年產(chǎn)6萬噸超高分子量PMMA項目設計

喜歡就充值下載吧。。資源目錄里展示的文件全都有，，請放心下載，，有疑問咨詢QQ:1064457796或者1304139763 ==================== 喜歡就充值下載吧。。資源目錄里展示的文件全都有，，請放心下載，，有疑問咨詢QQ:1064457796或者1304139763 ==================== 62 Ind.Eng.Chem.Prod.Res.Dev.1985,24,62-68 Kinetics of the Partial Oxidation of Isobutene over Silica-Supported Molybdenum-Uranium Oxide Catalyst Vlcente CortQs Corberbn,Avellno Corma,and Gojko Kremenld Institute de Catiilisis y Petroleoqdmica,C.S.I.C.,Serrano,119,28006 Madrid.Spain The kinetcs of the formation of the main primary products(Le.,methacrolein,acetone,CO,and COP)obtained during the oxidation of isobutene over a silica-supported molybdenum-uranium oxide catalyst have been studied over the temperature range of 320 to 380 OC in an isothermal flow reactor operated differentially.The data were fitted to a number of models based on Langmuir-Hinshelwood,Rideal-Eley,Mars-van Krevelen,and steady state of adsorption mechanisms.The preferred models for the formation of methacrolein and acetone are based on a Langmuir-Hinshelwood mechanism in which the rate-controlling step is the surface reaction between adsorbed isobutene and nondissociated oxygen.An enhanced oxygen adsorption has to be considered for acetone formation in order to explain experimental results.The best model for both CO and COP formations is consistent with a redox mechanism in which the catalyst reoxidation rate is second order in oxygen and catalyst reduction is first order in isobutene.Introduction The catalytic oxidation of isobutene over metal oxide catalysts has received considerably less attention in the literature than the closely related oxidation of propene.The available information is concerted mainly with the catalytic properties of determined catalytic systems and mechanistic and catalyst development studies(Hucknall,1974;Van der Wiele and van der Berg,1978;Cullis and Hucknall,1982).Comparatively little work on the reaction kinetics has been reported,and perhaps the reason is that the kinetics seems to be rather complex,as evidenced by the very different isobutene and oxygen dependencies reported and rate equations found applicable.Mann and KO(1973)oxidized isobutene over a cop-per-promoted bismuth molybdate catalyst in the range 300-560 C.The rate of formation of methacrolein was satisfactorily correlated by a Langmuir-Hinshelwood mechanism which assumes the rate-controlling step to be the surface reaction between adsorbed isobutene and ox-ygen species.Ray and Chanda(1976)studied the oxidation of iso-butene over bismuth molybdate in the range 350-500 O C and found that methacrolein and carbon oxides formation rates were independent of oxygen partial pressure and first order in isobutene when the 0 xygen:isobutene ratio was higher than 61.They found that methacrolein and carbon oxides were formed by parallel routes.Cartlidge et al.(1977)analyzed the kinetics of isobutene oxidation over bismuth molybdate by means of kinetic models.The model for methacrolein formation,found applicable only below 370 OC,contained no hydrocarbon term and the best model for carbon oxides formation contained no oxygen term and assumed that all of these carbon oxides were formed by degradation of methacrolein.Vinogradova et al.(1977)studied the oxidation of iso-butene in the range 320-400 C,over a molybdenum-co-balt oxide catalyst modified with Bi and Fe.They found that methacrolein,methacrylic acid,and carbon oxides are formed according to a parallel-consecutive scheme and that the total oxidation was first order in oxygen;meanwhile,selective oxidation is first order in oxygen for oxygen:iso-butene ratios below 3 and zero order for higher oxygen:isobutene ratios.Schuhl et al.(1980)oxidized isobutene over U Sb301,in the range 300-400 C and found that methacrolein formation is independent of oxygen and first order in isobutene while C02 formation is zero order in isobutene and first order in oxygen.Zhiznevskii et al.(1978)cor-0196-4321185111224-0062$01.50/0 Table I.Characteristics of the Mo03-U03/Si02 Catalyst particle size:0.42-0.59 mm BET surface area:78 m2 g-l pore volume:0.653 cm3 g-l mean pore radius:120 A crystalline phases:orthorhombic MOO,monoclinic U02Mo04 equivalent oxides area9 22.1 m2 g-a Determined by oxygen chemisorption.related the rate of formation of methacrolein,methacrylic and acetic acids,and carbon oxides in the oxidation of isobutene over a Te-V-Mo-0(Te:V:Mo=4:1.2:4.8)cat-alyst,with rather complex empirical expressions in which methacrolein formation was independent of hydrocarbon and dependent on oxygen and water,C02 was dependent on oxygen and isobutene and inhibited by water,and CO was dependent on oxygen and isobutene,having all the rate expressions fractional orders in each reactant.These observed oxygen and isobutene dependencies are not easily related to the reaction mechanism,since as we have just seen,different kinetics are observed even in catalysts in which the reaction mechanism is the same.For this reason,it seems convenient to analyze the rate data by means of kinetic models.In a previous paper(CortBs Corberin et al.,1984)we found that the reaction network of isobutene oxidation over silica-supported molybdenum-uranium oxide cata-lysts is complex,the main primary products being meth-acrolein,acetone,CO,and COz.In the present paper we have studied the kinetics of the formation of these prod-ucts and the data have been analyzed by means of kinetic models.Experimental Section The catalyst used was a Si0,-supported molybdenum uranium mixed oxide(20.8 wt%of active phase)with an atomic ratio Mo:U=8:l.It was prepared by a double impregnation method,which has been described elsewhere(KremeniE et al.,1982),and the main characteristics are indicated in Table I.The kinetic experiments were carried out in an iso-thermal stainless steel(i.d.13 mm)flow reactor,with a coaxial thermocouple for measuring the temperature inside the catalytic bed.Reactants and products were analyzed on line by gas chromatography.The apparatus and procedure were described in detail in a previous paper 0 1985 American Chemical Society Ind.Eng.Chem.Prod.Res.Dev.,Vol.24,No.1,1985 63 Table 11.Initial Rates of Formation of Primary Products r;X lo4(mol of i formed/h e of cat.),-temp,C Ph,atm 3,atm CH(CH3)CHO CH3COCH3 coz co 380 380 380 380 380 380 380 380 380 350 350 350 350 350 350 350 320 320 320 320 320 320 0.0245 0.098 0.196 0.294 0.490 0.196 0.196 0.196 0.196 0.0245 0.098 0.196 0.294 0.196 0.196 0.196 0.0245 0.098 0.196 0.294 0.196 0.196 0.294 0.294 0.294 0.294 0.294 0.049 0.098 0.196 0.392 0.294 0.294 0.294 0.294 0.098 0.196 0.392 0.294 0.294 0.294 0.294 0.196 0.392(Corti%Corberln et al.,1984).All experiments were carried out at conditions in which diffusion was not a significant effect.Contact times were varied by changing the catalyst weight and/or the molar flow of isobutene.Catalyst samples(0.05-0.5 g)were di-luted with carborundum bits in order to achieve a constant volume of the catalytic bed.Experiments were rejected if the errors in mass,carbon,or oxygen balances were greater than 5%or temperature differences along the catalytic bed were more than 3 OC.Results Kinetic Study.The kinetics for the formation of the primary products has been studied in a differential reactor by the initial rate technique.We have examined the in-fluence of the principal operation variables,Le.,contact time,temperature,and partial pressure of reactants.To study the influence of the composition of the reacting mixture we have used the usual technique of varying the concentration of one reactant while keeping constant the concentrations of the rest.All kinetic measurements were made within the following ranges:partial pressure,iso-butene 0.025-0.50 atm and oxygen 0.05-0.40 atm,tem-perature,320-380 OC,and atmospheric pressure.Water partial pressure was kept constant at 0.20 atm.Proper weights of catalyst samples and isobutene molar flows were chosen in order to maintain operation in a differential regime,i.e.,xT I 10%.In these experimental conditions the influence of the reaction products can be neglected.The primary reaction products are methacrolein,ace-tone,biacetyl,methallyl alcohol,CO,and COP(Corti%Corberh et al.,1984).At initial conditions we found that yields of biacetyl and methallyl alcohol were very low,inferior,or very close to the detection limit of our analytical method,and their kinetics have not been studied.We have centered our study in the four main products,i.e.,meth-acrolein,acetone,CO,and COP,which represent at least 95%of the total products of the reaction.Initial formation rates of each primary product have been calculated by linear regression of the typey=ax,of the yields(y)as a function of the contact time(x).The results are shown in Table 11.In all cases,the initial rate increases when the partial pressure of isobutene or oxygen increases.32.5 67.0 70.5 91.7 96.0 20.8 32.9 60.0 81.1 10.3 13.2 13.9 14.6 11.7 16.2 7.17 4.53 5.26 5.44 5.55 4.50 5.97 9.44 37.5 46.0 62.3 101 4.27 8.69 22.5 65.1 3.97 8.00 9.61 2.81 6.09 2.19 3.40 3.88 4.28 2.41 6.16 12.0 15.5 115 319 608 708 752 176 389 568 512 49.2 174 353 448 144 278 359 19.3 65.2 142 159 116 150 28.4 89.6 156 262 310 130 164 167 56.8 12.8 48.8 99.2 37.7 80.0 117 110 5.24 19.5 41.6 44.4 32.2 46.0 Kinetic Analysis.The kinetic analysis of the experi-mental data w a s made by kinetic models based on different possible mechanisms.Only models in which the reaction rate is dependent of both isobutene and oxygen partial pressures have been considered.Four different mechanisms have been proposed in the literature for the catalytic oxidation of olefins:those of Langmuir-Hinshelwood,Rideal-Eley(Mann and KO,1973),Mars-van Krevelen,or redox(Krenzke and Keulks,1980),and steady state of adsorption(Muller et al.,1976).According to the literature,the rate-controlling step in the catalytic oxidation of olefins can be the adsorption of reactants or the surface reaction between the adsorbed species.In the study of the oxygen adsorption on our catalyst,it has been found that,in the temperature interval 320-380 C,the rate of oxygen chemisorption is very high in such a way that the process can be considered instan-taneous(Salazar,1983)and consequently the oxygen ad-sorption can be neglected as a possible rate-controlling step.Assuming this,the models based on the Lang-muir-Hinshelwood mechanism(Hinshelwood,1940)that have been tested are the following.Models 1 to 3 suppose competitive adsorption of oxygen and isobutene on one type of center.The rate-controlling step is the adsorption of isobutene in models 1A and 1B and the surface reaction in models 2A,2B,and 3.Besides,models lB,2B,and 3 consider that oxygen adsorption is dissociative,and model 3 considers the possibility of the two dissociated oxygen species to be involved in the rate-controlling step.Models 4A and 4B assume that adsorption of reactants is not competitive on two types of centers,and surface reaction is the rate-controlling step being the oxygen ad-sorption nondissociative in model 4A and dissociative in model 1B.Models based on the Rideal-Eley mechanism(Bond,1962)suppose alternatively that the adsorbed reactant is isobutene(model 5)or oxygen(model 6).As the real kinetics of the reduction of our catalyst by isobutene,and its reoxidation by oxygen,are not known,the models based on the Mars-van Krevelen mechanism(Mars and van Krevelen,1954)that we have tested suppose that the catalyst reduction in first order in olefin and that catalyst reoxidation can be first(model 7A)or one-half order(model 7B)in oxygen.04 Ind.Eng.Chem.Prod.Res.Dev.,Vol.24,No.1,1985 Table 111.Rate Equations Derived from the Tested Models model mechanismn rate-controlling step rate eauation lA,lBb 2A,2Bb 3 4A,4Bh 5 L.H.(1 type of center)L.H.(1 type of center)L.H.(1 type of center)L.H.(2 types of center)R.E.6 R.E.7A,7B M.v.K.8A,8Bb S.S.A.9 adsorption of isobutene surface reaction surface reaction involving two dissociated 0 atoms surface reaction surface reaction between adsorbed isobutene and surface reaction between adsorbed oxygen and equilibrium between catalyst reduction and its equilibrium between catalyst reduction and oxygen power law rate expression oxygen in gas phase isobutene in gas phase reoxidation adsorption L.H.=Langmuir-Hinshelwood;R.E.=Rideal-Eley;M.v.K.=Mars-van Krevelen;S.S.A.=steady state of adsorption.bModels A assume nondissociative adsorption of oxygen(n=1);models B,dissociative adsorption(n=i2).Catalyst reoxidation is first order in model 7A(n=1)and 1/2 order in model 7B(n=I/*).Table IV.Influence of Isobutene Partial Pressure methacrolein formation acetone formation C.C.4 F 380 lA,lB,6 0.0254 0.752 1.399 1 0.0215 0.964 0.297 53 2A,2B 9.13 2.92 0.976 0.279 61 4.52 5.12 0.877 0.619 10 3 3.60 2.07 0.966 0.333 42 1.66 2.97 0.875 0.625 10 4A,4B,5,7A,7B,8A,8B 5.12 99.4 0.993 0.148 227 24.3 0.997 0.094 566 350 lA,lB,6 0.0063 0.284 4.233 3 0.0047 0.575 0.945 1 2A,2B 34.1 4.67 0.987 0.229 74 29.0 7.77 0.976 0.305 41 3 10.8 2.88 0.978 0.297 43 8.53 3.96 0.970 0.341 32 4A,4B,5,7A,7B,8A,8B 7.31 674 0.997 0.118 288 43.1 769 0.997 0.114 305 320 A,lB,6 0.0025 0.156 7.510 3 0.0018 0.747 1.930 2 2A,2B 57.1 7.00 0.988 0.223 79 57.2 10.3 0.985 0.245 65 3 15.6 3.79 0.978 0.296 44 146 4.82 0.977 0.302 42 4A,4B,5,7A,7B,8A,8R 10.5 1780 0.998 0.089 500 57.1 2260 0.996 0.128 243 9 0.08-7.38 0.983 0.261 57 0.27-7.40 0.994 0.151 173 temp,C type of model Ahn ha C.C.4 F Aha Bhn 9 0.36-4.31 0.973 0.296 54 0.76-4.07 0.987 0.205 117 9 0.14-6.34 0.986 0.234 71 0.44-6.19 0.994 0.152 170 C02 formation _ _ _.-CO formation temp,C type of model Ah Bha C.C.J.F Ah Bhn C.C.*F 380 lA,lB,6 0.0717 0.953 2A,2B 1.89 3.00 0.907 3 0.833 2.09 0.905 4A,4B,5,7A,7B,8A,8B 8.22 18.4 0.999 350 lA,IB,6 0.436 0.973 2A,2B 2.11 4.26 0.844 3 0.839 2.63 0.845 4A,4B,5,7A,7B,8A,8B 18.8 13.3 0.999 9 0.92-3.22 0.996 9 0.83-2.78 0.995 320 lA,lB,6 0.017 0.947 2A,2B 4.13 6.60 0.789 3 1.40 3.52 0.788 4A,4B,5,7A,TB,8A,8B 45.7 41.2 0.999 9 0.90-4.17 0.992 a Constant whose units depend on the kinetic equation involved.Models based on the steady-state adsorption mechanism(Sheltad et al.,1960;Juusola et al.,1970),which can be considered a s a variation of the redox mechanism,consider the cases where oxygen adsorption is dissociative(model 8B)or not(model 8A).The power law rate,which is widely used for reactor design purposes,does not give information about the mechanism of the reaction,but we have included it in our testing as model 9 for comparison purposes.The rate expression derived from these models are summarized in Table 111.The experimental data hwe been fitted to the above rate expressions conveniently linearized as a function of iso-butene(Ph)or oxygen(po)partial pressures.Simultane-0.340 0.543 0.549 0.065 0.125 0.264 0.758 0.756 0.029 0.125 0.372 0.870 0.871 0.046 0.179 39 14 14 1188 318 54 5 5 4698 259 26 3 3 1927 123 0.195 2.22 1.17 1.93 0.67 0.162 1.07 0.532 4.85 0.91 0.060 2.25 0.952 12.3 0.89 1.43 1.28 8.70-1.91 2.21 1.70 5.70-1.94 3.52 2.32-2.95 18.6 0.749 0.985 0.985 0.998 0.981 0.987 0.919 0.920 0.999 0.999 0.962 0.832 0.832 0.999 0.994 0.741 0.222 0.221 0.080 0.248 0.182 0.557 0.556 0.030 0.077 0.315 0.784 0.784 0.056 0.149 5 99 100 776 78 118 11 11 4396 678 37 5 5 1285 177 ously,several statistical tests were used to estimate the goodness of the fit and to discriminate between models.The tests used were:the correlation coefficients(c.c.),the F of Fisher(Fisher,1955),and the$of Exner(1966).Influence of the Isobutene Partial Pressure.The influence of ph was studied by fitting the experimental rates at constant po and pw into the linearized rate ex-pressions as a function of p,.The results indicate that for the formation of each one of the primary products,only the models whose linearlized rate expression is of the type(1)(i.e.,models 4A,4B,5,7A,7B,8A,and 8B)produce a statistically significant fit(Table IV).l/r=Ah(l/Ph)+Bh Ind.Eng.Chem.Prod.Res.Dev.,Vol.24,No.1,1985 65 Table V.Influence of Oxygen Partial Pressure(I)methacrolein formation acetone formation temp,C type of model Ao B O a C.C.$F 4 0 B O C.C.+F 380 4A,7A,8A 20.3 73.9 0.996 0.114 384 124 174 0.999 0.050 2034 4B,7B,8B 125-93.6 0.995 0.128 302 765-1190 0.994 0.141 250 5 0.024 0.898 0.492 17 0.015 0.975 0.251 76 9 0.67-4.12 0.994 0.143 244 1.34-3.82 0.996 0.114 381 350 4A,7A,8A 101 361 0.999 0.062 1035 377-280 0.999 0.034 3507 4B,7B,8B 498-192 0.993 0.164 147 1830-2360 0.999 0.066 338 5 0.0047 0.776 0.728 5 0.0036 0.975 0.255 58 9 0.59-5.86 0.994 0.156 162 1.20-5.41 0.997 0.115 298 320 4A,7A,8A 216 1120 0.999 0.067 665 1000-985 0.998 0.097 316 4B,7B,8B 337 320 0.996 09145 142 3880-4660 0.995 0.175 97 5 0.0017 0.904 1.651 1 0.0015 0.973 0.282 36 9 0.41-7.03 0.995 0.168 105 1.36-6.11 0.999 0.053 1052 Constant whose units depend on the kinetic equation involved.Table VI.Influence of Oxygen Partial Pressure(11)CO formation C02 formation temp,C type of model Aoa B O C.C.+F A O BOb C.C.#F 380 4A,7A,8A 6.48 33.5 0.958 0.371 33 2.22 8.71 0.968 0.324 44 4B,7B,8B 38.3-15.2 0.915 0.521 15 13.2-8.13 0.929 0.479 19 5 0.055 0.670 1.346 1 0.739 0.390 1.030 1 9 0.48 03.50 0.886 0.600 11 1.02-0.86 0.650 0.981 2 10 0.285 5.54 0.994 0.135 269 0.097 16.3 0.996 0.119 352 350 4A,7A,8A 23.4 20.9 0.991 0.194 104 5.69 10.1 0.990 0.200 98 4B,7B,8B 112-104 0.975 0.313 39 27.4-20.4 0.974 0.319 37 5 0.032 0.896 0.514 12 0.109 0.791 0.707 5 9 0.79-3.68 0.976 0.306 41 0.69-2.57 0.968 0.354 30 10 1.78 79.4 0,999 0.012 30137 0.433 24.3 0.999 0.053 1405 320 4A,7A,8A 33.3 130 0.998 0.112 238 7.89 45.4 0.989 0.261 43 4B,7B,8B 129 8.34 0.994 0.190 82 30.4 16.6 0.981 0.338 25 5 0.0132 0.342 1.294 1 0.044 0.830 1.917 1 9 0.48-4.92 0.993 0.207 69 0.38-3.82 0.975 0.381 20 10 4.27 190 0.999 0.028 3973 1.02 59.4 0.998 0.122 201 Constant whose units depend on the kinetic equation involved.Table VII.Methacrolein Formation temp,C Kh,atm-KO,atm-k X lo4 k,X k,(=kadJ x 380 19.4(6.5)3.6(1.9)194(33)493(66)194(80).,350 92(19)3.6(0.8)320 170(24)5.2(1.1)In mol of CjH,O/(h g of cat.).mol of C4H60/(atm h g of cat.).Influence of the Oxygen Partial Pressure.The influence of po was studied by fitting the experimental rates at constant P h and pw to the different rate expres-sions.The results are shown in Tables V and VI.From Table V it can be deduced that the most probable models for mechacrolein formation are models 4A,7A,and 8A.The same models produce the best fit for acetone,but in this case,negative values are found for one constant.None of the previously detailed models,even the power law rate model,produces an acceptable fit of the experi-mental data for the formation of CO and COz(Table VI).For this reason,and looking to the curves of rate vs.oxygen partial pressure,we have tried a semiempirical model,10,with a linear form of the rate expression as a function of oxygen partial pressure(2)and whose linear form as function of the isobutene partial pressure is that corresponding to models 4A,7A,and 8A,i.e.,eq 1.The fit of initial rate data for formation of Cc)and of COz to this new model is significant at the three temper-l/r=Ao(l/Po2)+Bo 137 i22j 99.i(8.5)28.9(4.3)95(12)46.3(5.8)9.34(0.94)atures,and therefore this will be the preferred one.Theoretical justification of this model will be seen in the Discussion.Discussion(a)Methacrolein Formation.The most probable models for methacrolein formation are models 4A,7A,and 8A,whose linearized expressions are equivalent although they are based on three different mechanisms.The cal-culated values of the constants corresponding to these models are shown in Table VII,with their 95%confidence intervals(in parentheses).The three models are acceptable from the chemical point of view,because the variations of the constants with temperature agree with the theoretical forecasts.In order to discriminate among them,the rate values calculated according to each one have been plotted vs.experimental values in Figure 1.Values calculated with model 4A fit quite well to the line of slope unity with a random distribution around it;meanwhile,those predicted by models 7A and 8A are considerably biased above that line.Then,the most probable model is model 4A,which as-sumes a Langmuir-Hinshelwood mec