PROCESS FOR PURIFICATION OF CRUDE POLYETHER POLYOLS

Abstract

A process for the purification of crude polyether polyols is provided, wherein the crude polyether polyols are prepared by anionic polymerization of alkylene oxides in the presence of basic catalysts, the process comprising the steps of: neutralization of the catalysts with mineral acid, addition of adsorption agents and/or filter agents and removing the resultant salts and added filter aids by filtration wherein a filter cake is formed, and wherein polyether polyols are recuperated from the filter cake in a subsequent step of pressing the filter cake.

Claims

1. A process for purifying a crude polyether polyol prepared by anionic polymerization of an alkylene oxide in the presence of a basic catalyst, the process comprising neutralizing the catalyst with a mineral acid to form a salt, adding an adsorption agent and/or a filter aid, removing the salt and the adsorption agent and/or the filter aid by filtration to form a filter cake, and subsequently pressing the filter cake to recuperate the polyether polyol from the filter cake.

2. The process of claim 1, comprising: heating the filter cake prior to pressing the filter cake.

3. The process of claim 2, wherein a temperature of the filter cake during said pressing ranges from 40 C. to 90 C.

4. The process of claim 1, wherein the filter cake is pressed using a pressure ranging from 20 to 200 bar.

5. The process of claim 1, wherein the filter cake is pressed for 1 minute to 10 minutes.

6. The process of claim 1, wherein a high-pressure chamber press filter is used for pressing the filter cake.

7. The process of claim 1, wherein the polyether polyol is recuperated from the filter cake by first drying the filter cake by blowing out residual polyether polyol using nitrogen and then subsequently pressing the filter cake.

8. The process of claim 1, wherein the process is conducted batch-wise, and said pressing is performed at the end of a batch.

9. The process of claim 8, wherein a cycle time of pressing the filter cake is less than half a cycle time of a batch and one filter press is shared between at least two production lines.

10. The process of claim 1, further comprising: blending the recuperated polyether polyol with filtered polyether polyol to obtain a blended polyether polyol such that viscosity and/or residual amounts of the basic catalyst of the blended polyether polyol are within specified ranges.

11. The process of claim 10, wherein the process is conducted batch-wise, the filter cake is pressed at the end of a batch, and the recuperated polyether polyol is blended with purified polyether polyol of a subsequent batch.

12. The process of claim 11, wherein the recuperated polyether polyol of the last batch of a production cycle is stored and blended with filtered polyether polyol of the first batch of a subsequent production cycle.

13. The process of claim 1, wherein a residual amount of polyether polyol in the filter cake after pressing is less than 40% by weight.

14. The process of claim 1, wherein the basic catalyst is potassium hydroxide.

15. The process of claim 1, wherein the alkylene oxide is ethylene oxide, propylene oxide, or a mixture of ethylene oxide and propylene oxide.

Description

EXAMPLES

[0040] Several polyether polyol products where produced as examples A to G using a main reactor for synthesis of the polyether polyols, a filtration reactor with filtration unit, a filtration press and a workup reactor.

[0041] In the main reactor a mixture consisting of starter molecules, basic catalyst and alkylene oxide is introduced. After completing the polymerization, the excess alkylene oxide is removed and the crude polyether polyol is fed into the filtration reactor.

[0042] In the filtration reactor water and ortho-phosphoric acid or CO.sub.2 are added to the crude polyether polyol.

[0043] The reaction mixture is stirred for a time. Water is removed to form the salts. After the salts are formed an adsorption agent and filter aid is added to the reaction mixture. The salts resulting from the neutralization and the solid additives such as the adsorption agent and optionally the filtration aids are removed by means of filtration. Water is removed from the purified polyether polyol produced as filtrate in the workup reactor.

[0044] The filter cake is removed from the filtration unit. For each example A to G several samples of wet filter cake have been taken. The weight of each of the wet filter cake samples is determined and for some samples the wet filter cake has been pre-dried by blowing nitrogen gas through the wet filter cake. The wet or pre-dried filter cake samples are then inserted into the chamber of a high pressure chamber filter press. The amount of recuperated polyether polyol is measured as well as the weight of the dry filter cake after pressing.

[0045] The amount of polyol which is recuperated is determined by subtracting the mass m.sub.dc of the wet filter cake from the mass m.sub.i of the wet filter cake. This can be expressed in percent of the initial wet filter cake mass m.sub.i using the relation

[00001]$\begin{array}{cc}{R}_{\mathrm{polyol}}=\frac{m_i-m_{\mathrm{dc}}}{m_i}.Math.100.& \left(1\right)\end{array}$

[0046] The mass flow of filter cake m.sub.f which can be processed by the filter press is dependent on the pressing time t.sub.p and the changeover time t.sub.c required to insert the wet filter cake and to remove the dry filter cake. The mass flow per hour can be expressed using the relation

[00002]$\begin{array}{cc}{\stackrel{.}{m}}_f=\frac{3600}{t_p+t_c}.Math.m_i& \left(2\right)\end{array}$

wherein the changeover time t.sub.c and the pressing time t.sub.p are given in seconds. Thus, the mass flow of recuperated polyol per hour is given by

[00003]$\begin{array}{cc}{\stackrel{.}{m}}_{\mathrm{polyol}}=\frac{3600}{t_p+t_c}.Math.m_i.Math.\frac{R_{\mathrm{polyol}}}{100}.& \left(3\right)\end{array}$

[0047] In the following, the changeover time t.sub.c is assumed to be half of the pressing time t.sub.p so that the mass flow of recuperated polyol per hour is given by

[00004]$\begin{array}{cc}{\stackrel{.}{m}}_{\mathrm{polyol}}=\frac{2400}{t_p}.Math.m_i.Math.\frac{R_{\mathrm{polyol}}}{100}.& \left(4\right)\end{array}$

[0048] In addition, the amount of polyether polyol in the wet filter cake and in the dry filter cake have been analyzed in laboratory measurements wherein the polyol was extracted from a known amount of the filter cake using dichloromethane. The dichloromethane was then evaporated using a nitrogen flow and the mass of the extracted polyol has been measured. The laboratory measurements were performed twice for each sample to account for inhomogeneity in the sample. The polyol content for the dry filter cakes was determined on the actual sample after pressing, while measurements for the wet filter cake were performed on a different sample.

[0049] The amount of recuperated polyol based on the lab measurements is then calculated as follows

[00005]$\begin{array}{cc}{R}_{\mathrm{polyol},\mathrm{lab}}=\frac{m_{\mathrm{pi}}-m_{\mathrm{pd}}}{m_i}.Math.100& \left(5\right)\end{array}$

wherein m.sub.pi is the mass of polyol in the wet filter cake and m.sub.pd is the mass of polyol in the dry filer cake and R.sub.polyol, lab is the recuperated polyol in percent.

[0050] Samples Used:

[0051] The filter cake samples have been prepared by producing at least one batch of polyether polyols for each example using the parameters listed in table 1 wherein at least two samples were taken. Table 1 also lists the molecular weight of the polyol. In case of example G, two batches have been produced and the samples were taken from those two batches. In case of examples E, F and G some samples were pre-dried using nitrogen.

TABLE-US-00001 TABLE 1 Starter Molecular Example # of samples molecule Alkylene oxide weight [g/mol] A 12 glycerin PO/EO 3410 B 6 glycerin PO/EO 5390 C 2 glycerin PO/EO 5170 D 7 glycerin PO/EO 3550 E 10 glycerin PO/EO 4360 F 6 glycerin PO 1070 G 12 glycerin PO/EO 5390

[0052] For all examples potassium hydroxide has been used as basic catalyst and CO.sub.2 has been used as acid. Further, for all examples magnesium silicate has been added as filtration aid.

[0053] The properties of the polyols of the examples are listed in table 2. The average residual value of potassium ions (Na/K level) has been averaged over the measured values for purified polyether polyols of several batches.

TABLE-US-00002 TABLE 2 Average Example Na/K [ppm] A 4.3 B 4.36 C 3.68 D 9.37 E 3.37 F 3.63 G 3.04

[0054] Pressing of the Samples

[0055] For examples A to D only wet filter cake samples were used. 3 samples of example E, 2 samples of example F and 6 samples of example G were pre-dried by blowing nitrogen gas through the filter cake.

[0056] The wet or pre-dried filter cake sample was inserted into the chamber of a high pressure chamber filter press. FIG. 1 shows a schematic view of a high pressure filter press 10 comprising a cylindrical housing 20. A chamber 12 is located inside the cylindrical housing 20. The chamber 12 is also of cylindrical shape with a diameter D. The chamber 12 is terminated on one side by a fixed end 22 and on the other side by a movable piston 14. The depth d of the chamber 12 can be adjusted by moving the piston 14. A filter cake inserted into the chamber 12 is pressed by applying a force F to piston 14. Polyether polyols are pressed out of the filter cake and leave the chamber 12 through filters 16 and drains 18. After pressing the piston 14 is moved back to its original position and the pressed filter cake is discharged from chamber 12. The recuperated polyols are collected and the amount of recuperated polyols is measured.

[0057] The chamber of the high pressure filter press used for the pressing of the samples has a diameter D of 135 mm and the maximum applied pressure was 100 bar. The depth of the chamber, temperature of the filter cake and pressing time were varied. After pressing, the height h of the pressed filter cake and the mass m.sub.dc of the dried filter cake was determined. The filter cakes were heated prior to insertion into the chamber and the initial temperature was determined. The temperature decreased during pressing and the temperature of the pressed dry filter cake was measured at the end of the pressing step. The diameter of the pressed dry filter cake corresponds to the diameter of the chamber.

Example A

[0058] The measurement results for the samples of example A are listed in table 3:

TABLE-US-00003 TABLE 3 Sample # 1 2 3 4 5 6 7 8 9 10 11 12 m.sub.i [g] 250 250 300 300 350 350 400 400 500 600 700 400 T [ C.] 52 52 45 53 56 51 55 54 51 60 58 54 (start) T [ C.] 24 24 28 27 26 26 26 25 30 25 (end) T.sub.p [s] 180 180 300 300 320 360 360 380 420 400 540 300 m.sub.dc [g] 36 81 89 98 127 117 112 129 160 206 238 135 d [mm] 30 35 40 50 30 h [mm] 3.4 3.9 4.1 4.37 6.6 4.5 5.0 6.0 7.6 10.5 11.3 6.7

Example B

[0059] The measurement results for the samples of example B are listed in table 4:

TABLE-US-00004 TABLE 4 Sample # 1 2 3 4 5 6 m.sub.i [g] 300 400 400 500 600 300 T [ C.] 64 64 65 66 65 64 (start) T [ C.] 25 24 29 28 28 28 (end) T.sub.p [s] 180 300 300 320 390 360 m.sub.dc [g] 121 163 163 199 246 115 d [mm] 20 25 25 30 40 20 h [mm] 5.7 8.0 8.0 9.8 11.8 6.0

Example C

[0060] The measurement results for the samples of example C are listed in table 5:

TABLE-US-00005 TABLE 5 Sample # 1 2 m.sub.i [g] 300 300 T [ C.] 63 64 (start) T [ C.] 23 24 (end) T.sub.p [s] 450 480 m.sub.dc [g] 165 179 d [mm] 30 30 h [mm] 8.8 9.4

Example D

[0061] The measurement results for the samples of example D are listed in table 6:

TABLE-US-00006 TABLE 6 Sample # 1 2 3 4 5 6 7 m.sub.i [g] 400 400 400 400 400 500 400 T [ C.] 58 58 58 52 54 51 50 (start) T [ C.] 23 23 23 25 25 29 25 (end) T.sub.p [s] 240 250 240 251 248 240 220 m.sub.dc [g] 217 209 234 224 228 276 233 d [mm] 30 30 30 30 30 40 30 h [mm] 9.2 9.6 9.9 9.6 9.8 12.0 10.0

Example E

[0062] The measurement results for the samples of example E are listed in table 7 and 8. The measurements listed in table 7 were performed by pressing wet filter cakes.

TABLE-US-00007 TABLE 7 Sample # 1 2 3 4 5 6 7 8 9 10 m.sub.i [g] 300 300 300 400 500 500 500 200 300 300 T [ C.] (start) 86 86 86 74 60 60 62 68 65 67 T [ C.] (end) 36 36 36 30 30 30 27 30 30 30 T.sub.p [s] 120 120 120 180 230 240 250 120 200 180 m.sub.dc [g] 149 147 148 203 259 255 270 142 263 256 d [mm] 20 20 20 25 30 30 30 15 25 25 h [mm] 6.7 7.4 7.0 9.5 12.0 12.3 12.5 7.6 12.0 11.5

[0063] The measurements listed in table 8 were performed by pressing pre-dried filter cakes:

TABLE-US-00008 TABLE 8 Sample # 1 2 3 m.sub.i [g] 200 300 300 T [ C.] 68 65 67 (start) T [ C.] 30 30 30 (end) T.sub.p [s] 120 200 180 m.sub.dc [g] 142 263 256 d [mm] 15 25 25 h [mm] 7.6 12.0 11.5

Example F

[0064] The measurement results for the samples of example F are listed in tables 9 and 10. The measurements listed in table 9 were performed by pressing wet filter cakes.

TABLE-US-00009 TABLE 9 Sample # 1 2 3 4 m.sub.i [g] 600 700 700 700 T [ C.] 65 65 43 65 (start) T [ C.] 45 45 43 50 (end) T.sub.p [s] 80 80 100 80 m.sub.dc [g] 150 219 192 201 d [mm] 40 45 45 45 h [mm] 6.8 9.8 8.3 9.0

[0065] The measurements listed in table 10 were performed by pressing pre-dried filter cakes:

TABLE-US-00010 TABLE 10 Sample # 1 2 m.sub.i [g] 400 500 T [ C.] 65 65 (start) T [ C.] 45 48 (end) T.sub.p [s] 60 60 m.sub.dc [g] 221 284 d [mm] 25 30 h [mm] 9.2 11.8

Example G

[0066] The measurement results for the samples of example G are listed in tables 11 and 12. The measurements listed in table 11 were performed by pressing wet filter cakes.

TABLE-US-00011 TABLE 11 Sample # 1 2 3 4 5 6 m.sub.i [g] 300 400 500 500 500 500 T [ C.] 65 67 67 69 62 60 (start) T [ C.] 34 36 36 48 53 (end) T.sub.p [s] 140 180 200 200 230 180 m.sub.dc [g] 88 109 143 185 165 155 d [mm] 20 30 40 40 40 40 h [mm] 4.0 5.7 6.8 7.25 6.9 7.3

[0067] The measurements listed in table 12 were performed by pressing pre-dried filter cakes:

TABLE-US-00012 TABLE 12 Sample # 1 2 m.sub.i [g] 300 400 T [ C.] 60 60 (start) T [ C.] 32 45 (end) T.sub.p [s] 120 150 m.sub.dc [g] 162 218 d [mm] 20 30 h [mm] 7.8 10.0

[0068] Measurement Results

[0069] For each example A to G an average value of recuperated polyether polyol R.sub.polyol in % of the initial mass m.sub.i of the filter cake calculated according to equation (1) is given. This average value is calculated over all samples for each example even though the measurement conditions are different for each sample. The results are listed in table 13. Further, table 13 gives the average value of recuperated polyether polyol R.sub.polyol, lab according to lab results according to equation (5).

TABLE-US-00013 TABLE 13 Standard deviation Example R.sub.polyol R.sub.polyol R.sub.polyol,lab A 67.57% 2.24% 57.14% B 59.84% 0.99% 51.43% C 42.67% 3.3% 47.42% D 44.08% 2.24% 65.11% E 49.21% 1.73% 50.13% E (with nitrogen) 18.67% 9.02% 6.97% F 71.89% 2.62% 91.58% F (with nitrogen) 43.98% 1.10% 26.91% G 68.97% 3.54% 69.09% G (with nitrogen) 46.69% 4.51% 42.13%

[0070] For most of the examples the lab results which have been obtained by extracting the residual polyols using dichloromethane compare well with the results obtained by pressing. The average value over the different examples shows that around 51% of the initial filter cake mass m.sub.i can be recuperated as polyol.

[0071] In FIG. 2 a bar diagram is shown wherein for each example a first bar (left) depicts the weight percent of the polyol in the wet/pre-dried cake and a second bar (right) depicts the weight percent of the polyol in the pressed cake. Additionally, FIG. 2 depicts the results for examples E to G wherein the samples have been pre-dried using nitrogen. These results have been marked with a * in the figure. It can be seen in FIG. 2 that there is some variation of polyol recuperation between products.

[0072] In order to assess the quality of the recuperated polyols, the residual amount of potassium ions (Na/K level) has been determined.

[0073] FIG. 3 shows a diagram wherein the recuperated polyol in % is plotted vs. Na/K level in ppm for each example including in addition the results for the pre-dried samples of examples E to G. A weak relation between Na/K levels in the final recuperated polyol can be seen. Na/K levels are higher for examples with worse polyol recuperation.

[0074] FIG. 4 shows a schematic flow diagram for a production campaign with N batches wherein polyether polyols are recuperated and blended with filtered polyols of a subsequent batch. For each batch the steps of synthesis, filtration and workup are performed. Each step requires a set amount of time which is in the diagram of FIG. 4 expressed as a product of total time h and a fraction X.sub.1, X.sub.2 and X.sub.3 respectively. The synthesis step performed in the main synthesis reactor takes up most of the time. Thus, X.sub.1>X.sub.2 and X.sub.1>X.sub.3. Preferably the total time required for pressing the filter cake in the filter press is equal to or less than X.sub.1/2 allowing a single filter press to be shared between two production lines.

[0075] In the diagram of FIG. 4 A tons of filtered polyols and B tons of filter cake are obtained for each batch. If it is assumed that 51% by weight of polyols can be recuperated from the wet filter cake and the recuperated polyols are blended with the filtered polyols, a total amount of A+0.51.Math.B tons of purified polyols can be obtained by the proposed method and only 0.49.Math.B tons of filter cake have to be disposed of as waste. Typically, the amount B of obtained filter cake is about 1% by weight of the filtered polyols A. The recuperated polyols are blended with the filtered polyols of the subsequent batch. For the last batch, the recuperated polyols may be stored in a tank.

[0076] In table 14 the measured Na/K level for recuperated polyols and the average Na/K value for filtered purified polyols are listed.

TABLE-US-00014 TABLE 14 Average Na/K for recuperated Na/K [ppm] of Example polyols [ppm] filtered polyols A 31.25 4.3 B 33.17 4.36 C 51.00 3.68 D 71.71 9.37 E 24.71 3.37 E (with nitrogen) 98.67 3.37 F 18.75 3.63 F (with nitrogen) 13.50 3.63 G 23.83 3.04 G (with nitrogen) 19.17 3.04

[0077] It can be seen that the Na/K levels of the recuperated polyols are higher than the respective Na/K levels of the filtered polyols. Preferably the recuperated polyols are blended with the filtered polyols in the work-up reactors.

[0078] In table 15 the average cake mass flow in grams per hour and the average mass flow of the recuperated polyols are listed. The flow rates have been calculated according to equations (2) and (3) respectively, assuming that the changeover time t.sub.c is half of the pressing time t.sub.p.

TABLE-US-00015 TABLE 15 Average cake Average polyol Example mass flow (g/h) mass flow (g/h) A 2849 1924 B 3307 1975 C 1550 663 D 4128 1819 E 5479 2703 E (with nitrogen) 3867 730 F 19200 13773 F (with nitrogen) 18000 7900 G 5727 3946 G (with nitrogen) 3900 1823