VALUE CHAIN RETURN PROCESS FOR THE RECOVERY OF PHOSPHOROUS ESTER-BASED FLAME RETARDANTS FROM POLYURETHANE RIGID FOAMS

Abstract

Polyurethane rigid foams containing at least one phosphorous ester-based flame retardant are returned to the value chain by hydrogenating the polyurethane rigid foam, to obtain a hydrogenation product containing a polyamine and a polyol from the polyurethane rigid foam, and recovering the flame retardant from the hydrogenation product. The hydrogenation is carried out in an organic aprotic solvent, in a hydrogen atmosphere, in the presence of at least one homogeneous transition metal catalyst complex, wherein the transition metal is selected from metals of groups 7, 8, 9 and 10 of the periodic table of elements according to IUPAC, and at a reaction temperature of at least 120 C.

Claims

1: A value chain return process for a polyurethane rigid foam comprising at least one phosphorous ester-based flame retardant, comprising: hydrogenating the polyurethane rigid foam: in an organic aprotic solvent, in a hydrogen atmosphere, in the presence of at least one homogeneous transition metal catalyst complex, wherein the transition metal is selected from metals of groups 7, 8, 9 and 10 of the periodic table of elements according to IUPAC, and at a reaction temperature of at least 120 C., to obtain a hydrogenation product comprising a polyamine and a polyol from the polyurethane rigid foams; and recovering the flame retardant from the hydrogenation product.

2: The process according to claim 1, comprising recovering the polyamine from the hydrogenation product via distillation.

3: The process according to claim 1, comprising recovering the flame retardant via distillation from the hydrogenation product, or from a distillation bottoms thereof.

4: The process according to claim 1, comprising recovering the polyol by extraction from the hydrogenation product, or from a distillation bottoms thereof, or as a distillation bottoms after removal of volatile components.

5: The process according to claim 1, wherein the polyurethane rigid foam comprises an aromatic isocyanate-based polyurethane rigid foam.

6: The process according to claim 1, wherein the at least one phosphorous ester-based flame retardant comprises at least one selected from the group consisting of tris(2-chloroethyl)phosphate, tris(chloroisopropyl)phosphate, tris(1,3-dichloro-2-propyl)phosphate, tris(2-ethylhexyl)phosphate, tricresylphosphate, tris-(2,3-dibromo)phosphate, tetrakis-(2-chlorethyl)-ethylenediphosphate, dimethylphosphonate, dimethylpropylphosphonate, diphenylcresylphosphate, and triethylphosphate.

7: The process according to claim 1, wherein the organic aprotic solvent comprises at least one selected from the group consisting of an ether and an aromatic hydrocarbon.

8: The process according to claim 7, wherein: the ether is selected from the group consisting of tetrahydrofuran, 1,4-dioxane, and anisole; and the aromatic hydrocarbon is selected from the group consisting of benzene, toluene, xylene, and mesitylene.

9: The process according to claim 1, wherein the at least one homogeneous transition metal catalyst complex comprises a transition metal selected from the group consisting of manganese, rhenium, ruthenium, iridium, nickel, palladium, and platinum.

10: The process according to claim 1, wherein the at least one homogeneous transition metal catalyst complex comprises at least one polydentate ligand having at least one nitrogen atom and at least one phosphorous atom which are capable of coordinating to the transition metal.

11: The process according to claim 10, wherein the at least one polydentate ligand comprises at least one ligand according to formula (I): ##STR00021## in which each R is independently H or C.sub.1-C.sub.4-alkyl, R.sup.1 and R.sup.2, independently of one another, are C.sub.1-C.sub.12-alkyl, cycloalkyl or aryl, which alkyl is unsubstituted or carries 1, 2, 3, 4 or 5 identical or different substituents R.sup.7, and which cycloalkyl and aryl are unsubstituted or carry 1, 2, 3, 4 or 5 identical or different substituents R.sup.8, R.sup.3 and R.sup.4, independently of one another, are H or C.sub.1-C.sub.12-alkyl, which is unsubstituted or carries 1, 2, 3, 4 or 5 identical or different substituents selected from heterocycloalkyl, aryl, hetaryl, alkoxy, cycloalkoxy, heterocycloalkoxy, aryloxy, hetaryloxy, hydroxyl, NE.sup.1E.sup.2 and PR.sup.1R.sup.2, R.sup.5 is H or C.sub.1-C.sub.12-alkyl, which is unsubstituted or carries 1, 2, 3, 4 or 5 identical or different substituents R.sup.7, R.sup.6 is H or C.sub.1-C.sub.4-alkyl, or R.sup.4 and R.sup.6 are absent and R.sup.3 and R.sup.5, together with the nitrogen atom to which R.sup.3 is bonded and the carbon atom to which R.sup.5 is bonded, form a 6-membered heteroaromatic ring, which is unsubstituted or carries 1, 2, 3, 4 or 5 identical or different substituents which are selected from C.sub.1-C.sub.12-alkyl, cycloalkyl, aryl and hetaryl, which alkyl is unsubstituted or carries 1, 2, 3, 4 or 5 identical or different substituents R.sup.7, and which cycloalkyl, aryl and hetaryl are unsubstituted or carry an alkyl substituent which is unsubstituted or carries a substituent selected from alkoxy, cycloalkoxy, heterocycloalkoxy, aryloxy, hetaryloxy, hydroxyl, NE.sup.1E.sup.2 and PR.sup.1R.sup.2, each R.sup.7 is independently cycloalkyl, heterocycloalkyl, aryl, hetaryl, alkoxy, cycloalkoxy, heterocycloalkoxy, aryloxy, hetaryloxy, hydroxyl or NE.sup.1E.sup.2, each R.sup.8 is independently C.sub.1-C.sub.4-alkyl, cycloalkyl, heterocycloalkyl, aryl, hetaryl, alkoxy, cycloalkoxy, heterocycloalkoxy, aryloxy, hetaryloxy, hydroxyl or NE.sup.1E.sup.2, and E.sup.1 and E.sup.2, independently of one another and independently of each occurrence, are radicals selected from H, C.sub.1-C.sub.12-alkyl, cycloalkyl and aryl.

12: The process according to claim 11, wherein the at least one polydentate ligand comprises at least one ligand according to formula (II): ##STR00022## in which D is H, C.sub.1-C.sub.12-alkyl, cycloalkyl, aryl or hetaryl, which alkyl is unsubstituted or carries 1, 2, 3, 4 or 5 identical or different substituents R.sup.7, and which cycloalkyl, aryl or hetaryl are unsubstituted or carry an alkyl substituent which is unsubstituted or carries a substituent selected from alkoxy, cycloalkoxy, heterocycloalkoxy, aryloxy, hetaryloxy, hydroxyl, NE.sup.1E.sup.2 and PR.sup.1R.sup.2, preferably NE.sup.1E.sup.2 and PR.sup.1R.sup.2.

13: The process according to claim 10, wherein the at least one polydentate ligand comprises at least one selected from the group consisting of compounds A to L, wherein Et is ethyl, .sup.iPr is isopropyl, .sup.tBu is tert-butyl, Cy is cyclohexyl, Ph is phenyl: ##STR00023##

14: The process according to claim 1, wherein the hydrogenation reaction is carried out at a pressure of 30 to 500 bar absolute.

15: The process according to claim 1, wherein the hydrogenation reaction is carried out in the presence of a base.

16: The process according to claim 1, wherein the polyurethane rigid foam comprises at least one selected from the group consisting of a methylenedi(phenylisocyanate)-based polyurethane rigid foam, a polymeric methylenedi(phenylisocyanate)-based polyurethane rigid foam, and a 1,5-naphthyldiisocyanate-based polyurethane rigid foam.

17: The process according to claim 1, wherein the at least one homogeneous transition metal catalyst complex comprises a transition metal selected from the group consisting of manganese, ruthenium, and iridium.

18: The process according to claim 1, wherein the hydrogenation reaction is carried out at a pressure of 40 to 300 bar absolute.

19: The process according to claim 1, wherein the hydrogenation reaction is carried out at a pressure of 50 to 200 bar absolute.

20: The process according to claim 1, wherein the hydrogenation reaction is carried out in the presence of at least one base selected from the group consisting of an alkali metal carbonate, and alkaline earth metal carbonate, an alkali metal hydroxide, an alkaline earth metal hydroxide, an alkali metal alcoholate, and an alkaline earth metal alcoholate.

Description

[0141] FIG. 1 shows a comparison of the .sup.31P NMR spectrum of a sample of flame retardant TCPP (top) and the flame retardant obtained after hydrogenation of PU rigid foam Index 100 (bottom).

EXAMPLES

Materials

[0142] All chemicals and solvents were purchased from Sigma-Aldrich or ABCR and used without further purification, unless otherwise specified. .sup.1H, .sup.13C and .sup.31P NMR spectra were recorded on Bruker Avance 200 or 400 MHz spectrometer and were referenced to the residual proton (.sup.1H) or carbon (.sup.13C) resonance peaks of the solvent. Chemical shifts () are reported in ppm. .sup.31P NMR spectra were referred to an external standard (ample of D.sub.3PO.sub.4).

[0143] In the hydrogenation examples that follow, PU rigid foam Index 100 was used. PU rigid foam Index 100 is based on 74 wt.-parts Lupranol 3422 (a commercial polyetherpolyol available from BASF SE, Germany based on sorbitol and propyleneoxide containing only secondary hydroxy groups), 20 wt.-parts Lupragen TCPP (flame retardant tris(2-chloroisopropyl)phosphate), 3 wt.-parts Tegostab B 842045 (silicone surfactant available from Evonik Industries AG), 0.5 wt.-parts Lupragen N600 (a tertiary amine available from BASF SE, Germany), 2.5 wt.-parts water, 5 wt.-parts cyclopentane, and 100 wt.-parts Lupranat MP 102 (short chain prepolymer based on pure 4,4-diphenylmethane diisocyanate available from BASF SE, Germany), containing therefore 9.4 wt.-% of the flame retardant TCPP in the final PU rigid foam. The results are summarized in table 1.

[0144] Hydrogenation catalysts P and Q were prepared according literature protocols: E. Balaraman, J. Am. Chem. Soc. 2010, 132, 16756-16758 and D. Srimani, Adv. Synth. Catal. 2013, 355, 2525-2530. Hydrogenation catalyst Mn-1 was prepared similar to literature protocols using Mn(CO).sub.5Br as metal precursor: W. Zhou, Chem Sus Chem 2021, 14, 4176-4180 (ligand synthesis) and V. Zubar, Angew. Chem. Int. Ed. 2018, 57, 13439-13443 (complex synthesis). Hydrogenation catalyst Mn-5 was prepared according to literature protocols: W. Zhou, Chem Sus Chem 2021, 14, 4176-4180 and U. K. Das, ACS Catal. 2018, 9, 479-484.

Reference Example 1: Synthesis of Hydrogenation Catalyst H

##STR00008##

[0145] First step: In a 50 mL Schlenk tube, 6-methyl-2,2-bipyridine (511 mg, 3.00 mmol) was dissolved in 15 mL Et.sub.2O, cooled to 0 C. and LDA (3.50 mL, 1 M in THF/hexanes) was added dropwise. After stirring at 0 C. for 1 h, the system was cooled to 80 C. by .sup.iPrOH/liquid N.sub.2 and CIPCy.sub.2 (815 g, 3.50 mmol) in 5 mL Et.sub.2O was added slowly. The cooling bath was removed after 1 h and the mixture was recovered to r.t. gradually and stirred overnight. The reaction mixture was quenched by adding 10 mL of degassed water to the yellow slurry. The organic phase was separated and the aqueous phase was extracted with ether (25 mL). The combined organic phase was dried over Na.sub.2SO.sub.4, filtered and the solvent was removed to give the crude ligand as a sticky orange oil. 52% purity based on .sup.31P NMR. It was used directly for the next step without further purification.

[0146] Second step: The ligand obtained in the first step was dissolved in 20 mL THF. RuHCl(CO)(PPh.sub.3).sub.3 (952 mg, 1.00 mmol) was added, the mixture was stirred at 70 C. for 5 hours and then cooled to r.t. The solvent was reduced to ca. 10 mL under vacuum and 20 mL of Et.sub.2O were added to the remaining red-orange dispersion. The solution was removed via cannula and the solid was washed with Et.sub.2O (210 mL) and dried under vacuum to give 465.2 mg of the orange product (87% yield based on Ru).

[0147] .sup.31P {.sup.1H} NMR (122 MHz, CD.sub.2Cl.sub.2) 83.68.

[0148] .sup.1H NMR (301 MHz, CD.sub.2Cl.sub.2) 9.22-9.13 (m, 1H), 8.07-7.97 (m, 1H), 7.93 (d, J=8.0 Hz, 1H), 7.86 (td, J=8.0, 1.6 Hz, 1H), 7.82 (td, J=8.0, 0.9 Hz, 1H), 7.49 (d, J=7.7 Hz, 1H), 7.45-7.39 (m, 1H), 3.82-3.56 (m, 2H), 2.46-2.27 (m, 2H), 2.08-0.99 (m, 20H), 14.83 (d, J=23.6 Hz, 1H).

[0149] .sup.13C {.sup.1H} NMR (126 MHz, CD.sub.2Cl.sub.2) 207.71 (d, J=14.9 Hz), 161.70 (d, J=5.1 Hz), 156.38, 154.78 (d, J=2.7 Hz), 153.51 (d, J=1.7 Hz), 137.30, 136.51, 126.42 (d, J=1.9 Hz), 123.13 (d, J=9.6 Hz), 122.76 (d, J=1.6 Hz), 119.73, 40.59 (d, J=22.2 Hz), 38.59 (d, J=23.4 Hz), 35.76 (d, J=28.9 Hz), 31.01 (d, J=2.9 Hz), 29.60 (d, J=4.2 Hz), 28.61 (d, J=4.5 Hz), 28.20 (d, J=13.6 Hz), 27.73, 27.56 (d, J=9.2 Hz), 26.82 (d, J=4.4 Hz), 26.74 (d, J=3.5 Hz), 26.71 (d, J=2.0 Hz), 26.35 (d, J=1.5 Hz). HRMS (ESI): m/z calcd. for C.sub.24H.sub.32N.sub.2OPRu [M-Cl].sup.+: 497.1296, found: 497.1291.

Reference Example 2: Synthesis of Hydrogenation Catalyst Mn-1

##STR00009##

[0150] First step: The Cy-PNN ligand was synthesized similarly to the above-described procedure for hydrogenation catalyst H and used without further purification in the next step.

[0151] Second step: To a solution of the Cy-PNN ligand (406 mg, 1.11 mmol) in 4 mL THF was added under argon atmosphere an orange solution of Mn(CO)Br (240 mg, 0.88 mmol) in 10 mL THF and the reaction mixture was kept stirring at room temperature for 24 h (Note: The CO gas liberated needs to be removed occasionally in vacuo). The solution was evaporated in vacuo. The solid residue was washed with pentane (10 mL), which on evaporation gave a dark brown solid product. The brown crude product was dissolved in THF (15 mL), the solution was filtered, concentrated, layered with pentane and kept in the refrigerator (30 C.) to obtain a dark red solid in 71% (350 mg, based on Mn) yield.

[0152] .sup.31P {.sup.1H} NMR (162 MHz, CDCl.sub.3) 89.77.

Example 1: Hydrogenation of a Polyurethane Rigid Foam Containing a Phosphorous Ester-Based Flame Retardant

[0153] A stainless steel autoclave (Premex) fitted with a Teflon insert was charged with PU rigid foam Index 100 (1.00 g). Inside a glovebox, catalyst and base were added. The walls were rinsed with the indicated solvent and the autoclave was closed. Outside the glovebox, the autoclave was flushed with hydrogen gas (25 bar) and charged with hydrogen gas (50 bar). The autoclave was heated for 21 h to 200 C. under stirring (pre-heated metal block, 750 RPM). After cooling to room temperature (ice-bath), the residual pressure was carefully released. The suspension was filtered over a suction filter and the remaining solid was washed with dichloromethane (35 mL) and EtOH (35 mL). The solid, residual polymer was dried under reduced pressure (r.t., <5.0 10.sup.2 mbar) and used for the determination of the conversion (conversion=[(polymer usedpolymer recovered)/polymer used]100). After removal of the solvent of the filtrate under reduced pressure (45 C., min. pressure 80 mbar), the residue was redissolved in CDCl.sub.3 (2 mL). An aliquot of 1,1,2,2-tetrachloroethane (50.0 L) as internal standard was added and the solution was homogenized by swirling. The samples were analyzed by .sup.1H and .sup.31P NMR spectroscopy. In the .sup.1H NMR, the amount of TCPP in the corresponding sample was determined by integration of the TCPP signal (=4.67 ppm) against the 1,1,2,2-tetrachloroethane signal (=6.00 ppm). The samples described in table 1 contained undecomposed flame retardant.

TABLE-US-00001 TABLE 1 Hydrogenation of PU rigid foam Index 100 (polyurethane rigid foam containing phosphorous ester flame retardants). [00010] [00011] [00012] # PU rigid foam catalyst base H.sub.2 pressure organic solvent T t conversion flame ret. .sup.[1] polyol amine [g] [mol] [mol] [bar] volume [mL] [C] [h] [%] [g] [g] [g] 1 1 Mn-5 KO.sup.tBu 50 THF/toluene 200 21 65 0.066 .sup.[2] n.d. .sup.[3] n.d. .sup.[3] 40.0 80 8/8 2 1 Ru-8 KO.sup.tBu 50 THF 200 21 78 0.057 .sup.[2] n.d. .sup.[3] n.d. .sup.[3] 20 80 10 3 1 Ru-10 KO.sup.tBu 50 THF 200 21 88 0.084 .sup.[2] n.d. .sup.[3] n.d. .sup.[3] 20 80 10 .sup.[1] flame retardant = TCPP .sup.[2] determined by quantitative .sup.1H NMR by integration of corresponding signals against 1,1,2,2-tetrachloroethane (50.0 L ) .sup.[3] not determined

[0154] The phosphorous ester-based flame retardant of table 1 remained undecomposed as evidenced from .sup.31P NMR as can be seen from FIG. 1. In FIG. 1, a comparison is shown of the .sup.31P NMR spectrum of a sample of flame retardant TCPP (FIG. 1, top) and the flame retardant obtained after hydrogenation of PU rigid foam Index 100 (FIG. 1, bottom). All relevant signals in the range of from 4.10 to 3.55 ppm are detected in both the top and the bottom spectrum of the .sup.31P NMR.

Example 2: Hydrogenation of a Polyurethane Rigid Foam (5 g) Containing a Phosphorous Ester-Based Flame Retardant

[0155] A stainless steel autoclave (Premex) fitted with a Teflon insert was charged with PU rigid foam Index 100 (5.12 g). Inside a glovebox, Mn-5 (200 mol) and KOtBu (400 mol) were added. The walls were rinsed with THF (40 mL) and toluene (40 mL) and the autoclave was closed. Outside the glovebox, the autoclave was flushed with hydrogen gas (215 bar) and finally charged with hydrogen gas (50 bar). The autoclave was heated for 21 h to 200 C. under stirring (pre-heated metal block, 750 RPM). After cooling to room temperature (ice-bath), the residual pressure was carefully released. The suspension was filtered over a suction filter and the remaining solid was washed with dichloromethane (320 mL). The solid, residual polymer was dried under reduced pressure (r.t., <5.0.Math.10.sup.2 mbar) and used for the determination of the conversion (conversion=[(polymer usedpolymer recovered)/polymer used]100). The conversion of the hydrogenation of PU rigid foam Index 100 was 62%. After removal of the solvent of the filtrate under reduced pressure (45 C., min. pressure 80 mbar), the residue was purified by distillation in a Kugelrohr oven at 150 to 170 C. under reduced pressure (1.4.Math.10.sup.1 mbar). The obtained fractions were analyzed by .sup.1H and .sup.31P NMR spectroscopy. TCPP was obtained as light yellowish oil (404 mg).

Comparative Example 1: Hydrolytic Depolymerization of a Polyurethane Rigid Foam Containing a Phosphorous Ester-Based Flame Retardant

[0156] As comparative example, hydrolytic depolymerization of a polyurethane rigid foam was carried out instead of hydrogenative depolymerization. For this purpose, a sample of a PU rigid foam containing a phosphorous ester-based flame retardant (Index 100, composition see above) was hydrolyzed with a mixture of pyridine and water at 160 C. in a sealed glass pressure tube.

##STR00013##

[0157] After the reaction, a reaction mixture in the form of a dark brown solution was obtained which was free of solids. By analyzing said reaction mixture via GC/MS, 4,4-methylenedianiline was detected. However, the reaction mixture did not contain a phosphorous ester-based flame retardant according to GC/MS or .sup.31P NMR data. Thus, the phosphorous ester-based flame retardant was hydrolyzed under these conditions.

[0158] Experimental details: In air, a stainless steel autoclave (Premex) fitted with a Teflon insert was charged with PU rigid foam Index 100 (1.00 g). The walls were rinsed with pyridine (20 mL) and water (2 mL). The autoclave was closed and heated for 16 h to 160 C. After cooling to room temperature (ice-bath), the brown solution was filtered over a suction filter and the filter was rinsed with EtOH (35 mL). No solid remained on the filter. The solvent was removed under reduced pressure (45 C., min. pressure 60 mbar). According to .sup.1H and .sup.31P NMR analysis, no flame retardant was detected in neither of the crude or isolated material.

Comparative Example 2: Hydrogenative Depolymerization of a Polyurethane Rigid Foam Containing a Phosphorous Ester-Based Flame Retardant Using MACHO Catalysts in Iso-Propanol

[0159] The results and conditions of the following procedure are summarized in table 2.

[0160] A stainless steel autoclave (Premex) fitted with a Teflon insert was charged with PU rigid foam Index 100 (1.00 g). Inside a glovebox, catalyst and base were added. The walls were rinsed with iso-propanol and the autoclave was closed. Outside the glovebox, the autoclave was flushed with hydrogen gas (215 bar) and finally charged with hydrogen gas (50 bar). The autoclave was heated for 21 h to 180 C. under stirring (pre-heated metal block, 750 RPM). After cooling to room temperature (ice-bath), the residual pressure was carefully released. The suspension was filtered over a suction filter and the remaining solid was washed with dichloromethane (35 mL) and EtOH (35 mL). The solid, residual polymer was dried under reduced pressure (r.t., <5.0 10.sup.2 mbar) and used for the determination of the conversion (conversion=[(polymer usedpolymer recovered)/polymer used]100). After removal of the solvent of the filtrate under reduced pressure (45 C., min. pressure 80 mbar), the residue was redissolved in CDCl.sub.3 (2 mL). An aliquot of 1,1,2,2-tetrachloroethane (50.0 L) as internal standard was added and the solution was homogenized by swirling. The samples were analyzed by .sup.1H and .sup.31P NMR spectroscopy. In the .sup.1H NMR, the amount of TCPP in the corresponding sample was determined by integration of the TCPP signal (=4.67 ppm) against the 1,1,2,2-tetrachloroethane signal (=6.00 ppm).

[0161] The reaction mixture of entry 2 was additionally purified by flash column chromatography (EtOAc-hexane) to determine the amount of amine. In the end, the silica pad was flushed with EtOH to elute the polyol fraction.

TABLE-US-00002 TABLE 2 Hydrogenation of PU rigid foam Index 100 (polyurethane rigid foam containing phosphorous ester flame retardants). [00014] [00015] [00016] PU rigid foam catalyst base H.sup.2 pressure organic solvent T t conversion flame ret. .sup.[1] polyol amine # [g] [mol] [mol] [bar] volume [mL] [ C.] [h] [%] [g] [g] [g] 1 1 Mn-2 KOH 50 .sup.iPrOH 180 21 93 .sup.[2] n.d. .sup.[3] n.d. .sup.[3] 2.1 21 2.5 2 1 Mn-2 KOH 50 .sup.iPrOH 180 21 >99 0.010 .sup.[4,5] 0.461 .sup.[4,6] 0.488 [.sup.4,7] 16.0 160 19.2 3 1 Ru-5 KOH 50 .sup.iPrOH 180 21 >99 .sup.[2] n.d. .sup.[3] n.d. .sup.[3] 16.0 160 19.2 .sup.[1] flame retardant = TCPP .sup.[2] no flame retardant detected .sup.[3] not determined .sup.[4] quantification after flash column chromatography (EtOAc-hexane) .sup.[5] the isolated fraction contained decomposed flame retardant .sup.[6] contains aromatic side products as well as solvent (EtOH) according to .sup.1H NMR .sup.[7] contains solvent (EtOAc) as well as alkylated (mono or diiso-propyl) amine

Comparative Example 3: Stability of TCPP Under Hydrogenating Conditions (Mixture of Isomers, Flame Retardant) Using Mn- or Ru-Catalysts in Various Solvents

[0162] The results and conditions of the following procedure are summarized in table 3.

[0163] Inside a glovebox, a stainless steel autoclave (Premex) was charged with catalyst, base and TCPP (mixture of isomers, flame retardant). The walls were rinsed with solvent and the autoclave was closed. Outside the glovebox, the autoclave was flushed with hydrogen gas (215 bar) and finally charged with hydrogen gas (50 bar). The autoclave was heated for 21 h to the indicated temperature under stirring (pre-heated metal block, 750 RPM). After cooling to room temperature (ice-bath), the residual pressure was carefully released. The obtained mixture was filtered over a short pad of Celite in a Pasteur pipette and the pad was washed with dichloromethane (35 mL). After removal of the solvent under reduced pressure (45 C., min. pressure 80 mbar), the residue was weighed and analyzed by .sup.1H and .sup.31P NMR to determine the amount of recovered flame retardant (recovery=[(amount of TCPP detected)/(amount of TCPP used)]100).

TABLE-US-00003 TABLE 3 Hydrogenation of TCPP (mixture of isomers, flame retardant). [00017] [00018] [00019] [00020] # flame ret. .sup.[1] catalyst base H.sub.2 pressure organic solvent T t recovery flame ret. .sup.[1] [g] [mol] [mol] [bar] volume [mL] [ C.] [h] [%] [g] 1 0.164 Mn-5 KO.sup.tBu 50 THF/toluene 200 21 84 0.137 40 80 8/8 2 0.164 Ru-10 KO.sup.tBu 50 THF 200 21 82 0.134 20 80 10 3 0.164 Mn-2 KOH 50 .sup.iPrOH 180 21 <1 .sup.[2,3] 16 160 19.2 .sup.[1] flame retardant = TCPP .sup.[2] no flame retardant detected .sup.[3] very broad, flat signal detected in the .sup.31P NMR spectrum

[0164] Under hydrogenation conditions in the presence of an aprotic solvent, the phosphorous ester-based flame retardant (see table 3) remained undecomposed as evidenced from .sup.31P NMR (4.10 to 3.55 ppm).

[0165] However, under hydrogenation conditions in the presence of a protic solvent such as iso-propanol, only an unspecific and very broad, flat signal ( 31.76 to +20.82 ppm) was detected in the .sup.31P NMR spectrum.