PERSICO DANIEL F (US)
PERSICO DANIEL F
US3502619A | 1970-03-24 | |||
FR1433649A | 1966-04-01 | |||
US3530097A | 1970-09-22 | |||
US4523039A | 1985-06-11 | |||
DE2632837A1 | 1977-02-03 |
Journal of Polymer Science, Volume 18, No. 1, January 1980, (US), G.E. GERHARDT et al.: "Synthesis of Highly Branched Perfluoroethers by direct Fluorination. Promosing new Materials based on the Hexafluoroacteone-Ethylene Copolymer", see page 157-168
Journal of the Chemical Society, Perkin Transactions, Volume 1, 1981 (GB), G.E. GERHARDT et al.: "Synthesis of Perfluoropolyethers by Direct Fluorination: A Novel Preparation for Perfluoro (Polypropylene Oxide) Ethers and Perfluor(Polymethylene Oxide) Ethers", pages 1321-1328
1. | Perfluorinated hexafluoroacetone copolymers of the formula: wherein R^l and Rc ' are F, CF, , C,FC , C,F_ , CC..F. f fl 3 2 5 3 7 6 5 or cC6Fl ι ; Rf2, Rf '' Rf3 and Rf4 are F' CF3 or 2F5; Rf3' to Rf6* if F or CF3; RfA is CFRflRf2CRf3Rf4, CFRf3 f4, CFRf1'Rf2'CRf3 'Rf4'CRf5'Rf6' ' CFRf_3, 'Rf4'CRf5 *Rf6' orCPRf5. 'Rf6,* ; Rf is CF3, C_F , C3F_., COF and COOH; and x and y are integers greater than 1 such that x+y=850. |
2. | Perfluorinated hexafluoroacetone/epoxide copolymers of the formula: wherein Rf, is F, CF3, CF,, 3F7, cC8F.. or C"C6F11; , R£„2, Rf,.3, and Rf_4. are F, CF3, or C_2F5.; RfA is CFRflRf2CRf3Rf4 or CFRf3Rf4; RfB is CF3, C2F5 or C3F?; and n = 850. |
3. | Perfluorinated hexafluoroacetone/ethylene oxide copolymer of the formula: CF, 0 4 0 CF2CF2 wherein n = 850. |
4. | Perfluorinated hexafluoroacetone/propylene oxide copolymer of the formula: ?F3 CF 0 (■ C OCF2CF O CF(CF3)2 wherein n = 850. |
5. | Perfluorinated hexafluoroacetone/oxetane copolymer of the formula: wherein Rfl is F, CF , C2Fg, C.F.., cC8F5 or C"C6F11; Rf2 is CF3 or 2F5; RfA is CFRflRf2CRf3Rf4CRf5Rf6' CFRf3Rf4CRf5Rf6 or CFRf5Rf6; Rfβ is CR3, C„ 2F.5 or C3F?; and n=850. |
6. | Perfluorinated hexafluoroacetone/oxetane copolymer of the formula: ?P3 CF30 j C 0 CF2CF2CF20lCF(CF3)2 CP3 wherein n = 850. |
7. | A method of producing perfluorinated hexafluoro¬ acetone copolymers, comprising the steps of: a. copolymerizing hexafluoroacetone and an oxygencontaining cyclic comonomer selected from ethylene oxide, propylene oxide, epoxybutane, trimethylene oxide or substituted derivatives thereof; and b. perfluorinating the resulting copolymer by a direct fluorination procedure wherein the copolymer is exposed to gradually increasing concentrations of fluorine gas. |
Description
Field of the Invention
This invention is in the field of polymer chemistry.
Background of the Invention
Perfluoropolyethers are highly regarded in the specialty lubricant field because of their long liquid range, (i.e. liquid over a wide range of temperatures) low vapor pressure and high thermal and oxidative stability. Because of these proper¬ ties, they are excellent high performance lubricants and have many other uses as oils, waxes, greases, hydraulic fluids and heat transfer fluids. Commercially available perfluoropolyethers are produced by the polymerization of hexafluoro- propylene oxide (DuPont KRYTOX TM) and by photo- oxidative polymerization of perfluoro-olefins, utilizing oxygen and ozone (Montedison FOMBLIN TM) . Each product has an advantage over the other. The
TM KRYTOX fluids which have a regularly branched structure have superior high temperature thermal o TM stability (10 - 15 F) over the FOMBLIN fluids.
The FOMBLINTM fluids, random polymers having O-C-O units in their backbone, have a slight advantage (10 15 F) in their longer liquid range at low tempera¬ ture.
Two very interesting analytical themes seem to be developing in the analysis of lubrication pro¬ perties and thermal stabilities of perfluoro¬ polyethers. Many researchers feel that the unusual liquid properties of the random Fomblin Z copolymer is a direct result of the inclusion of difluoro- methylene oxide (CF_0) linkages in the polyether which provide hinge-like flexibility. Another view is that the high.er thermal stability of the perfluoro- propylene oxide polymers results in some way from the branching (trifluoromethyl group) in the poly¬ mer. An emerging view is that the difluoromethy- lene unit causes the lower thermal stability of the Fomblin Z material with respect to the Krytox structure.
Disclosure of the Invention
This invention pertains to a new class of perfluorinated polyether fluids comprising per- fluorinated copolymers of hexafluoroacetone (HFA) and cyclic oxygen-containing compounds. The new perfluoroethers are produced by the copolymerization of hexafluoroacetone with the oxygen-containing cyclic comonomers selected from ethylene oxide, ' propylene oxide, epoxy-butane and/or trimethylene oxide (oxetane) or substituted derivatives thereof and subsequent perfluorination of the resulting
copolymer. The perfluorohexafluoroacetone copolymers have from 1% up to 50% O-C-O linkages in their backbone chain and branched (trifluoromethyl substi¬ tuted) methylene units in a regular copolymer sequence. The hexafluoroacetone copolymers of this invention are useful as lubricants (greases and oils) at temperatures ranging from about -50°C to about 300°C.
The combined features of the O-C-O linkages and of the branching confer dual properties of a long liquid range and a high thermal stability. For example, a 1:1 HFA/epoxide alternating copolymer structure contains approximately 50% of the O-C-O linkage in the structure. This linkage is believed responsible for low temperature liquidity. In contrast, FOMBLIN TM fluids contain approximately 10
- 20% of this linkage. In preliminary tests, the perfluorinated hexafluoroacetone copolymers of this invention exhibited thermal stability comparable to
TM a higher molecular weight KRYTOX fluid.
The hexafluoroacetone copolymers are copolymers of hexafluoroacetone and oxygen-containing cyclic compounds. The oxygen-containing cyclic compounds can be the epoxides such as ethylene oxide, propy- lene oxide and epoxy-butane and/or oxetane (tri- methylene oxide) or substituted derivatives of these compounds.
Perfluorinated hexafluoroacetone/epoxide
' copolymers are represented generally by the formula:
wherein R fl is F, CF 3 , C F 5' C.-F-,, c-CgF-. or
C"C 6 P 11 ; R f2 R f3 and R f4 are F ' F 3 ° r C 2 F 5 ; R fA 'is CFR fl R f2 CR f3 R f4 or C R^R^; R fB is CF 3 , C 2 5 or C 3 F ? ; and n = 8-50.
Hexafluoroacetone/oxetane copolymers are represented by the formula:
wherein R fl is F, CF 3 , C_, 5 , C 3 F_,, c-CgF 5 or
C"C 6 F 11 ;
R f2 is F, CF 3 or C.P.;
Rf-A_ is CFRf__lRf_2_C f_3_ f_4.CRfj__5Rfj_6-.,
CFR f3 R f4 CR f5 R f g or CFR f5 R f6 ; R fβ is CF 3 , C 2 F 5 or C 3 F ? ; and n=8-50.
Perfluorinated hexafluoroacetone/epoxide/ oxetane copolymers are represented by the formula:
wherein R fl and R f 1 ' are P, CF-,, C 2 F_., C_,F 7 , c-C 8 F-. or c-C 6 F n ;
R f2' R f2 * ' R f3 and R f4 are F ' CF 3 or C 2 F 5 7
^ 3 R^^, ' ' 5 'R f6 « ;
R fB is CF 3 , C 2 F 5 and C 3 F ? ; and x+y are integers greater than 1 such that x+y=8-50.
Best Mode of Carrying Out the Invention
The perfluorinated hexafluoroacetone copolymers are produced by perfluorinating, in a controlled, stepwise manner, copolymers of hexafluoroacetone and oxygen-containing cyclic comonomer(s) .
Several techniques exist for synthesizing hexafluoroacetone/epoxide and hexafluoroacetone/ oxetane copolymers. See e.g., U.S. Patent 3,475,376;
U.S. Patent 3,485,792; U.S. Patent 3,502,619; U.S. Patent 3,530,097; and U.S. Patent 3,624,039. The polymerization reaction of HFA with an epoxide is illustrated by the following equation:
CF 3 -
wherein R__ is H, CH, , C 2 H g , C.,H 7 , c-CgH, . or C-CgH l ι ;
R f2 , R f3 and R f . are H, CH 3 or C 2 H_. ; and x=500-1000 .
The polymerization reaction of HFA with oxetane is exemplified by the following formula :
wherein R-... is H, CH 3 , C 2 H_., C 3 H 7 , c-CgH_. or c-C 6 H n ;
R f2 , R f3 and R f . are H, CH 3 or C 2 H_.;
R_H_o, to R_H_o_. is H or CH3,; and x=10-50.
A preferred method of producing a 1:1 alter¬ nating hexafluoroacetone/epoxide copolymer is a method of bulk polymerization with an anionic catalyst such as tetraethylamine chloride, cesium fluoride or potassium fluoride.
The polymerization reaction is performed under a vacuum in an inert environment (e.g. under
a nitrogen atmosphere) . The anionic catalyst is added to a reaction vessel in an inert atmosphere. The vessel is evacuated and cooled either with liquid nitrogen to -196°C or to -78°C with a dry ice acetone slush. HFA and the epoxide or oxetane is condensed into the vessel which is then sealed under a vacuum. The vessel is then warmed to about -78 C (e.g. by placing it into a dry ice bath) if liquid nitrogen cooling has been used. Thereafter the reaction vessel is warmed to room temperature and maintained for about 1-6 hours. After synthesis is complete, volatile components are removed by evaporation under a vacuum. The solid polymer product is recovered from the vessel.
The hexafluoroacetone polymers are perfluori- nated by elemental fluorination. Fluorination is performed in a suitable fluorination apparatus. A preferred apparatus is that described by Lagow, R. j., Progress in Inorganic Chemistry, 26, 161
(1979) . Another suitable fluorination apparatus is described in United States Patent Number 4,510,335, the teachings of which are incorporated by reference herein. Fluorine gas mixed with helium as the diluent is added to the reaction chamber containing the hexafluoroacetone/epoxide or hexafluoracetone/ oxetane polymer. To avoid extensive chain degra¬ dation during the fluorination process, fluorina- tion is performed in a controlled, stepwise manner by initially contacting the polymer with ' a dilute
fluorine- mixture (approximately 1-2% fluorine gas) and then increasing the fluorine concentration in a stepwise fashion over an interval of about 48 hours. Temperature is also an important parameter which is generally increased over the course of fluorination (from room temperature to about 60°C).
The specific conditions of fluorination vary widely for the different polymers. Conditions are dependent upon the stability of the polymer being fluorinated. For example, as detailed in the Exemplification section below, optimal conditions of temperature and fluorine concentration are significantly different for HFA/EO copolymers and HFA/TMO polymers (see Table 1) . Optimal condi¬ tions of fluorination for the various types of polymers can be ascertained experimentally. Further, conditions of fluorination can be regu¬ lated to control degradation of the polymer during fluorination and as a result, the size of the resulting perfluorinated product.
The- fluorination of three hexafluoroacetone copolymers is depicted below:
0 - CF 2 CF 2 - 0}
Hexafl uσroacetone/Ethyl ene Oxide Copolymer
T έ 3 - 0 - T C 3 - CH 2 - 0> χ " /F 2 2 → C 3 - 0 - T C 3 - CF 2 - 0 \ j 2 x amb/Δ j j
CF 3 H CF 3 F
Hexafl uoroacetone/Propyl ene Oxide Copol ymer
0 - CF 2 CF 2 CF 2 - 0) y
In general, perfluorinated copolymers having from about 10 to about 50 repeat units are de- sirable. These copolymers represent those having the optimum combination of thermal stability and liquid range.
The perfluorinated hexafluoroacetone/epoxide and perfluorinated hexafluoroacetone/oxetane polymers have features which confer high thermal stability and low temperature liquidity. They represent a combination of the advantageous properties of the FOMBLIN TM and KRYTOXTM systems.
The perfluorinated hexafluoroacetone copolymers of the invention contain 0-C-O linkages which provide flexibility in the backbone chain, a property which is believed to provide a long liquid range. As pointed out, a 1:1 alternating copolymer contains 50% of this O-C-O linkage, which is more
TM than is in the FOMBLIN fluids.
The polymers also contain branched methylene units. This branching is believed to be respon- sible for the thermal stability of perfluorinated polyethers. The extent of branching, of course, depends on the kind and number of pendant groups on the backbone in the perfluorinated hexafluoro¬ acetone copolymers. This can be modified by altering the epoxide or oxetane monomers incor¬ porated into the polymers and thus the polymers can be tailored for specific end uses.
In initial tests, perfluorinated hexafluoro- acetone/propylene oxide, perfluorinated hexa- fluoroacetone/ethylene oxide and perfluorinated hexafluoroacetone/oxetane oligomers having an average number of repeat units of about 8-16 (Mn
2,800-6,000) compared favorably with a higher molecular weight KRYTOX TM fluids (AZ grade) and
TM FOMBLI fluids in thermal stability, as deter¬ mined by thermogravimetric analysis. Perfluori¬ nated hexafluoroacetone copolymers of greater average number of repeat units up to 50 would have superior thermal stability (due to pendant groups) , yet exhibit low liquid ranges (due to O-C-O units) .
The- invention is further illustrated by the following exemplification.
Exemplification
All gaseous reactants were measured by using vacuum line techniques and assuming ideal be¬ havior. The reactants were degassed at liquid nitrogen temperatures and expanded into a 2.07 liter PYREX TM bulb connected to, a mercury mano-
TM meter. A PYREX tube which had been charged, under nitrogen, with the selected catalyst, was used as the reaction container. The reactants were condensed into the Pyrex tube at -196 C, degassed again, and the tube was sealed. The tube was then brought to -78 C in a dry ice-acetone slush and slowly warmed to ambient temperature.
The reactants were agitated at ambient temperature for six hours and then the temperature was raised to 65 C for 36 hours. When the reaction was complete, the resulting product was held under vacuum to remove any unreacted monomer.
The fluorination apparatus used is described by Lagow, R. J. , Progress in Inorganic Chemistry, 26, 161 (1979) , incorporated by reference herein. The reaction conditions to preserve each polymer structure varied significantly. Optimized condi¬ tions are given under each synthesis below. Volatile products from the reaction were collected in a glass trap maintained at dry ice temperature. These products were washed with a KOH/H-0 mixture
to remove any functional components, dried on molecular sieves and separated by gas chromato- graphy to obtain pure compounds for analysis. The resultant nonvolatile oils from the reaction were dissolved in Freon 113 (CF-.C1CC1F) and filtered through a CELITE TM/NORITTM b~ed to remove any foreign matter. The solvent was then removed under vacuum at 80 C for 24 hours.
Physical analysis of the isolated compounds and the oligomeric oils were obtained in the following manner. Elemental analyses were per¬ formed by Schwarzkopf Microanalytical Laboratory, Inc., Woodside, New York, and Galbraith Labora¬ tories, Inc., Knoxville, Tennessee. Infrared spectra were obtained with a Perkin Elmer 467 spectrometer utilizing a gas cell with KBr win¬ dows, or by casting a film (oils) from Freon 113 onto KBr plates. 19F NMR spectra were run on a
Varian EM-390 spectrometer operating at 84.67 MHz.
13C[19F] (19F decoupled) NMR spectra were run on a
Bruker WH-100 spectrometer. Mass spectra were obtained with a Bell and Howell 21-490 mass spectrometer with the ion source cooled to ambient temperature. Gas phase chromatography (GPC) was done on a Bendix 2300 programmable chromatograph equipped with a cryogenic controller and a thermal conductivity detector. The column used for sample separation was 3/8 in. x 24 ft. packed with 10%
FOMBLIN Z-25 TM on Chromosorb PTM (60/30 mesh) , with a helium flow of 100 cc/min. GPC analyses
were done using a Waters Associates Inc. system equipped with a R-401 differential refractometer and an IBM 10 micron GPC/SEC column. Thermo- gravimetric analyses were obtained on a DuPont Model 990 Thermal Analyzer equipped with a DuPont Model 951 Thermogravimetric Analyzer.
Hexafluoroacetone/Ethylene Oxide Copolymer (HFA/EO)
Five milligrams of (C 2 H 5 ) .NCI were added to 4.65 x 10~ moles each of HFA and EO. The resul- tant polymer weighed 8.6 grams and this corre¬ sponds to a yield of 88%. The white crystalline powder had a melting range of 182-184°C (lit. 188-192°C, G. E. Gerhardt and R. J. Lagow, J. Chem. Soc, Perkin Trans. 1_, 1321 (1980); G. E. Gerhardt t jal. , J. Poly . Sci. Polym. Chem. Ed.
18, 157 (1979). Elemental analysis gave C, 28.65; F, 54.50; H, 1.96 (Theor. C, 28.57; F, 54.29; H, 1.90); confirming a 1:1 copolymer.
Due to the crystalline nature of the polymer ambient conditions were first studied for the fluorination procedure. These conditions proved to be unsatisfactory as significant cleavage to hydrogen-containing volatile products occurred before pure fluorine conditions could be reached. Subsequently low temperatures were found to be necessary early in the reaction to assure per¬ fluorinated oil. Initial conditions were there¬ fore set at -78 C and the total sequence of reaction conditions can be seen in Table 1.
Ko no 12 0 2.0 • 55
Hυ'.VJ_0 12 . 0 2.0 65
IiOH 30 ' 12 0 2.0 ■ 70
Roπiio 12 . 0 2.0 70
HO 1S0 12 60 0 70
.«;HDI60 HFA/PO and HFA/TMO
ROW 170 60 100 1.0 arnb
ROWIΛ0 24 0 1.0 arnb
ROW190 24 0 1.0 35
ROW200 24 0 1.0 45
RO A 210 12 0 1.0 50
ROW220 12 0 1.0 50
These conditions proved to be advantageous as degradation yielding extremely volatile products (those which are volatile at -78°C) was kept to an acceptable level and a reasonable amount of nonvolatile perfluoropolyether oil was also obtained. After each temperature increment the volatile products were weighed and analyzed by gas chromatography. The relative amount of volatile products at each temperature was the same indica- ting that the same fragmentation processes were responsible for the volatile species over the entire temperature range. There was also no
threshold temperature (up to 70°C) above which all of the polymer degraded to volatile species. Infrared analysis of the nonvolatile oil proved it contained functional end groups, as would be 5 expected by the cleavage route (see Discussion section) . To obtain a nonfunctional product the oil was allowed to air hydrolyze thus allowing the carbonyl fluoride end groups to be hydrolyzed to carboxylic acid end groups. These acid groups 0 were then removed by further reaction with elemen¬ tal fluorine which causes decarboxylation leaving behind perfluoro nonfunctional end groups. Infrared analysis of the volatile by-products revealed that they also contained a substantial 5 quantity of functionality. The volatile fractions were washed with a KOH/H_0 solution leaving behind the nonfunctional polyethers. These were subse¬ quently separated and characterized. The results are shown in Table II.
Table II P e rfl u oio Ethers Isolated Crom Hcxanuoroacetone/Ethyk ne Oxide Pol> inei Fluorina ti on compd "F shift (ppm),'re. inlcnsUy hi hest mic in mass s
( 1) ( CF,).CFOCF, ' (a) 83.3/6. (b) 148.2/1. 235 ( C.F.O.. a " b e (c) 56.6/3 p " F
( 2) CF,OC(CF,)_OCF, (a) 57.3/1. (b) 78.7/1 301 (C,F„0.).
P - F
(3) ( CF,)_CFOCF.CF.OCF(CF J )- (a) 83.3/6, (b) 146.9/1. 451 tC,F„0,). a b c e b a (c) 88.5/2 P - F
(4) CF » OCιCF l ) 4 OCF-CF 1 OCFiCF,) l (a) 56.7/3, (b) 78.4/6. 467 (C.F.,0,). a b c d β f (c) 89.7/2. (d) 87.7/2. P - CF, .
(e) 147.1/1, (0 83.2/6
(5 ) CF,OC(CF,).OCF.CF. C(CF,. ς OCF, (a) 57.1/3. (b) 78.7/6, 545 (C„F„O t ), ' a b c c b a (c) 89.7/2 P - F,
(6 ) ( CF, ) 1 CFOCF.CF l OC(CF,) i OCF J CF 1 OCF(CF,), (a) 83.2/6, (b) 147.1/1. 733 (C u F„OΛ a b e d e d c b a (c) 87.7/2, (d) 88.1/2. P - F
(e) 78.3/3
(7) CF J OC(CF 1 ).OCF.CF.OC(CF 1 ) 1 OCF J CF-OCF(CF J ). (a) 56.6/3.1, (b) 78.3/11.6, 799 (C,.F„0,). a b e c b e d e f (c) 89.0/5.9, (d) 87.6/2.0. P - F
(e) 147.1/1.0, (f) 83.2/6.0 .
(8) (CF,) 4 CFU[CF a CF ; OC(CF,)0] i OCF I (a) -81.3, (b) -145.7. a b c c d e (c) -87.7. (d) -76.6.
(e) -54.9
The. infrared spectra for the seven isolated compounds were all similar in showing a strong broad absorption in the carbon-fluorine and ether carbon-oxygen region (1350-1050 cm ) as well as strong, sharp characteristic absorptions in the region at 1000-980, and 730-660 cm " . Additio¬ nally a peak at 890-870 cm " can be found in all molecules containing a perfluoro-methoxy end group. The infrared spectrum of the resultant oil contains the same basic features with no trace of functional groups.
Mass spectrometry was extremely useful for determining the molecular, weights of the volatile components. By running the ion source of the spectrometer at ambient the extensive fragmen¬ tation and rearrangements inherent in mass spectra of fluorocarbons can be kept to a minimum. This method resulted in the observation of parent minus fluorine ions for most of the volatile compounds (Table II) along with the characteristic fragmen¬ tation pattern.
The most useful method for structural identi- fication proved to be 19F NMR. The perfluoro- ethers 19F NMR signals, assignments and relative intensities are listed in Table II. The average chemical shifts of the various fluorine nuclei were (in ppm from external CFC1,) : CF, of methyoxy end groups, -56.9; internal pendant groups, -78.5; CF 3 of isopropyl end groups, -83.2; all CF 2 0 groups, -88.5; CF of isopropyl end groups, -147.3. The observed coupling patterns are good examples
of the unusual coupling characteristics often
19 observed in F NMR spectroscopy. The CF 3 and CF groups of the isopropyl group do not couple with one another; however, they both couple with the
05 CF 2 "across" the oxygen linkage yielding average coupling constants of 5 and 21 Hz, respectively.
All other coupling characteristics were routine.
13 19 The C [ F] NMR data are in agreement with the assigned structure and are listed in Table
(4) CF,OC(CFa,) J OCF,CF J OCF(CF-) 1 (a) 120.2. (b) 95.9, (c) 118.7. -98 (bp 144 * C) a b e d - d e f (d) 115.2, (e) 302.6, (f) 118.0
(5) CF,OC(CF,),OCF 1 CF i OClCF,) 1 OCF, - (a) 120.3, (b) 96.0, (c) 118.8, 0 a b e d d b c a (d) 115.3
(βMCFjkCFOCF.C .OCCCFO j OCF j CF j OCFCCF-), (a) 118.0, (b) 102.6, (c) 115.1, -74 a c c d e e c b a (d) 96.6, (e) 118.8
(7) CF,OC(CF,) α OCF I CF l OC(CF-).OCF J CF 2 OCF(CF 1 ) I (a) 120.1, (b) 95.8. (c) 118.8, -58 a . b e d d b e d d e f (d) 115.2, (e) 102.6, (f) 118.1
(8) (CF;) J CFOICF.CF-OC(CF,).0] X OCF, (a) 118.2, (b) 102.7, (c) 115.3, a b c d e f (e) 118.9
The average chemical shifts of the various carbon nuclei were (in ppm downfield from TMS) : carbon of ethoxy end groups, 119.8; carbon of pendant internal methyl groups, 118.8; carbon of isopropyl
methyl groups, 118.0; carbon of CF 2 0 groups, 115.2; tertiary carbon of isopropyl groups, 102.5; quaternary carbon, 96.0.
The physical characterization involving melting point determination was complicated by the problem of the liquids freezing to glasses rather than crystalline powders. Repeated analysis, however, gave reproducible results and these are found in Table III along with the boiling points of the two perfluoroethers which were obtained in sufficient amounts.
Table IV lists the yields of volatile pro¬ ducts and oil obtained from a typical reaction relative to 1.0 gram of starting materials. From
19
F NMR end group analysis the average molecular weight of the perfluoropolyether oil was found to be 2,800.
TTL20 . Table IV
Yields of Corπpoun ds Produced by Fluorination of
Hexa uorαacetoπe/Elhylene Oxide Copolymer*
HDB 0 GC retentn yield. -78 frac. starting polymer. compd time, min rng wt % mol %
ROW50 1 1 57 10.9 3-3
ROW60 2 3. 69 13.2 4.5
ROW70 3 11 122 23.2 8.7
RO S0 4 21 149 28.3 10.5
RO 90 5 29 45 8.6 3.1
ROW.00 6 40 54 ' 10.4 3-9
ROW110 7 52 24 4.5 17
ROW.20 uπseparated 5 03 0.4* material
ROWI30 total 525 100 36.6
F.NTI40 "Yield of nonvolatile oi 1, 0.450 g; starting po lymer, 33.8 mol%.
Calculations based on 1.00 g of starting polymer. GC temperature program; 50 "C for 5 min; 75 β C for 20 min; 100 "C for 40 min. * Assumes average molecule is a seven-carbon ether.
Hexafluoroacetone/Propylene Oxide Copolymer (HFA/PO)
Add
4.5 x 10 15 cc) of PO. The polymer produced was a slightly yellow, amorphous solid with a melting point of 45-50 C. The resultant polymer weighed 5.2 grams which corresponded to a yield of 52%.
Previous experience with oils and waxy solids has shown that such materials would be fluorinated best if they were coated on a support thus increas¬ ing the surface area available for fluorination. The polymer was dissolved in Freon 113 and Teflon TM powder was added. The solution was evaporated leaving behind a free running powder, much like the HFA/EO copolymer.
The structural similarity to the HFA/EO copolymer led us to use the previous fluorination conditions. It was found, however, that upon elevating the temperature to ambient much of the polymer readily degraded to volatile products, some of which were partially fluorinated and that upon reaching 35 C extensive fragmentation oc¬ curred. The subsequent fluorination conditions selected are listed in Table I.
As with the HFA/EO copolymer the volatile products were collected and analyzed prior to each temperature change. It was found that if a nonvolatile product was desired it was essential that the temperature be maintained no higher than 50 C. The oil was allowed to hydrolyze prior to each elevation in temperature, thus assuring nonfunctional products.
The-volatile products showed some functional- ization by infrared analysis and the functional- ization was removed in the same manner as for the HFA/EO copolymer. The resultant nonfunctional volatile portion differed only in quantity from fraction to fraction but not in composition. Chromatographic analysis proved the volatile fraction to be composed of predominately one compound with extremely small amounts of other components. Mass spectral results and 19F NMR results were again very useful in ascertaining the struc¬ ture of the volatile compound, a perfluoro diether containing two isopropyl end groups and the internal branched propyl group. The mass spectrum gave a large parent minus fluorine peak while 19F
NMR analysis was not so straightforward owing to the extent of coupling and the proximity of the
CF, and CF 2 resonances which were broadened and unresolved. Also, since the propylene group is unsymmetrical, the tertiary fluorines on the isopropyl groups are unequivalent and emerge as a sextet while the propylene tertiary fluorine appears as a broadened triplet. The assignment of the CF- and CF_ groups was based on chemical shift information and coupling.
Infrared analysis of the volatile and nonvola¬ tile components were very similar. The volatile product showed a strong, broad absorption in the carbon-fluorine and C-O-C region (1375-1025 cm ) . Also there was a strong, sharp absorption in the region from 980-960 cm and a medium,
broad peak in the 720-650 c region. The nonvolatile component showed a strong, broad absorption (1350-1030 cm "1 ) in the C-F, C-O-C region as well as medium intensity sharp peak from
05 980-960 cm " and a weak, broad absorption in the region 730-670 cm " .
Complete analysis of both the volatile and nonvolatile products is given in Table V. The average molecular weight of the perfluoropolyether
10 oil was determined to be 3,000 by 19F NMR end group analysis.
Table V Products Isolated from Hexafluoroacetone Propylene Oxide Fluorination** compd "F shift (ppm)/rel intensity mp/bp, * C
(9) (CF,),CFOCFlCF,)CF-OCF(CF,) J (a) -82.2/6, (b) -141.7/2, (c) -146.4/1, -107/134 a b c d d b e (d) -81.3/5, (e) -82.5/6
(10) (CF,).CFLOC(CF,) 3 OCF(CF,)CF..O] x (a) -80.4, (b) -145.7, (c) -77.1, a b c d e e (d) -144.3. (e) -81.1
»C{"F} shift, ppm elem anal
(9) (CF,).CFOCF(CF,)CF..OCF(CF,) J (a) 118.2, (b) 103.1, (c) 303.7, Actual: (d) 118.0, (e) 136.8, (f) 102.7 C, 20.86; F, 73.02
(10) (CF,).CF_OCF(CF,)CF α OC(CF.) 1J; - (a) 118.3, (b) 303.0, (c) 103.8, • Actual: a b c d e f g (d) 118.0, (e) 136.8, (f) 96.8, (g) 118.0 C, 21.78; F, 69.02 Theory:
C, 21.56; F, 69.30
° Yield of volatile component = 520 mg = 87. 5 wt % of volatile components. This is equivalent to 34.6 mol To of sta polymer based on 1.00 g. Yield of nonvolatile oil = 175 mg - 12.2 mol % of starting polymer.
Hexafluoroacetone/Trimethylene Oxide Copolymer (HFA/TMO)
To 5 milligrams of Et.NCl were added 3.21 x
15 10 -2 moles (2.1 cc) of TMO 4 . The resultant white powdered polymer weighed 5.4 grams (75% yield) and had a melting point of 118-121°C.
Considering the crystalline nature of the polymer, it was decided that conditions similar to those used for the HFA/EO copolymer would be utilized. Later it was found that conditions more similar to those used for the HFA/PO polymer produced better results. Finally it was found advantageous to combine the two techniques; once the polymer turned to a gel early in the fluori¬ nation process, it was dissolved in Freon 113 and coated on a Teflon powder support.
The infrared analysis of the volatile and nonvolatile products were very similar and corre¬ sponded closely to those of the two previous polymers. The carbon-fluorine and ether linkages showed a strong absorption at 1350-1030 cm . In addition, a sharp absorption of medium intensity was seen in the region 990-960 cm as well as a weak, broad characteristic absorption in the
730-680 cm " region.
19F NMR and mass spectral analysis again proved quite useful in establishing the structures of the three isolated compounds. Although only three volatile compounds were isolated, a variety of end groups were present, as seen by the 19F data. Mass spectral results gave parent minus fluorine or parent minus three fluorine as the highest molecular weight ion. The second is uncommon in perfluoropolyether fragmentation patterns. On the two most abundant volatile compounds and on the oil, 13C [19F] spectra were obtained. The average molecular weight of the oil
was determined by 19F NMR end group analysis and was found to be 2,800. Table VI and Table VII show the compiled analytical and spectral data for the products.
Table VI Volatile Perfluoro Ethers Isolated from Hexafluoroacetone/Oxetane Polymer Fluorination
.--' • •'- ' compd '*F shift (ppm)/rel intensity highest mle in mass
(11) (CF,).CFOCF 1 CF a CF.OCF(CF,) I ~ (a) 83.0/8, (b) 1 3.4/1, (c) 130.6/1 501 lC,F..O.). a b a e a b a _.. " - ' P- F
(12)CF,OC(CF,) J OCF 1 CF l CF l OCF(CF,) ϊ (a) 57.3/3, (b) 78.7/6. (c) 84.9/4, ~ . . 529 (C„F„0,) a b e d c e f (d) 130.8/2, (e) 147.0/1, (f) 83.6/6 " • P-F,
_13)(CF,).CFOCF.CF 2 CF : OC(CF,).OCF.CF, (a) 83.0/6, (b) 146.3/1. (c) 83.8/4, . 617 (C.Fa.O,) a b c d c e f f (d) 330.4/2, (e) 78.2/6, (f) 89.4/5 -. P-F
(14) (CF,) J CFlOCF.CF J CF i OC(CF,) J ] x OCF 3 CF, (a) -81.1, (b) -145.0, (c) -82.5,
. (d) -129.7, (e) -76.4, (f) -87.2 - '
Table VII Volatile Perfluoro Ethers Isolated from Hexafluoroacetone/Oxetane Polymer Fluorination compd "C{"F} shift, ppm mp, 'C
(11) (CF .CFOCF FiCF j OCFtCF, , (a) 118.0. (b) 102.6, (c) 116.3, (d) 107.5 -305 a b c d c b a
(32) not enough material to obtain spectra -95
(33) (CF,) a CFOCF.,CF.CF-OC(CF,)..OCF_CF, (a) 133.0, (b) 302.6, (c) 116.4, -85 a b e c e f g f . (d) 107.6, (e) 96.0, (f) 318.7, (g) 134.7
(34) (CF,).CFlOCF.CF 3 CF-OC(CF,).] x (a) 138.0. (b) 102.6, (c) 336.5. (d) 307.8, a b c d c e f (e) 96.1. (0318.6
Discussion
Fluorination conditions were adjusted very carefully for each polymer system in a manner to preserve the structure. The HFA/EO system was the easiest to handle. The HFA/TMO was the most
delicate. It is, however, common in the labora¬ tory for yields and reaction techniques to improve significantly on further development of reactions. The absence of end groups due to carboncarbon bond cleavage along with the large relative amount of isopropyl end groups (especially in the HFA/PO system) is noteworthy. A mechanism consistent with chain scission at the branched carbon is illustrated.
F 0
I 2 II
-C(CF,), - OCHRO- ■>- -CF(CF,), + F-C-R-O-
Further elimination of carbonyl fluoride would result in methyl end groups, in the case of the HFA/EO polymer:
F 2 -0-CF 2 -CF 2 -0-C(CF 3 ) 2 -0 > -0-CF 3 + CF(CF 3 ) 2 "0- + C0F 2
The fact that an equal amount of perfluoroisopropyl end groups and trifluoromethyl end groups were observed in the HFA/EO experiment supports such a view.
After obtaining average molecular weights, the new oils produced in this study were compared with commercial materials. The comparisons were ' done on a DuPont Thermogravimetric Analysis Differential
Scanning Calorimeter to obtain values for T..._..
Separate runs were done on each sample, one using helium as a purge gas and one using air as a purge
gas. The tests with air would provide a measure of the polyether's stability in a moderately oxidative atmosphere. The fluids used for comparison were the
TM
Krytox AZ (Mn 4000) and AA(Mn 6500) and the Fomblin z-25 and Z-45 fluids. Numbers for the Krytox molecular weights were obtained by 19F NMR end group analysis. The variety and type of pendant groups on the Fomblin fluids made it difficult to ascertain Mn by this method. Further the physical property data on the Fomblin fluids are not really comparable to the new copolymers or to the Krytox data since they are of much greater molecular weight (10,000-15,000). The average M of the synthesized fluids were also obtained by NMR end group analysis and are as follows: HFA/EO, M 2,800; HFA/PO, M 3,000;
HFA/TMO, M 2,800. Analysis by GPC using Freon 113 as the liquid phase also showed the fluids to be approximately the same M . These values correspond to a DPn of about 8.
Extrapolation of the T. ,_ data (Table VIII) indicates that the new fluids may have a lower vapor pressure than commercial fluids of comparable molecular weight and similar stability in oxidizing environments. The synthesizing of these three new perfluoro¬ polyethers, along with the characterization of eleven new volatile perfluoroethers, ■ extends the range of perfluoropolyether structures available for study. The highly branched nature of these new oils, along with their morphologies, should prove interesting in exending the knowledge of lubrication
properties and of many other types of physical properties of perfluoropolyethers. Particularly valuable will be the structure versus property content of such information.
Industrial Applicability
The perfluorinated hexafluoroacetone/epoxide and hexafluoroacetone/oxetane copolymers of this invention are useful as high performance lubricants (oils and greases) .
Equivalents
Those skilled in the art will recognize, or be able to ascertain using no more than routine experi¬ mentation, many equivalents to the specific embodi¬ ments of the invention described herein. Such equivalents are intended to be encompassed by the following claims.
Next Patent: FUNCTIONALIZED POLYPHENYLENE ETHERS AND METHODS OF PREPARATION AND USE