Patent Law

Ex Parte Matsuoka: the case for prima facie non-obviousness

By Alloy Comprising

On October 29, 2007, the Board of Patent Appeals and Interferences decided Ex Parte Matsuoka , upholding obviousness rejections of the claims of U.S. Patent Application No. 10/181,810.

The claims are directed to a steel sheet.  The important limitations of independent claims 3 and 27 are listed as follows (ignore the equation for now):

Claim 3

• C: 0.01 - 0.032%
• Mn: about 1.43 - 2.02%
• V: 0.01 - 0.5%
• V, Nb, Ti and C satisfy equation 0.55xC/12 =< V/51+2xNb/93+2xTi/48 =< 3xC/12
• microstructure of ferrite primary phase and at least 1% martensite
• tensile strength: 450 MPa - 630 MPa (i.e. 46 - 64 kg/mm2), and
• Lankford-value: at least 1.4

Claim 27

• C: 0.038 - 0.05%
• Mn: 2.49 - 2.8%
• V: 0.01 - 0.5%
• V and C satisfy equation 0.55xC/12 =< V/51 =< 3xC/12
• microstructure of ferrite primary phase and at least 1% martensite
• tensile strength: at least 660 MPa (i.e. at least 67 kg/mm2), and
• Lankford-value: at least 1.1.

Claims were rejected over JP 55-158226 to Matsuo in view of JP 57-067131 to Hashiguchi.  Matsuo discloses (note that the translation of Table 1 is erroneous in that four working examples A are listed, which should instead list four working examples A-D):

• C: 0.005 - 0.15% (page 3)
• Mn: 1.6 to 3.0% (page 3)
• V: can be added to further improve the mechanical properties (page 12)
• microstructure of ferrite and martensite
• claim 3 (examples 1, 2, 3 and 7 have claimed tensile strength; examples 3, 4, 5, 6 and 8 have claimed Lankford-value)
• claim 27 (example 8 has claimed tensile strength; examples 2-8 have claimed Lankford-value)

Hashiguchi discloses:

• 0.01 - 0.10% Nb, Ti and V to strengthen the steel (page 12)

Accordingly, the rejection relied upon Matsuo for teaching all but the claimed amount of vanadium, which was supplied by Hashiguchi.

Applicants arguments against the rejections the were ineffective.  Applicants argued (1) vanadium is optional in the steels of Matsuo and Hashiguchi, (2) the present invention avoids the complicated annealing processes of Matsuo and Hashiguchi, and (3) as to claim 27, none of the Matsuo examples disclose carbon and manganese amounts within the claimed ranges.  IMO, the first argument supports the rejection in that it is an admission and the second argument is irrelevant.  The third argument is evidence of non-obviousness but not strong enough to counter that Matsuo broadly discloses overlapping ranges.

Yet, IMO the appeal was winnable as will be explained.

Foremost, the Matsuo and Hashiguchi do not make obvious the claimed combinations of carbon concentrations and tensile strengths in claims 3 and 27.

Regarding claim 3, none of the examples of Matsuo and Hashiguchi (see Tables above) has 0.032% or less carbon and a tensile strength of at least 46 kg/mm2.  This is not a coincidence.  Matsuo teaches at page 5, "addition of 0.05% or less (carbon) is favorable for low yield stress, high tensile strength steel sheets for automotive use with a tensile strength in the 40 kg/mm2 range.  Hashiguchi teaches at page 10 that carbon "controls the tensile strength of the steel plate... To increase the strength, the amount of C should be as high as possible."  Accordingly, it would not be obvious to PHOSITA to achieve the claimed combination of 0.032% or less carbon and a tensile strength of at least 46 kg/mm2.  In fact, Matsuo and Hashiguchi both teach away from the claimed combinations of carbon concentration and tensile strength.

The same rationale would apply to claim 27, with the additional evidence that no working examples of Matsuo or Hashiguchi have a tensile strength of at least 67 kg/mm2.  The closest working example has 65.8 but with a 0.100% C content that far exceeds the claimed range of 0.050% or less.

Additionally, the claimed amount of Mn of about 2.49 - 2.8% of claim 27 would not be obvious from Matsuo and Hashiguchi.  Matsuo discloses at a broad range of 1.6 to 3.0% Mn.  This is evidence of obviousness.  However, this must be weighed against the evidence of non-obviousness.  Specifically, the working examples of Matsuo (see Table above) include a range of 1.62 to 2.08% Mn.  Matsuo discloses no utility in further increasing the amount of manganese.  On the other hand, Hashiguchi teaches at pages 10 and 11 that increasing the manganese content worsens the Lankford-value.  One of the goals of Matsuo is high Lankford value.  Accordingly, a PHOSITA motivted by achieving a high Lankford value would (1) read that Matsuo teaches that the Mn content of the working examples may be increased, (2) have no disclosed reason to increase the Mn content and(3) learn that increasing the Mn content would decrease the Lankford-value, which is desired to be high.  Accordingly, the weight of the evidence would be that it is not obvious to increase the Mn content to about 2.49 - 2.8%.

Finally, there is a challenge to the modification of Matsuo to add vanadium.  The case for obviousness is that Matsuo teaches an alloy having C: 0.005 - 0.15%, Mn: 1.6 to 3.0%, Si: up to 1.5%, and Al: 0.020 to 0.100% (page 3).  Matsuo teaches that vanadium can be added to further improve the mechanical properties (page 12).  Hashiguchi teaches that 0.01 - 0.10% of V can be added to strengthen the steel (page 12).  Accordingly, the Examiner and Board conclude that it would be obvious to add 0.01 - 0.10% of V.

However, issue is not whether it would be obvious to add 0.01 - 0.10% vanadium, but instead whether it  would be obvious to add 0.01 - 0.10% vanadium while maintaining the claimed tensile strength, the claimed Lankford-value, and the claimed microstructure of ferrite primary phase and at least 1% martensite.  The experimental results of the instant application shows that there would be no expectation of success in adding the claimed amount of vanadium while maintaining the claimed properties and microstructure.

In particular, the instant application discloses seven examples (see Table above) where an alloy having C: 0.005 - 0.15%, Mn: 1.6 to 3.0%, Si: up to 1.5%, and Al: 0.020 to 0.100% (such as Matsuo) also has from 0.01 - 0.10% vanadium.  In each instance, either the claimed Lankford-value is not met or the microstructure of ferrite primary phase and at least 1% martensite is not met.  Accordingly, this shows a lack of expectation of success of merely adding 0.01 - 0.10% vanadium to the steel of Matsuo.  Accordingly, it not be obvious to modify Matsuo by adding 0.01 - 0.10% vanadium while maintaining the claimed tensile strength, the claimed Lankford-value, and the claimed microstructure of ferrite primary phase and at least 1% martensite.

In conclusion, the above arguments have shown (1) Matsuo and Hashiguchi do not make obvious the claimed combinations of carbon concentrations and tensile strengths, (2) the claimed amount of Mn of about 2.49 - 2.8% of claim 27 would not be obvious from Matsuo and Hashiguchi, and (3) it not be obvious to modify Matsuo by adding 0.01 - 0.10% vanadium while maintaining the claimed tensile strength, the claimed Lankford-value, and the claimed microstructure of ferrite primary phase and at least 1% martensite.  Even if any of these reasons alone would not be sufficient, together the evidence of non-obviousness of claims 3 and 27 is overwhelming.

Seems Applicant should have won.  Are there flaws in this analysis?

Full post as published by Benefit of Hindsight on October 13, 2008 (boomark / email).