BE537452A - - Google Patents

Description

       

   <Desc / Clms Page number 1>
 



   The invention relates to the conversion of glycerol epihalohydrienes, i.e. 1-halo-2,3-epoxypropanes, into glycerol.



   A number of different methods have already been proposed for the preparation of glycerol. A suitable method is the hydrolysis of mono- and
 EMI1.1
 dichlorohydrins, in which aqueous α, 3-hydroxide solutions, preferably buffered with a salt of a strong base and a weak acid, for example sodium carbonate, were used as hydrolysis agent. In the known methods there was a serious disadvantage, namely the formation of by-products, especially high molecular weight products, such as polyglycerols, which also complicate the preparation of pure glycerol.



  A further drawback was a relatively slow reaction rate, which required the use of large reactors. Another drawback was the use of very dilute. solutions requiring evaporation of large amounts of water.



     According to the invention, the above drawbacks can be overcome.
 EMI1.2
 When the hyclrolyte of 1-halo-2,3-epoxypropane is carried out by contacting these compounds for a short time at high temperature and under pressure with an aqueous inorganic carbonate solution: 4 g. The reaction is carried out at a temperature between 130 ° C and 200 ° C, preferably between 150 ° C and 180 ° C, and in the presence of carbon dioxide, under such a pressure, for example under a pressure of 7 to 35 atm.,
 EMI1.3
 that the reaction components for 55td: lr 2 (} \ tidJ:

  m. The liquid phase. The optimum temperature Q. 1 ½ time varies slightly depending on the glycerol concentration of the fully hydrolysed product ultimately obtained, as shown by the following numbers:
 EMI1.4
 Glycerol mononentation (wt%) 5 10-15 20-25 Temperature, 00 154-166 157-180 16 '/ 80 YerbliµQ time, minutes 9-15 7-9 8-10 Temperatures and residence times that are above these optimal limits (promote the formation of polymeric products, while at temperatures and residence times below, incomplete hydrolysis takes place
 EMI1.5
 The temperature and residence timeline are interdependent variables.

   Usually the residence time is determined by the volume of the reactor and the required product cap rate. An optimal yield is then obtained by controlling the reaction temperature.



   In this way, one can obtain substantially complete conversion of the 1-halo-2,3-epozypropane with the formation of very few by-products. The reaction can be advantageously carried out with halo-epoxypropane
 EMI1.6
 Solutions containing, for example, about 15 to 30% by weight of 1-chloro-2,3-epoxypropane, at which concentrations little polymeric products are formed. In these concentrates, correspondingly concentrated glycerol solutions are obtained, reducing the amount of water when the product has to be evaporated, it is greatly reduced.



   An important advantage of the present process over the methods used hitherto is that not only an improved yield of glycerol is obtained, but that the alkali consumption can also be considerably reduced. Preferably, the method is carried out with such a
 EMI1.7
 any amount of inorganic arbona + ± that an excess of about 2% -20%, preferably about 15% - 20%, of the required stoichiometric amount remains at the end of the reaction. Smaller amounts of carbonate may be However, less complete hydrolysis is usually obtained,

  especially when an excess of less than 10% is used When using larger amounts of inorganic carbon
 EMI1.8
 After all, the yield of glycerol does not increase and the removal becomes more difficult. Furthermore, it is advantageously high concentrations of 1-halo-2,3-epoxy resin.

 <Desc / Clms Page number 2>

 
 EMI2.1
 propane. Thus, it has been found that when the glycerol concentration of the instant solution is increased from 15 to 25% by weight and the optimum excess of sodium carbonate in the product, of 15-20%, is maintained, a reduction is achieved. From the consumption of sodium carbonate of 79 kg per 100 kg glycerol to 71 kg per 100 kg glycerol is obtained when the other conditions are kept constant.
 EMI2.2
 



  Carbonates, such as sodium carbonate, are considered to be substances which are equivalent to other basic substances, such as sodium hydrode, etc. in the hydrolysis of hydroxyhalides.
 EMI2.3
 halo-2,3-epoxypropanes, however, this is not the case, since the carbonates or bicarbonates to be used according to the invention, hereinafter referred to as "inorganic carbonates", cannot be replaced by other bases without significant advantages of the invention. The carbon dioxide formed during the reaction can act as a catalyst and a buffer, avoiding too high a pH so that excessive polymer formation does not take place. The use of small amounts of other basic
 EMI2.4
 fabrics,

   so that the reaction is not adversely affected, can be used, if desired, in conjunction with the inorganic carbonates. However, using, for example, 10-15 wt.% sodium hydroxide in the total alkali feed, undesirably greater polymer formation occurs. can be advantageously prepared by allowing the carbon dioxide liberated in the reaction to be absorbed by a sodium hydroxide solution, for which a small excess of sodium hydroxide is desirable.



   The preferred method used in the inventive method
 EMI2.5
 inorganic carbonates are alkali metal carbonates and bdarbonate, especially sodium carbonate and bicahbonate, as they are cheap and dissolve well in the reaction medium.



   The method has been found to be particularly useful in
 EMI2.6
 the hydrolysis of 1-chlor-2,3-epoxypropane to glycerol. However, it can also be used with good results for the hydrolysis of librgom-2,3-epoxypropane or other 1-halo-2,3-epoypropane. The starting 1-halo-2,3-epoxypropane will be formed in the reaction mixture from the corresponding dihydroxypropanes. It is necessary only to use the amount of sodium carbonate used for each additional equivalent halogen of half a mole present. Suitable starting materials in the present process are mixtures of dichlorohydrin and epichlorine
 EMI2.7
 hydrines, for example those which are "in bob" lmenikaaiée, oétrécr = 1 script na = 6052g3a The method according to the invention can be either discontinuous:

  can be performed continuously. The preferred continuous process can use a reactor optionally with Raschig rings
 EMI2.8
 or other suitable inert fill material can be flattened. The reactor can be a vessel of the horizontal or vertical type.

   In the latter case, the halo-epoxypropane or the dihalohydroxypropane can pass with the aqueous sodium carbonate solution upward or downward through the reactor.
 EMI2.9
 19 It is preferable to use a tubular, non-insulating reactor, which makes it possible to properly control the reaction temperature, such as a gradual increase of the temperature as the temperature progresses.
 EMI2.10
 gradual reaction. When using such reactors, the carbonate is used very effectively,

   which is apparently due to the fact that less "backmixing" occurs, which prevents side reactions between the sodium carbonate ex-impurities in the starting material. Rather, the formation of undesired high-boiling by-products is counteracted by reducing "bac3mixing". "of glycerol with incompletely hydrolysed reaction mixture. A combination of a tubular reactor and a tank reactor can be used mpk, where the first part of the

 <Desc / Clms Page number 3>

 
 EMI3.1
 reaction (that is, that portion which proceeds, for example, in the first 1/5 - 1/3 of the total reactor volume, in which part the hydrolysis mainly takes place) is carried out in heated reaction tubes. schematically shows one suitable reaction system of this type.



   In the drawing 1 is a storage tank for 1-halo-2,3-epoxypropane or dihalohydroxypropane or a mixture thereof. This feedstock is collected via line 2, together with the aqueous sodium chloride supplied through line 3.
 EMI3.2
 carbonate solution, passed through pump 4 to the preheater 5. The mixture is heated in this preheater by indirect heat transfer with the aid of steam supplied through line 6, the condensate of which is removed through line 7. If desired, other heating means can also be used. The reaction mixture discharged via line 8 can be further heated by directly injecting high-pressure steam via line 9. The reaction mixture can be brought to the above desired temperature of 130 ° C to 200 ° C by this direct injection of steam.
 EMI3.3
 ,

  The thus preheated mixture, which is under a pressure of 7 to 35 kg / cm2, keeping the system in the liquid phase, is fed to reactor 10, in which so many tubes 11 are arranged that approximately 1/2. 4 of the total required reactor volume has been obtained. The product flowing out of the reactor 10, which is still at the reaction temperature and pressure, passes through line 12 to the tank-reactor 13, where the hydrolysis is completed. the reactors 10 and 13, which are preferably provided with heat-insulating material to prevent heat loss and which can optionally be provided with additional heating
 EMI3.4
 The medium is between 5 and 15 minutes. The reacted mixture 1 is fed via the rescue 14 provided with a reduction valve 15 to the receiver 16,

  wherein the carbon dioxide is separated from the aqueous glycerol solution discharged via line 18. The carbon dioxide is removed via line 17. The excess sodium carbonate in the glycerol solution is neutralized in the neutralizer with acid supplied via line 19.
 EMI3.5
 from where the solution is fed through line 21 to the evaporation and distillation stills 22 and 23. The salt formed is removed by the usual way and discharged through line 24, while the water and low-boiling impurities are passed through line 25 and 23. the high-boiling impurities are removed via line 26. The glycerol is obtained from line 27.
 EMI3.6
 



  The carbon dioxide discharged from the condenser 16 via line 17 is passed, together with the carbon dioxide discharged from the neutralizer 20 via line 28, to the absorber 29. The carbon dioxide is fed into the absorber in countercurrent to the introduced via line 30. sodium hydroxide solution. The sodium carbonate solution obtained is recycled to the reaction via line 31 and line 3, along with a make-up stream of sodium carbonate, which is returned via line 32, in an amount sufficient to bathe one equivalent amount.
 EMI3.7
 rasqaqLent amount of reactive halogen of the starting material to be delivered,

   preferably in an excess of 5% - 20% with respect to the stoichiometric amount required for the hydrolysis.



   When one proceeds in this way, the required amount is
 EMI3.8
 base, d, w, z, the sodium hydroxide supplied via line 30 and the sodium carbonate supplied via line 32, 20-35% lower than when the hydrolysis is carried out using a buffered sodium hydromide solution. acid and the amount of the salt to be removed (tlycerol) become proportionately lower. Also the yield of glycerol is considerably higher, in particular because the
 EMI3.9
 undesired high-boiling products in a much smaller amount

 <Desc / Clms Page number 4>

 formed, namely in an amount that is only 1/10 to 1/4 of the amount previously formed.



   The invention will be further illustrated by the examples below.
 EMI4.1
 



  EXAMPLE 11: LD I 1-chloro-2,3-epoxyW: Ropane, which was analyzed to contain only about 75% by weight of 1-chloro-2,3-epoxypropane, was fed into a tube reactor consisting of a,? mril .ribxis ^ zanr1-cfsr cut, which was curved such that a vertical structure consisting of
 EMI4.2
 of four horizontal layers connected in series. The 1-c. 2,3-propylene oxide was mixed in a manifold with an aqueous sodium carbonate solution, which was placed on the suction side of a feed pump and the resulting mixture was pressurized to 10.5 atm
 EMI4.3
 The heat required for the reaction was obtained by approximately 1.20 m downstream from the feed stream entry point, steam at 12.5 atm pressure.

   The product exiting the reactor was brought to atmospheric pressure and the aqueous glycerol was separated from the carbon dioxide.
 EMI4.4
 was absorbed in a sodium hydroxide solution, thereby forming sodium carbonate for further hydrolysis, and the aqueous glycerol solution was neutralized, evaporated and distilled, yielding pure glycerol. Below are some of the results obtained in tests that were carried out. performed at various 1-chloro-2,
 EMI4.5
 3-epoxypropane concentrates (expressed as equivalent glycerolone concentrations) in the reaction mixture and under optimal reaction conditions for these concentrations.

   
 EMI4.6
 
<tb> Equivalent
<tb>
 
 EMI4.7
 (Hyouronic interest -2 15 25%
 EMI4.8
 
<tb> Temprature, <SEP> c <SEP> 154-166 <SEP> 157-180 <SEP> 163-180
<tb>
<tb> Stay <SEP> time, <SEP> minutes <SEP> 8-15 <SEP> 7-9 <SEP> 8-10
<tb>
 
 EMI4.9
 Excess Na2CO in product, wt.% 15-20 15-20 15-20 Glycerol concentration in product, wt.% 4-6 13-18 20-25 Alkali consumption, kg Na2C0 -feed / 81-84 76- 79 64-72
 EMI4.10
 
<tb> 100 <SEP> kg <SEP> obtained <SEP> glycerol
<tb>
<tb>
<tb>
<tb>
<tb> Ratio <SEP> NaCl / <SEP> glycerol <SEP> in <SEP> product <SEP> 1.10 <SEP> 0.71 <SEP> 0.67
<tb>
 
 EMI4.11
 Ratio H 0 / glycerol in product 18.0 5.0 2.5
 EMI4.12
 
<tb> Yield <SEP>% 2 <SEP> on <SEP> basis <SEP> from <SEP> the <SEP> from <SEP> 50
<tb>
<tb> raw material <SEP> of <SEP> the <SEP> glycerol:

  
<tb>
<tb> Glycerol <SEP> 98. <SEP> 5 <SEP> 98.5 <SEP> 97.1
<tb>
<tb> Polymer <SEP> 1.3 <SEP> 1.0 <SEP> 2. <SEP> 6
<tb>
<tb> Unhydrolyzed <SEP> 0. <SEP> 2 <SEP> 0. <SEP> 5 <SEP> 0. <SEP> 3
<tb>
 The temperatures quoted in this table are the maximum temperatures reached during the reaction. Some reaction occurs at the lower temperatures at which the reactants are mixed and pumped to the reactor, before finally heating to the indicated reaction temperature.



   The residence times indicated in the table are calculated by dividing the reactor volume, in liters, by the feed rate of the raw material, in liters per minute. The actual residence time of the liquid in the reactor is somewhat shorter, because the carbon dioxide released during the reaction partially fills the reactor.

 <Desc / Clms Page number 5>

 



  EXAMPLE II
The hydrolysis was carried out batchwise in a high pressure reactor
 EMI5.1
 Stainless steel was equipped with a high speed mixer and a heating coil. The reactor was charged with a warm, muddy sodium carbonate solution. Then the 1-chloro-2,3-propylene oxide (78.9% potential glycerol) was added and closed the reactor. The mixture, which had a temperature of 80-90 ° C, was then heated to reaction temperature.

   When the pressure in the reactor had risen sufficiently high due to the evolving carbon dioxide, the pressure was held constant for the remainder of the reaction period by venting gas through a valve and passing it to a scrubbing tower where the carbon dioxide was passed through. a 20% sodium hydroxide solution was absorbed. The glycerol yield was determined by analysis of the feedstock, product formed in the reactor, its wash water, and the solution emerging from the washing tower. Then the hydrolyzed product was treated in the usual manner. processed into pure glycerol.



   The following are the results of some tests that were performed under different reaction conditions.
 EMI5.2
 
<tb> max.
<tb>
<tb>
<tb>



  Composition <SEP> nutrition <SEP> stoe- <SEP> max. <SEP> reac- <SEP> reac- <SEP> no-ge <SEP> high- <SEP> glyce-
<tb>
<tb>
<tb> chio- <SEP> reac- <SEP> tie- <SEP> tie- <SEP> hydro- <SEP> cooking-roll
<tb>
<tb>
<tb> 1-chlorine- <SEP> sodium- <SEP> water <SEP> metri- <SEP> tie- <SEP> press <SEP> time, ¯ <SEP> reads <SEP> the <SEP> up-
<tb>
 
 EMI5.3
 2, r i; p carbo temp.

   (atm.) ') output yield
 EMI5.4
 
<tb> propane <SEP> naat <SEP> excess <SEP> (C) <SEP> (Min.) <SEP> materi- <SEP> duc <SEP> (wt%)
<tb> on <SEP> al <SEP> for <SEP> ten
<tb>
 
 EMI5.5
 Na2CO3 glycerol (dl) ¯¯¯¯¯¯¯ - ¯¯¯¯ Zen.%) Ew.%) ¯¯¯¯ 1 Ci; 1 6.4 83.7 5% 162 470 15 1.8 1 i 3 97, 8
 EMI5.6
 
<tb> 15.0 <SEP> 8.6 <SEP> 76.0 <SEP> 2% <SEP> 160 <SEP> 450 <SEP> 15 <SEP> 3.6 <SEP> 0.4 <SEP> '97, 2
<tb>
<tb>
<tb>
<tb>
<tb>
<tb>
<tb> 20.0 <SEP> 11.5 <SEP> 68.5 <SEP> gean <SEP> 165 <SEP> 400 <SEP> 15 <SEP> 0.4 <SEP> 2.8 <SEP> 97 , 4
<tb>
<tb>
<tb>
<tb>
<tb>
<tb> 20.1 <SEP> 11.7 <SEP> 68.2 <SEP> 2% <SEP> 1.65 <SEP> 420 <SEP> 15 <SEP> none <SEP> 2.2 <SEP> 97, 3
<tb>
<tb>
<tb>
<tb>
<tb>
<tb>
<tb> 25.0 <SEP> 14.3 <SEP> 60.7 <SEP> none <SEP> 158 <SEP> 360 <SEP> 15 <SEP> 1.7 <SEP> 0.5 <SEP> 98 , 0
<tb>
<tb>
<tb>
<tb>
<tb>
<tb>
<tb>
<tb> 24.7 <SEP> 14,

  9 <SEP> 60.4 <SEP> 5% <SEP> 163 <SEP> 450 <SEP> 15 <SEP> none <SEP> 0; 3 <SEP> 98.8
<tb>
<tb>
<tb>
<tb>
<tb>
<tb>
<tb>
<tb> 24.6 <SEP> 15.4 <SEP> 60.0 <SEP> 10% <SEP> 161 <SEP> 445 <SEP> 15 <SEP> 1.0 <SEP> 1.1 <SEP> 98.6
<tb>
<tb>
<tb>
<tb>
<tb>
<tb>
<tb>
<tb> 24.5 <SEP> 14.9 <SEP> 60.6 <SEP> 5% <SEP> 153 <SEP> 200 <SEP> 15 <SEP> none <SEP> - <SEP> 99.0
<tb>
<tb>
<tb>
<tb>
<tb>
<tb>
<tb>
<tb> 24.6 <SEP> 15.5 <SEP> 59.9 <SEP> 10% <SEP> 152 <SEP> 410 <SEP> 16 <SEP> none <SEP> - <SEP> 99.2
<tb>
   @ Response time is the total residence time in the reactor, including the heating time.

   of EXAMPLE III
 EMI5.7
 The influence of variations in the maximum working temperature 0 'on the results of the hydrolysis of mixtures of 1-chloro-2,3-epoxypropane (75% -80%) with about 20-25% glycerol dichlorohydrins was measured under various conditions. determined in two series of tests. One series of tests was carried out in the tubular reactor used in Example I and the other tests were carried out in a tubular reactor with a diameter of 60 cm and a height of 5.6 m.

   The feed consisting of 1-chloro-2,3-epoxypropane was fed into this reactor
 EMI5.8
 (78-84% glycerol starting material; the remainder consisted of organic impurities, including trichloropropane, etc.) in a manifold with after-
 EMI5.9
 triumcarbonate, which nozzle was fitted to the suction side of a single stage centrifugal pump capable of delivering a pressure of 75-80 kg.



  The pump was used to mix and disperse the reactants and to pressurize the installation. The pump was connected to it

 <Desc / Clms Page number 6>

 lower end of the reactor, which was operating at a pressure of about 10.5 a-bm. The heat supplied to the reactor was provided by steam at a pressure of 12 atm. to the bottom of the reactor and by the heat released during the reaction. The product discharged via the tqp of the reactor - was passed to a phase separator, where the aqueous glycerol was separated from the carbon dioxide. sodium hydroxide to form a sodium carbonate solution for further hydrolysis.



   Vertical tank reactor
 EMI6.1
 
<tb> Response time <SEP> 12-16 <SEP> minutes
<tb>
<tb> Excess <SEP> sodium carbonate: 16 <SEP> - <SEP> 23 <SEP> wt.%
<tb>
<tb> Maximum <SEP> Glycerol <SEP> Loss <SEP> to <SEP> yield <SEP> *) <SEP> from <SEP> the
<tb>
<tb> temperature <SEP> concentration <SEP> starting material <SEP> for <SEP> glycerol,
<tb>
<tb> c <SEP> (wt%) <SEP> in <SEP> wt%
<tb>
<tb> non-hydro <SEP> polymer
<tb>
<tb> lysed <SEP> substance
<tb>
<tb>
<tb> 135 <SEP> 5.6 <SEP> 3.3 <SEP> 1.7
<tb>
<tb> 149 <SEP> 5.0 <SEP> 1.3 <SEP> 1.9
<tb>
<tb> 163 <SEP> 5. <SEP> 5 <SEP> 0.0 <SEP> 1. <SEP> 5
<tb>
<tb> 173 <SEP> 5. <SEP> 4 <SEP> 0. <SEP> 2 <SEP> 2.

   <SEP> 3
<tb>
 Tube reautor
 EMI6.2
 
<tb> Response time <SEP> 6.9-9.3 <SEP> minutes
<tb>
<tb>
<tb> Excess <SEP> sodium carbonate <SEP> 15-32 <SEP> wt.%
<tb>
<tb>
<tb>
<tb>
<tb> Maximum- <SEP> Glycerol- <SEP> Loss <SEP> to <SEP> yield *) <SEP> from <SEP> the
<tb>
<tb>
<tb>
<tb> temperature <SEP> concentration <SEP> starting material <SEP> for <SEP> glycerol,
<tb>
<tb>
<tb> c <SEP> (Wt. <SEP> in <SEP> wt.%
<tb>
<tb>
<tb>
<tb>
<tb> non-hydro <SEP> polymer
<tb>
<tb>
<tb> ¯¯¯¯¯¯¯¯ <SEP> ¯¯¯¯¯¯¯¯¯¯¯¯ <SEP> lysed <SEP> substance <SEP> ¯¯¯¯¯¯¯¯¯¯¯
<tb>
<tb>
<tb>
<tb>
<tb>
<tb> 151 <SEP> 12. <SEP> 8 <SEP> 0. <SEP> 5 <SEP> 0.9
<tb>
<tb>
<tb>
<tb> 164 <SEP> 13.2 <SEP> 0 <SEP> .5 <SEP> 1.5
<tb>
<tb>
<tb>
<tb> 177 <SEP> 15.7 <SEP> 0.3 <SEP> 1.6
<tb>
<tb>
<tb> 179 <SEP> 23. <SEP> 6 <SEP> 0.2 <SEP> 2.8
<tb>
<tb>
<tb>
<tb> 186 <SEP> 22.4 <SEP> 0.

   <SEP> 3
<tb>
 *) yield of glycerine = 100 non-hydrolyzed material + polymer) EXAMPLE IV
The influence of residence time changes on yield losses in hydrolyzing the crude 1-chloro-2,3-propylene oxide used in Example III can be seen from the following table for tests run at temperatures of 157 ° C. -177c have been carried out with the tubular reactor described in Example I and the vertical tank reactor used in Example III.

 <Desc / Clms Page number 7>

 
 EMI7.1
 
<tb>



  Stay <SEP> time <SEP> Glycerol <SEP> Excess <SEP> Loss <SEP> to <SEP> yield
<tb>
<tb>
<tb> in <SEP> reactor <SEP> concentration <SEP> consumes <SEP> from <SEP> the <SEP> starting material
<tb>
<tb>
<tb> (min.) <SEP> in <SEP> product <SEP> sodium <SEP> for <SEP> glycerol <SEP> in <SEP> wt%
<tb>
 
 EMI7.2
 (wt%) carbonate non-cyclopolymer
 EMI7.3
 
<tb> wt <SEP>% <SEP> lysed <SEP> substance
<tb>
<tb>
<tb> Tube reactor
<tb>
<tb>
<tb>
<tb> 6.7 <SEP> 5.2 <SEP> 9.7 <SEP> 2.9 <SEP> 0.0
<tb>
<tb>
<tb>
<tb>
<tb>
<tb> Tank reactor
<tb>
<tb>
<tb>
<tb> '<SEP> 10 <SEP> 3.6 <SEP> 16.8 <SEP> 0.0 <SEP> 1.5
<tb>
<tb>
<tb>
<tb> 15 <SEP> 4.2 <SEP> 39.0 <SEP> 0.0 <SEP> 1.0
<tb>
<tb>
<tb>
<tb> 25 <SEP> 5.8 <SEP> 7.5 <SEP> 0.0 <SEP> 2.5
<tb>
<tb>
<tb>
<tb>
<tb> Tube = reactor
<tb>
<tb>
<tb>
<tb> 5.4 <SEP> 15.5 <SEP> 13.7 <SEP> 0.6 <SEP> 1,

  9
<tb>
<tb>
<tb>
<tb> 7.1 <SEP> 15.7 <SEP> 14.8 <SEP> 0.6 <SEP> 1.6
<tb>
<tb>
<tb>
<tb> 10.4 <SEP> 16.6 <SEP> 11.4 <SEP> 0.8 <SEP> 2.7
<tb>
<tb>
<tb>
<tb> 13.0 <SEP> 17.8 <SEP> 13.4 <SEP> 0.4 <SEP> 3.0
<tb>
 
 EMI7.4
 Tube reactor
 EMI7.5
 
<tb> 6.5 <SEP> 21.5 <SEP> 9.4 <SEP> 0.8 <SEP> 3.2
<tb>
<tb>
<tb> 7.2 <SEP> 10.2 <SEP> 29.4 <SEP> 0.2 <SEP> 2.8
<tb>
<tb>
<tb> 11.5 <SEP> 26.6 <SEP> 13.3 <SEP> 0.3 <SEP> 3.6
<tb>
<tb>
<tb> 12.3 <SEP> 24.8 <SEP> 21.4 <SEP> 0.1 <SEP> 3.4
<tb>
 
 EMI7.6
 1JOR IMAGE V
The effect of the amount of excess sodium hydroxide used in carrying out the hydrolysis of the type shown in Example III.
 EMI7.7
 The crude 1-chloro-2,3-epoxypropane in the reactors described in Examples I and III obtained results.



   Vertical tank reactor.
 EMI7.8
 
<tb>



  Maximum temperature: <SEP> 157 <SEP> - <SEP> 166 C
<tb>
<tb> Glycerol concentration <SEP> in <SEP> descending <SEP> product: <SEP> 4 <SEP> - <SEP> 6%
<tb>
 
 EMI7.9
 dwell time 10 - 15 min.
 EMI7.10
 
<tb>



  Excess <SEP> sodium <SEP> Loss <SEP> to <SEP> brings <SEP> from <SEP> to the <SEP>
<tb>
<tb> carbonate <SEP> wt <SEP>% <SEP> raw material <SEP> for <SEP> glycerol <SEP> in <SEP> wt%.
<tb>
<tb>
<tb> non-hydrolyzed <SEP> polymer
<tb>
<tb> material
<tb>
 
 EMI7.11
 -4> ..... .dil ....... ..... d> .. -., ...--. "'" 1> 1 -. ".--" "" "-
 EMI7.12
 
<tb> 3.7 <SEP> 0.1 <SEP> 2.7
<tb>
<tb>
<tb> 6.2 <SEP> 0.0 <SEP> 1.9
<tb>
<tb>
<tb> 16.8 <SEP> 0.0 <SEP> 1.5
<tb>
<tb>
<tb> 19.8 <SEP> 0.0 <SEP> 1.5
<tb>
<tb>
<tb> 20.3 <SEP> 0.0 <SEP> 1.5
<tb>
<tb>
<tb> 44.6 <SEP> 0.0 <SEP> 1.6
<tb>
 Tube reactor.
 EMI7.13
 
<tb>



  Maximum temperature: <SEP> 162 <SEP> - <SEP> 182 C
<tb>
<tb> Glycerol concentration <SEP> in <SEP> obtained <SEP> .product! <SEP> 20 <SEP> 27%
<tb>
<tb> Stay <SEP> time: <SEP> 7 <SEP> - <SEP> 11.5 <SEP> min.
<tb>
 

 <Desc / Clms Page number 8>

 
 EMI8.1
 
<tb>



  Excess <SEP> sodium <SEP> Expired <SEP> to <SEP> yield <SEP> from <SEP> the <SEP> initial
<tb>
<tb>
<tb>
<tb> carbonate <SEP> wt <SEP>% <SEP> material <SEP> for <SEP> glycerol <SEP> in <SEP> wt <SEP>%
<tb>
<tb>
<tb>
<tb> non-hydrolyzed <SEP> polymer
<tb>
<tb>
<tb> product <SEP> ¯¯¯¯¯¯¯¯¯¯ <SEP> ¯¯¯¯¯¯
<tb>
<tb>
<tb>
<tb> 5.5 <SEP> 0.9 <SEP> 5.0
<tb>
<tb>
<tb>
<tb> 7.4 <SEP> 1.3 <SEP> 4.1
<tb>
<tb>
<tb>
<tb> 9.1 <SEP> 0.4 <SEP> 4.1
<tb>
<tb>
<tb>
<tb> 11.3 <SEP> 0.6 <SEP> 3.4
<tb>
<tb>
<tb>
<tb> 13.3 <SEP> 0.3 <SEP> 2.8
<tb>
<tb>
<tb>
<tb>
<tb> 23.0 <SEP> 0.4 <SEP> 2.6
<tb>
<tb>
<tb>
<tb> 25.2 <SEP> 0.2 <SEP> 2.8
<tb>
<tb>
<tb>
<tb> 26.5 <SEP> 0.2 <SEP> 2.8
<tb>
<tb>
<tb>
<tb> 29.4 <SEP> 0.2 <SEP> 2.8
<tb>



    

Claims (1)

CONCLUSIONS. 1. A process for the preparation of glycerol by hydrolysis of 1-halo-2,3-epoxypropanes, characterized in that the hydrolysis is carried out by contacting the 1-halo-2,3-epoxypropane at a temperature of 130-200 ° C. with an aqueous solution of an inorganic carbonate, in the presence of carbon dioxide and at a pressure such that the reaction components remain in the liquid phase for 5 to 20 minutes. Process according to Claim 1, characterized in that the aqueous inorganic carbonate solution is used in an excess of 2-20% with respect to the stoichiometric amount required for the hydrolysis. Process according to Claim 2, characterized in that the aqueous inorganic carbonate solution is used in an excess of about 10-20% with respect to the amount required stoichiometrically for the hydrolysis. Process according to one or more of Claims 1 to 3, characterized in that an aqueous alkali metal, preferably sodium; solution of a carbonate or bicarbonate is used. 5. Process according to one or more of claims 1 to 4, characterized in that the ratio of the reactants is selected such that a glycerol solution with a concentration of about 15-25% is obtained. Process according to one or more of claims 1 to 5, characterized in that during the entire reaction the pressure is at about 7 to 35 atm. is being held. Process according to one or more of claims 1 to 6, characterized in that the reaction is carried out at a temperature between 150 ° C and 180 ° C. @ 8. Process according to one or more of claims 1 to 7, characterized in that 1-chloro-2,3-epoxypropane is used as the 1-halo-2,3-epoxypropane. 9. Process substantially as described above with reference to Examples. Glycerol prepared according to the method of one or more of claims 1 to 9.