Process for the epoxidation of propene to propylene oxide

Abstract

A continuous process for the preparation of propylene oxide, comprising providing a liquid feed stream comprising propene, hydrogen peroxide, methanol, water, at least one dissolved potassium salt of hydroxyethylidenediphosphonic acid, and optionally propane; passing the liquid feed stream provided in (i) into an epoxidation reactor comprising a catalyst comprising a titanium zeolite of structure type MFI, and subjecting the liquid feed stream to epoxidation reaction conditions in the epoxidation reactor, obtaining a reaction mixture comprising propylene oxide, methanol, water, and the at least one dissolved potassium salt of hydroxyethylidenediphosphonic acid, and optionally propane; removing an effluent stream from the epoxidation reactor, the effluent stream comprising propylene oxide, methanol, water, at least a portion of the at least one potassium salt of hydroxyethylidenediphosphonic acid, and optionally propane.

Claims

1. A continuous process for preparing propylene oxide, the process comprising: (i) passing a liquid feed stream, comprising propene, hydrogen peroxide, methanol, water, at least one dissolved potassium salt of hydroxyethylidene diphosphonic acid, and optionally propane, into an epoxidation reactor comprising a catalyst comprising a titanium zeolite of structure type MFI, and subjecting the liquid teed stream to epoxidation reaction conditions in the epoxidation reactor, to obtain a reaction mixture comprising propylene oxide, methanol, water, and the at least one dissolved potassium salt of hydroxyethylidene diphosphonic acid, and optionally propane; and removing an effluent stream from the epoxidation reactor, the effluent stream comprising propylene oxide, methanol, water, at least a portion of the at least one potassium salt of hydroxyethylidenediphosphonic acid, and optionally propane.

2. The process of claim 1, wherein the molar ratio of potassium relative to phosphorus in the at least one potassium salt of hydroxyethylidene diphosphonic acid ranges from 1:2 to 2:1.

3. The process of claim 1, wherein the at least one potassium salt of hydroxyethylidenediphosphonic acid comprises a dipotassium salt of hydroxyethylidenediphosphonic acid.

4. The process of claim 1, wherein: in the liquid feed stream, the molar ratio of potassium comprised in the at least one potassium salt of hydroxyethylidenediphosphonic acid relative to the hydrogen peroxide ranges from 510.sup.6:1 to 100010.sup.6:1; and in the liquid feed stream, the molar ratio of potassium in the liquid feed stream relative to the potassium comprised in the at least one potassium salt of hydroxyethylidenediphosphonic acid ranges from 1.2:1 to 1:1.

5. The process of claim 1, wherein, in the liquid feed stream, the molar ratio of phosphorus in the liquid feed stream relative to phosphorus comprised in the at least one potassium salt of hydroxyethylidenediphosphonic acid ranges from 1.2:1 to 1:1.

6. The process of claim 1, wherein: the liquid feed stream passed into the epoxidation reactor has a temperature ranging from 0 to 60 C.; and the liquid feed stream passed into the epoxidation reactor is at a pressure ranging from 14 to 100 bar.

7. The process of claim 1, wherein: the temperature of the reaction mixture is controlled using a heat transfer medium; the epoxidation reaction conditions comprise an epoxidation reaction temperature ranging from 10 to 100 C., wherein the epoxidation reaction temperature is defined as the temperature of the heat transfer medium prior to controlling of the temperature of the reaction mixture; and the epoxidation reaction conditions comprise an epoxidation reaction pressure ranges from 14 to 100 bar, wherein the epoxidation reaction pressure is defined as the pressure at the exit of the epoxidation reactor.

8. The process of claim 1, wherein: the effluent stream further comprises hydrogen peroxide and optionally propene; and the process further comprises: (iii) separating propylene oxide from the effluent stream, obtaining a stream being depleted in propylene oxide and comprising hydrogen peroxide, methanol, water, at least a portion of the at least one potassium salt of hydroxyethylidenediphosphonic acid, optionally propene and optionally propane; (iv) passing the stream being depleted in propylene oxide and comprising hydrogen peroxide, methanol, water, at least a portion of the at least one potassium salt of hydroxyethylidenediphosphonic acid, optionally propene and optionally propane, obtained in (iii) into an epoxidation reactor comprising a catalyst comprising a titanium zeolite of structure type MFI, and subjecting the stream to epoxidation reaction conditions in the epoxidation reactor, to obtain obtaining a reaction mixture comprising propylene oxide, methanol, water, the portion of the at least one dissolved potassium salt of hydroxyethylidenediphosphonic acid, and optionally propane; and (v) removing an effluent stream from the epoxidation reactor of (iv), the effluent stream comprising propylene oxide, methanol, water, at least a portion of the portion of the at least one potassium salt of hydroxyethylidenediphosphonic acid, and optionally propane.

9. The process of claim 1 wherein the epoxidation reaction conditions comprise a hydrogen peroxide conversion ranging from 90 to 100%, wherein the hydrogen peroxide conversion is calculated based on the amount of hydrogen peroxide comprised in the effluent stream removed in (ii), relative to the amount of hydrogen peroxide comprised in the liquid feed stream in (i).

10. The process of claim 1, wherein: the catalyst comprising a titanium zeolite of structure type MFI is present in the reactor as fixed-bed catalyst; and the titanium zeolite of structure type MFI comprises titanium silicalite-1.

11. The process of claim 1, wherein: the oxygen selectivity of the epoxidation reaction is at most 1.2%, wherein the oxygen selectivity is defined as the molar amount of oxygen comprised in the effluent stream removed in (ii), relative to the molar amount of hydrogen peroxide comprised in the liquid feed stream; and the organic by-product selectivity of the epoxidation reaction is at most 9.0%, wherein the organic by-product selectivity is defined as the molar amount of hydrogen peroxide consumed to produce the molar amount of organic by-products comprised in the effluent stream removed in (ii), relative to the total molar amount of hydrogen peroxide consumed.

12. The process of claim 1, wherein the molar ratio of potassium relative to phosphorus in the at least one potassium salt of hydroxyethylidene diphosphonic acid ranges from 1:2 to 1.5:1.

13. The process of claim 1, wherein the molar ratio of potassium relative to phosphorus in the at least one potassium salt of hydroxyethylidene diphosphonic acid ranges from 0.75:1 to 1.25:1.

14. The process of claim 1, wherein the molar ratio of potassium relative to phosphorus in the at least one potassium salt of hydroxyethylidene diphosphonic acid ranges from 0.9:1 to 1.1:1.

15. The process of claim 1, wherein, in the liquid feed stream, the molar ratio of potassium comprised in the at least one potassium salt of hydroxyethylidene-diphosphonic acid relative to the hydrogen peroxide ranges from 1010.sup.6:1 to 70010.sup.6:1.

16. The process of claim 1, wherein, in the liquid feed stream, the molar ratio of potassium comprised in the at least one potassium salt of hydroxyethylidene-diphosphonic acid relative to the hydrogen peroxide ranges from 1010.sup.6:1 to 50010.sup.6:1.

17. The process of claim 1, wherein, in the liquid feed stream, the molar ratio of phosphorus in the liquid feed stream relative to phosphorus comprised in the at least one potassium salt of hydroxyethylidenediphosphonic acid ranges from 1.1:1 to 1:1.

18. The process of claim 1, wherein, in the liquid feed stream, the molar ratio of phosphorus in the liquid feed stream relative to phosphorus comprised in the at least one potassium salt of hydroxyethylidenediphosphonic acid ranges from 1.05:1 to 1:1.

19. The process of claim 1, wherein the liquid feed stream passed into the epoxidation reactor has a temperature ranging from 25 to 50 C.

20. The process of claim 1, wherein the liquid feed stream passed into the epoxidation reactor is at a pressure ranging from 15 to 25 bar.

Description

EXAMPLES

Reference Example 1: Preparation of a Titanium Containing Zeolite (TS-1)

(1) In a reaction vessel, 550 kg DI water were provided and stirred. 400 kg TPAOH (tetra-n-propylammonium hydroxide) were added under stirring. Stirring was continued for 1 h. The resulting mixture was transferred in a suitable vessel. The reaction vessel was washed twice with 2000 l DI water in total. In the washed reaction vessel, 300 kg TEOS (tetraethoxysilane) were provided and stirred. A mixture of 80 kg TEOS and 16 kg TEOT (tetraethyl orthotitanate) was added to the 300 kg TEOS. The remaining 340 kg TEOS were added.

(2) Subsequently, the TPAOH solution was added, and the resulting mixture was stirred for another hour. Then, the reaction vessel was heated and the ethanol obtained was separated by distillation. When the internal temperature of the vessel had reached 95 C., the reaction vessel was cooled. 1143 kg water were added to the resulting suspension in the vessel, and the mixture was stirred for another hour. Crystallization was performed at 175 C. within 24 h at autogenous pressure. The obtained titanium silicalite-1 crystals were separated, dried, and calcined at a temperature of 500 C. in air.

(3) The obtained powder and Walocel were mixed in a muller and mixed for 5 min. Within 10 min, the polystyrene dispersion was continuously added. Subsequently, 15 l Ludox AS-40 were continuously added. The resulting mixture was mixed for 5 min, and polyethylene oxide was continuously added within 15 min, followed by mixing for 10 min. Then, water was added. The formable mass was extruded through a matrix having circular holes with a diameter of 1.5 mm. The obtained strands were dried in a band drier at a temperature of 120 C. for 2 h and calcined at a temperature of 550 C. in lean air (100 m.sup.3/h air/100 m.sup.3/h nitrogen). The yield was 89 kg extrudates.

(4) For the subsequent water treatment of the extrudates, 880 kg DI water were filled in a respective stirred vessel, and the extrudates were added. At a pressure of 84 mbar, the vessel was heated to an internal temperature of from 139 to 143 C. The resulting pressure was in the range of from 2.1 to 2.5 bar. Water treatment was carried out for 36 h. The extrudates were separated by filtration, dried for 16 h at 123 C. in air, heated to a temperature of 470 C. with 2 C./min and kept at a temperature of 490 C. in air for 5 h. The yield was 81.2 kg.

Reference Example 2: Experimental Setup for Example 1 and Comparative Example 1

(5) A TS-1 catalyst as obtained according to Reference Example 1 above was loaded into a reaction tube with a length of 180 cm and a volume of 300 ml. The tube diameter was 0.75 inch (1.905 cm), with a wall thickness of 0.07 inch (0.19 cm). In the center of the reaction tube a smaller (0.125 inch (0.3175 cm)) tube was installed, containing the thermoelements for measuring the temperature over the catalyst bed. Feed-materials: 54 g/h Propene (liquid) 94 g/h 40% H.sub.2O.sub.2 Solvent: 390 g/h Methanol Buffer-solution: 4-8 g/h 0.3 weight-% K.sub.2HPO.sub.4/K.sub.2HEDP (flow adjusted to maintain 130-238 micromole K.sup.+/(mole H.sub.2O.sub.2))

(6) Propene was stored in 50 l gas bottles, containing dip tubes, facilitating the transfer to the mini-plant by means of 25 bar nitrogen pressure. The precise amount was measured using a Brunkhorst flow meter with a 0-500 g/h range and the flow is controlled by means of a Flowserve control-valve. Hydrogen peroxide was transferred into the reactor using a Grundfos pump DME2. The amount was determined using a balance. The measurement showed liters/minute. The respective buffer solution was fed to the reactor, using an HPLC pump. The precise amount was determined using a balance. For feeding the methanol a Lewa pump with a range of 0-1500 ml/h was used. Feed control was accomplished using a Lewa KMM. Nitrogen was fed using a Flowserve control-valve. The amount was measured using a Brunkhorst flow meter with a range of 0-200 Nl/h.

(7) The experiments were carried out at an absolute pressure of 20 bar. The temperature in the reactor was controlled to ensure a H.sub.2O.sub.2 conversion of approximately 90%. Typical start temperature was approximately 43 C. Then the temperature was slowly ramped up to approximately 60 C. final over a run-time of 600 hours. At the beginning of the run the reactor was cooled as the exothermic heat would overheat the reactor otherwise. Towards the end of the run the reactor was heated to reach a temperature of 60 C. All feed-materials were entering the reactor tube via a 0.25 inch (0.635 cm)-mixer. Feed direction was bottom to top.

(8) The reactor effluent was passed through a 2 micrometer filter to remove fine (catalyst) particles before it was passed into the first separator. The bottom level valve controlled a level of 25% in the first separator, while the upper pressure valve set a pressure of 20 bars over the entire upstream reaction system. The second separator was also operated at a liquid level of 25%, while the upper pressure valve reduced the pressure to 2 bars.

(9) This lower pressure served for allowing the flashing of unconverted propene, allowing a safe sample taking, and having an additional safety buffer. The two separators had a volume of 2 liters each and were kept at a temperature of 5 C., using cooling water. A nitrogen stream of 5 Nl/h was fed through the entire system (reactor.fwdarw.1.sup.st separator-2.sup.nd separator.fwdarw.vent-system) to maintain a sufficient gas flow in the direction of the vent to ascertain that traces of oxygen, formed by partial decomposition of H.sub.2O.sub.2 were flashed out and could be analyzed at the end of the vent pipe.

Example 1: Using HEDP as Buffer

(10) As a buffer, the di-potassium salt of hydroxyethane diphosphonic acid (K.sub.2HEDP (the trade name of the material is Dequest 2014 from Italmatch)) was used:

(11) ##STR00002##

(1-hydroxyethane-1,1-diyl)bis(phosphonic acid)

(12) At the first 620 hours on-stream time, a K.sub.2HEDP concentration of 208 micromole K.sup.+/mole-H.sub.2O.sub.2 was adjusted, and then reduced to 138 micromole K.sup.+/mole-H.sub.2O.sub.2. The following results were obtained (see Table 1 below):

(13) TABLE-US-00001 TABLE 1 Results of Example 1 Time T X-HP S-PO AA PM2 PM1 Ac PG O.sub.2 [h] [ C.] [%] [%] [%] [%] [%] [%] [%] [%] 23 43 98.6 94.6 0.41 2.09 2.09 0.14 0.72 0.1 47 43 93.2 96.1 0.43 1.30 1.36 0.14 0.66 0.2 71 43 94.3 96.0 0.44 1.38 1.40 0.22 0.58 0.2 95 43 93.2 95.6 0.55 1.49 1.43 0.31 0.57 0.25 119 44 92.6 95.6 0.57 1.47 1.44 0.29 0.59 0.25 143 44 92.6 95.6 0.54 1.56 1.47 0.26 0.53 0.25 167 44 89.9 95.6 0.58 1.48 1.45 0.35 0.56 0.3 191 45 90.6 96.0 0.65 1.29 1.23 0.40 0.47 0.3 215 47 90.5 95.7 0.64 1.35 1.35 0.41 0.52 0.3 239 47 83.0 96.1 0.59 1.21 1.18 0.45 0.48 0.3 263 49 85.1 95.8 0.66 1.30 1.24 0.49 0.51 0.3 287 51 89.0 95.5 0.72 1.37 1.37 0.43 0.57 0.35 311 51 90.6 95.3 0.78 1.47 1.44 0.43 0.57 0.4 335 51 92.6 95.0 1.01 1.57 1.51 0.41 0.53 0.35 359 51 91.2 95.0 1.20 1.42 1.36 0.47 0.54 0.35 383 51 86.6 94.8 1.19 1.43 1.49 0.47 0.60 0.4 407 53 88.8 94.4 1.19 1.58 1.62 0.51 0.65 0.45 431 53 88.4 94.2 1.31 1.62 1.65 0.57 0.68 0.5 455 54 89.5 94.3 1.29 1.60 1.61 0.56 0.67 0.55 479 55 87.2 94.2 1.28 1.63 1.56 0.59 0.69 0.55 503 57 88.1 94.4 1.23 1.57 1.54 0.57 0.68 0.6 527 58 89.2 94.4 1.20 1.58 1.53 0.55 0.70 0.6 551 58 89.2 94.4 1.24 1.56 1.49 0.60 0.69 0.65 575 58 92.6 94.2 1.21 1.63 1.69 0.58 0.73 0.7 599 58 92.1 93.9 1.26 1.74 1.77 0.59 0.78 0.75 623 58 89.2 93.8 1.24 1.78 1.76 0.60 0.79 0.8 647 58 91.9 92.0 1.42 2.22 2.57 0.69 1.06 0.6 671 58 91.9 91.4 1.44 2.54 2.81 0.65 1.21 0.6 695 58 91.8 91.1 1.43 2.64 2.89 0.65 1.25 0.65 Time Length of catalyst cycle/on-stream time in hours T [ C.] Cooling water temperature in degree Celsius X-HP [%] Conversion of hydrogen peroxide (HP) in weight percent S-PO [%] Hydrogen peroxide based selectivity to PO in mole-% AA [%] Hydrogen peroxide based selectivity to acetaldehyde in mole-% PM2 [%] Hydrogen peroxide based selectivity to 1-methoxy-2-propanol in mole-% PM1 [%] Hydrogen peroxide based selectivity to 2-methoxy-1-propanol in mole-% Ac [%] Hydrogen peroxide based selectivity to hydroxyacetone in mole-% PG [%] Hydrogen peroxide based selectivity to propylene glycol in mole-% O.sub.2 [%] Selectivity to oxygen in mole-% due to hydrogen peroxide decomposition HEDP Hydroxyethane diphosphonic acid DKP Di-potassium-phosphate K.sub.2HPO.sub.4 K.sub.2HEDP Di-potassium salt of HEDP

Comparative Example 1: Using K.SUB.2.HPO.SUB.4 .as Buffer

(14) The following results were obtained (see Table 2 below):

(15) TABLE-US-00002 TABLE 2 Results of Comparative Example 1 Time T X-HP S-PO AA PM2 PM1 Ac PG O.sub.2 [h] [ C.] [%] [%] [%] [%] [%] [%] [%] [%] 23 43 97.3 94.3 0.45 2.19 2.14 0.26 0.67 0.1 47 44 89.2 95.8 0.54 1.51 1.37 0.32 0.47 0.15 71 46 86.5 95.7 0.55 1.55 1.42 0.32 0.50 0.2 95 48 87.9 95.5 0.58 1.60 1.48 0.32 0.55 0.25 119 50 91.2 95.4 0.60 1.62 1.51 0.30 0.59 0.3 143 50 91.2 95.2 0.64 1.65 1.62 0.31 0.60 0.3 167 50 89.7 95.0 0.66 1.68 1.68 0.34 0.67 0.35 191 51 91.9 94.8 0.68 1.76 1.69 0.37 0.68 0.35 215 52 90.8 95.0 0.70 1.62 1.65 0.36 0.71 0.35 239 52 91.5 94.7 0.72 1.74 1.73 0.40 0.69 0.35 263 52 88.9 95.0 0.74 1.63 1.58 0.44 0.65 0.35 287 54 89.5 94.9 0.76 1.63 1.55 0.47 0.66 0.4 311 54 90.3 94.5 1.24 1.63 1.55 0.40 0.63 0.4 335 54 91.2 94.4 1.14 1.70 1.69 0.44 0.65 0.4 359 52 90.8 94.2 1.29 1.68 1.68 0.47 0.69 0.4 383 55 89.2 94.2 1.32 1.68 1.65 0.49 0.68 0.45 407 56 93.0 93.4 1.43 1.88 1.95 0.53 0.79 0.5 431 56 91.1 93.4 1.36 1.91 1.98 0.57 0.81 0.55 455 56 89.2 93.4 1.39 1.93 1.90 0.62 0.79 0.6 479 57 89.4 93.1 1.44 1.92 2.00 0.63 0.86 0.6 503 58 88.3 93.2 1.34 1.94 1.97 0.66 0.93 0.6 527 59 89.5 93.1 1.34 1.97 1.97 0.67 0.93 0.7 551 59 89.9 93.2 1.34 1.92 1.97 0.68 0.89 0.7 575 59 90.5 92.9 1.40 1.99 2.02 0.72 0.92 0.75 599 59 90.5 92.8 1.46 2.04 2.02 0.72 0.92 0.8 623 59 88.9 92.7 1.53 2.12 2.06 0.68 0.93 0.8 647 60 90.3 92.7 1.53 2.12 2.04 0.71 0.91 0.85 671 60 89.9 92.5 1.57 2.18 2.06 0.73 0.92 0.9 695 61 91.2 92.5 1.62 2.18 2.06 0.73 0.94 0.95 Time Length of catalyst cycle/on-stream time in hours T [ C.] Cooling water temperature in degree Celsius X-HP [%] Conversion of hydrogen peroxide (HP) in weight percent S-PO [%] Hydrogen peroxide based selectivity to PO in mole-% AA [%] Hydrogen peroxide based selectivity to acetaldehyde in mole-% PM2 [%] Hydrogen peroxide based selectivity to 1-methoxy-2-propanol in mole-% PM1 [%] Hydrogen peroxide based selectivity to 2-methoxy-1-propanol in mole-% Ac [%] Hydrogen peroxide based selectivity to hydroxyacetone in mole-% PG [%] Hydrogen peroxide based selectivity to propylene glycol in mole-% O.sub.2 [%] Selectivity to oxygen in mole-% due to hydrogen peroxide decomposition HEDP Hydroxyethane diphosphonic acid DKP Di-potassium-phosphate K.sub.2HPO.sub.4 K.sub.2HEDP Di-potassium salt of HEDP

Results of Example 1 and Comparative Example 1

(16) The graphs according to FIGS. 1 to 5 clearly illustrate the advantage of using K.sub.2HEDP in comparison with the standard buffer K.sub.2HPO.sub.4 in terms of decreased by-product selectivities and increased main product PO selectivity as well as extended catalyst activity/lower deactivation based on the cooling water temperature required to maintain a conversion of H.sub.2O.sub.2 of about 90%.

Reference Example 3: Experimental Setup for Example 2

(17) A TS-1 catalyst as obtained according to Reference Example 1 above (3497 g) was loaded into a reaction tube with the length of 1200 cm and a volume of 12.2 l. The tube outside diameter was 1.315 inch (3.34 cm), with a wall thickness of 0.065 inch (0.1651 cm). In the center of the reaction tube a smaller ( inch (0.3175 cm)) tube was installed, containing the thermoelements for measuring the temperature over the catalyst bed. Feed-materials: 2.06 kg/h Propene (liquid) 3.25 kg/h 40 wt % H.sub.2O.sub.2 Solvent: 11.97 kg/h Methanol Buffer-solution: 0.76 ml/min/h 1.0 weight % K.sub.2HPO.sub.4 (baseline)

(18) Propene was sourced by pipeline. The precise amount was measured using a MicroMotion Coriolis mass flow sensor (model CMF010M324NQBUEZZZ). Hydrogen peroxide flow was measured by a MicroMotion Coriolis mass flow sensor (model CMF010M324NQBUEZZZ). The measurement was reported in lb/hr. The nominal concentration of the hydrogen peroxide solution was 40 weight-%. The buffer solution was fed to the reactor, using an HPLC pump. The precise amount was determined using a balance. The flow of methanol to the reactor was measured with a MicroMotion Coriolis mass flow sensor (model CMF010M324NQBUEZZZ). Nitrogen was fed as a block utility at a specification of 99.999 mol-%. The flowrate was measured using a Brooks thermal mass flow sensors (model SLAMF60S1BAA0K2A3) and reported in units of standard cubic feet per hour (scfh).

(19) The experiments were conducted at an absolute pressure of 20 bar. The temperature in the reactor was controlled to ensure a H.sub.2O.sub.2 conversion of approximately 90%. Typical start temperature was approximately 35 C. Then the temperature was slowly ramped up to approximately 70 C. final over a run-time of 550 hours. At the beginning of the run the reactor was cooled as the exothermic heat would overheat the reactor otherwise. Towards the end of the run the reactor was heated to reach a temperature of 70 C. The buffer solution was injected into the methanol feed stream, and the resulting mixture passed over a 21-element 0.25 inch (0.635 cm) Koflo static mixer. Hydrogen peroxide and propylene were injected into the methanol stream, followed by another 21-element 0.25 inch (0.635 cm) static mixer. Feed direction was bottom to top.

(20) The reactor effluent was fed to a first separator, where vapor and liquid were allowed to disengage. The vapor stream from the condenser was diluted with nitrogen and passed through a condenser to remove the more readily condensable species in order to prevent condensation in the lines. The liquid from the separator was fed to a distillation column consisting of a packed section (3-schedule 40 pipe129) and sump (6-schedule 40 pipe by 50). A constant level was maintained in the sump of this column by constant draw-off of liquid. The sump contained an electrical heater for heat input to the column. Vapor from the top of the column was condensed in the same cooler as the vent from the separator drum. Condensed material was collected in a drum and either returned to the column as reflux or forwarded for further processing. The pressure for the entire upstream reaction system was set by means of a pressure control valve on the overhead vapor stream from the column.

(21) The mass flow rate of the vapor stream from the column was measured using a MicroMotion Coriolis meter (model CMF010M324NQBUEZZZ). The flow was reported in units of pounds per hour. The composition of this stream was analyzed by a gas chromatograph (GC) and reported in on a molar basis. The mass flow rate of liquid distillate from the column was measured using a MicroMotion Coriolis mass sensor (model CMF010M324NQBUEZZZ). This stream was sampled routinely. The concentration of the organic components was determined by GC analysis. The water concentration was determined by Karl Fischer titration.

(22) The mass flow rate of the bottoms stream from the column was measured using a MicroMotion Coriolis mass sensor (model CMF010M324NQBUEZZZ). This stream was sampled routinely. The concentration of the organic components was determined by GC analysis. The water concentration was determined by Karl Fischer titration. The concentration of unreacted hydrogen peroxide was determined by a colorimetric method.

(23) The flow rates and composition analyses of all the streams were used to determine conversions and selectivities of species of interest.

Example 2

(24) During this experiment, dipotassium hydrogen phosphate (K.sub.2HPO.sub.4) was used as the buffer for the baseline experiments, while mixtures of K.sub.2HPO.sub.4 and 1-hydroxyethane 1,1-diphosphonic acid (HEDP) were used (K.sub.2HPO.sub.4, USP grade, from Fisher Scientific, HEDP sourced as a 60 wt % aqueous solution from Spectrum Laboratory Products) to investigate the effect of HEDP as part of the buffer system. See Table 3 below for the specific buffer combinations that were tested. All buffer solutions fed to the reactor contained 1 wt % K.sub.2HPO.sub.4.

(25) TABLE-US-00003 TABLE 3 Specific buffer combinations tested HEDP Additive Time K.sub.2HPO.sub.4 [HEDP] on Ratio Solution Period on K.sub.2HPO.sub.4 Ratio [mol HEDP a 40% H.sub.2O.sub.2 [mol Buffer flow rate Line Flow K.sup.+/mol Flow feed basis HEDP/mol Solution [ml/min] [hrs] [gmol/hr)] H.sub.2O.sub.2] [gmol/hr] [ppmw] H.sub.2O.sub.2] 1% 0.76 0 to 242, 0.00262 137 0 0 0 K.sub.2HPO.sub.4 425 to 547 1% 0.90 246 to 0.00310 162 0.00066 42 17 K.sub.2HPO.sub.4 + 353 0.25% HEDP 1% 1.1 353 to 0.00379 198 0.00160 102 42 K.sub.2HPO.sub.4 + 384 0.50% HEDP 1% 1.25 384 to 0.00430 226 0.00182 117 48 K.sub.2HPO.sub.4 + 425 0.50% HEDP

(26) Data was not reported for the time period from 353 to 384 hours on line because the process did not have time to achieve steady state. The following results were obtained (see Table 4 below):

(27) TABLE-US-00004 TABLE 4 Results of Comparative Example 1 Time T X-HP S-PO AA PM2 PM1 Ac PG O2 ROOH [h] [ C. ] [%] [%] [%] [%] [%] [%] [%] [%] [%] Buffer 13 34.9 88.6 96.9 0.24 1.02 1.27 0.04 0.45 0.10 1.37 137 mol K.sup.+/mol H.sub.2O.sub.2 25 44.5 90.6 97.6 0.03 0.90 1.04 0.02 0.25 0.15 0.76 137 mol K.sup.+/mol H.sub.2O.sub.2 37 47.0 88.9 97.2 0.09 1.04 1.21 0.02 0.32 0.14 0.96 137 mol K.sup.+/mol H.sub.2O.sub.2 49 48.9 89.3 97.1 0.09 1.06 1.23 0.02 0.32 0.15 0.95 137 mol K.sup.+/mol H.sub.2O.sub.2 61 50.5 88.7 96.9 0.08 1.16 1.35 0.02 0.34 0.17 1.03 137 mol K.sup.+/mol H.sub.2O.sub.2 73 52.0 89.0 96.8 0.07 1.19 1.39 0.02 0.37 0.22 1.09 137 mol K.sup.+/mol H.sub.2O.sub.2 85 52.5 89.0 96.7 0.07 1.20 1.41 0.02 0.37 0.21 1.12 137 mol K.sup.+/mol H.sub.2O.sub.2 97 53.5 90.4 96.7 0.07 1.20 1.41 0.02 0.38 0.21 1.15 137 mol K.sup.+/mol H.sub.2O.sub.2 110 55.0 90.2 96.6 0.07 1.24 1.46 0.02 0.40 0.22 1.19 137 mol K.sup.+/mol H.sub.2O.sub.2 121 55.5 90.6 96.4 0.06 1.29 1.53 0.02 0.43 0.25 1.36 137 mol K.sup.+/mol H.sub.2O.sub.2 134 56.5 90.4 96.4 0.07 1.31 1.55 0.02 0.44 0.24 1.39 137 mol K.sup.+/mol H.sub.2O.sub.2 145 56.5 91.1 96.3 0.07 1.33 1.56 0.02 0.44 0.25 1.40 137 mol K.sup.+/mol H.sub.2O.sub.2 158 57.3 90.4 96.4 0.07 1.33 1.57 0.02 0.44 0.22 1.42 137 mol K.sup.+/mol H.sub.2O.sub.2 169 58.0 90.5 96.1 0.11 1.39 1.64 0.02 0.48 0.26 1.50 137 mol K.sup.+/mol H.sub.2O.sub.2 182 58.2 90.2 95.8 0.10 1.52 1.79 0.02 0.52 0.28 1.51 137 mol K.sup.+/mol H.sub.2O.sub.2 193 59.0 89.6 96.1 0.08 1.42 1.67 0.02 0.48 0.26 1.51 137 mol K.sup.+/mol H.sub.2O.sub.2 205 59.0 89.9 96.2 0.08 1.39 1.62 0.02 0.47 0.26 1.44 137 mol K.sup.+/mol H.sub.2O.sub.2 205 59.0 89.9 96.2 0.08 1.39 1.62 0.02 0.47 0.26 1.44 137 mol K.sup.+/mol H.sub.2O.sub.2 217 60.0 89.1 96.0 0.10 1.44 1.70 0.02 0.51 0.27 1.62 137 mol K.sup.+/mol H.sub.2O.sub.2 229 60.0 89.4 96.0 0.08 1.42 1.67 0.02 0.50 0.32 1.58 137 mol K.sup.+/mol H.sub.2O.sub.2 241 61.0 89.0 95.8 0.10 1.50 1.75 0.02 0.53 0.30 1.63 137 mol K.sup.+/mol H.sub.2O.sub.2 253 61.5 89.7 95.8 0.09 1.54 1.84 0.02 0.57 0.16 1.74 (162 mol K.sup.+ + 17 mol HEDP)/mol H.sub.2O.sub.2 274 61.5 91.0 95.9 0.09 1.48 1.79 0.02 0.55 0.18 1.68 (162 mol K.sup.+ + 17 mol HEDP)/mol H.sub.2O.sub.2 286 61.5 89.8 95.7 0.10 1.53 1.89 0.02 0.59 0.21 1.70 (162 mol K.sup.+ + 17 mol HEDP)/mol H.sub.2O.sub.2 298 62.6 89.2 95.4 0.10 1.61 1.98 0.02 0.63 0.22 1.85 (162 mol K.sup.+ + 17 mol HEDP)/mol H.sub.2O.sub.2 310 63.0 88.7 95.4 0.10 1.64 2.02 0.02 0.62 0.22 1.82 (162 mol K.sup.+ + 17 mol HEDP)/mol H.sub.2O.sub.2 322 64.0 89.7 95.3 0.10 1.67 2.03 0.02 0.63 0.22 1.81 (162 mol K.sup.+ + 17 mol HEDP)/mol H.sub.2O.sub.2 335 64.0 89.7 95.4 0.10 1.65 2.02 0.02 0.64 0.18 1.86 (162 mol K.sup.+ + 17 mol HEDP)/mol H.sub.2O.sub.2 347 64.5 89.3 95.4 0.09 1.65 2.00 0.02 0.63 0.18 1.86 (162 mol K.sup.+ + 17 mol HEDP)/mol H.sub.2O.sub.2 392 66.0 88.5 94.8 0.11 1.83 2.32 0.03 0.76 0.19 1.38 (226 mol K.sup.+ + 48 mol HEDP)/mol H.sub.2O.sub.2 404 67.0 89.7 94.7 0.10 1.86 2.32 0.02 0.76 0.23 1.39 (226 mol K.sup.+ + 48 mol HEDP)/mol H.sub.2O.sub.2 416 67.5 89.5 94.7 0.10 1.87 2.34 0.02 0.77 0.19 1.41 (226 mol K.sup.+ + 48 mol HEDP)/mol H.sub.2O.sub.2 427 67.9 89.9 94.4 0.11 1.98 2.52 0.02 0.82 0.18 1.51 137 mol K.sup.+/mol H.sub.2O.sub.2 439 68.0 89.0 94.7 0.13 1.83 2.20 0.03 0.76 0.36 1.38 137 mol K.sup.+/mol H.sub.2O.sub.2 451 69.0 90.3 94.9 0.10 1.79 2.08 0.02 0.72 0.36 1.49 137 mol K.sup.+/mol H.sub.2O.sub.2 463 69.1 89.5 94.8 0.15 1.78 2.09 0.03 0.74 0.40 1.54 137 mol K.sup.+/mol H.sub.2O.sub.2 475 69.5 89.8 94.9 0.11 1.77 2.04 0.03 0.70 0.40 1.45 137 mol K.sup.+/mol H.sub.2O.sub.2 487 70.0 90.2 94.7 0.14 1.83 2.11 0.03 0.74 0.45 1.53 137 mol K.sup.+/mol H.sub.2O.sub.2 499 70.0 89.9 94.8 0.12 1.79 2.05 0.03 0.72 0.47 1.48 137 mol K.sup.+/mol H.sub.2O.sub.2 511 70.0 89.3 94.6 0.13 1.84 2.11 0.03 0.75 0.48 1.57 137 mol K.sup.+/mol H.sub.2O.sub.2 523 70.5 90.8 94.8 0.10 1.79 2.05 0.02 0.73 0.45 1.54 137 mol K.sup.+/mol H.sub.2O.sub.2 535 70.5 89.2 94.6 0.15 1.84 2.12 0.03 0.76 0.48 1.60 137 mol K.sup.+/mol H.sub.2O.sub.2 547 71.0 88.9 94.4 0.13 1.94 2.22 0.03 0.79 0.51 1.65 137 mol K.sup.+/mol H.sub.2O.sub.2 Time Length of catalyst cycle/on-stream time in hours T [ C.] Cooling water temperature in degree Celsius X-HP [%] Conversion of hydrogen peroxide (HP) in weight percent S-PO [%] Hydrogen peroxide based selectivity to PO in mole-% AA [%] Hydrogen peroxide based selectivity to acetaldehyde in mole-% PM2 [%] Hydrogen peroxide based selectivity to 1-methoxy-2-propanol in mole-% PM1 [%] Hydrogen peroxide based selectivity to 2-methoxy-1-propanol in mole-% Ac [%] Hydrogen peroxide based selectivity to hydroxyacetone in mole-% PG [%] Hydrogen peroxide based selectivity to propylene glycol in mole-% O.sub.2 [%] Selectivity to oxygen in mole-% due to hydrogen peroxide decomposition HEDP Hydroxyethane diphosphonic acid DKP Di-potassium-phosphate K.sub.2HPO.sub.4 K.sub.2HEDP Di-potassium salt of HEDP

Results of Example 3

(28) The graphs according to FIGS. 6 to 10 illustrate the advantage of using HEDP in comparison with the standard buffer K.sub.2HPO.sub.4 in terms of decreased by-product selectivities to oxygen. No negative effects were observed on other process measures.

Reference Example 4: Experimental Setup for Comparative Examples 2, 3 and 4 (Micro Plant)

(29) A TS-1 catalyst as obtained according to Reference Example 1 above (140 g) was loaded into a reaction tube with the length of 180 cm and a volume of 300 ml. The tube outside diameter was inch (1.905 cm), with a wall thickness of 0.07 inch (0.19 cm). In the center of the reaction tube a smaller ( inch (0.3175 cm)) tube was installed, containing the thermoelements for measuring the temperature over the catalyst bed. Feed-materials: 54 g/h Propene (liquid) 94 g/h 40% H.sub.2O.sub.2 Solvent: 390 g/h Methanol Buffer-solution: 0.054 g/min 1.2 weight % K-ATMP, 0.027 g/min 1.2 weight % K.sub.2-ATMP (comparative example 2); 0.052 g/min 0.5 weight % (NH.sub.4).sub.4-HEDP (comparative example 3); 0.051 g/min 0.5 weight % (NH.sub.4).sub.6-ATMP (comparative example 4); (flow adjusted to maintain 130-238 mole K.sup.+/mole-H.sub.2O.sub.2 or 130-238 mole NH.sub.4.sup.+/mole-H.sub.2O.sub.2 depending on which buffer salt was used)

(30) Propene was stored in 50 l gas-bottles, containing dip tubes, facilitating the transfer to the mini-plant by means of 25 bar nitrogen pressure. The precise amount was measured using a Brunkhorst flow meter with a 0-500 g/h range and the flow was controlled by means of a Flowserve control-valve. Hydrogen peroxide was transferred into the reactor using a Grundfos pump DME.sub.2. The amount was determined using a balance. The measurement showed liters/minute. The buffer solution was fed to the reactor using an HPLC pump. The precise amount was determined using a balance. For feeding the methanol a Lewa pump with a range of 0-1500 ml/h was used. Feed control was accomplished using a Lewa KMM. Nitrogen was fed using a Flowserve control-valve. The amount was measured using a Brunkhorst flow meter with a range of 0-200 Nl/h.

(31) The experiments were conducted at an absolute pressure of 20 bar. The temperature in the reactor was controlled to ensure a H.sub.2O.sub.2 conversion of approximately 90%. Typical start temperature was approximately 43 C. Then the temperature was slowly ramped up to approximately 60 C. final over a run-time of 600 hours. At the beginning of the run the reactor was cooled as the exothermic heat would overheat the reactor otherwise. Towards the end of the run the reactor was heated to reach a temperature of 60 C.

(32) All feed-materials were introduced into the reactor tube via a inch (0.6 cm)-mixer. Feed direction was bottom to top.

(33) The reactor effluent was passed through a 2 micrometer filter to remove fine (catalyst) particles before it was passed into the first separator. The bottom level valve controlled a level of 25% in the first separator, while the upper pressure valve set a pressure of 20 bars over the entire upstream reaction system. The second separator was also operated at a liquid level of 25%, while the upper pressure valve reduced the pressure to 2 bars.

(34) This lower pressure served for allowing the flashing of unconverted propene, allowing a safe sample taking, and having an additional safety buffer. The two separators had a volume of 2 liters each and were kept at a temperature of 5 C., using cooling water. A nitrogen stream of 5 Nl/h was fed through the entire system (reactor.fwdarw.1.sup.st separator.fwdarw.2.sup.nd separator.fwdarw.vent-system) to maintain a sufficient gas flow in the direction of the vent to ascertain that traces of oxygen, formed by partial decomposition of H.sub.2O.sub.2 were flashed out and could be analyzed at the end of the vent pipe.

(35) The samples were analyzed routinely. The concentration of the organic components was determined by GC analysis. The water concentration was determined by Karl Fischer titration. The concentration of unreacted hydrogen peroxide was determined by a colorimetric method. The flow rates and composition analyses of all the streams were used to determine conversions and selectivities of species of interest.

Reference Example 5: Synthesis of Potassium Salts of Aminotris (Methylenephosphonic Acid) (ATMP)

(36) K-ATMP was prepared by adding 10-g of ATMP into 10-g of demineralized water, adding under continuous stirring 1.87 g of solid KOH and diluting this solution with additional demineralized water until a weight of 100-g was achieved.

(37) K.sub.2-ATMP was prepared by adding 10-g of ATMP into 10-g of demineralized water, adding under continuous stirring 3-g of solid KOH and diluting this solution with additional demineralized water until a weight of 100-g was achieved.

Reference Example 6: Synthesis of Ammonium Salts of HEDP

(38) NH.sub.4HEDP was prepared based on adding 10 g of a 60 weight-% solution of etidronic acid (HEDP) to 100 g of water, and then slowly dosing 5.1 g of a 20% NHOH solution under continuous stirring.

Reference Example 7: Synthesis of the Ammonium Salt of ATMP

(39) A quantity of 20 g of ATMP was dissolved in 20 g of water, and a 33 weight-% solution of NH.sub.4OH was added until a pH of 6 was achieved. The resulting solution was evaporated and 24.3 g of a solid residue were obtained.

Comparative Example 2: Using K-ATMP as Buffer

(40) During this experiment, K-ATMP was used as the buffer. A K-ATMP concentration of 130 mole K.sup.+/mole-H.sub.2O.sub.2 was adjusted until 120-hrs on-stream time, when the K-ATMP was replaced by the less acidic K.sub.2-ATMP. The results are shown in Table 5 and FIGS. 11 to 13.

(41) TABLE-US-00005 TABLE 5 Results of Comparative Example 2-Experimental data for a potassium salt of ATMP Time X-HP [%] S-PO [%] AA [%] PM2 [%] PM1 [%] Ac [%] PG [%] O.sub.2 [%] 24 100 59.3 59.3 1.03 12.21 22.9 0.23 4.34 48 96 70.8 67.9 1.4 9.32 13.99 0.31 4.2 72 89.4 78.4 70.1 1.75 6.64 9.69 0.47 3.07 96 84 79.7 66.9 1.79 6.2 8.84 0.54 2.92 120 89 79.2 70.5 1.66 6.2 9.09 0.5 3.35 144 88.1 89.9 79.2 1.25 2.95 4.09 0.4 1.38 168 86.8 89.4 77.6 1.19 3.18 4.39 0.38 1.46 192 83.9 88.4 74.1 1.26 3.38 4.88 0.4 1.7 216 86.6 86.1 74.5 1.23 4.31 5.64 0.51 2.25 240 90.5 84.4 76.4 1.46 4.76 6.46 0.46 2.48 264 88.6 84 74.4 1.48 4.87 6.7 0.48 2.5 288 87.9 82.6 72.6 1.76 5.25 7.23 0.61 2.59 312 86.3 81.7 70.5 1.87 5.5 7.52 0.66 2.75 336 86.3 81.8 70.6 1.83 5.45 7.32 0.66 2.91 360 88 81.8 72 1.88 5.51 7.34 0.67 2.83 384 87.6 81.2 71.1 1.93 5.6 7.61 0.65 3.04 Time Length of catalyst cycle/on-stream time in hours X-HP [%] Conversion of hydrogen peroxide (HP) in weight percent S-PO [%] Hydrogen peroxide based selectivity to PO in mole-% AA [%] Hydrogen peroxide based selectivity to acetaldehyde in mole-% PM2 [%] Hydrogen peroxide based selectivity to 1-methoxy-2-propanol in mole-% PM1 [%] Hydrogen peroxide based selectivity to 2-methoxy-1-propanol in mole-% Ac [%] Hydrogen peroxide based selectivity to hydroxyacetone in mole-% PG [%] Hydrogen peroxide based selectivity to propylene glycol in mole-% O.sub.2 [%] Selectivity to oxygen in mole-% due to hydrogen peroxide decomposition

Results of Comparative Example 2

(42) The graphs according to FIGS. 11 to 13 clearly illustrate the disadvantage of using the K-ATMP as well as K.sub.2-ATMP in terms of increased by-product selectivities and decreased main product PO selectivity, i.e. K-ATMP as well as K.sub.2-ATMP resulted in substantially lower PO-selectivities, and consequently higher PO by-product selectivities when comparing to the results of K.sub.2HEDP as of Example 1 (due to the large deviation of the results, the K.sub.2HEDP results were not graphically represented in FIGS. 11 to 13).

Comparative Example 3: Using NH.SUB.4.HEDP as Buffer

(43) During this experiment, (NH.sub.4).sub.4-HEDP was used as the buffer. An (NH.sub.4).sub.4-HEDP concentration of 208 mole NH.sub.4.sup.+/mole-H.sub.2O.sub.2 was adjusted and the results are summarized in the table 6 below and graphically represented in FIGS. 14 to 16.

(44) TABLE-US-00006 TABLE 6 Results of Comparative Example 3-Experimental data for NH.sub.4HEDP Time T [ C.] X-HP [%] S-PO [%] AA [%] PM2 [%] PM1 [%] Ac [%] PG [%] O.sub.2 [%] 23 43 99.9 94.6 0.41 2.12 2.60 0.07 0.72 0.10 47 43 95.9 96.1 0.55 1.77 1.91 0.12 0.59 0.15 71 43 90.1 96.0 0.61 1.51 1.66 0.15 0.52 0.15 95 45 91.5 95.6 0.62 1.53 1.67 0.16 0.51 0.15 119 46 91.5 95.6 0.62 1.54 1.68 0.16 0.54 0.15 143 48 87.2 95.6 0.74 1.58 1.82 0.21 0.57 0.15 167 51 90.7 95.6 0.73 1.76 1.98 0.22 0.70 0.15 191 53 88.5 96.0 0.81 1.86 2.14 0.25 0.77 0.15 215 57 87.1 95.7 0.92 1.94 2.24 0.28 0.82 0.15 239 56 87.1 96.1 0.94 2.10 2.33 0.28 0.84 0.20 263 58 92.4 95.8 1.11 2.20 2.37 0.30 0.88 0.20 287 58 90.5 95.5 1.14 2.32 2.66 0.28 0.96 0.20 311 59 86.5 95.3 1.36 2.24 2.63 0.36 0.96 0.20 Time Length of catalyst cycle/on-stream time in hours X-HP [%] Conversion of hydrogen peroxide (HP) in weight percent S-PO [%] Hydrogen peroxide based selectivity to PO in mole-% AA [%] Hydrogen peroxide based selectivity to acetaldehyde in mole-% PM2 [%] Hydrogen peroxide based selectivity to 1-methoxy-2-propanol in mole-% PM1 [%] Hydrogen peroxide based selectivity to 2-methoxy-1-propanol in mole-% Ac [%] Hydrogen peroxide based selectivity to hydroxyacetone in mole-% PG [%] Hydrogen peroxide based selectivity to propylene glycol in mole-% O.sub.2 [%] Selectivity to oxygen in mole-% due to hydrogen peroxide decomposition

Results of Comparative Example 3

(45) The graphs according to FIGS. 14 to 16 clearly illustrate the disadvantage of using NH.sub.4HEDP in terms of increased by-product selectivities and decreased main product PO selectivity, i.e. NH.sub.4HEDP resulted in substantially lower PO-selectivities, and consequently higher PO by-product selectivities when comparing to the results of K.sub.2HEDP as of Example 1.

Comparative Example 4: Using an Ammonium Salt of ATMP as Buffer

(46) During this experiment, [NH.sub.4].sub.6ATMP was used as the buffer. An [NH.sub.4].sub.6ATMP concentration of 208 mole NH.sub.4.sup.+/mole-H.sub.2O.sub.2 was adjusted to obtain the results shown in Table 7 and graphically represented in FIGS. 17 to 20.

(47) TABLE-US-00007 TABLE 7 Results of Comparative Example 4-Experimental data for [NH.sub.4].sub.6ATMP Time X-HP [%] S-PO [%] AA [%] PM2 [%] PM1 [%] Ac [%] PG [%] O.sub.2 [%] 24 100.0 91.2 0.41 2.65 4.88 0.08 0.78 0.1 48 100.0 94.7 0.58 1.85 2.22 0.10 0.54 0.1 72 92.2 95.5 0.88 1.47 1.53 0.16 0.47 0.1 96 88.5 95.2 1.07 1.46 1.55 0.21 0.47 0.1 120 89.2 95.1 1.10 1.51 1.58 0.22 0.49 0.1 144 89.2 95.2 0.97 1.54 1.56 0.25 0.51 0.1 168 89.5 94.5 1.10 1.70 1.85 0.26 0.64 0.1 192 89.4 94.3 1.09 1.77 1.92 0.25 0.67 0.1 216 91.2 94.0 1.04 1.88 2.05 0.25 0.74 0.1 240 91.2 93.7 1.13 1.97 2.21 0.24 0.79 0.1 264 91.2 93.6 1.09 2.02 2.25 0.24 0.81 0.1 288 91.2 93.6 1.09 2.02 2.25 0.24 0.81 0.1 312 91.2 93.5 1.14 2.03 2.26 0.26 0.81 0.1 336 88.5 93.3 1.17 2.06 2.29 0.31 0.85 0.1 360 89.2 93.2 1.20 2.09 2.35 0.29 0.87 0.1 384 89.2 92.7 1.25 2.21 2.56 0.33 0.93 0.1 408 88.6 92.1 1.26 2.40 2.80 0.33 1.07 0.1 432 88.0 92.0 1.24 2.47 2.91 0.31 1.10 0.1 456 88.0 91.8 1.25 2.51 2.95 0.32 1.12 0.15 480 87.6 92.0 1.25 2.47 2.91 0.31 1.10 0.15 504 88.5 91.8 1.22 2.50 2.97 0.33 1.12 0.15 528 90.5 91.6 1.25 2.56 3.09 0.33 1.12 0.15 Time Length of catalyst cycle/on-stream time in hours X-HP [%] Conversion of hydrogen peroxide (HP) in weight percent S-PO [%] Hydrogen peroxide based selectivity to PO in mole-% AA [%] Hydrogen peroxide based selectivity to acetaldehyde in mole-% PM2 [%] Hydrogen peroxide based selectivity to 1-methoxy-2-propanol in mole-% PM1 [%] Hydrogen peroxide based selectivity to 2-methoxy-1-propanol in mole-% Ac [%] Hydrogen peroxide based selectivity to hydroxyacetone in mole-% PG [%] Hydrogen peroxide based selectivity to propylene glycol in mole-% O.sub.2 [%] Selectivity to oxygen in mole-% due to hydrogen peroxide decomposition

Results of Comparative Example 4

(48) The graphs according to FIGS. 17 to 20 clearly illustrate the disadvantage of using [NH.sub.4].sub.6ATMP in terms of increased by-product selectivities and decreased main product PO selectivity, i.e. [NH.sub.4].sub.6ATMP resulted in substantially lower PO-selectivities, and consequently higher PO by-product selectivities when comparing to the results of K.sub.2HEDP as of Example 1.

SHORT DESCRIPTION OF THE FIGURES

(49) FIG. 1 shows, on the Y axis, the selectivity of 2-methoxy-1-propanol (solid circles: K.sub.2HPO.sub.4, rectangles: K.sub.2HEDP as buffer). The x axis shows the run time in h.

(50) FIG. 2 shows, on the Y axis, the H.sub.2O.sub.2-based selectivity to PO (solid circles: K.sub.2HPO.sub.4, rectangles: K.sub.2HEDP as buffer). The x axis shows the run time in h.

(51) FIG. 3 shows, on the Y axis, the oxygen selectivity (solid circles: K.sub.2HPO.sub.4, rectangles: K.sub.2HEDP as buffer). The x axis shows the run time in h.

(52) FIG. 4 shows, on the Y axis, the cooling water temperature (solid circles: K.sub.2HPO.sub.4, rectangles: K.sub.2HEDP as buffer). The x axis shows the run time in h.

(53) FIG. 5 shows, on the Y axis, the H.sub.2O.sub.2 conversion (solid circles: K.sub.2HPO.sub.4, rectangles: K.sub.2HEDP as buffer). The x axis shows the run time in h.

(54) FIG. 6 shows, on the Y axis, the selectivity of 2-methoxy-1-propanol. The x axis shows the run time in h.

(55) FIG. 7 shows, on the Y axis, the H.sub.2O.sub.2-based selectivity to PO. The x axis shows the run time in h.

(56) FIG. 8 shows, on the Y axis, the oxygen selectivity. The x axis shows the run time in h.

(57) FIG. 9 shows, on the Y axis, the cooling water temperature. The x axis shows the run time in h.

(58) FIG. 10 shows, on the Y axis, the H.sub.2O.sub.2 conversion. The x axis shows the run time in h.

(59) FIG. 11 shows, on the Y axis, the selectivity of 2-methoxy-1-propanol from Comparative Example 2 (rectangles: K-ATMP as buffer, triangles: K.sub.2-ATMP as buffer). The x axis shows the run time in h (HOS, hours on stream).

(60) FIG. 12 shows, on the Y axis, the selectivity of 1-methoxy-2-propanol from Comparative Example 2 (rectangles: K-ATMP as buffer, triangles: K.sub.2-ATMP as buffer). The x axis shows the run time in h (HOS, hours on stream).

(61) FIG. 13 shows, on the Y axis, the H.sub.2O.sub.2-based selectivity to PO from Comparative Example 2 (rectangles: K-ATMP as buffer, triangles: K.sub.2-ATMP as buffer). The x axis shows the run time in h (HOS, hours on stream).

(62) FIG. 14 shows, on the Y axis, the selectivity of 2-methoxy-1-propanol from Example 1, Comparative Example 1 and Comparative Example 3 (solid circles: K.sub.2HPO.sub.4, crosses: K.sub.2HEDP, rectangles: NH.sub.4HEDP as buffer). The x axis shows the run time in h (HOS, hours on stream).

(63) FIG. 15 shows, on the Y axis, the selectivity to oxygen from Example 1, Comparative Example 1 and Comparative Example 3 (solid circles: K.sub.2HPO.sub.4, crosses: K.sub.2HEDP, rectangles: NH.sub.4HEDP as buffer). The x axis shows the run time in h (HOS, hours on stream).

(64) FIG. 16 shows, on the Y axis, the H.sub.2O.sub.2-based selectivity to PO from Example 1, Comparative Example 1 and Comparative Example 3 (solid circles: K.sub.2HPO.sub.4, crosses: K.sub.2HEDP, rectangles: NH.sub.4HEDP as buffer). The x axis shows the run time in h (HOS, hours on stream).

(65) FIG. 17 shows, on the Y axis, the selectivity of 2-methoxy-1-propanol from Example 1, Comparative Example 1 and Comparative Examples 3 and 4 (solid circles: K.sub.2HPO.sub.4, crosses: K.sub.2HEDP, rectangles: NH.sub.4HEDP as buffer, triangles: [NH.sub.4].sub.xATMP as buffer). The x axis shows the run time in h (HOS, hours on stream).

(66) FIG. 18 shows, on the Y axis, the selectivity of 1-methoxy-2-propanol from Example 1, Comparative Example 1 and Comparative Examples 3 and 4 (solid circles: K.sub.2HPO.sub.4, crosses: K.sub.2HEDP, rectangles: NH.sub.4HEDP as buffer, triangles: [NH.sub.4].sub.xATMP as buffer). The x axis shows the run time in h (HOS, hours on stream).

(67) FIG. 19 shows, on the Y axis, the selectivity to oxygen from Example 1, Comparative Example 1 and Comparative Examples 3 and 4 (solid circles: K.sub.2HPO.sub.4, crosses: K.sub.2HEDP, rectangles: NH.sub.4HEDP as buffer, triangles: [NH.sub.4].sub.xATMP as buffer). The x axis shows the run time in h (HOS, hours on stream).

(68) FIG. 20 shows, on the Y axis, the H.sub.2O.sub.2-based selectivity to PO from Example 1, Comparative Example 1 and Comparative Examples 3 and 4 (solid circles: K.sub.2HPO.sub.4, crosses: K.sub.2HEDP, rectangles: NH.sub.4HEDP as buffer, triangles: [NH.sub.4].sub.xATMP as buffer). The x axis shows the run time in h (HOS, hours on stream).

CITED LITERATURE

(69) U.S. Pat. No. 4,833,260 U.S. Pat. No. 4,824,976 EP 0 757 045 A