Application of Kenneth W. Doak and Michael Erchak, Jr, 418 F.2d 1387 (C.C.P.A. 1969)

RICH, Acting Chief Judge.

This appeal is from a decision of the Patent Office Board of Appeals¹  affirming the examiner's rejection under 35 U.S.C. § 103 of claims 31-34 in application serial No. 211,355, filed July 20, 1962 for "Ethylene Polymerization Peroxide-Initiator and Process."

The invention relates to a continuous process for preparing low-density polyethylene by polymerizing ethylene in a tubular reactor at temperatures of 225-600°F and pressures of at least 1000 atmospheres in the presence of certain free radical initiators or catalysts. Appellants' specification comments on certain problems involved in prior art processes:

* * * In polymerization reactions at these high pressures oxygen or a peroxide, that is a free radical initiator, is usually employed at temperatures ranging from about 225° to about 600° F. The polymerization reaction of ethylene in high pressure processes of this nature is highly exothermic and heat control methods have to be employed to avoid an undue rise in reaction temperature which can result in run-away reactions and explosions. * * *

* * * it is known that in polymerization reactions of ethylene erratic uncontrolled temperature fluctuations in the reaction zone directly affect the physical properties of the polymer and as a consequence various schemes have been proposed in the art for effecting a reaction without large fluctuations in the reactor temperature curve (profile). The temperature profile can be constructed by simply plotting the readings from all of the thermocouples inserted along the reactor. Thus when ethylene and peroxide initiator are introduced into a tube and polymerization begins, usually at a temperature above 225° F., due to the exothermicity of the reaction and the initial high concentration of the initiator, a temperature peak occurs in the reaction which affects conversion and the properties of the polymer. The properties of the polymer are affected if such a peak is followed by uncontrolled erratic peaks or valleys along the reactor temperature profile. * * * To obviate some of those difficulties, appellants employ in their process a multicomponent, free radical-generating, peroxide initiator composition which is said to effect both a controlled, uniform rate of change of temperature between the initiation temperature and the peak reaction temperature as well as an increased conversion of ethylene to polymer. The nature of the particular peroxide initiator composition is best reflected in representative claim 31 (emphasis ours):

31. A process for polymerizing ethylene in a tubular reactor at pressures of at least 15,000 psi and at temperatures which rise, due at least in part to the heat generated by the polymerization reaction, from about 225° F to as high as 600° F, which comprises maintaining a controlled temperature rise during said polymerization reaction by providing a substantially controlled and continuous supply of free radicals during the polymerization temperature rise by employing during said reaction a series of free radical forming peroxide initiators of distinct but overlapping decomposition temperature ranges, at least three such peroxide initiators being employed and at least one of each of said peroxides being selected from each type of the group consisting of Types I, II and III, said types having 10 hour half-lives at the following temperatures: Type I — from 110° to 175° F; Type II — from 175° to 250° F; and Type III — from 250° to 320° F.

Appellants' specification explains the result of their use of three initiators having the decomposition properties defined in the above claim as follows:

* * * in the prior art, in order to obtain a high conversion with a high temperature peroxide, for example a Type III initiator, it was essential to inject it at a low temperature. However, in order to obtain the required high rate of free radical formation at the low temperature, a large quantity of peroxide was required. As the temperature increased, the rate of radical formation would become so high that the temperature would rise above the decomposition point of ethylene. This invention overcomes this difficulty by the use of a lesser amount of a high temperature peroxide, that is a Type III initiator, and the introduction of a controlled amount of a Type II peroxide initiator which has a half-life of 10 hours at 175° to 250° F. This Type II peroxide supplies the required amount of free radicals at the lower temperature. Therefore, as the temperature increases, the Type II peroxide is dissipated and the Type II peroxide furnishes the required (but smaller) amount of free radicals. The use of Type I peroxide permits the use of a still lower initiation temperature, this lower initiation temperature being limited only by practical considerations involving kinetic chain length at low temperatures. The use of several peroxide initiators thus makes it possible to initiate the reaction at a lower temperature and obtain a controlled (although not necessarily uniform) temperature rise across the reaction zone. * * *

In a process, as the instant, employing a multicomponent peroxide initiator, at any given temperature range, the quantity of the useful initiator present will determine how many polymer chains will be started while at that temperature. Also, since the amount of polymerization in any given section determines the quantity of heat liberated and thus the temperature rise of the reactants, the amount of useful initiator present in its useful range will determine the slope of the reactor temperature profile within its useful range. It is thus possible in accordance with this invention to adjust the slope of portions of the reactor temperature profile without drastically altering other conditions. * * *

The useful range of the initiators employed should preferably overlap each other. That is, at any given temperature range one particular initiator Type is predominantly useful; however at the beginning of the range the preceding initiator is still somewhat useful and at the end of the range the next initiator begins to be useful (i. e. to decompose). Such overlaps are preferred as they contribute related benefits. The overlap of free radicals generated from the different peroxides will prevent wide or sharp fluctuations in the free radical concentration in the reaction zone, and this prevents sharp changes in rate of change of temperature which results in keeping a smooth temperature profile. This results in better control over the reaction and leads to better control of polymer properties. [Emphasis ours.]

The references are:

 Overbaugh 2,909,513 Oct. 20, 1959 Deex et al. 3,142,666 July 28, 1964 (Deex) (filed Mar. 5, 1959)

The examiner rejected the claims under 35 U.S.C. § 103 as unpatentable over Deex alone or in view of Overbaugh.² 

Deex relates to "methods for controlling the highly exothermic polymerization of ethylene as carried out continuously by a non-solvent process in a tubular reactor so as to obtain high conversions to polymers of ethylene having good density and tensile properties." He accomplishes those results by employing (1) a mixture of organic peroxide initiators having "significantly different" decomposition or initiation temperatures or, alternatively, (2) a mixture of "one or more" peroxide catalysts with oxygen wherein there is a "significant difference"³  in the effective initiating temperature of "at least two" of the initiators. According to Deex, with such an initiator combination, the ethylene polymerization procedure involves:

* * * utilizing a high temperature peroxygen or oxygen catalyst under high temperature conditions suitable for obtaining high conversion with some loss of desirable polymer properties, and the additional use therein of a low temperature peroxygen catalyst to obtain a substantial portion of polymerization while being heated up to the high temperature, thereby improving the properties of the resulting polymer and possibly also improving conversion.

In Example 3, Deex employs a mixture of benzoyl peroxide and t-butyl perbenzoate⁴  ("decomposition temperatures" as defined by Deex of 133°C and 166°C, respectively) as polymerization initiators in amounts of 6 and 11 micromols/mol of ethylene, respectively, to obtain a conversion of 15.8% over a reaction temperature range of 120-220°C, whereas increasing the benzoyl peroxide amount from 6 to 8½ micromols/mol raised the conversion to 19.5%.

In Example 5, in which benzoyl peroxide and oxygen were employed as polymerization initiators ("decomposition temperatures" as defined by Deex of 133°C and 160°C, respectively), Deex states:

* * * A substantial part of the foregoing polymerization, as determined from the internal temperatures measured in the reaction tube, took place at reaction temperatures of 120 to 170°C., and the peak reaction temperature in the latter half of the tube was only 215°C. In contrast to this, a single catalyst would ordinarily have no effective polymerization occurring in large parts of the reaction tube, as indicated by temperature data; for example, with t-butyl perbenzoate as a single catalyst, there will often be practically no polymerization in the upstream portion of the reaction tube, with temperatures of the order of 50 to 70°C., while in the downstream portion of the tube there is danger of the peak temperature going high enough to cause substantial carbonization. In one aspect, the present invention can be considered as utilizing catalyst mixtures to effect a more smooth reaction temperature-reactor time curve, i. e., to have a substantial portion of the polymerization occur at the lower temperatures around 120 to 170°C., and to also have a substantial portion at temperatures about 170°C. but without any sharp, high peaks in the temperature curve. In the present reaction the temperature has been controlled to the extent that the peak temperature is only 215°C., while polyethylene can be polymerized at temperatures as high as 320°C. It will be realized that it would be feasible to conduct the polymerization at considerably higher temperatures while retaining a smooth reaction temperature curve without reaching undesirably high peak temperatures, and to obtain the even higher conversions associated with higher reaction temperature. * * * [Emphasis ours.]

In the examiner's view, appellants and Deex recognized many of the same problems existing in the ethylene polymerization art and have attempted to solve those problems in much the same manner. He acknowledged that Deex did not expressly disclose the use of three peroxide initiators, but nevertheless maintained that the language of Deex's disclosure suggests that two or more peroxides were contemplated. The examiner considered that it would be obvious to one of skill in the art to use three peroxides having different, but overlapping, decomposition temperature ranges since Deex specifically discloses that a smoother reaction temperature-reactor time curve free of sharp high peaks can be obtained using multiple, rather than single, free radical initiators.

The board agreed, adding:

* * * the reference utilizes catalyst mixtures to effect a more smooth reaction temperature-reactor time curve without any sharp high peaks which is the result appellants wish to attain.

Thus, the process of the reference accomplishes substantially the same results as appellants' process, and we are reinforced in this position by reference to * * * [appellants' specification] which states:

"A minimum of two peroxide initiators is required in accordance with this invention."

* * * since Deex et al. clearly suggest the use of a plurality of peroxide intiators, it is considered that appellants have merely followed this teaching and have employed three or more such peroxide initiators for the same purpose in view. We consider this to be the ordinary and expected advance in the art and within the ordinary skill of those working in this field.

Appellants urge that the examiner and board failed to consider either the background of their invention or the state of the prior art at the time they made their invention. As a result, appellants contend, the examiner and board erroneously concluded that the respective objectives of Deex and appellants were the same, and that Deex suggests the process appellants carry out wherein a reaction temperature reactor time curve containing what they term a sharp, high reaction temperature peak is obtained.

The alleged state of the prior art, to which appellants refer, was set forth in their petition for reconsideration before the board and is reiterated here. It relates to so-called "constant environment" ethylene polymerization processes in which polymerization occurs in autoclaves or tubular reactors at substantially constant temperature or over a relatively narrow temperature range employing either a single catalyst or two catalysts decomposing at substantially the same temperature. Appellants correctly view their process as a departure from such prior art processes, inasmuch as it involves polymerization over a wide temperature range of 225°F-600°F with the assistance of three or more polymerization initiators having the different but overlapping decomposition temperature ranges recited in the claims. Implicit in appellants' argument here is their relegation of Deex to the category of "constant environment" or constant temperature polymerization processes. As evidence of that, appellants rely on the heretofore quoted Example 5 of Deex where the reaction is carried out in a manner to avoid "any sharp, high peaks" in the reaction temperature-reactor time curve.

We cannot agree with appellants' characterization of Deex, or with their assertion that they are the first to polymerize ethylene over a wide, "dynamic" temperature range. It is evident from Deex as a whole — Example 5 in particular — that he did not wish to conduct his polymerization at temperatures approaching isothermal, constant environment conditions or to avoid a reaction temperature peak,⁵  but that he, like appellants, desired to polymerize ethylene over a broader temperature range than had theretofore been possible with the use of a single catalyst. To accomplish that result, rather than using a single catalyst, most of which decomposes at a relatively high temperature to result in little polymerization in the upstream portion of the reactor but extensive polymerization in the downstream portion with danger of substantial carbonization, Deex employs two peroxygen catalysts having significantly different decomposition temperatures to obtain substantial polymerization in the upstream portion of the reactor initiated by the low temperature catalyst which continues over a smooth temperature curve to a peak temperature, thus obtaining higher conversion and more desirable polymer properties. With the two initiators employed in Examples 3 and 5, the peak temperature was approximately 215-220°C, although Deex suggests that it is possible to retain a smooth temperature curve and obtain even higher conversions by conducting the polymerization at still higher reaction temperatures while avoiding undesirably high peak temperatures. As one way of accomplishing that result, we see no reason why one of ordinary skill in this art would not, as a matter of course, employ the requisite amount of yet another initiator compound having a still higher decomposition temperature.

Contrary to appellants' arguments, it seems to us that both the general and more specific disclosure of Deex reasonably suggests the use of at least three initiators for effecting polymerization over a broad temperature range as the examiner has found.

The question remains whether Deex would suggest the use of three peroxide initiators having the overlapping decomposition properites which provide the substantially continuous supply of free radicals during the polymerization temperature rise as recited in the claims. That issue was recognized by the examiner, though he did not discuss it to any great extent other than to observe that Deex, like appellants, desires to obtain a smooth reaction temperature-reactor time curve over a wide temperature range. It appears to have been the examiner's position that one of ordinary skill would realize that he would necessarily have to employ at least two, perhaps three or more, peroxides having overlapping decomposition temperature ranges in order to obtain such a smooth curve free of erratic changes.

On that issue, appellants contend:

* * * The true concept of Deex et al is clearly disclosed in the specification of their patent * * *. This concept did not involve initiators with overlapping decomposition temperature ranges as is obvious from the list presented * * *.

Appellants then give an illustration of their contention. The general selection of initiators which appellants seem to urge would be made from the list contained in the Deex disclosure ignores, however, the more specific selection criteria imposed by Examples 3 and 5 to permit obtention of a smooth temperature curve in accordance with the patentee's suggestion. We note that appellants do not argue that one of ordinary skill would be unaware that the initiators suggested by Examples 3 and 5 of Deex do in fact have overlapping decomposition properties.

Appellants' arguments have not convinced us of error in the position taken by the examiner and affirmed by the board. The decision is affirmed.

Affirmed.