BRANCHED ORGANOSILOXANES USED AS HEAT TRANSFER FLUID

Abstract

The invention relates to a method for operating a system at an operating temperature of between 300° C. and 500° C., using a heat transfer fluid comprising branched siloxanes of general formula (I) (R.sub.3SiO.sub.1/2), (SiO.sub.4/2) in which w represents integral values of between 4 and 20, z represents integral values of between 1 and 15, and R represents a methyl group, the sum of the fractions of all siloxanes of general formula (1) being at least 95 mass %, in relation to the whole heat transfer fluid.

Claims

1. A method of operating a system at an operating temperature of 300° C. to 500° C. using a heat transfer fluid comprising branched siloxanes of general formula I
(R.sub.3SiO.sub.1/2).sub.w (SiO.sub.4/2).sub.z,  (I) where w represents integer values from 4 to 20, z represents integer values from 1 to 15, R represents methyl, wherein sum total of proportions of all siloxanes of general formula I is not less than 95% by mass, based on the heat transfer fluid as a whole.

2. The method as claimed in claim 1 wherein w represents values from 4 to 10 and z represents values from 1 to 4.

3. The method as claimed in claim 1 wherein w represents 4 and z represents 1.

4. The method as claimed in claim 1 wherein the heat transfer fluid contains dissolved or suspended or emulsified nanofluids as well as siloxanes of general formula I.

5. The method as claimed in claim 1 wherein the sum total of the proportions of all siloxanes of general formula I is not less than 99.5% by mass, based on the heat transfer fluid as a whole.

6. The method as claimed in claim 1 wherein the system is a solar thermal device.

7. The method as claimed in claim 6 wherein the system is a concentrated solar power (CSP) power plant.

8. The method as claimed in claim 3 wherein the heat transfer fluid contains dissolved or suspended or emulsified nanofluids as well as siloxanes of general formula I.

9. The method as claimed in claim 8 wherein the sum total of the proportions of all siloxanes of general formula I is not less than 99.5% by mass, based on the heat transfer fluid as a whole.

10. The method as claimed in claim 9 wherein the system is a solar thermal device.

11. The method as claimed in claim 10 wherein the system is a concentrated solar power (CSP) power plant.

Description

EXAMPLES

[0037] General working techniques, solvents and chemicals Syntheses were all carried out under Schlenk conditions. The protective gas used was argon (99.998%, Westfalen AG). The chlorosilanes used were obtained from Wacker Chemie AG or from Sigma-Aldrich, fractionally distilled under reduced pressure before use, and stored under protective gas. The solvents used were conventionally dried and stored over activated molecular sieve (3 Å, VWR) and under argon. The drier used was sodium/diphenylmethanone for THF (tetrahydrofuran) and sodium for methanol. The other reagents and solvents were all acquired from ABCR, Sigma-Aldrich, Fluka or VWR and, unless otherwise stated, used without further purification. The water used was completely ion-free. KDS-200-CE syringe pump from KD Scientific was used for metered additions. To carry out the heat treatments, either 100 mg of the siloxane were fused under vacuum into a 1.5 ml prescored borosilicate glass ampoule (from Wheaton), or 1.5 g of the siloxane were introduced into a pressureproof 14 ml steel flask. The filling of the steel flasks was carried out in a suitably equipped glovebox (argon) from MBraun, into and out of which all the components were imported and exported via a conventional load lock system. The flasks were closed with a screw lid. A sealing paste (WS 600, from WEKEM) was used as additional sealing aid for the screw thread. The final tightening of the lid was done using a spanner after exporting the flasks out of the glovebox. The thermal aging of the siloxanes was effected by storing the filled glass ampoules or steel flasks in Carbolite LHT ovens from Carbolite GmbH at different temperatures in order to show the effects of the method according to the invention.

[0038] Samples were taken and analyzed by NMR spectroscopy, gas chromatography or dynamic viscometry. Comparative analyses of NMR measurements (mol %) and GC measurements (GC area %) display very good agreements, which is why data is entirely reported in % in the experimental section.

[0039] Analysis

[0040] Nuclear magnetic resonance spectra were recorded on a Bruker Avance I 360 (.sup.1H: 360.1 MHz, .sup.29Si: 71.6 MHz) spectrometer or a Bruker Avance III HD 500 (.sup.29Si: 99.4 MHz) spectrometer using a BBO 500 MHz S2 probe head. Chemical shifts are reported in 6 values (ppm) relative to the residual proton signal of the deuterated solvent used (Euriso-top) (CDCl.sub.3, δ=7.26 ppm; CD.sub.2Cl.sub.2, δ=7.24 ppm). The internal reference used for the .sup.29Si spectra was tetramethylsilane (.sup.29Si: δ=0.00 ppm). Multiplicity of signals was abbreviated as follows: s (singlet), t (triplet) and q (quadruplet). The .sup.29Si spectra were recorded using the INEPT pulse sequence for monomeric products of synthesis and using the inverse gated pulse sequence (NS=3000; 150 mg of siloxane in 500 μl of a 4×10.sup.−2 molar solution of Cr(acac).sub.3 in CD.sub.2Cl.sub.2) for oligomeric products of synthesis. The glass hump was removed from the spectrum using the cryoprogld software package from Bruker. The GC/MS analysis was carried out on a Varian GC-3900 gas chromatograph (column: VF-5ms, 30 m×0.25 mm×0.25 μm, carrier gas: helium, flow rate: 1 ml/min, injector: CP-1177, split: 1:50) coupled to a Varian MS Saturn 2100 T (EI, 70 eV) mass spectrometer. The GC analysis was carried out using a Varian GC-3900 gas chromatograph (column: VF-200ms, 30 m×0.32 mm×0.25 μm, carrier gas helium, flow rate: 1 ml/min, injector: CP-1177, split: 1:50, detector: FID 39X1, 250° C.). Elemental analyses were carried out using a Vario EL analyzer from Elementar. Viscosity was determined at a temperature of 25° C. using a pVISK viscometer from RheoSense Inc. A B-580 ball tube oven from Buchi was used for the kugelrohr ball tube distillations. Reported temperatures correspond to the internal temperature of the oven.

[0041] Measurement of D and T Group Contents (.sup.29Si NMR)

[0042] The extent of disproportionation (D or T group content) was determined using nuclear magnetic resonance spectroscopy (.sup.295i NMR; Bruker Avance III HD 500 (.sup.295i: 99.4 MHz) spectrometer using a BBO 500 MHz S2 probe head; inverse gated pulse sequence (NS=3000); 150 mg of siloxane in 500 μl of a 4×10.sup.−2 molar solution of Cr(acac).sub.3 in CD.sub.2Cl.sub.2. The glass hump was removed from the spectrum using the cryoprogld software package from Bruker. For this, the integral of the D and T values was set in relation to the overall sum of the integral values for S groups (free silanes (e.g., Me.sub.4Si or polymer-incorporated groups C.sub.4Si), M groups (chain ends Me.sub.3SiO.sub.1/2— or C.sub.3SiO.sub.1/2), D groups (chain members —Me.sub.2SiO.sub.2/2— or C.sub.2SiO.sub.2/2), T groups (branch points MeSiO.sub.3/2) and Q groups (crosslink points SiO.sub.4/2) to give % D and % T respectively. Disproportionation leads to the conversion of 2M.fwdarw.S+D and of 2D.fwdarw.M+T. Therefore, the degree of disproportionation was judged by the proportion of T units on proceeding from DM systems and by the proportion of D units on proceeding from TM and QM systems.

[0043] Syntheses

[0044] Methyltris(trimethylsiloxy)silane (TM.sub.3) (Not Inventive)

[0045] Methyltris(trimethylsiloxy)silane (TM.sub.3) was prepared as described in EP1481979. Boiling point: 145° C., 20 mbar; .sup.1H NMR (360.1 MHz, CDCl.sub.3): δ[ppm]=0.12 (s, 27 H, (.sub.CR-3).sub.3SiO.sub.1/2), 0.01 (s, 3 H, CH.sub.3SiO.sub.3/2); .sup.29Si NMR (71.6 MHz, CDCl.sub.3): δ [ppm]=7.6 (s, (CH.sub.3).sub.3SiO.sub.1/2), −64.0 (s, CH.sub.3SiO.sub.3/2); MS: m/z (%)=295.0 (100) [M.sup.+ —CH.sub.3]; elemental analysis: reckoned (%) for C.sub.10H.sub.30O.sub.3Si.sub.4: C 38.66, H 9.73; observed: C 38.42, H 9.86; viscosity: 1.39 mPa*s (25° C.)

[0046] MT siloxane T.sub.1M.sub.1.2 (Not Inventive)

[0047] In an argon-inertized 1 1 2-neck Schlenk flask equipped with septum and gas outlet, 50.0 g (335 mmol) of trichloromethyl-silane and 116 g (1.07 mol) of chlorotrimethylsilane are dissolved in 400 ml of acetonitrile. A syringe pump is used to add 56.0 g (3.11 mol) of water at 0.20 ml/min and 20° C. under agitation by vigorous stirring. On completion of the addition, the reaction solution is stirred at 20° C. for 2 h, admixed with 168 ml of water and extracted with methyl tert-butyl ether (3×200 ml). The combined organic phases are washed neutral with 150 ml of saturated aqueous sodium bicarbonate solution, with water (3×200 ml) and with 50 ml of saturated sodium chloride solution, dried over sodium sulfate (˜40 g) and filtered. The solvent is removed under reduced pressure and the residue obtained is admixed with 5.00 g (31.0 mmol) of hexamethyldisilazane and stirred at 20° C. for 12 h. Volatile compounds are subsequently removed under reduced pressure (3.10.sup.−2 mbar, 25° C.) to obtain 54.0 g of the product as a colorless oil. .sup.29Si NMR (99.4 MHz, CD.sub.2Cl.sub.2): M to T ratio 1.2:1; viscosity: 6.73 mPa*s (25° C.)

[0048] Tetrakis(trimethylsilyloxy)silane (QM.sub.4)

[0049] An argon-inertized 2 1 4-neck round-bottom flask equipped with reflux condenser, thermometer and stirrer is initially charged with 649 g (4.02 mol) of hexamethyldisiloxane and 64.0 g (2.00 mol) of methanol, followed by cooling to below 10° C. with an icebath. 9.80 g (100 mmol H.sub.2SO.sub.4) of concentrated sulfuric acid are added over 15 min, followed by stirring at 10° C. for 30 min. Then, 152 g (999 mmol) of tetramethyl orthosilicate are added over 30 min and the reaction solution is stirred at 10° C. for 1 h. Following addition of 105 g (5.83 mol) of water, the reaction solution is stirred for 3 h at room temperature and then for 3 h under reflux. The reaction solution is admixed with 500 ml of 1 M aqueous sodium bicarbonate solution and stirred at 20° C. for 10 min and the aqueous phase is separated off. The organic phase is washed neutral with water (3×200 ml). Then, the organic phase is desolventized under reduced pressure and the residue is subjected to fractional distillation in vacuo to obtain 201 g (522 mmol, 52.3% yield) of product as colorless oil. Boiling point: 180° C., 42.0 mbar; .sup.1H NMR (360.1 MHz, CDCl.sub.3): δ [ppm]=0.10 (s, 36 H, (CH.sub.3).sub.3SiO.sub.1/2); .sup.29Si NMR (71.6 MHz, CDCl.sub.3): δ [ppm]=8.5 (s, (CH.sub.3).sub.3SiO.sub.1/2), −104.7 (s, SiO.sub.4/2); MS: m/z (%)=369.2 (100) [M.sup.+ —CH.sub.3] ; elemental analysis: reckoned (%) for C.sub.22H.sub.360.sub.4Si.sub.5: C 37.45, H 9.43; observed: C 37.26, H 9.56; viscosity: 2.71 mPa*s (25° C.)

[0050] Hexakis(trimethylsiloxy)disiloxane (Q.sub.2M.sub.6)

[0051] Step 1: Synthesis of Hexaethoxydisiloxane

[0052] In an argon-inertized 500 ml 2-neck Schlenk flask equipped with septum and reflux condenser, 62.5 g (300 mmol) of tetraethyl orthosilicate, 1.35 ml (74.9 mmol) of water and 1.25 ml of 6 N hydrochloric acid are dissolved in 200 ml of tetrahydrofuran and refluxed for 3 h. The solvent is removed under reduced pressure and the residue obtained is fractionally distilled under reduced pressure to obtain 3.80 g (11.1 mmol, 7.33% yield) of the product as colorless oil. Boiling point: 125° C., 10.0 mbar; .sup.1H NMR (360.1 MHz, CDCl.sub.3): 5 [ppm]=3.88 (q, .sup.3J(H,H)=7.0 Hz, 12 H, OCH.sub.2CH.sub.3), 1.25 (t,.sup.3J(H,H)=7.0 Hz, 18 H, OCH.sub.2CH.sub.3); .sup.29Si NMR (71.6 MHz, CDCl.sub.3): δ [ppm]=−88.6 (s, (OEt).sub.3SiO.sub.1/2); MS: m/z (%)=297.1 (100) [M.sup.+ —OCH.sub.2CH.sub.3]; elemental analysis: reckoned (%) for C.sub.12H.sub.30O.sub.7Si.sub.2: C 42.1, H 8.83; observed: C 35.5, H 9.03.

[0053] Step 2: Synthesis of hexakis(trimethylsiloxy)disiloxane (Q.sub.2M.sub.6)

[0054] In a 25 ml round-bottom flask fitted with reflux condenser, 1.71 g (4.99 mmol) of hexaethoxydisiloxane and 4.16 g (31.5 mmol) of trimethylsilyl acetate are initially charged and admixed with 710 μl of concentrated hydrochloric acid. The reaction solution obtained heats up to about 50° C. Once the reaction solution has cooled down to 20° C., it is admixed with 50.0 ml of diethyl ether, washed with saturated sodium bicarbonate solution (2×50.0 ml) and water (50.0 ml), dried over sodium sulfate (˜5 g) and filtered. The solvent is removed under reduced pressure and the residue obtained is admixed with 15.0 ml of methanol. The precipitate formed is filtered off, washed with cold methanol and purified by sublimation (60° C., 0.02 mbar) to obtain 670 mg (1.10 mmol, 22.0% yield) of the product as colorless solid. .sup.1H NMR (360.1 MHz, CDCl.sub.3): δ [ppm]=0.12 (s, 54H, (CH.sub.3).sub.3SiO.sub.1/2); .sup.29Si NMR (71.6 MHz, CDCl.sub.3): δ [ppm]=8.8 (s, (CH.sub.3).sub.3SiO.sub.1/2), −106.8 (s, SiO.sub.4/2); MS: m/z (%)=591.2 (100) [M.sup.+ —CH.sub.3]; elemental analysis: reckoned (%) for C.sub.18H.sub.54O.sub.7Si.sub.8: C 35.6, H 8.96; observed: C 35.55, H 9.07.

[0055] MQ Silicone Oil Q.sub.3M.sub.3.7

[0056] In a 1 1 3-neck flask fitted with reflux condenser and dropping funnel, 20.0 g (118 mmol) of tetrachlorosilane are dissolved in 200 ml of tetrahydrofuran. A solution of 63.7 g (706 mol) of trimethylsilanol in 200 ml of tetrahydrofuran is added over 2 h. The reaction solution obtained is stirred under reflux for 3 h and at 20° C. for 17 h. 500 ml of 1 M aqueous sodium bicarbonate solution and 500 ml of diethyl ether are added, the resulting phases are separated and the aqueous phase is extracted with diethyl ether (3×50.0 ml). The combined organic phases are washed neutral with water, dried over sodium sulfate (˜40 g) and filtered. The solvent, excess trimethylsilanol and hexamethyldisiloxane are removed under reduced pressure and the residue obtained is admixed with 5.00 g (31.0 mmol) of hexamethyldisilazane and stirred at 20° C. for 12 h. Volatile compounds are subsequently removed under reduced pressure (3.10.sup.−2 mbar, 25° C.) to obtain 27.0 g of the product as a colorless oil. .sup.29Si NMR (99.4 MHz, CD.sub.2Cl.sub.2): M to Q ratio 3.7:1; viscosity: 5.97 mPa*s (25° C.)

[0057] Heat Treatment: Siloxanes and Siloxane Mixtures with Minimal Rearrangement (Equilibration)

EXAMPLE 1

Heating of WACKER AK 5 (D.SUB.4.M.SUB.1.) (Not Inventive)

[0058] 1.5 g of WACKER AK 5 MD siloxane (viscosity: 5.33 mPa*s at 25° C.) is heated in a steel ampoule for 7 days.

[0059] Low boilers (M.sub.2, MDM, D.sub.3, D.sub.4) before heating (.sup.29Si NMR): 0%

[0060] Low boilers (M.sub.2, MDM, D.sub.3, D.sub.4) after heating at 400° C. (.sup.29Si NMR): 13.4%

[0061] Low boilers (M.sub.2, MDM, D.sub.3, D.sub.4) after heating at 425° C. (.sup.29Si NMR): 13.4%

EXAMPLE 2

Heating of TM.SUB.3 .(Not Inventive)

[0062] 1.5 g of TM.sub.3 MT siloxane (viscosity: 1.39 mPa*s at 25° C.) is heated in a steel ampoule for 7 days.

[0063] Low boilers (M.sub.2) before heating (.sup.29Si NMR): 0%

[0064] Low boilers (M.sub.2) after heating at 400° C. (.sup.29Si NMR): 4.7%

[0065] Low boilers (M.sub.2) after heating at 425° C. (.sup.29Si NMR): 4.9%

EXAMPLE 3

Heating of T.SUB.1.M.SUB.1.2.(Not Inventive)

[0066] 1.5 g of T.sub.1M.sub.1.2 MT siloxane (viscosity: 6.73 mPa*s at 25° C.) is heated in a steel ampoule for 7 days.

[0067] Low boilers (M.sub.2) before heating (.sup.29Si NMR): 0%

[0068] Low boilers (M.sub.2) after heating at 400° C. (.sup.29Si NMR): 7.1%

[0069] Low boilers (M.sub.2) after heating at 425° C. (.sup.29Si NMR): 9.0%

EXAMPLE 4

Heating of QM.SUB.4

[0070] 1.5 g of QM.sub.4 MQ siloxane (viscosity: 2.71 mPa*s at 25° C.) is heated in a steel ampoule for 7 days.

[0071] Low boilers (M.sub.2) before heating (.sup.29Si NMR): 0%

[0072] Low boilers (M.sub.2) after heating at 400° C. (.sup.29Si NMR): 0.1%

[0073] Low boilers (M.sub.2) after heating at 425° C. (.sup.29Si NMR): 0.3%

EXAMPLE 5

Heating of Q.SUB.2.M.SUB.6

[0074] 100 mg of Q.sub.2M.sub.6MQ siloxane is heated in a glass ampoule for 7 days.

[0075] Low boilers (M.sub.2) before heating (GC): 0%

[0076] Low boilers (M.sub.2) after heating at 400° C. (GC): 0.1%

[0077] Low boilers (M.sub.2) after heating at 425° C. (GC): 0.2%

EXAMPLE 6

Heating of Q.SUB.1.M.SUB.3.7

[0078] 1.5 g of Q.sub.1M.sub.3.7 MQ siloxane (viscosity: 5.97 mPa*s at 25° C.) is heated in a steel ampoule for 7 days.

[0079] Low boilers (M.sub.2) before heating (.sup.29Si NMR): 0%

[0080] Low boilers (M.sub.2) after heating at 400° C. (.sup.29Si NMR): 0.3%

[0081] Low boilers (M.sub.2) after heating at 425° C. (.sup.29Si NMR): 0.4%

[0082] A comparison of Examples 1 to 6 demonstrates the present invention (table 1): TM siloxanes (Examples 2 and 3) show that the degree of rearrangement is less pronounced as compared with MD siloxanes (Example 1), resulting in the formation of significantly less by way of low boilers, which are largely responsible for the vapor pressure increase and the viscosity reduction. In the case of QM siloxanes (Examples 4 to 6), rearrangement is almost fully eliminated, which consequently results in the formation of almost no low boilers.

TABLE-US-00001 TABLE 1 Comparison of heat treatment experiments Low boilers after Low boilers after Siloxane 7 d at 400° C. [%] 7 d at 425° C. [%] D.sub.4M.sub.1 (WACKER AK 5)* 13.4 13.4 TM.sub.3* 4.7 4.9 T.sub.1M.sub.1.2* 7.1 9.0 QM.sub.4 0.1 0.3 Q.sub.2M.sub.6 0.1 0.2 Q.sub.1M.sub.3.7 0.3 0.4 *not inventive

[0083] It transpires that rearrangement is reduced by avoidance of D units and use of T units and almost fully eliminated by avoidance of D and T units and use of Q units. The consequence is that the physical properties of the heat transfer oil, for example its viscosity—as evidenced by table 2—or its vapor pressure, change much less, if at all.

TABLE-US-00002 TABLE 2 Comparison of heat treatment experiments Viscosity after Viscosity after 7 d at 400° C. 7 d at 425° C. absolute [mPa*s] absolute [mPa*s] and the relative and the relative change from change from initial viscosity initial viscosity Viscosity [%] (in [%] (in Siloxane [mPa*s] parentheses) parentheses) D.sub.4M.sub.1 (WACKER 5.33 3.15 (40.9) 3.15 (40.9) AK 5)* QM.sub.4 2.71 2.71 (0) 2.69 (0.7) Q.sub.1M.sub.3.7 5.97 6.01 (0.7) 6.05 (1.3) *not inventive

[0084] As a result, there is no need for additional control engineering requirements during operation of a heat transfer system or even extra capital expenditure to design and construct the device, or for but limited utility or even complete inutility of the device over this period. A further advantage of QM siloxanes is that, as compared with DM siloxanes such as WACKER AK 5, no cyclic D siloxanes (inter alia D4) are formed, which is of substantial advantage with regard to environmental and safety engineering aspects in particular.

[0085] Heat Treatment: Siloxanes and Siloxane Mixtures with Minimal Disproportionation

EXAMPLE 7

Heating of WACKER AK 5 (Not Inventive)

[0086] 1.5 g of WACKER AK 5 MD siloxane (viscosity: 5.33 mPa*s at 25° C.) is heated in a steel ampoule at 450° C. for 42 days.

[0087] Fraction of disproportionation product (T groups) before heating (.sup.29Si NMR): 0%

[0088] Fraction of disproportionation product (T groups) after heating (.sup.29Si NMR): 4.9% and 1.2% on standardization to the D starting ratio (D/M start ratio=4/1).

EXAMPLE 8

Heating of QM.SUB.4

[0089] 1.5 g of QM.sub.4 MQ siloxane (viscosity: 2.71 mPa*s at 25° C.) is heated in a steel ampoule at 450° C. for 42 days.

[0090] Fraction of disproportionation product (D groups) before heating (.sup.29Si NMR): 0%

[0091] Fraction of disproportionation product (D groups) after heating (.sup.29Si NMR): 1.9% and 0.5% on standardization to the M starting ratio (M/Q start ratio=4/1).

EXAMPLE 9

Heating of Q.SUB.1.M.SUB.3.7

[0092] 1.5 g of 4.sub.1M.sub.3.7 MQ siloxane (viscosity: 5.97 mPa*s at 25° C.) is heated in a steel ampoule at 450° C. for 42 days.

[0093] Fraction of disproportionation product (D groups) before heating (.sup.29Si NMR): 0%

[0094] Fraction of disproportionation product (D groups) after heating (.sup.29Si NMR): 2.2% and 0.6% on standardization to the M starting ratio (M/Q start ratio=3.7/1).

[0095] A comparison of Examples 7 to 9 demonstrates the present invention (table 3): QM siloxanes (Examples 8 and 9) show that the extent of disproportionation is less pronounced as compared with MD siloxanes (Example 7) and as a result the numbers of chain ends (M), the linear chain members (D), the branch points (T) and the crosslink points (Q) change less, the consequence of which is that the physical properties of the heat transfer oil, for example its vapor pressure or its viscosity, change to a significantly smaller degree. As a result, there is no need for additional control engineering requirements during operation of a heat transfer system or even extra capital expenditure to design and construct the device, or for but limited utility or even complete inutility of the device over this period. A reduced extent of disproportionation extends the service life and reduces the exchange rate of the heat transfer oil, as a result of which the operation of heat transfer systems becomes significantly more economical. The lower rate of disproportionation means that the heat transfer oil can also be used at higher operating temperatures than MD siloxanes, as a result of which higher heat transfer efficiencies are attained in solar thermal systems for example. This leads to a distinctly enhanced efficiency and more economical operation of heat engines.

TABLE-US-00003 TABLE 3 Comparison of heat treatment experiments Fraction of disproportionation Fraction of product after 42 d disproportionation at 450° C. [%] as product after 42 d at standardized to the Siloxane 450° C. [%] starting ratio D.sub.4M.sub.1 (WACKER AK 5)* 4.9 1.2 QM.sub.4 1.9 0.5 Q.sub.1M.sub.3.7 2.2 0.6 *not inventive