Coal Gasification

4

Arc Synthesis of Hydrocarbons CHARLES

SHEER

and

SAMUEL

KORMAN

Downloaded by UNIV LAVAL on July 13, 2016 | http://pubs.acs.org Publication Date: June 1, 1974 | doi: 10.1021/ba-1974-0131.ch004

Chemical Engineering Research Laboratories, Columbia University, 632 W. 125th St., New York, N. Y. 10027

A convective arc featuring a novel cathode injection system has been studied on a laboratory scale producing hydrocarbons. Injection gases included H , CO-H mixture, and steam projected against a carbon anode. Also injected was a powdered solid, (CH ) , entrained in argon. The arc effluent was withdrawn via a hole in the anode through an intermediate hot zone to sampling equipment. A spectrum of operating parameters was studied whereby the hydrocarbon in the product could be varied from pure methane to pure acetylene. A catalytic surface effect on effluent composition within the intermediate hot zone was also observed involving wall temperature, contact surface material, and residence time. The results indicate that this technique may ultimately be applied to coal gasification. 2

2

2

X

' T ' h i s report summarizes some results of an investigation using a novel type of arc for hydrocarbon synthesis. The principal reacting system was carbon and hydrogen. Substitutions for hydrogen were also employed in a limited way, including a 50-50 vol % mixture of C O and H , steam, and a solid petroleum residue, essentially ( C H ) # . The type of arc employed was developed i n our laboratory and is characterized b y a number of unique features; an important one is the high rate of continuous through-put of feed material including fluids. The feed is raised to high temperatures and comprises the plasma envi ronment of the discharge, particularly the arc-conduction column main tained between the electrodes. The composition of the plasma is derived from that of the feed; however, the atomic, molecular, or free radical plasma species differ significantly from the molecular composition of the feed. Conventional arcs, consisting of a gaseous, electrical conducting col umn joining a positive anode and a negative cathode, are used as a 2

2

42 Massey; Coal Gasification Advances in Chemistry; American Chemical Society: Washington, DC, 1974.

4.

SHEER A N D KORMAN

Arc Synthesis of Hydrocarbons

43


source of heat, for example, as i n electric smelting processes. Heat is transferred to feed materials by radiation and conduction from the hot column. This is the zone of primary energy dissipation i n which the electrical energy is converted to radiant energy and sensible heat which flow out i n a l l directions through the intervening layer of atmosphere. The maximum temperature which may thereby be maintained i n the surrounding charge is limited to about 2500 °C. In such arcs, little if any of the material treated enters the energy dissipating region within the arc-conduction column. The first opportunity for treating materials continuously to tempera tures higher than 2500°C arose with the discovery by Beck ( I ) i n 1910 of the high intensity arc. W i t h this type of discharge, the substance of the anode can be vaporized and passed through the conduction column where the temperature is raised to 10,000°K or more (2). When the feed material is incorporated into the anode, a major fraction can be exposed to the high energy density of the column (3). This is suitable however only for treating solids. Problems are encountered when one or more of the reactants is a gas. Foremost of these is the instability induced i n the arc column by appre ciable forced convection. A considerable effort has been expended during the past two decades i n stabilizing arcs subject to vigorous convection (4). A number of stabilization techniques have evolved including con finement of the column i n a water-cooled channel ( 5 ) , vortex stabiliza tion (6), and magnetic stabilization (7). In several techniques (8, 9) the problem is avoided to some extent by mixing a gaseous reactant with the arc effluent; then the column proper is not subject to strong convec tion. None of the above, however, achieves a high degree of penetration of the gas into the primary energy dissipation zone within the column. Recent work i n this laboratory showed that large quantities of gases can be injected into the arc column i n a practical manner. The gas is injected from the cathode end by means of a specially designed annular nozzle surrounding the cathode. This device is called the fluid convec tion cathode ( F C C ) . Basis of the Fluid Convection Cathode The arc column converges to a small tip at the cathode surface (see Figure 1). This convergence, representing an inhomogeneous electric current flux, defines a zone of inhomogeneity i n the accompanying mag netic field that produces a fluid mechanical thrust away from the cathode toward the anode, thus causing a pressure gradient away from the cathode tip. T o stabilize this gradient, gas is aspirated into the arc

Massey; Coal Gasification Advances in Chemistry; American Chemical Society: Washington, DC, 1974.

44

COAL

GASIFICATION


column i n the region of inhomogeneity and is propelled away from the cathode, creating the cathode plasma jet (10). This region is the only portion of the arc other than the anode crater through which appreciable quantities of gas may be injected without disturbing the stability of the discharge.

SHROUD

GAS

FLOW

Figure I . Sketch of FCC showing compressive effect of gas flow on base of arc column The F C C was developed during a study of the influence of gas convection into the base of the arc column near the cathode of the arc ( I I ) . The conical tip of the cathode is surrounded by an annular nozzle which terminates upstream from the tip and which directs the gas i n a converging high-speed layer into the column of the arc close to the point where it originates on the cathode. It was found that if this were done so that the gas impinged on the arc column in the contraction zone, the gas would preferentially enter the column. Further, the gas could be so injected at 10-20 times the natural aspiration rate. If an attempt is made to force the gas into the arc column elsewhere, the degree of penetration is far less and the injected gas tends to unstabilize and blow out the arc. In confined arcs, where stability is achieved b y enclosing the arc discharge within a water-cooled channel, it has been shown that over 70% of the injected gas never enters the column (12) and receives considerably less than the maximum possible activation energy from the arc.


4.

SHEER

A N D KORMAN


45


Experimental Figure 2 is a diagram of the arc apparatus, showing an F C C cathode through which the hydrogen gas is injected into the conduction column and a 1-inch diameter cylindrical carbon anode. The anode has a 1/4-inch hole along its longitudinal axis. The anode is connected at its back end by 3/16-inch i d metal tubing to a 11/16-inch diameter, type 304 stainless steel tube surrounded by an electrically heated laboratory tube furnace. Leaving the furnace is a water-cooled heat exchanger following a tee connection valved to permit the gas stream to vent i n either of two directions: (1) to a flowmeter and laboratory pump, or (2) to a manifold of a vacuum gauge and several valved 500-ml gas-sampling bottles. These are evacuated before use. The carbon anode can thus serve as a combi nation source of solid carbon and of carbon vapor issuing from the anodic arc terminus into the plasma column, as well as an arc crater gas-sampling probe. Depending on the pumping flow rate or timed pressure rise i n the sampling branch, it is possible to draw an arc flame effluent gas stream from the reaction zone through the tube furnace to vary the residence time at any temperature up to about 1000 °C and thence through the heat exchanger and into the sample bottles i n sequence. In operation with diametrically opposed electrodes, the F C C arc column bears directly on the carbon anode which is completely covered by the arc crater at 150 amp or more. The pump valve is opened suffi ciently to meter the plasma down through the anode hole, to purge and TO PUMP

D.C +

D.C.

GAS IN

'

FCC

>

1 CARBON ANODE TUBE FURNACE

HEAT EXCHANGER

SAMPLE B O T T L E S

Figure 2.

lb

VACUUM GAUGE

Experimental arrangement


F

46

COAL

GASIFICATION

equilibrate the effluent hot zone i n the tube furnace, and then to meter samples into the gas sampling bottles in sequence at timed rates, measur ing the rise period of the vacuum-gauge pressure. The samples were analyzed by gas chromatography using helium carrier gas and an air-hydrogen flame in a Model 609 F & M Scientific Corp. flame ionization chromatograph with a Poropak Q column. A test gas mixture containing the aliphatic compounds C H , C H , C H , C H , C H , and C H was used to calibrate the analytical procedure. 4

3

6

4

2

2

2

4

2

6

8


Results The early objectives of the program were primarily exploratory, so the results reported here are essentially qualitative although within a given test series weight can be given to concentration ratios of com ponents of a test sample mixture for comparison purposes. Series I. Hydrogen Flow Rate Through the FCC. This was effected by operating the arc at standard conditions of 150 amp, maintaining the effluent hot zone at 800°C, sampling at about 30-60 sec per sample, and varying the hydrogen flow rate into the F C C . Comparison of hydro carbon composition is shown in Table I in terms of the relative distribu tion of the volume concentration of the products found. The distribution was obtained by calculating the per cent contribution which each chromatographic amplitude recording made to the sum of all, i n arbitrary scale divisions. There appears to be a significant dependence of efHuent hydrocarbon composition upon the amount of hydrogen fed into the F C C . Table I.

Relative Distribution in Effluent vs. H

2

Flow Through FCC

H , moles/min

CH*

CH

C #e

3.4 6 8.5 14.1

0 4.4 45.6 85

100 66.2 45.6 15

0 29.4 8.8 0

2

2

2

3

Series II. Time Factor. Standard conditions of 150 amp and 8.5 moles hydrogen/min were used with varying sampling rates through the 800°C effluent hot zone. Results are given i n Table II. N o other hydrocarbons were observed. These data suggest that methane and acetylene are produced and disappear at different rates. Series III. Hot Zone Temperature. Standard conditions included 150 amp, 8.5 moles hydrogen/min through the F C C , and sampling rate through the effluent hot zone at 2% min with variation of the hot zone temperature. Results are shown in Table III. It is evident that the hot zone temperature has a significant effect on the hydrocarbon composition of the effluent.


4.

SHEER

AND KORMAN

Table II.


Relative Distribution vs. Sampling Flow Time Through Hot Zone

Time 10 sec 12 20 30 2M 4 12


Table III.

47

C#4

C2H2

44 75.5 91.5 95.5 93 100 9 0

56 24.5 8.5 4.5 7 0 0 0

Relative Distribution vs. Hot Zone Temperature

Temp., °C

CH

CH

800 500 400 200 25

100 35 40 2 0.5

0 65 60 87.5 99.5

A

2

2

C # 3

6

0 0 0 10.5 0

Series IV. Hot Zone Surface Area. W e noted the result of increased time of flow in the sampling rate shown above in Series II and assumed that 8.5 moles hydrogen/min through the F C C creates a steady state for carbon and hydrogen i n the plasma at the arc crater (the sampling source), then the time of exposure to the hot zone wall of type 304 stainless steel was observed. This was accomplished at 150 amp, 8.5 moles hydrogen/min, and 800°C hot zone temperature, i n two diameters of hot zone tubes, 11/16 and 9/16 inch—a cross-sectional area ratio of 4:1. To equate the sample residence times, the sample flow periods were adjusted to this ratio. Results are shown in Table IV. N o acetylene was found in the 9/16-inch diameter samples while the small amounts in the 11/16-inch samples were consistent with the distribution for 20 and 30 sec checked with similar times observed i n the earlier Series II above. Table IV.

Hydrocarbon Ratio vs. Hot Zone Surface

Diam.

Sampling Rate, sec

CH4 Ratio

9/16:1-1/16

80:20 120:30

1:11 1:8

Series V. Further Effect of Hot Zone Area. The result of Series IV was followed by further observations comparing hydrocarbon yields and ratios in two cases and at two temperatures, as follows: A . 11/16-inch diameter at 800° and 500°C B. 11/16-inch diameter, into which tube a section of stainless steel wool was added,

fa&fflQg^

Society Library 1155 16th St. N. W. Massey; Coal Gasification Washington, C. 20036 Advances in Chemistry; American D. Chemical Society: Washington, DC, 1974.

48

COAL

Table V .

Hydrocarbon Distribution vs. Surface Area 800°C

c . Sampling lime, sec 30 150


7

g L

Without $

CH, 94 100

L

GASIFICATION

W

o

d

CH 6 0

S L

2

2

500°C With

g t

W

CH, 100 100

o

d

S

CH 0 0 2

2

t

Without $ L

CH, 18 35

W

o

$

d

CH 82 65

L

2

2

With

g L

W

o

CH 84 100

d

CH 16 0

A

2

2

These conditions were compared because any effect resulting from exposure to a large surface of stainless steel would not require comparably rapid sampling flow. In other words, the effect of increased surface area alone could be observed. The hydrocarbon distribution at each tempera ture with and without added stainless steel wool is shown in Table V . Table V indicates that the preponderance of methane and absence of acetylene is not affected at 800 °C. To interpret the apparent shift, how ever, at 500°C it is necessary to compare the relative concentrations of all samples. This is shown i n terms of their ratios in Table V I . It w i l l be noted that the effect of the stainless steel hot zone is relatively constant and appreciable for the times and temperatures of exposure. A t 800 °C, no acetylene is present, as expected for temperature as shown in Table III above while at 500°C the time-related suppression of acetylene previously observed in Table II above is also more strongly enhanced by the increased stainless steel surface area. W e interpret this to mean that acetylene disappears much more rapidly than methane under these conditions and that the disappearance is related to the surface area of the stainless steel hot zone. Table V I . . Sampling Time, sec 0

7

Ratio of Hydrocarbon Concentrations in Samples 800°C Ratio, +/— : . CH, CH

30 1/1.3 150 1/4 ° -f = with stainless steel wool; — = without stainless steel wool.

a

2

500°C Ratio, +/— . CH CH a

2

0/0 0/0

A

1/1.7 1/4.8

2

2

1/27 1/200

Series V I . Effect of Hot Zone Surface Composition. W e noted that the time-related suppression of methane and acetylene suggested a hot zone surface effect when stainless steel was used, so this mate rial was replaced by several others, using 11/16-inch diameter tubes. Arc crater gas samples taken under otherwise identical conditions (viz., 150 amp, 8.5 moles hydrogen/min, hot zone temperature 800°C, parallel sampling flow rates) produced hydrocarbon compositions as shown i n Figures 3-6.



4.

SHEER

A N D KORMAN


49

6 TIME

(MIN

Figure 3.

TIME

)

Silica

(MIN)

Figure 4. Iron


50

COAL

GASIFICATION


For fused silica (Figure 3 ) , the time of exposure i n the 800°C hot zone has roughly parallel effects on the presence and disappearance of both methane and acetylene. This is i n contrast with iron (Figure 4) and type 304 stainless steel (Figure 5) where acetylene disappears rapidly while methane persists. Increasing the nickel content by using Incoloy 800 (32% N i , 46% F e ) and nickel-200 ( 99.5% N i ) appears to produce further suppression of both acetylene and methane—an effect which does not appear to be especially time-sensitive.

T I M E (MIN

Figure 5.

)

Stainless steel type 304 (8% Ni, 74% Fe)

Some Results with Other F C C Gas Feeds. Preliminary tests were carried out i n which substitution was made for hydrogen as the F C C injected gas. The first substitute was a H - C O , 50-50 v o l % mixture. The usual standard test conditions were employed, except that the gas volume flow rate was set to a value which included a relatively small amount of hydrogen (0.6 m o l e / m i n ) . The results (Table V I I ) show a comparison of the (interpolated) analog distribution resulting from the same amount of pure hydrogen alone with the distribution using the mixture with C O . The absolute amount of acetylene i n the C O - H mix ture tests also increased with time whereas the acetylene i n the low-rate pure hydrogen analog diminished rapidly and no hydrocarbon was found after 10 sec. 2

2


4.

SHEER


AND KORMAN

51

20 | -

z o

CO

> 15 Q

i

CH (INC0L0Y), C H « 0

10

4

2

2

>-

Coal Gasification

Recommend Documents