Recovery of Alkanol from Bleed Streams
in Chloroprene Production
The present invention relates to the production of chloroprene from 34-dichlorobutene-l in a two phase mixture of water and alcohol and to the recovery of alcohol froni other organic materials present in organic phases from the dehydrochlorination reaction.
The production of chloroprene by a dehydrochlorination 'of 3,4-dichlorobutene-l with alkali in a two phase mixture of water and an alcohol at elevated temperatures is disclosed in United States patent specification 3 079 446. The chloroprene is distilled off from the reaction mixture as it is formed. The dehydrochlorination of 3,4-dichlorobutene-l to give chloroprene at temperatures in the range 40 to 70"C is disclosed in German patent specification 2 707 G73.
In the course of the dehydrochlorination there is an accumulation of impurities in the organic phase. The above mentioned German specification states that it may be desirable to discard some of the organic phase and to replace it by fresh alcohol to prevent an accumulation of undesirable higher boiling materials in the system. In a continuous process this is conveniently carried out by taking a bleed-stream from the organic phase once it has settled in a separator and before recycle to the reactor. This bleed may be taken from an organic phase which contains substantially no chloroprene, which may be the case if chloroprene is distilled off as it is formed or if a chloroprene removal step is carried out on the organic phase outside the dehydrochlorination reactor before the bleed is taken.
Alternatively the-bleed may be taken from an organic phase from which chloroprene has not been removed. In either case it is desirable for economic reasons to recover the alcohol from the organic bleed.
Our copending application (Case 4723/ (Case 4727) is concerned with carrying out an aqueous extraction of the organic bleed stream by carrying out the initial extraction step using a volume ratio of water to organic phase not less than certain defined values, in order to obtain a satisfactory extraction of the alcohol. It is desirbale to recover the maximum quantity of alcohol from the bleed stream because of the expense of replacing alcohol which is discharge in the effluent.
In the operation of a process for the dehydrochlorination of 3,4-dichlorobutene it is desirbale to feed as little water as possible to the process as a whole because any water added, whether in the reaction stage or to the recovery stage will finally appear in the aqueous effluent and so increase the total volume of water to be disposed of. If the objective is to avoid introducing additional water it is obvious that a suitable source of water will be the water rejected from the base of the distillation column or columns in which it is separated from organic materials such as chloroprene.
This water will be warm because of the heat introduced by the distilation step. The organic bleed taken from the dehydrochlorination step will also be warm as the dehydrochlorination step is generally carried out at temperatures above ambient temperature. The net result will be that the extraction step will take place at temperatures well above ambient temperatures. A man skilled in the art would see no objections to this however.
It has now been found that operating the aqueous extraction in what would appear to be the obvious way of carrying out such a process leads to a reduction in the efficiency with which alkanol is recovered by extraction with water from the organic bleed.
According to the present invention a process for the dehydrochlorination of 3,4-dichlorobutene-l to give chloroprene in a reaction medium which is a two phase liquid mixture of water and an alcohol in the presence of alkali in which a bleed is taken from the organic phase is characterised in that the bleed is extracted with water and the temperature at which the extraction takes place is not more than 32"C.
The dehydrochlorination of 3,4-dichlorobutene-l to give chloroprene in a reaction medium which is a two phase liquid mixture of water and an alcohol is well known in the art.
The reaction medium must consist of two liquid phases and the alcohol employed must be such as to give with the water two liquid phases under the reaction conditions used. Examples of suitable alcohols are higher alkoxyalkanols, and higher alkanols e.g. these containing 3 to 8 carbon atoms. An example of a preferred alkoxyalkanol is butoxyethanol. Examples of suitable unsubst.tuted alkanols are those having 3 to 5 carbon atoms in the molecule e.g.
propanol-l, propanol-2, butanol-l. The ability to form separate liquid phase will depend on the nature of the alcohol, the relative quantities of alcohol and water, and the concentration of dichlorobutenes -and chloroprene in the organic phase and elf alkali metal hydroxide and alkali metal chloride in the aqueous phase, but for any given reaction mixture a suitable alcohol can be selected by simple test. Thus, it is preferred to use an alcohol which, when shaken with a saturated brine solution containing 22% wt/wt NaOH at a volume ratio of alcohol to aqueous phase corresponding to that to be used in the dehydrochlorination process, forms a separate phase, which phase contains at least 0.1% wt/wt NaOH. An example of a volume ratio of alcohol to aqueous phase at which the test may be carried out is 3:1.
It is particularly preferred to use alcohols whic'hgi've a separate phase containing at least 0.5% wt/wt Na0H in order to obtain fast reaction rates.
Thus, when various alcohols were shaken with a saturated brine solution containing 22% wt/wt NaOH at a volume ratio of alcohol to aqueous phase 3:1, the values obtained for the NaOH content of the alcohol phase (at room temperature) were:
Alcohol NaOH Content
-sec-butanol (butanol-2) 0.05% wt/wt
iso-butanol (trimethyl methanol) 0.19% wt/wt
n-butanol (butanol-l) 0.6% wt/wt
2-butoxyethanol 3.6% wt/wt
It will be seen that butanol-2 is not a preferred solvent, while n-butanol -and 2-butoxyethanol are preferred solvents. It is particularly preferred to use 2-butoxyethanol.
The quantity of organic phase taken as a bleed varied with the purity of the feed and with the reaction conditions. Typically the bleed is in the range 2 tolO%, e.g. 2 to 4 wt/wt of the organic phase present.
It is particularly preferred to carry out the extraction step at a temperature of not more than 200C. In order to obtain sufficiently low temperatures it may be necessary to cool the organic bleed and/or the water fed to the extraction step. If the water fed to the extraction step comes from a distillation step it will be particularly important to cool it to obtain the desired low extraction temperature Whatever the source of the water it may be necessary to cool it to well below normal ambient temperature (e.g. about 20"C).
Thus it may be necessary to pass the water through a cooler to reduce its temperature, preferably below 15"C, more preferably 12"C. Suitable cooling devices are well known to those skilled in the art, and may make use of a chilled heat exchange liquid.
The process of the present invention is preferably used together with the process of copending application (Case 4723/4727), in which the bleed is subjected to extraction with water and the volume ratio of water to organic phase at the first extraction step is not less than 15¯1. If. the bleed contains no substantial quantity of chloroprene, organic phase volume ratio may be reduced to not less than 6:1
It may be possible to operate with lower volume ratios of water to organic phase if phase separation is obtained by the use of suitable filters or membrances, but this is not preferred.
The extraction step is a step of bringing into contact and then separating the organic and aqueous phases. This may be carried out by any of the well known methods. A suitable method is by mixing the water and the organic phase together in a stirred vessel or by using an in-line static mixer and then feeding the mixture to a decanter where the organic phase separates from the aqueous phase under gravity.
Depending on the degree of recovery of alkanol required it may be necessary to use only one extraction step but if higher recoveries of alkanol are necessary additional extraction steps may be carried out.
The accompanying drawings are diagrammatic representations of apparatus or apparatus suitable for carrying out the process of the present invention.
Figure 1 is a diagrammatic representation of apparatus suitable for carrying out the process of the invention when dehydrochlorination is carried out under conditions such that chloroprene is distilled form the reaction mixture as it is formed.
Figure 2 is a diagrammatic representation of apparatus suitable for carrying out the process of the invention when the dehydrochlorination is carried out at temperatures such that chloroprene is retained in the liquid withdrawn from the reactor, and the chloroprene is subsequently recovered by distillation from an organic phase separated from the reaction mixture.
The operation of the embodiment shown in Figure 1 will now be described in more detail. A mixture of nitrogen and nitric oxide, sodium hydroxide solution and 3,4-dichlorobuten-l are introduced into reactor 1, provided with stirrer 2, -through pipes 3, 4 and 5 respectively.
The reactor contains a mixture of water and alcohol (e.g. 2-butoxy- ethanol). The reactor 1 is heated by heating means (not shown) to maintain the contents at a temperature such that chloroprene will distill off from the reaction mixture as it is formed. (Examples of suitable temperatures are those in the range 80 to 1050C). Vapour passes through pipe 6 into distillation column 7 and condensate returns through pipe 8. Chloroprene is removed overheat at 9. The reactor 1 is provided with a stand pipe 10 through which the livid contents of reactor 1 overlow into decanter 11, where they separate into a lower aqueous phase containing dissolved NaCl, and an upper organic phase. The aqueous phase is discarded through pipe 12.
The organic phase consiting of unreacted dichlorobutenes, alcohol, chloroprene, and high boiling impurities is recycled through pipe 13 to the reactor.
A bleed is taken through pipe 14 to a distillation column 70 in which the-residual chloroprene is distilled overheat at 90. The material from the base of the distillation column is passed through; line 140 to a stirred vessel 15 to which water is introduced through pipe 16. The resulting mixture passes through pipe 18 to decanter 17 from which a lower organic phase is separated and discarded through pipe 19. The upper aqueous phase containing the extracted alkanol is returned through pipe 20 to the reactor 1.
The water fed through pipe 16 can be subjected to cooling in cooler 21.
In Figure 2 those items which are identified by the same numbers as those in Figure 1 perform the same function as in Figure 1.
However, the dehydrochlorination reactor is operated at a temperature such that the chloroprene produced does not boil off from the reaction mixture (Examples of suitable temperatures are 50"C to 700C). All the chloroprene therefore remains in the organic phase from decanter 11 which passes through pipe 130 to distillation column 71 from which chloroprene is recovered overhead at 91. The product from the base of the column is returned through pipe 131 and mixed with the 3,4-dichlorobutene-l feed to the reactor (although of course it can be added separately). A bleed is taken through pipe 14 and subjected to extraction with water as in the apparatus disclosed in Figure 1 and the aqueous extract containing alkanol is returned to the reactor through pipe 20.
The water fed through pipe 16 can be subjected to cooling in cooler 21.
Example 1
3,4-Dichlorobutene was dehydrochlorinated continuously with sodium hydroxide in a mixture of water and 2-butoxyethanol at a temperature (e.g. 90" to 105"C) such that chloroprene was distilled out of the reaction mixture and was recovered overhead from a distillation column attached to the reactor. The liquid reaction mixture was passed to a decanter from which an organic phase containing unreacted dichlorobutenes, 2-butoxyethanol, residual chloroprene and high boiling impurities was returned to the reactor. A bleed was taken from this organic phase and subjected to distillation to smove
residual chloroprene. (The procedure upto here corresponds to that
of Figure 2 of our copending application (Case 4723/4737).
The
composition of the bleed after this distillation was 2-butoxyethanol
(67.2% wt) dichlorobutenes (7.7% wt), high boiling impurities (12.9% wt)
and chloroprene (0.3% wt). A sample of the bleed was extracted
batchwise with 8 columns of water at a temperature such that the
temperature of the resulting mixture was 30.1 C. The distribution
coefficient of alkanol between organic raffinate and aqueous extract
was 5.15 and 85.9% wt of the butoxyethanol was extracted into the
aqueous phase.
Examples 2 and 3
Experiments was carried out as in Example 1 but with the temperature
of the water used to carry out the extraction such that the temperature
of the mixture was as given in Table 1.
Table 1
EMI7.1
<tb> Example <SEP> Extraction <SEP> Distribution <SEP> Alkanol
<tb> <SEP> Temperature <SEP> Coefficient <SEP> Extracts <SEP> (% <SEP> wt)
<tb> <SEP> 2 <SEP> 22"C <SEP> 3.81 <SEP> 89.1
<tb> <SEP> 3 <SEP> 17.2"C <SEP> 3.14 <SEP> 91.9
<tb>
Examples 4 to 6
An organic bleed similar to that subjected to batch extraction in Example 1 was subjected to continuous extraction in a mixer/decanter system at a water to organic phase volume ratio of 8:1. The temperature of the water fed to the extraction was adjusted so as to give the temperature in the mixture set out in Table 2.
Table 2
EMI7.2
<tb> Example <SEP> Temperature <SEP> OC <SEP> Alkanol <SEP> Extracted
<tb> <SEP> % <SEP> wt
<tb> <SEP> J
<tb> <SEP> 4 <SEP> 22 <SEP> 1 <SEP> 81,5
<tb> <SEP> 5 <SEP> 17 <SEP> 85.6
<tb> <SEP> 5 <SEP> 17 <SEP> 85.6
<tb> <SEP> 6 <SEP> 14 <SEP> 86
<tb>