The product distribution and kinetics of the thermal cracking of propane and propane‐propylene mixtures were investigated for a temperature range of 700° to 850°C, exit total pressures from 1.2 to 2 atm. absolute, and steam dilutions of 0.4 and 1.0. The reaction kinetics were determined by two methods. The first uses the equivalent reactor volume concept to reduce the data to isothermicity before attempting the kinetic analysis. The second determines the parameters directly from the nonisothermal data by means of a search routine. Both methods led to a reaction order of 1. The effect of propylene addition was also investigated. Finally, a detailed molecular reaction scheme was derived from a nonisothermal and nonisobaric simulation of the cracking experiments. This scheme was used in the simulation of an industrial cracker and led to excellent agreement, in particular of the product distribution.

A novel experimental setup allowed both the kinetics of the thermal cracking of ethane and of the coke formation to be studied over a temperature range extending from 750 to 870°C. The overall kinetics of ethane disappearance are in excellent agreement with previously reported pilot results. Coking rates are initially high but rapidly decrease to reach an asymptotic value. The initial product distribution also differs from the asymptotic product distribution. A kinetic model for the coking is derived from the experiments and used in conjunction with a set of conservation equations for the simulation of industrial ethane cracking. The predicted run length, thickness of the coke layer and evolution of the tubeskin temperatures are in agreement with industrial Observations. SCOPEThermal cracking of hydrocarbons around 800°C is the most important process for the production of feedstocks for the petrochemical industry. Side reactions always lead to carcycle has to be interrupted tor decoking by burning off the coke. Little is known about the rate of formation of coke. Yet, this is required for a better planning of the production and for the coking during ethane cracking in a novel experimental set up and makes use of this information to industrial operation and predict the evolution of the process parameters and the coke layer with time.bonaceous material that deposits on the wall ofthe cracking coil more optimal operation. This Paper Presents a kinetic Study of thus reducing heat transfer, requiring critical tubeskin temPeratures, increasing the Pressure drop so that after 20-60 days, depending upon the severity of operation, the production CONCLUSIONS AND SIGNIFICANCESpecially designed equipment consisting of a completely mixed reactor in which a hollow cylinder is suspended at the arm of an electrobalance allows the kinetics of the main reaction(s) and of the coking to be determined. The reactor, with a volume of some 5 mL yields kinetic parameters for the main reactions in agreement with those obtained in a pilot plant. Accurate rates of coking a r e obtained which are used in the simulation of an existing industrial cracking unit. In this way, it becomes possible to predict the evolution of conversion, product distribution, pressures, and coke layer in the cracking coil. The predicted results are in line with industrial observations.

Science, Columbia University, 1979. Literature CitedAggenbach, P. De Rovij, A. H. Neth. Appl. 7009685, 1972. APHA, "Standard Methods for the Examination of Water and Waste Water," 13th ed.; 1975. pp 237 for a, 211 for b, 96 for c. ; Regnier, J.; Miquel, P.: Talllard, D. Report CEACONF-1903CEACONF- (1974). Patigny, P.; Regnbr, J.; Miquel, P.; Taillard, D. Roc. Inst. Solvent Extr. Conf. 1974, 3, 2019. Ralney, R. H. Nucl. Appl. 1965, 7(4), 310. Richardson, G. L.; Swanson. J. L. Report HEW-TME-7531 (1975). Schlee. C. S.; Caverly, M. R.; Henry, H. E.; Jenklns, W. J. USAEC Report Schulre, R. German Offen. 692412, 1940. Swanson, J. L.The paper presents experimental data on naphtha cracking in a pilot plant and discusses three ways of approach for scaling up naphtha-cracking coils. The first, direct experimental simulation, is not sufficiently accurate if the conversion is to be predicted, since complete similarity between industrial-, pilot-, and benchgcale units is bnpossible. The second approach is based upon the equivalent space time VE/Fo. It is shown that, for a given dilution, there is a unique relation between VE/Fo and the naphtha conversion. The corresponding product distribution is obtained from graphs or correlations derived from essentially isobaric small-scale experimentation. Since the product yiekls depend upon partial and total pressure and since in commercial cracking these vary along the coil, the selection of an average value for the total and partial pressure inevitably leads to some error In the predicted product distribution. The third approach makes use of a detailed mathematical model to simulate the cracking operation. The model contains a number of rate equations which can be derived from bench and pilot-scale experimentation. The model generates the conversion, product yiekls, temperatwe, and pressure profiles along the reactor. Exce#ent agreement between simulated and measured values is arrived at for the cracking of a heavy naphtha in an Industrial reactor.The conversion of dibenzothiophene to the corresponding sulfoxide and sulfone by the action of chlorine and water was studied. The effects of temperature, reaction time, and solvent on the yield of these products and on the concurrent ring chlorination of dibenzothiophene were examined. Slurry-phase chlorination of dibenzothiophene in water at 70 O C was found to be an excellent method for the preparation of the sulfone.

In thermal cracking, use is made of severity factors to characterize the process conditions and to relate these in a concise way to the product slate. A fundamental, generalized severity factor, kV,IF,, based upon the equivalent reactor volume concept is derived from the model equations representing the cracking coil operation.It relates a modified exit conversion in a unique way to operating conditions (temperature and pressure profile, steam dilution ratio), no matter what the feedstock may be. In commercial units temperature and pressure profiles are in general insufficiently defined, so that only easily accessible exk yields can be used as a severity factor. From an inspection of a large data set on naphtha, kerosene, and gas oil cracking, it follows that the C3to propylene yield ratio is the best measure of the severity of operation.Evaporation and Wling heat-transfer coefficients are presented for thin water HLms flowing over the outsii of horizontal, electrically heated brass tubes. Tests were conducted with a 5.08-cm-dlameter smooth tube, a 5.08-cmdlameter clrcumferentially grooved tube, and a 5.08-cm-diameter axially grooved tube. Both local and average heat-transfer data were obtelned for nonboillng and bolilng conditions correspondi to feed-water temfrom 1.16 to 3.79 cm3/s per centimeter length of tube. Correlations of the average heat-transfer coefficients for nonboiling and boiling conditions were developed and compared. The results indicate that both enhanced tubes provlded hlgher heat-transfer coefficients than the smooth tube.peratures ranglng from 49 to 127 O C end heat-flux values ranging from 30 000 to 80 000 W/m 9 . Flow rates ranged

The product distribution and the kinetics of the thermal cracking of kerosene were investigated in a pilot plant under conditions of residence time, temperature, total pressure, and dilution close to those used in industrial operation. The influence of feed composition, total pressure, and inlet partial pressure on the product distribution were determined. Kinetics of the cracking of individual components in the kerosene mixture were calculated.