Opinion
Patent Appeal No. 8130.
June 12, 1969.
Barry A. Bisson, Wilmington, Mass., attorney of record, for appellants.
Joseph Schimmel, Washington, D.C., for the Commissioner of Patents. Raymond E. Martin, Washington, D.C., of counsel.
Before WORLEY, Chief Judge, RICH, ALMOND and BALDWIN, Judges.
This appeal is from the decision of the Patent Office Board of Appeals affirming the rejection of all claims, 1-10, of application serial No. 144,219, filed October 10, 1961, entitled "Preparation of Unsaturated Chlorohydrocarbons."
The Invention
Appellants' invention is a method of preparing unsaturated chlorohydrocarbons containing two carbon atoms, e.g., vinyl chloride, trichloroethylene, and tetrachloroethylene. This method, which involves reacting ethane and/or ethylene with ferric chloride in the vapor phase at a temperature in the range of 315-500°C., is based on the discovery that ferric chloride vapors will react directly with ethane and ethylene to effect chlorination thereof and will also catalyze the dehydrochlorination of the chlorinated products to the corresponding unsaturated compounds. The following equations depict the successive reactions that are said to occur in appellants' process to produce vinyl chloride from ethane, equations 1 and 3 representing chlorination and 2 and 4 representing dehydrochlorination:
"Dehydrochlorination" refers to the removal of hydrogen and chlorine atoms from adjacent carbon atoms and the concomitant introduction of a carbon-carbon double bond.
As will be noted from equations 1 and 3, that portion of the ferric chloride which serves as a chlorination agent is reduced to ferrous chloride which is a solid under the conditions of appellants' process. After it has been removed from the reaction zone, this ferrous chloride can be reconverted to the ferric form by reacting it with oxygen and hydrogen chloride recovered as a reaction by-product, as well as with make-up hydrogen chloride from an external source. The ferric chloride thus obtained may be reacted, as before, with the ethane and ethylene. Appellants' specification explains the advantages of this regeneration procedure as follows:
From the foregoing description it can be seen that the source of chlorine for the present process is the make-up HCl which is used in converting the ferrous chloride back to the ferric state. This is an advantage over conventional processes which use free chlorine for chlorination. In the latter HCl is a by-product of the process and presents a disposal problem, whereas in the present process all by-product HCl is utilized by being consumed in converting FeCl2 to Fe2Cl6. In view of this and the fact that HCl is a cheaper commodity than free chlorine, the present process offers decided economic advantages over conventional chlorinations.
Claim 1, the broadest claim, reads:
1. Method which comprises contacting a hydrocarbon selected from the group consisting of ethane and ethylene with a chlorinating agent consisting essentially of ferric chloride in vapor phase at a temperature in the range of 315-500°C. and separating from the reaction mixture an unsaturated chlorohydrocarbon having two carbon atoms.
Independent claim 3 is similar to claim 1 but specifically recites vinyl chloride as the product. Two other independent claims, 5 and 6, in addition to the limitations of claim 1, include further limitations relating to (1) a stage of "subjecting the reaction to reaction at a temperature in the range of 500-750°C. to dehydrohalogenate saturated components," (2) separating the reaction product into four fractions, and (3) recycling two of the four fractions. The dependent claims recite either that the feed is ethane or that the reaction temperature is 350-425°C.
The References
The sole issue is whether the claimed method is obvious ( 35 U.S.C. 103) in view of two United States patents:
Cook 2,838,577 June 10, 1958 Borkowski 3,172,915 Mar. 9, 1965 (filed April 13, 1961)
Cook, the primary reference, discloses a process for producing ethyl chloride and vinyl chloride involving the chlorination of ethane and the dehydrohalogenation of dichloroethanes. The reaction procedes in two stages: (1) a highly exothermic chlorination stage in which the ethane is reacted with chlorine and thereby converted essentially to ethyl chloride and 1,1-dichloroethane, with hydrogen chloride being formed as a by-product, and (2) an endothermic dehydrochlorination stage in which the 1, 1-dichloroethane from the chlorination stage, and additional dichloroethane from a recycle or other external source, are converted to vinyl chloride, hydrogen chloride again being formed as a by-product. Cook utilizes a fluidized bed of particulate solids extending between the two reaction zones so that heat released in the chlorination stage is "stored" in the bed and transferred by it to the dehydrochlorination stage. It is disclosed that this fluidized bed may consist merely of an inert particulate material, such as sand or graphite, or may consist in whole or in part of a catalyst which will promote one or both of the reactions in the process. The catalysts of one group disclosed by Cook are "the chlorides, oxides, and oxy chlorides of metals having variable valences," which are said to have "dual catalytic properties" in that they will catalyze both the dehydrochlorination of dichloroethanes to vinyl chloride and the oxidation of hydrogen chloride to chlorine. Ferric chloride is mentioned as one of the preferred catalysts of this type. Cook explains how this oxidation of hydrogen chloride may be effected (emphasis ours):
* * * hydrogen chloride and oxygen are contacted with the reduced form of the catalyst in a first reactor to form the higher valence form of the catalyst which is then transferred to the chlorination reaction zone of this process. Under the reaction conditions employed for the process, chlorine is liberated with the formation of the reduced valence form of the catalyst which is recycled to the first reactor to be regenerated as a chlorination agent.
Cook also discloses that the oxidation of hydrogen chloride may be effected entirely within the chlorination zone by introducing the oxygen and hydrogen chloride into that zone along with the other gaseous reactants, i.e., chlorine, ethane, dichloroethanes, and ethyl chloride.
There has been some controversy as to whether, under the reaction conditions disclosed by Cook (i.e., temperatures of from 300° to 600°C. and pressures of from 2 to 30 atmospheres), ferric chloride would inherently be in the vapor phase. Suffice it to say that we do not think Cook would suggest to one skilled in the art to employ reaction conditions under which ferric chloride could or would be in the vapor phase and that, in any event, the Patent Office brief indicates that it now relies only on Borkowski for such a suggestion.
Borkowski discloses a process for the oxidation of methane (CH4) to methanol (CH3OH) and/or dimethyl ether ((CH3)2O), which includes an initial reaction of the methane with ferric chloride (FeCl3) to produce methyl chloride (CH3Cl). That initial reaction is shown thus:
Ferric chloride is alternatively referred to in this opinion as FeCl3 and Fe2Cl6, depending on the source quoted.
and Borowski discloses:
* * * the FeCl 3 can be solid, liquid or gaseous form. Preferably the temperature is above the boiling point of the FeCl 3 so that the latter is in vapor phase. Reaction between the CH4 and FeCl3 occurs with the formation of CH3Cl and simultaneous reduction of the iron compound to FeCl2. Generally some of the other chloromethanes are formed but the yield of CH3Cl can be regulated by appropriate adjustment of conditions, particularly the ratio of FeCl3 to CH4. Low FeCl3:CH4 molar ratios give CH3Cl as the main reaction product, while high ratios favor the formation of more highly chlorinated methanes. [Emphasis ours.]
The Borkowski patent attributes invention of the process of the just-quoted passage to one Blair who described it in his abandoned application. Appellants argue that the teaching attributed by Borkowski to Blair, being the subject of an abandoned application, is not available as of the 1961 filing date of the Borkowski patent in view of our decision in In re Lund, 376 F.2d 982, 54 CCPA 1361, (1967). Appellants, however, apparently fail to appreciate the significance of the different circumstances of Lund. There we were concerned with matter appearing only in an abandoned application and not appearing in an issued patent. Here, however, the abovequoted passage stands on its own feet as a disclosure in the issued Borkowski patent. The board was therefore correct in stating:
We see no reason why Borkowski et al. cannot be employed as a reference in this case, because the patent has a different inventive entity, and has a prior effective date.
The inventors named in the patent are Walter L. Borkowski (one of the applicants here), Paul E. Oberdorfer, Jr., and Walter H. Seitzer.
The Rejection
Turning now to the merits of the § 103 rejection, the examiner stated his position as follows (emphasis ours):
Cook et al. disclose a process for the chlorination of ethanes and dehydrochlorination of the chlorination product to form vinyl chloride. In * * [one] embodiment * * * the patentees employ a variable valent metal chloride chlorinating agent which is contacted with the ethane * * *. The reduced metal chloride is then passed to a separate reactor, where it is regenerated with HCl and oxygen. The chlorinated products are then passed to a dehydrochlorination zone. The variable valent metal chloride may be ferric chloride * * *. The primary difference of the instant claims over the Cook et al. reference resides in the use of ferric chloride in the vapor phase. However, this difference is considered to be obvious within the meaning of 35 U.S.C. 103 because Borkowski et al. teach that ferric chloride may be advantageously used in the vapor phase as a chlorinating agent. The use of this technique in the process of the primary reference is deemed obvious. It is similarly deemed obvious to employ higher ratios of chlorinating agent to hydrocarbon to obtain more highly chlorinated ethylenes such as trichloroethylene and perchloroethylene.
We are, of course, not concerned with the obviousness of that difference once it has been observed. Rather, we are concerned with the obviousness of appellants' invention as a whole. 35 U.S.C. § 103.
The board, in affirming the examiner, stated:
We will sustain this rejection for the reasons given in the Examiner's Answer. We agree with the Examiner that the Cook et al. patent not only shows the employment of ferric chloride as a chlorinating agent, or catalyst, but also shows this agent as a preferred material for oxidative chlorination * * *. A person skilled in this chlorination art, familiar with the properties of ferric chloride [,] would realize that the ferric chloride in the process, particularly as set forth in col. 4 of Cook et al., would, or could, be a vapor at the indicated temperatures of chlorination. Accordingly, a person skilled in this art would have had to depend very little on the disclosure of the Borkowski et al. patent for a suggestion of the employment of the ferric chloride as a chlorinating agent in vapor phase in the process of Cook et al. Borkowski et al. make it clear, however, that the ferric chloride may be used in solid, liquid, or gaseous form * * * but that the ferric chloride is preferably in a vapor phase.
Opinion
Appellants' case for the unobviousness of their invention is bottomed on two contentions. The first is that the use of ferric chloride vapor in the Cook process would render that process inoperative because doing so would cause the ferric chloride part of the fluidized bed to be consumed, after which, appellants say, the Cook process would cease to work. We find this argument unconvincing since the Cook process is disclosed as being operative without any catalyst at all. An additional reason why we find this argument unconvincing is that we are concerned with whether the references render appellants' process obvious, not with whether some feature of the Borkowski process can be physically incorporated into that of Cook.
Appellants' second contention is stated as follows:
The employment of ferric chloride in vapor phase as a chlorinating agent for methane does not teach or suggest to the man skilled in the art the application of such chlorination method to hydrocarbons having more than one carbon atom, such as ethane, since the man skilled in the art has no way of predicting the effect that the carboncarbon bond will have on the chemical reaction.
In support of this contention, appellants point to a statement in Cook that under certain reaction conditions some tar formation can result when ethane is reacted with elemental chlorine. Appellants also state:
In general, when reactions involving solid versus liquid versus vapor phase are compared, the man skilled in the art realizes that the conditions required for vaporization at atmospheric or superatmospheric pressure provide more severe reaction conditions than do the conditions required for liquid phase or solid phase contact. There is nothing in the present record, apart from appellants' disclosure, which would lead the art to contact ethane or ethylene with ferric chloride in the form (vapor) which would be likely to provide the most severe reaction conditions.
At most, what can be attributed to Blair from the citation in Borkowski et al. is a teaching that methane can be chlorinated by ferric chloride in solid, liquid or vapor phase. Neither Blair nor Borkowski et al. teach that ethane can be so chlorinated. The reactions of methane, a hydrocarbon containing only one carbon atom, cannot be said to teach the man skilled in the art what will occur with ethane, a hydrocarbon containing more than one carbon atom. In as far as the present art teaches, it would be as reasonable to expect to produce acetylene or tars under the vapor phase conditions of the present claims as to expect that ferric chloride vapor will cause ethane to both chlorinate and dehydrochlorinate to produce vinyl chloride. [Emphasis quoted.]
It may be true, as appellants contend, that one of ordinary skill in the art being in possession of the teachings of Cook and Borkowski, as he is assumed to be, but not being in possession of appellants' disclosure, would think it less than likely that ethane could be chlorinated and the products dehydrohalogenated using ferric chloride vapors. The record here, however, does not provide any real basis for concluding that that would be the reasonable expectation of one skilled in the art. Appellants have told us, in effect, that there are differences between methane and ethane and that there are differences in the severity of the conditions required to effect a reaction in vapor phase as opposed to the solid or liquid phase. Appellants, however, have not given us any basis for judging what degree of significance one of ordinary skill in the art would attach to the just-mentioned differences in predicting the likelihood of success. In short, we think that the examiner, by his citation of the art of record and by his reasoning based thereon, has made out a prima facie case of obviousness which appellants have not adequately rebutted.
The decision of the board is affirmed.
Affirmed.