporated herein.

BACKGROUND

[0002] This invention relates to copper alloys, and in particular to copper-tin-nickel-phosphorus alloys with improved strength and formabiiity.

[0003] There is a continued need for high strength copper alloys of good formabiiity and reasonable cost for use in electrical connectors, and in particular for use in automotive electrical connectors. Current connector alloys in the low cost Cu-Sn-Ni-P family lack the combination of properties of practical strength (77 KSl), intermediate conductivity (37 %IACS), excellent formabiiity, and decent stress relaxation (65% at 150⁰C). Formabiiity in the document is measured by forming a strip by roller bending it 90^s about a die of known radii. The ratio of the smallest die radii that the strip can be formed without cracking is divided over the strip thickness. Bends were measured both parallel (bad way, BW) and perpendicular (good way, GW) to the direction of rolling. Table 1 shows currently available Cu-Sn-Ni-P alloys: Table 1 : Available connector alloys in the Cu-Sn-Ni-P family

[0004] C19025 comes close to achieving the desired properties but lacks the strength with acceptable formability; C40820 has the strength and superior formability but does not have the electrical conductivity. SUMMARY

[0005] Embodiments of the present invention provide a copper-tin- nickel-phosphorus alloy with an improved combination or properties, and in particular improved combination of yield strength and formability. In one preferred embodiment the alloy comprises between about 1 % and about 2% Sn; between about 0.3% and about 1 %Ni; between about 0.05% and about 0.15% P, and at least one of between about 0.01% and about 0.20% Mg and about 0.02% and about 0.4% Fe, the balance being copper. The addition of iron can be used as a low cost substitute for of Mg if good stress relaxation is not required for the application. More preferably the alloy comprises between about 1.1% and about 1.8% Sn, between about 0.4% and about 0.9% Ni, between about 0.05% and about 0.14% P, and between about 0.05 and about 0.15 Mg. Fe may be substituted for some of the Mg. Most preferably the alloy comprises, between about 1.2% and about 1.5% Sn; between about 0.5% and about 0.7%Ni; between about 0.09% and about 0.13% P, and between about 0.02% and about 0.06% Mg, the balance being copper. The alloy is preferably processed to have a yield strength of at ieast about 77 KSI, electrical conductivity of at least about 37 %IACS, and formability (90° GW/BW) of 1.0/1.0. The alloy preferably also has a stress relaxation of 65% at 150⁰C.

[0006] The Sn gives the alloy solid solution strengthening. Ni and Mg are added to form precipitates of phosphorus with the added benefit of Mg increasing strength without lowering the electrical conductivity. The metal (Ni+Mg) to P ratio (the M/P ratio) is preferably controlled to a range of 4 to 8.5. If the ratio falls below 4 strengthening is not obtained and if is greater that 8.5 the material does not achieve 40% IACS.

[0007] In accordance with the preferred embodiment of this invention, the alloy is processed by melting and casting, hot rolling from about 850 -C to about 1000^QC cold rolling up to about 75% annealing between about 450 ^QC - about 600^QC, cold roiling up to about a 60% reduction followed by annealing at 425 ²C to about 600⁵C, cold rolling to about 50% prior to the final anneal between about 400 ^SC and 550^eC. A final cold roll reduction is given to achieve the desired thickness and mechanical strength prior to a thermal stress relief treatment. In another preferred embodiment the processing includes a double final anneal treatment and the eiimination of an upstream anneal which improves formability and strength respectively. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Fig. 1 is a photomicrograph of the alloy in Exarnpie 1 ;

[0009] Fig. 2 is a graph showing the relationship between YS and M/P ratio, and illustrating the preferred M/P ratio for a Cu-Sn-Ni-P-Mg alloy;

[0010] Fig. 3 is a graphs showing the relationship between %IACS and M/P ratio, and illustrating the preferred M/P ratio of 4-8.5 ratio for a Cu- Sn-Ni-P-Mg alloy;

[0011] Fig. 4A is a flow chart of a preferred embodiment of a method of processing alloys in accordance with the principles of the present invention;

[0012] Fig. 4B is a flow chart of an alternate preferred embodiment of processing alloys in accordance with the principles of this present invention;

[0013] Fig. 4C is a flow chart of an alternate preferred embodiment of processing alloys in accordance with the principles of this present invention; and

[0014] Fig. 5 is a photomicrograph of an alloy 4 after double anneal, showing a grain size of between 6 - 7 μm, with some areas appearing to have not fully recrystailized grains; and

[0015] Fig. 6 is a photomicrograph of an alloy 4 from the process 3 after strip anneal, showing a grain size of 4 - 5 μm. DETAILED DESCRIPTION

[0016] Embodiments of the present invention provide a copper-tin- nickel-phosphorus alloy with an improved combination or properties, and in particular improved combination of yield strength and formabiiity. In one preferred embodiment the alloy comprises between about 1% and about 2% Sn; between about 0.3% and about 1%Ni; between about 0.05% and about 0.15% P, and at least one of between about 0.01% and about 0.20% Mg and about 0.02% and about 0.4% Fe, the balance being copper. The addition of iron can be used as a low cost substitute for of Mg if good stress reiaxation is not required for the application.

[0017] More preferably the alloy comprises, between about 1.2% and about 1.5% Sn; between about 0.5% and about 0.7%Ni; between about 0.09% and about 0.13% P, and between about 0.02% and about 0.06% Mg, the balance being copper. The alloy is preferably processed to have a yield strength of at least about 77 KSI, electrical conductivity of at least about 37 %IACS, and formabiiity (90° GW/BW) of 1.0/1.0. The alloy preferably also has a stress relaxation of 65% at 150⁰C.

[0018] The Sn gives the alloy soiid solution strengthening. Ni and Mg are added to form precipitates of phosphorus with the added benefit of Mg increasing strength without lowering the electrical conductivity. The M/P ration is preferably controlied to a range of 4 to 8.5. If the ratio falls below 4 strengthening is not obtained and if is greater that 8.5 the material does not achieve 40% IACS.

[0019] In accordance with the preferred embodiment of this invention, the alloy is processed by melting and casting, hot rolling from 850- 1000^sC cold rolling up to about 75% annealing between 450- 600⁵C, cold roliing about 60% followed by annealing at 425-600^QC, cold rolling about 50% prior to the final anneal between 400-550^sC. A final coid roll reduction is given to achieve the desired thickness and mechanical strength prior to a thermal stress relief treatment. In another preferred embodiment the processing includes a double final anneal treatment and the elimination of an upstream anneal which improves formability and strength respectively.

Example 1

[0020] A series of 10 pound laboratory ingots with the compositions listed in Table 2 were melted in silica crucibles and cast into steel molds which were after gating 4"x4"x1.75". After soaking for 2 hours at 900-C they were hot rolled in three passes to 1.1" (1.671.35/1.1"), reheated at 900⁵C for 10 minutes, and further reduced by hot rolling in three passes to 0.50" (0.970.770.5"), followed by a water quench. After trimming and milling to remove the surface oxide, the alloys were cold rolled to 0.120" and annealed at 570^QC for 2 hours. The alioys were cleaned and coid rolled to 0.048" and annealed at 525-C for 2 hours. The alloys were cold rolied to 0.030" and annealed at 500^sC for 2 hours. The final coid roll was 60% to 0.012" and a stress relief heat treatment was performed at 250^sC for 2 hours.

Table 2: Ailo s and ro erties from Exam le 1

*for this Table and throughout this document YS means Yield Strength and is given in units of KSI

From the data in Example 2, it was determined that the Ni level is preferably at least 0.5 and the best overall alloys had a Ni/P ratio of 7-9. All the bends were poor due to the presence of contamination of sulfur forming long stringers as shown in Figure 1.

Example 2

[0021] A series of 10 pound laboratory ingots with the compositions listed in Table 3 were melted in silica crucibles and cast into steel molds which were after gating 4"x4"x1.75". After soaking for 2 hours at 900^gC they were hot rolled in three passes to 1.1" (1.671.35/1.1"), reheated at 900^sC for 10 minutes, and further reduced by hot rolling in three passes to 0.50" {0.970.770.5"), followed by a water quench. After trimming and milling to remove the surface oxide, the alloys were cold rolled to 0.120" and annealed at 57O³C for 2 hours. The alloys were cleaned and cold rolled to 0.048" and annealed at 525^SC for 2 hours. The alloys were cold rolled to 0.024" and annealed at 450^sC for 8 hours. The final cold roll was 50% to 0.012" and a stress relief heat treatment was performed at 250-C for 2 hours.

Table 3- Alloys from Example 2.

In general the strengths are low with the exception of alloys K293 and K294. Both these alloys contained more Sn than any of the others by about 0.5% correlating higher Sn levels to higher strength. The strengths of K286, K287 and K288 indicate the benefit of Mg as opposed to alloys of very close composition but without Mg, K282 and K284. It is notable that there is no drop in conductivity (the %IACS) accompanying the increase in yield strength. There was an increase in strength with the addition of iron to K291 and Mg in K289 both without Ni. The conductivity for the iron containing alloy is lower than the Mg containing alloy by about 4 %iACS. Both of these alloys are almost perfectly balanced; Mg/P ratio is 1.81 for K289 close to the ideal of 1.2 and the Fe/P ratio for K291 is 4.00 which is also close to the ideal of 3.6. Iron is a more effective strengthener but leads to lower conductivity. Example 3

[0022] A series of 10 pound laboratory ingots with the compositions listed in Table 4 were melted in silica crucibles and cast into steel molds which were after gating 4"x4"x1.75". After soaking for 2 hours at 900^ΩC they were hot rolled in three passes to 1.1" (1.671.35/1.1"), reheated at 900^eC for 10 minutes, and further reduced by hot rolling in three passes to 0.50" (0.970.770.5"), followed by a water quench. After trimming and milling to remove the surface oxide, the alloys were cold rolled to 0.120" and annealed at 570-C for 2 hours. The alloys were cleaned and cold rolled to 0.048" and annealed at 525^SC for 2 hours. The alloys were cold rolled to 0.024" and annealed at 450⁶C for 4 hours only for the single anneal condition and for 450⁹C for 4 hours plus 375°C for another 4 hours constituting the double anneal condition. The final cold roll was 50% to 0.012" and a stress relief heat treatment was performed at 250^QC for 2 hours for both conditions.

Table 4- Alloys from example 3 including both annealing conditions.

Higher Sn levels helped the strength levels considerably but at lower conductivities. Compare alloys K320 and K319; 7KSI difference in YS and 3%IACS in conductivity. The trend holds for those alloys with iron (K312 and K313) and those with magnesium (K 314 and K315) although the impact on strength is less than those without any other addition. There was no overall advantage of zinc K311 in contrast to K310; strength is increased but with lower conductivity. The double anneal showed an increase in formability (i.e., a decrease in the 90° bend radii that can be achieved). Slight increases in the conductivities are also noted. Example 4

[0023] A series of 10 pound laboratory ingots with the compositions listed in Table 4 were melted in silica crucibles and cast into steel molds which were after gating 4^Hx4^ax1.75". After soaking for 2 hours at 900⁹C they were hot rolled in three passes to 1.1" (1.671.35/1.1"), reheated at 900-C for 10 minutes, and further reduced by hot rolling in three passes to 0.50" (0.970.770.5"), followed by a water quench. After trimming and milling to remove the surface oxide, the alloys were cold rolled to 0.120" and annealed at 570-C for 2 hours. The alloys were cleaned and cold rolled to 0.048" and annealed at 525^QC for 2 hours. The alloys were cold rolled to 0.024" and annealed at 450⁹C for 4 hours only plus 375⁰C for another 4 hours. The final cold roll was 50% to 0.012" and a stress relief heat treatment was performed at 250^aC for 2 hours.

Table 5. Data from example 4

Thirteen of the twenty-two alloys in this group had yield strengths of 75 KSI or above. Six contained iron (K338, K339, K345, K346, K355 and K361 ) none of which made electrical conductivity of 40%lACS, although K338 is the closest at 38%IACS. Four contained Mg (K350, K351 , K352 and K356) and 3 of these 4 exceeded 40%IACS. Note that K350 which did not achieve 40%iACS had a metal to phosphorus ratio of 9, greater that the recommended 8.5. Three of the alloys with yield strengths of 75 ksi or greater contained neither iron nor Mg (K343, K345, and K348), but none of these alloys had conductivities of 40%!ACS.

Example 5

[0024] All the data for Mg containing alloys and Mg-free ailoys are combined in Tables 6 and 7. These data are from example 2, (Table 3 ailoys which were double annealed and included in Tables 6 and 7), Example 3 (Table 4), and Example 4 (Table 5), and include data from Example 3. The process used for all the alloys is identical to the process used in the final double anneal of 4 (or 8 hours; see note) at 450^sC+ 4 hours at 375^ςC.

Ξ 5? Φ σ> ω

O

T3 Φ a. a, a

O

3 m X

(D Φ

Ip

SL

Φ

CD Q.

Table 7. Grouped data from all Examples without Mg, all double annealed

* Alloys K 319 and K320 are similar to C19020 and C19025, but with lower P.

Alloys in highlighted in light gray had a slightly different final double anneal 450^sC for 8 hours -f 4 hours at 375⁹C

[0025] Overall the YS in Table 6 with Mg are higher than those in Table 7 without Mg. Only a few Mg-free alioys reach a minimum YS of 75 KSI: K293, K294, K310, K326, K343, K345, and K348, with corresponding electrical conductivities of: 42.2, 38.5, 38.5, 38.5, 38.4, 38.6 and 32.8 %IACS respectively. Note with the exception of K293 none of the alloys achieve 40%IACS. Alloys K293, K294 and K326 all have properties of YS and conductivities close to C19025 but have better bends. In contrast the Mg alloys in Table 6 all have YS of at least 75 KS! with the exception of K289 and K290 (which had no Ni and an M/P ratio below 4). The electrical conductivities of all the alloys are at or above 40%IACS except for K318 {38.7 %!ACS) with an M/P of 7.66 and K350 (38.1 %IACS) with an M/P ratio of 9.02. As the metal to phosphorus ratio increases the conductivity decreases and the combination of desirable properties becomes more difficult to reach. The addition of Mg enables the combination of yield strength over 75 KSl and conductivity of at least 40 %IACS achievable when employing appropriate processing and maintaining an M/P ratio between 4 and 8.5. Figures 2 and 3 illustrate the relationships between the ratios and YS and %IACS respectively. The vertical lines in Figs, 2 show the preferred M/P ratio of 4-8.5. Example 7

[0026] A series of 10 pound laboratory ingots with the compositions listed in Table 8 were melted in silica crucibles and cast into steel molds which were after gating 4"x4"x1.75". After soaking for 2 hours at 900⁹C they were hot rolled in three passes to 1.1" (1.671.35/1.1"), reheated at 900^QC for 10 minutes, and further reduced by hot rolling in three passes to 0.50" (0.970.770.5"), followed by a water quench. After trimming and milling to remove the surface oxide, the alloys were cold roiled to 0.080" and annealed at 550⁹C for 2 hours. The alloys were cleaned and cold rolled to 0.036" and annealed at 450⁹C for 4 hours only plus 375°C for another 4 hours. The final cold roll was 60% to 0.012" and a stress relief heat treatment was performed at 250⁹C for 2 hours. Table 8, Data from Example 7

increased cold work improved strength for all alloys. However, the Mg containing alloy with an M/P ratio below 9 (K352) was the only one to improve YS whiie maintaining or improving conductivity.

Example 8

[0027] A series of 10 pound laboratory ingots with the compositions listed in Table 3 were melted in silica crucibles and cast into steel molds which were after gating 4"x4"x1.75". After soaking for 2 hours at 900^sC they were hot rolied in three passes to 1.1" {1.671.35/1.1"), reheated at 900^sC for 10 minutes, and further reduced by hot rolling in three passes to 0.50" (0.970.770.5"), followed by a water quench. After trimming and milling to remove the surface oxide, the alloys were cold rolled to 0.120" and annealed at 570^sC for 2 hours. The alloys were cleaned and cold rolled to 0.048" and annealed at 525^aC for 2 hours. The alloys were cold rolied to 0.024" and annealed at 450^ςC for 4 hours minimum. The final cold roll was 50% to 0.012" and a stress relief heat treatment was performed at 250^sC for 2 hours. The samples were subjected to stress relaxation testing at 15O⁰C for 1000hrs. The results are given in Table 9 below: Table 9. Data from Example 8

Alloys K291 and K312 with iron did not maintain 60% of the initial stress. The results are similar between the two despite the presence of Ni in K312. K314 with Ni and Mg combination maintained more than 65% of the initial stress.

Example 9

[0028] A set of Mg and Mg-free ailoys were processed using the indicated schedules. Tables 10 and 11 summarize the results. Both sets of alloys achieved yield strengths over 80 KSl. The Mg-containing alloys, ail exceeded the target conductvity of 38% IACS, whereas the Mg-free alloys, with the exception of K412, did not. In addition, the formability of the Mg-containing alloys was generally better.

Table 10 Summary of Results for Mg-Containing alloys

YS is in KSI

Process Details: HRP + CR to 0.060 gage + 500°C/8hrs + CR 50% to 0.030 gage +

450°C/4hrs + 375°C/4hrs + CR 60% to 0.012 gage + 250°C/2hrs

Table 11 Summary of Results for the Mg-f ree alloys

YS is in KSI

Process Details: HRP + CR to 0.060 gage + 475°C/16hrs + CR 50% to 0.030 gage + 450°C/4hrs + 375°C/4hrs + CR 60% to 0.012 gage + 250°C/2hrs Example 10

[0029] Plant processing was conducted on six alloys whose nominal compositions are set forth in Table 12, The processes are detailed in Table 13, where Process 1 is a laboratory process for comparison purposes, and Processes 2, 3, and 4 are plant processes.

Table 12 Chemistry of Plant- Processed Bars

[0030] The chemistry given in the Table 12 is the analyzed chemistry for the cast bars. Alloy 6 lies within the CDA range for C19025 and is present as a comparative example. Ail alloys were processed the same way: They were all hot rolled from 900⁰C, coil milled and then cold rolled to 0.125 or 0.100 gauge.

Table 13 Definition of Processes for Example 10

The resulting properties at final gage are shown in Table 14. Alloy 6 processed using Processes 3 and 4 possessed the expected properties for this alloy, having higher yield strength and poorer bends for Process 4 versus Process 3. Alloy 5 had a Sower yield strength (YS) and poorer bad way bends when processed according to Process 2 in contrast to the Process 3 metal. Alloy 3 had comparable yield strength and conductivity for both the Process 2 and Process 3 processing but meta! processed according to Process 3 had better bad way bends. Table 14 Results from the Plant Trial as Compared to the Laborator Processed Metal

These results are from process 4. Processes 3 and 4 generally gave the best results. The results for Processes 1 and 2 on alloys 1 and 4 show slightly different results if the process is conducted in the plant (Process 2) rather than in the lab (Process 1) may have caused grain growth. Table 15 shows that the double anneal process (Process 2) gives good bends when simulated in

the lab.

Table 15 Additional results for Alloy 4

TS (KSi) YS (KSl) Elong. (%) %IACS GW90 BW90

86.5 83.7 10.27 40.4 0.09 0.52

Plant processed alloys were subjected to stress relaxation testing at 15O⁰C.

Results for the transverse direction only are shown below in Table 16. All alloys except for alloy 2 had at least 65% stress remaining after 100Oh at 150⁰C.

Table 16 Stress Relaxation Data from the Plant Trial

Claims

What is claimed Is:

1. A copper base alloy comprising between about 1 % and about 2% Sn; between about 0.3% and about 1%Ni; between about 0.05% and about 0.15% P, and at least one of up to about 0.20% Mg and between about 0.1 and about 0.4% Fe, the balance being copper.

2. The copper base alloy according to ciaim 1 containing Mg but no Fe.

3. The copper base alloy according to claim 1 containing Fe but no Mg.

4. The copper base alloy according to claim 1 containing both Mg and Fe.

5. The copper base alloy according to claim 1 containing up to about 0.06% Mg.

6. The copper base alioy according to claim 1 processed to have a yield strength of at least about 77 ksi, while maintaining bend formability (90° GW/BW) of 1.0/1.0.

7. The copper base alloy according to claim 6 wherein the ailoy is processes to have a conductivity of at least about 37% IACS.

8. The copper base alloy according to claim 6 wherein the alloy is processed to have a conductivity of at least about 40% IACS

9. The copper base alloy according to claim 1 wherein the Ni:P ratio is less than about 9.

10. The copper alloy according to claim 1 wherein the (Ni+Mg):P ratio is between about 4 and about 8.5.

11. The copper base alloy according to claim 1 wherein the Sn is between about 1.2% and about 1.5%, the Ni is between about 0.5% and 0.7%, and the P is between about 0.09% and about 0.13%.

12. The copper base alloy according to claim 11 wherein the Ni:P ratio is less than about 9.

13. The copper alloy according to claim 11 wherein the (Ni+Mg):P ratio is between about 4 and about 8.5.

14. The copper base alloy according to claim 11 processed to have a yield strength of at least about 77 ksi, while maintaining bend formabiiity (90° GW/BW) of 1.0/1.0.

15. The copper base alloy according to claim 14 wherein the alloy is process to have a conductivity of at least about 37% IACS.

16. The copper base alloy according to claim 14 wherein the alloy is process to have a conductivity of at least about 40% IACS.

17. A copper base alloy comprising between about 1.2% and about 1.5% Sn; between about 0.5% and about 0.7%Ni; between about 0.09% and about 0.13% P, and at least one of up to about 0.20% Mg and between about 0.1 and about 0.4% Fe, the balance being copper, the alloy processed to have a yield strength of at least about 77 ksi, and an electrical conductivity of at least about 37% IACS.

18. The copper base alloy according to claim 17 wherein the alloy is processed to have a conductivity of at least about 40% IACS.

19. The copper base alloy according to claim 17 processed to have a bend formabiiity (90° GW/BW) of 1.0/1.0.

20. The copper alloy according to claim 11 wherein the (Ni+Mg):P ratio is between about 4 and about 8.5.

21. A method of processing a copper base alloy comprising between about 1 % and about 2% Sn; between about 0.03% and about 1%Ni; between about 0.05% and about 0.15% P, and at least one of up to about 0.20% Mg and between about 0.1 and about 0.4% Fe, the method comprising: casting the alioy; hot rolling the alloy at about 850 to about 1000^sC; subjecting the alloy to at least one cold roiling and annealing to substantialiy recrystalize the alioy; cold rolling the alloy to the desired thickness and mechanical strength; and subjecting the alloy to a thermal stress relief treatment, to provide an alioy with a yield strength of at least about 77 ksi and an electrical conductivity of at least about 37% IACS.

22. The method according to ciaim 21 wherein there are at least three cold rollings and annealings.

23. The method according to claim 22 wherein the at least three cold rollings and annealings comprise: a first cold rolling up to about a 75% reduction followed by annealing between about 450 and about 600⁹C for 1 to 48 hours; a second cold rolling up to about a 60% reduction followed by annealing at about 425 and about 600^QC for 1 to 48 hours; and a third cold rolling up to about a 50% reduction followed by an annealing at between about 400 and about 550⁹C for 1 to 48 hours.

24. The method according to claim 22 wherein one of the annealings comprises a step anneal.

25. The method according to claim 24 wherein the step anneal comprises a first anneal at between about 400 and about 500 ^SC followed by a second anneal at between about 300 and about 400 ^BC.