2 This invention relates to producing higher

₃ hydrocarbons from methane. More particularly, this

4 invention relates to the high temperature conversion of

5 methane and methane containing gases to C2 and higher

6 hydrocarbons, including benzene, in the presence of

7 minor amount of oxygen added to the reaction zone as a

8 free radical initiator. Specifically, it relates to

9 apparatus and method for introduction of the free 0 radical initiator into the reaction zone.

1 BACKGROUND OF THE DISCLOSURE

2 For over fifty years scientists have been 3 attempting to find efficient processes for producing 4 useful liquid hydrocarbon products from natural gas. In 5 October of 1931, the United States Bureau of Mines

16 published a report by Smith et al. titled "The Pro-

17 auction of Motor Fuels From Natural Gas" Report No.

18 R.I. 3143. In this report. Smith and coworkers sum- I9marized the results of extensive investigations

20 directed toward optimizing the production of benzene

21 and C2 unsaturates from methane by the high temperature 22 yrolysis of natural gas. Their process employed

furnace temperatures ranging from about 1150 to 1240°C and, under their optimum conditions, they obtained a feed conversion of 29% with a selectivity towards benzene production of 18.5 wt.% of the feed converted. Other products produced by the Smith et al process were ethylene, acetylene, hydrogen and tar and carbonaceous solids. The selectivity of feed converted to tar was 21 wt.%. Selectivity means the amount of product produced from the converted feed. Thus for every pound of methane that was converted to higher hydrocarbons, 0.21 pound of the product was tar and 0.185 pound was benzene. During their experiments. Smith and coworkers first purged the reactor of air using a stream of nitrogen to insure the absence of oxygen in the reac- tion zone, after which a methane or methane containing natural gas was introduced directly into the reactor. Shortly thereafter. Smith and coworkers were granted U.S. Patent 2,061,597 directed to optimization of the- reaction time for maximizing benzene production over the cracking range of 1000-1200°C. At the optimum conversion temperature of 1150°C, the reaction time was determined to be 42 milliseconds. It should be noted that Smith employed relatively short reaction times in order to avoid the formation of tar and carbonaceous materials. In his article. Smith disclosed that at a furnace temperature of 1200°C, a benzene to tar yield ratio of about 1.6 was obtained (tar does not include solid, carbonaceous materials). One can calculate that the reaction time at 1200°C was about 270 milliseconds.

in U.S. 2,063,133, Hans Tropsch disclosed that liquid products could be obtained from paraffinic and olefinic hydrocarbon gases at relatively low tem- perature conditions of 500-1000°C. The pyrolysis occured in the presence of from about 0.1 to 1.0% of chlorine or chlorine-containing compounds, with 1.0%

being regarded as the upper limit. Unfortunately, details of the Tropsch process are not disclosed. That is, Tropsch did not discuss the amount of tar and carbon that was generated using his process. Another attempt to produce benzene from methane or low mole- cular weight hydrocarbons is disclosed in U.S. 2,875,148 which used a two-stage catalytic process to form liquid products such as benzene.

Much of the art relating to the pyrolysis of low molecular weight hydrocarbon gases such as methane, ethane, etc. relates to the production of acetylene therefrom as is disclosed in U.S. 2,721,227 and 2,912,475. It should be noted that U.S. Patent 2,875,148 and 2,912,475 both teach processes which specifically exclude the presence of oxygen in the reaction zone. Thus, the reactions disclosed in these patents relate strictly to thermal pyrolysis in the absence of materials added to the reaction zone which could initiate or promote free radical reactions. Also, they do not disclose the formation of benzene and the hydrocarbon feed must contain at least two carbon atoms in the molecule. Suitable feeds disclosed include ethylene, benzene and a naphtha stream.

U.S. Patent 2,608,594 to Robinson discloses a two-stage methane cracking process for producing ben- zene. In this process, the methane feed is heated to about 1367 K and then mixed with an oxygen-free, hot combustion gas containing free hydrogen to produce a mixture of feed and hydrogen rich gas at a temperature of about 1900 K. This hot mixture is held at 1900 K for about 0.01 seconds which produces an acetylene containing gas rich in hydrogen. The acetylene con- taining gas is then quenched with additional, cooler hydrogen rich gas to a temperature of about 1422 K and

held at this temperature for about 0.8 seconds to produce a product rich in benzene. It should be noted that this process also produces substantial quantities of tar and solid carbonaceous products.

U.S. Patent No. 3 _/ 542,894 discloses a process for the production of acetylene by the partial oxida- tion of hydrocarbons with oxygen, the flame formed in the oxidation being stabilized by auxiliary oxygen which is injected into the gas mixture at an angle from more than 0° to 90° in the direction of flow of the gas mixture. The auxiliary oxygen burns part of the hydro- carbon forming "jets" which stabilize the hydrocarbon flame. A honeycomb structure is used as the burner block.

Despite more than fifty years of activity directed towards trying to pyrolyze low molecular weight gases to liquid hydrocarbons such as benzene, there is still a need for a process that will accom- plish this with negligible or relatively low carbon and tar formation and with relatively high selectivity of the feed converted to liquid hydrocarbons rich in benzene.

SUMMARY OF THE INVENTION

A process has been discovered for producing useful higher hydrocarbon products, including liquids rich in benzene, by a free radical initiated, thermal conversion of methane or methane containing gas feeds, with a relatively high selectivity of feed conversion to benzene and negligible production of tar and solid carbonaceous materials. In this process the methane containing gas feed contacts oxygen, in a reaction zone, wherein the oxygen acts as the free radical

initiator. An important part o this process is that the oxygen and gas feed should not be premixed, but should be separately introduced into the reaction zone. Thus, this process relates to producing C2 and higher gaseous hydrocarbons and hydrocarbon liquids rich in benzene from methane containing gas feeds by a process which comprises contacting said feed with oxygen at a temperature of at least about 1300 K for a time suf- ficient to convert at least a portion of said feed to benzene, wherein the oxygen is separately introduced into the reaction zone and is present in the reaction zone in an amount greater than 0.5 volume % of the methane.

Liquid hydrocarbon means, of course, hydro- carbons that are liquid at 25°C and one atmosphere pressure. By methane containing gas feed is meant natural gas, methane containing synthesis gas produced by the partial combustion of coal, coke or other car- bonaceous material, and the like. By negligible tar and solid carbonaceous materials is meant less than about 2 wt.% of the total product.

The present invention relates to a reactor for said free radical initiated methane pyrolysis and method for injecting said free radical initiator into said reactor, wherein said reactor comprises a reaction vessel having:

(a) at least one gas feed inlet located at an inlet end of said reaction vessel;

(b) a honeycomb gas straightener downstream of said gas inlet and spacially posi- tioned at said inlet end of said reac- tion vessel; and

(σ) at least one means for injecting a free radical initiator into said reaction vessel transverse to the direction of flow of the gas feed.

The free radical initiator is injected into the methane stream after the methane has been heated to the reaction temperature. The free radical ini- tiator is injected into the methane stream transverse to the methane stream flow either by radial or tangential injection. The radial injection may be either outward from the center or inward from the reactor wall.

It has also been found that the methane can be heated to relatively high temperatures of 1300 K or more in the presence of alumina without the formation of carbon on the alumina surface. This is surprising in view of the fact that those skilled in the art know that methane starts to decompose and cause fouling of surfaces at temperatures as low as about 923 K. Thus, it has also been found that alumina may be used as a heat exchange medium for preheating methane without incurring decomposition of the methane i-nto carbon- aceous materials.

DETAILED DESCRIPTION OF THE INVENTION

As stated above, the present invention re- lates to a reactor for said free radical initiated methane pyrolysis and method for injecting said free radical initiator into said reactor, wherein said reactor comprises a reaction vessel having:

- 7 - (a) at least one gas feed inlet located at an inlet end of said reaction vessel;

(b) a honeycomb gas straightener downstream of said gas inlet and spacially posi- tioned at said inlet end of said reac- tion vessel; and

(c) at least one means for injecting a free radical initiator into said reaction vessel transverse to the direction of flow of the gas feed.

In one embodiment, the means for injecting the free radical initiator into the reaction vessel will comprise an injection head centrally located in said reaction vessel in juxtaposi ion with and down- stream of the honeycomb; said injection head having circumferentially spaced injection ports located therein, said injection ports facing in a radially out- ward direction from the center of the reaction vessel. In another embodiment, the means for injecting the free radical initiator into the reaction vessel will σom- prise a multiplicity of injection ports located in a wall surface of the reaction vessel, including an internal wall surface of the reactor, said injection ports being inwardly facing and spaced ci cumferan- tially about the reaction vessel wall in at least one plane, said plane being perpendicular to a longi- tudinal axis of the reaction vessel. These injection ports may comprise a multiplicity of spray heads; said spray heads being capable of spraying the free radical initiator in a substantially flat spray configuration. The injection ports located in a surface of the reac- tion vessel wall may be directed inwardly at an angle of about 5 to about 45 degrees with reference to a

tangent to said reaction vessel, at the injection port location. Alternatively, the free radical initiator injection means or ports may direct the free radical initiator at an angle of about 45° to 90° relative to the direction of flow of the gas feed or at a 90° angle to the direction of flow of the gas feed.

Thus, one method for injecting a free radical initiator into a reactor for the pyrolysis of a methane containing gas stream will comprise injecting a jet of the free radical initiator or oxygen in a radial direction from an axially oriented injection head in a direction transverse to a flow line of the gas stream and another method will comprise injecting the free radical initiator in a radial direction inward from at least one port located in a wall surface of the reactor.

In one embodiment it will be advantageous for the methane containing gas stream to be introduced into the reactor at a predetermined first velocity and the jet of free radical initiator to be injected at a predetermined second velocity, whereby the ratio of the velocity of the free radical initiator jet to the velocity of the methane containing gas stream is about 102 to about 10 ⁴ .

Another aspect of the present invention com- prises a staged injection of the free radical ini- tiator. This staged injection may be accomplished by utilizing an axially oriented injection head for a first stage free radical injection and a second stage injection of free radical initiator being accomplished by injecting the free radical initiator in a radial direction inward from at least one port located down- stream of the injection head with reference to the

direction of flow of the methane containing gas stream, said at least one port being located in a wall surface of the reactor. Further, an additional injection stage may be utilized, said additional stage being specially oriented upstream or downstream of said first stage with reference to the direction of flow of the methane containing gas stream, said additional injec- tion stage being accomplished by injecting free radical initiator in a tangential direction from at least one injection port located in an internal wall surface of the reactor said injection being at an angle of about 5 to about 45 degrees with reference to a tangent to said reactor wall surface at the injection port location. The staged injection may also comprise a radially inward injection as a first stage and an additional stage injection wherein said additional stage comprises injecting the free radical initiator in a tangential direction from at least one injection port located in an internal wall surface of the reactor said injection being at an angle of about 5 to about 45 degrees with reference to a tangent to said reactor wall surface at the injection port location; said additional stage being spacially oriented upstream or downstream of said first stage with reference to the direction of flow of the methane containing gas stream.

The reaction time, temperature and amount of oxygen required in the process of this invention are interrelated. Higher reaction temperatures require less reaction time and smaller amounts of oxygen and vice-versa. In general, the reaction temperature will range from about 1300 to 1800 K. Preferred and optimum reaction temperatures will depend on the reaction pressure. At atmospheric pressure the reaction tem- perature will preferably range between about 1400 to 1700 , and more preferably from about 1400 to 1600 K.

Under these conditions the reaction time will broadly range from about 0.1 to 1 second, preferably 0.2 to 0.5 second, and still more preferably from about 0.2 to 0.3 second. If the reaction is allowed to continue for too long a time, the selectivity for benzene production will decrease, and undesirable tarry and carbonaceous materials will be formed.

It is important to the process of this invention for the oxygen and methane not to be mixed until the methane has reached the reaction temperature and then to mix them at the reaction temperature as rapidly as possible in order to achieve a free radical reaction initiated by the oxygen and thereby minimize undesirable reactions and concomitant formation of undesirable compounds. Similarly, the methane should be heated to the reaction temperature as rapidly as possible.to avoid degradative pyrolysis of the methane. The methane can, if desired, be preheated to a tem- perature as high as about 975 to 1075 K in the absence of oxygen for relatively short periods of time without cracking or polymerizing to carbonaceous materials or precursors thereof. Thus, it may be advantageous to preheat the methane by any convenient means to such temperature in order to minimize the heat duty of the reactor or reactor feed heater. In the Examples, infra, methane was heated in one step from room te - perature to the reaction temperature at a rate of from about 10 ⁴ to 10 ⁵ K/sec. If desired, the methane or methane-containing gas feed may be at least partially heated by burning some of the feed, mixing unburned feed with the combustion products and introducing the mixture into the reaction zone wherein it contacts the oxygen or oxygen precursor.

It is also important to rapidly cool or quench the reaction products formed by the process of this invention down to temperature levels sufficiently low to stop further reaction and concomitant loss of desired products to tar and carbonaceous materials. Suitable low temperatures will broadly range from about 500 to 1,000 K depending on (a) the products desired (i.e., C2 and higher saturated or unsaturated hydroca- rbon gases, or liquids such as benzene and toluene) , (b) the time that the products are held at such te - perature, and (σ) the secondary cooling rate from such temperature to temperature where no degradation occurs such as ambient temperatures. It is understood of course that degradation of benzene and other reaction products (especially unsaturates such as acetylene, ethylene and other olefins) may occur even at temper- atures as low as 500 K, the extent of such degradation being a function of time and temperature. The quench rate 'employed will depend on the reaction products desired. Thus, it will be appreciated that a faster quench rate will be needed if acetylene is a desired product than if the desired product is benzene and acetylene is not desired. As an illustrative, but non-limiting example, quench rates ranging between 10 ⁴ K/sec and 10 ⁶ K/sec have been successfully employed in the process of this invention when quenching methane reaction products from a reaction temperature of about 1500 K down to about 500 K. A quench rate of 10 ⁴ K/sec and 10 ⁵ K/sec, respectively, will cool from 1500 K down to 500 K in 100 milliseconds and 10 milliseconds, respectively.

As previously stated, more than about 0.5 volume percent of oxygen based on the methane content of the feed gas is required for the process of this invention. Preferably, at least 0.7 volume percent and

more preferably at least about 1% of oxygen will be used. This oxygen content is based on molecular oxygen. However, the oxygen may be present as either molecular oxygen or compounds which on heating yield oxygen containing free radicals wherein one or more unpaired electrons are on the oxygen atom, such as ROO _" , peroxy compounds, RO-, etc. While not wishing to be held to any particular theory, it is believed that the process of this invention is initiated by free radicals such as 0-, OH, and hydrocarbon free radi- cals formed by the reaction of oxygen with methane. The maximum amount of oxygen employed as a free radical initiator will depend on considerations of yield and product selectivity, but in general it is preferred not to exceed about 10 volume % and more preferably 5 volume % oxygen based on the methane content of the feed. Thus, those skilled in the art will appreciate that the process of this invention is not a σonven- tional combustion process or partial oxidation process.

Although benzene is a particularly desirable product of the process of this invention, other useful C2 and higher gaseous and liquid hydrocarbon products are also formed. The following table, obtained using the procedure in Example 2, below, gives a breakdown of a typical product slate from a methane feed contacted with 2.0 volume percent oxygen for 250 milliseconds at a reaction temperature of 1425 K. The extent of methane conversion was about 25 wt.%. No solid σar- bonaceous materials or tarry materials were detected.

As previously stated, negligible means less than about 2 wt.% based on the total product. In general it has been found that from about 0.4 to 2 wt.% of tarry and solid carbonaceous materials, based on total product (or 0.1 to 0.5 wt.% based on methane

feed) will be produced by the process of this inven- tion. Another way of expressing this is the ratio of tar and solid carbonaceous materials produced to the amount of benzene produced which is 1.1 to 5.7 wt.%. This is in marked contrast to prior art processes such as those of Smith et al. which produced 21 wt.% tar and 18.5 wt.% benzene based on the methane converted which can be expressed as a tar/benzene ratio of 113.5 wt.%.

Hydrocarbon Product PRODUCT Selectivity, % C ₂ 's 48 acetylene (75% of total C2's) ethylene ethane ^c 3' ^s 6 propylene methyl acetylene C4's 4 butadiene vinyl acetylene C6 35 benzene c ₇ 6 toluene Tar and solid carbonaceous products 1 Total 100.0

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 is a schematic drawing of the apparatus used in the Examples.

- 14 - Figure 2 is a graph illustrating percent methane converted to higher hydrocarbon products as a function of oxygen content at a reaction temperature of 1425 K and a reaction time of 250 milliseconds.

Figure 3 is a plot of hydrocarbon product distribution as a function of reaction temperature at a reaction time of 250 milliseconds with 2% oxygen.

Figure 4 is a plot of hydrocarbon product distribution as a function of reaction time with 2% oxygen at a reaction temperature of 1425 K.

Figure 5 shows a method of injecting free radicals initiator radially inward into the gas stream.

Figure 6 shows a tangential injection method for the free radical -initiator.

The invention will be more readily under- stood by the reference to the following examples.

EXPERIMENTAL PROCEDURE

The experimental reactor apparatus used is schematically shown in Figure 1. It comprised alumina tube 10 which was 61.0 centimeters long and had an I.D. of 7.0 centimeters surrounded by graphite heating element 12. Graphite heating element 12 was fitted over the alumina tube such that a space, 11, of roughly about 0.3 centimeters existed between it and the exterior wall of the alumina tube. Thus, the graphite heating element did not touch the alumina tube. The space, 11, in between element 12 and tube- 10 was con- tinuously purged with an inert gas such as helium or argon. About 6.4 centimeters of graphite felt insu-

lation 13 were then placed over heating element 12. A water-cooled, aluminum jacket, 15, was placed over graphite insulation 13. Tube 10 was fitted with an alumina honeycomb 14 and capped at one end by aluminum end plate 16. The other end of tube 10 was fitted with a warm-water cooled assembly 34 and capped with aluminum end plate 18. During operation, the methane containing feed gas entered the reaction chamber via inlet ports 20 and 22 and from there passed through honeycomb 14 which served to both straighten out the gas flow and heat same to the reaction temperature. After passing through honeycomb 14 the feed gas then entered reaction zone 24. Oxygen was admitted i.nto reaction zone 24 via line 26 and injector head 28. The free radical initiator (oxygen) is injected radially outward from the center of the reactor into the gas stream at right angles to the flow of gas. The oxygen and hydrocarbon feed streams were separately introduced into reaction zone 24 in order to insure that the oxygen initiated a free radical reaction of the methane at the desired temperature and not before. Reaction zone 24 was defined by the distance between honeycomb heater 14 and the tip of moveable sample probe 30.

Reaction products were quenched and removed at various axial distances from honeycomb 14 using a hot water cooled sample probe, 30, which comprised three concentric stainless steel tubes. In sampling the gas products, probe 30 was inserted into reaction zone 24 from the bottom of the furnace to a predeter- mined axial position. A sampling pump (not shown) connected to the probe was then turned on and regulated so that isokenetic gas samples were extracted through the probe from the tip thereof. The quenched sample was then passed into a gas σhromatograph (not shown) equipped with flame ionization and thermal conductivity

detectors for analysis. The reaction time for a par- ticular run was determined by the distance between the tip of probe 30 and honeycomb 14 and could be varied by adjusting the axial position of probe 30 in order to decrease or lengthen the distance between it and honey- comb 14.

Cooling water used for probe 30 was pre- heated to about 75°C in order to avoid both external and internal condensation of product thereon. It should be noted that hot water and not steam was dis- charged from probe 30. The quenching rate of the reaction products provided by this probe ranged from about 10 ⁴ K/sec to 10 ⁵ K/sec. A Teflon line (not shown) connected probe 30 to the gas chromatograph and was heated to 110°C to prevent adsorption and con- densation of product in the line. During operation, gaseous product that was not removed by sample probe 30 passed down through tube 10 and cooling assembly 34 and was withdrawn via vent line 32.

The temperature of the reactor was con- trolled and monitored by a boron-graphite/graphite thermocouple inserted through insulation 13 and located next to heating element 12. The exterior wall te - perature of alumina tube 10 was checked using an optical pyrometer aimed through sight windows in the wall of cooling jacket 15 and graphite insulation 13. The temperature in reaction zone 24 was determined using a zirconium oxide-coated platinum/platinum-13% rhodium thermocouple inserted into the reaction zone through the bottom of the furnace.

Ceramic honeycomb 14 was 2.54 cm thick, perforated with a number of straight, axially aligned- and radially spaced holes having a nominal pore dia-

meter of 0.318 cm and was cut into a cylindrical shape to just fit the inside diameter of alumina tube 10 in order to maximize the heat transfer between it and the reactor wall. The honeycomb served to both heat and straighten out the feed gas flow. Under the experi- mental conditions employed, heat transfer calculations showed that the honeycomb provided a heating rate to the gas feed of from about 10 ⁴ to 10-5 κ/sec. These calculations indicated that the gas temperature approached that of the temperature of the honeycomb itself upon exiting therefrom. These calculations were confirmed by measuring the temperature of the reaction zone with the zirconium oxide-coated, platinum/plati- num-13% rhodium thermocouple. Under typical run con- ditions at a wall temperature of 1500 K for tube 10, the reaction time-averaged temperature of the gases in reaction zone 24 was about 1425 K.

hile the thickness of the honeycomb, 1_4, and the diameter of the axially aligned radially spaced holes are not critical, it is preferred that the relationship between the thickness of the honeycomb and the hole diameter is such that the L/D ratio is about 4:1 to about 16:1; preferably about 6:1 to about 9:1. The hole diameter (D) is a function of the size of the reactor system and will depend on the material and desired rate of introduction. Those skilled in the art can readily size the inlet ports without undue experi- mentation.

In order to minimize resistance to flow caused by the positioning of the honeycomb, _14_, in the reactor inlet area it is preferred that the ratio of the cross-sectional area of the honeycomb to the total cross-sectional area of all of the axial holes in the-

honeycomb, 1__, is about 1.1 to about 2.0; preferably about 1.25 to about 1.75. The lower limit of this ratio is subject only to structural considerations.

The oxygen injector assembly was made of a feeder tube, 26, attached to a cylindrical head, 28. Head 28 contained six radially-drilled holes 0.022 cm in diameter which were evenly distributed near its closed end. The injector assembly was inserted through a central passage of honeycomb 14 and was positioned in a manner such that the holes in head 28 were just beneath the bottom surface of honeycomb 14. During the runs, oxygen was injected through the holes as radial jets outward from the center of the reaction zone. Cross flow jet mixing calculations revealed that, under the conditions used in the experiments, the time required for complete jet mixing of the oxygen with the hydrocarbon feed gas stream was negligible compared to the reaction time in the reaction chamber.

While the injector assembly head, 2_8, is located at the. center of the reactor to provide radially outward initiator injection the injector may be located in the walls of the reactor and permit either radially inward injection or tangential injection.

Referring now to Figure 5, a radially inward injection profile is depicted. Methane is introduced into the reaction chamber via inlet ports, 2_0_, and, 22, and passes through the honeycomb 14. Initiator is injected into the methane stream through injector, 35, located downstream of the honeycomb, 1_4, immediately adjacent to the honeycomb,

The initiator is preferably injected as a multiplicity of streams. While it is preferred that all initiator which is injected into the methane stream be injected radially toward the center, the initiator may be injected using a spray head having a spray angle, 3_6_, of not more than 45°, preferably the spray pattern is in a substantially flat plane which is at right angles to the line of flow of the feed rather than in the conventional conical spray pattern. As used in the specification and claims the term "spray angle" means the included angle enclosed by the outer- most spray streams of a particular spray head, the bisector of the included angle being a radial line with respect to the cross-section of the reaction chamber.

in the radial injection method (inward or outward) it is preferred that the initiator be injected at right angles to the flow. However, the angle of injection may vary from about 45° to about 90° relative to the flow direction, where the 90° angle is at right angles to the flow stream and 0° is in the direction of flow along a flow line.

Referring now to Figure 6 a tangential injec- tion scheme is depicted. While the injector inlet 37 is depicted as being tangential to the reactor tube, l_0, it is evident that if the initiator injection was at right angles to the radius, 39, of the tube in a true tangential fashion, the initiator would enter the reaction chamber along the reactor tube wall and not be mixed with the main gas stream. Hence, it is necessary that in the "tangential injection" method the initiator be injected into the methane stream at some oblique angle from the tangent. The injection angle should be at least -5°. Preferably the injection angle is about

10° to about 45° as measured from ^" the tangent, 3_8_, to a radius, 3_9_, the angle along the tangent being taken as 0 degrees.

The diameter of the initiator injection port is not critical. It will be determined by the ini- tiator flow rate requirement, the methane gas stream velocity and the available initiator injection pressure and the total number of injection ports. The diameter of each port can be readily determined by those skilled in the art to obtain the desired jet velocity into the methane feed stream for a given reactor size. To ensure that there is a uniform and rapid mixing between a jet stream of free radical initiator and the methane gas feed about 1 to about 12 injection ports are utilized; preferably about 4 to about 12 ports; more preferably about 6 to about 12 ports are utilized. While a greater number of injection ports than 12 may be utilized no apparent advantage is found in doing so. The critical parameter for efficient mixing of the free radical initiator and methane gas feed stream is the ratio of the velocity of the free radical initiator stream to the velocity of the methane gas feed stream. This ratio should be about 10^ to about 10 ⁴ . Wherever any injection ports are used they are preferably equally spaced in equal arc segments around the injec- tion head or reactor wall. The radial injection ports of head, 2_8_, of the experimental equipment shown in Figure 1 are 0.022 cm. in diameter.

As used in the specification and claims the term "reaction vessel" means the reaction tube, 10, and the end plates, 1^ and 1_8_. While the location of the initiator injection ports relative to the honeycomb is not critical they should be located as close to the

honeycomb as practical since the methane stream will have reached its reaction temperature by the time it exits the honeycomb.

The term "honeycomb" as used in the speci- fiσation and claims means a block having holes through its thickness, said holes being radially spaced with respect to the center of the face of the block. While the holes can be axially oriented, that is at right angles to a face of the block it is within the scope of this invention to have the holes at an oblique angle to the axial orientation while being positioned in a plane parallel to an axis of the block which is perpendicular to a face of the block. This orientation at an oblique angle will impart an angular velocity to the methane stream. The aforesaid oblique angle can be about 60 degrees to about 90 degrees as measured from a face of the block, 90° being the angle of a hole which is per- pendicular to a face of the block.

The honeycomb may be of solid monolithic material or may be cored. Where it is cored, hot gases can be passed through the honeycomb structure around the radial axial holes, which will appear as tubes within the cored structures, the hot gases thereby heating the methane stream.

While each of the three initiator injection techniques, radially outward, radially inward and tan- gential are independently described those skilled in the art will recognize that any combination of these techniques or all three techniques may be used simul- taneously in the same reactor. In large diameter reactors (a foot or more in di-ameter) it is advanta-

- 22 - geous to combine the radially outward method of in- jeσtion of initiator with at least one of the other techniques.

In a preferred embodiment a staged injection of free radical initiator is used. The term "staged injection" means the use of a multiplicity of injection ports, at least one of said ports being spacially oriented downstream, with respect to the methane feed, of at least one other injection port. Preferably several sets of ports are arranged to inject free radical initiator in different planes each plane being located further downstream than the others. The total quantity of free radical initiator to be injected into the reactor is distributed among the various ports. In a preferred embodiment the distribution among ports is equal. By way of illustration a first stage injection of initiator can be through an axially located injec- tion head as depicted in Figure 1. Subsequent stages of injection can be inwardly directed as depicted in Figures 5 and 6. It will be evident from this dis- closure to those skilled in the art that "staged injection" means that the injection is "staged" in terms of its location downstream of the methane feed inlet and not sequentially with respect to time.

^' Those skilled in the art will recognize that the sizing of the inlet port diameters is readily determined from a consideration of the total quantity of free radical initiator to be injected, the available injection head pressure and the desired jet velocity of the free radical initiator. This determination can readily be accomplished with reference to the foregoing disclosure by those skilled in the fluid dynamics arts.

Where the injection means of this invention comprises spray heads located in the reaction tube walls, the spray heads are preferably designed to pro- duce a substantially flat spray pattern rather than the customary conically shaped spray pattern. The term "substantially flat spray pattern" means a spray pat- tern which is substantially in a single plane and coincident with the plane in which the spray head is positioned.

The surface to volume ratio of reaction zone 24 was about 0.57 cm~l which was relatively small in order to insure that wall effects would be insignifi- cant. Experiments were run in order to verify this. To determine this, a gas flow restrictor was placed on top of honeycomb 14. This restrictor was a toroidal shaped alumina plate which just fit inside alumina tube 10 and had a hole therein 2.77 cm in diameter. The methane gas flow was thus restricted to the central portion of reaction zone 24, which amounted to about 15% of the entire cross-sec ional area of the reaction chamber, before it expanded again downstream of reaction zone 24. Comparing the product species distributions at the same reaction times revealed little difference between the products from the reactor both with and without the gas flow restrictor. This thus substan- tiated that there was no wall effect under the experi- mental conditions employed.

EXAMPLE 1

This experiment determined the effect of oxygen concentration on the extent of conversion of methane to higher hydrocarbons. The temperature of the reaction zone was 1425 K and the reaction time was 250 milliseconds. Ten liters per minute of methane were

mixed with one liter per minute of argon and the mixture fed into the reactor shown in Figure 1. The argon was used as a tracer for carbon balance deter- minations which, on average, gave carbon balances ranging between 98 and 100%. This range was within experimental error since there was no visible evidence of solid carbonaceous materials having been formed in the reaction zone. At the same time 0.2 liters per minute of a mixture of oxygen and helium were fed into the reactor through inlet 26 and head 28. The oxygen- helium stream was introduced in such a manner that its momentum was always kept constant to ensure the same mixing pattern with the methane-argon stream. To this end,, the amount of He was varied so that the total flow rate of the oxygen-helium mixture was maintained con- stant at 0.2 1/min at different oxygen concentrations. After about 1-2 minutes the reaction reached steady state conditions. During the reaction, product was continuously removed from reaction zone 24 via sample probe 30 and fed to the gas chromatograph. The volume percent of oxygen present based on the methane content of the feed gas was varied from 0-2%.

Figure 2 is a plot of the results of this experiment which clearly demonstrate the unexpected and unanticipated effect of a minimum amount of at least about 0.5 volume percent of oxygen required in the reaction zone in order to achieve free radical ini- tiated methane conversion. In all cases the selec- tivity to benzene formation from methane was about 35 wt.% and about 75% of the C2's were acetylene. It should be noted that at the end of each experiment significant amounts of carbonaceous products were found to have accumulated at the bottom of the reaction chamber. This was due to the fact that most of the reaction products were not removed from the reaction

chamber by the sample probe but continued to pass down to vent 32. As they continued to proceed down the hot reaction chamber they continued to react and poly- merize, ultimately forming tar and carbonaceous products.

However, it should be emphasized that, in this and in the examples below, there was no evidence of solid carbonaceous or tarry materials having been formed in the reaction zone, in the sample probe or in the products removed from the reaction zone by the sample probe.

EXAMPLE 2

This experiment was similar to that of Example 1 except that the oxygen concentration was maintained at 2 volume percent of the methane and the reaction time was maintained at 250 milliseconds. In this experiment, a series of runs were made varying the temperature in order to determine its effect on the methane conversion and the product distribution of the converted methane. The results of this experiment are shown in Figure 3 and illustrate an increasing amount of C2 hydrocarbon (75% acetylene) formation as well as increasing benzene production with increasing temper- ature.

EXAMPLE 3

This experiment was similar to Example 1 except that the reaction zone temperature was main- tained at 1425 K and the oxygen at 2 vol. % of the methane feed. In this experiment the reaction time was varied in order to determine its effect on the extent of methane conversion and the product distribution of

the converted methane . The results are shown in Figure 4 and show gradually increas i ng C2 and benzene p ro- duction as the reacti on time * i ncreased ( aga in , about 75% of the C2 * s were acetylene) .

EXAMPLE 4

This experiment demonstrates that premixing the methane and oxygen before the methane has reached the reaction temperature does not result in the process of this invention. In this experiment, two separate reactors were used. Reactor A was used to demonstrate the separate addition case and reactor B was used to demonstrate the premixed case. Both reactors comprised a 1.4 cm I.D. quartz tube axially fitted through a σyclindiσal electrically heated muffle furnace. The reaction products were analyzed by gas chromatography and the reaction temperature was 1373 K.

in reactor A, a 0.6 to 1 volume ratio of oxygen to argon was preheated and fed into the reaction zone via a quartz tube having an I.D. of 7 mm and an I.D. of 9 mm axially located inside the reactor tube. The preheat zone was 12.5 cm long and the reaction zone was 94 cm long. Methane was separately added into the reaction zone through the annular space between the inside of the reactor tube and the outside of the oxygen/argon injection tube. The preheat zone for the methane was also 12.5 cm long. The overall volume ratio of CH4:θ2: fed into the reaction zone was 1:0.06:0.1.

In reactor B, a 1:0.06:0.1 volume ratio of CH4:02:AE was premixed and fed into the reaction zone via a quartz preheating tube having a 3 mm I.D. The preheating zone or length of the 3 mm tube inside the furnace was 28 cm. The reaction zone was 79 cm long.

In both cases reactions were conducted at reaction times of about 250, 380 and 500 msec. It was found that the presence of the oxygen increased the methane conversion and the benzene yield only for the separate addition case, reactor A and not for the pre- mixed case, reactor B. In the premixed case no enhanσe- ment was observed with the oxygen. That is, in the premixed case essentially the same results were ob- tained without oxygen.

The results obtained for the separate in- jection case showed essentially the same level of im- prove ent in methane conversion as in the apparatus used for obtaining the data in Examples 1, 2 and 3, but at the lower temperature of 1373 K.