<<

1 LUBRIICABING GREASES

FATTY ACIDS IMPROVE UNIFORMITY Reducing The Variables If lubricating performance is to be con- composition can make or break a lubricating sistent, five or more variables must be kept in line: grease. However, the full effect of the com- 1. Variation in the soap cation (metallic component) ponents upon grease performance is not yet scientifi- cally defined. Many practical findings about fats and 2. Variation in the fatty-acid radicals fatty acids guide the greasemaker, but they do not 3. Variation in the oil show up consistently in purchase specifications for 4. Variation in additives - free alkali, free fatty these materials, even when intended for use in the acids, , , anti-oxidants, modifiers, same type of grease. etc. 5. Variation in the processing method and process With the help of the electron microscope, we now control. know that the structural framework of a typical grease So far as base for are concerned, is composed of fibrous crysralline aggregates of their relation m structure and performance have been soap, acting as a kind of colloidal sponge upon the explored in great detail (1) and will not be reviewed mineral oil. This must retain its oil-retentive structure here. under shear, temperature change, and time. Still, two greases which appear quite different under the micro- At the same time, the cation and the fatty- scope, as well as in their composition, may prove to acid radical are so interrelated in their effects upon be equal in performance. There is no single optimum the soap structure that generalizations as to the soap structure in grease. It is specifically interrelated effect of a particular metal (say, ) on perform- to the other constituents, and to the type of service ance also depend on the particular fatty acid combina- into which the grease goes. tion with which it is linked. Some metal soaps are critically affected by the degree of saturation of the Today's grease manufacturer, as he works to acid, others far less so. improve his product in a scientific way, faces he RATIO OF STEARATE task of reducing the variables, so that the effect of TO OLEATE IN SOAP any particular ingredient on performance can be pre- A-IM):o determined. This will direct his choice of materials 8-75: 25 toward more completely purified and uniform composi- C-5030 tion. One of the important milesmnes on the road to APPARENT scientific greasemaking is the increasing availability of fatty acids (and their pre-formed soaps) of well- defined and uniform analysis. Of course, there are many non-critical jobs for which minimumprice greases appear to be "good enough". But the whole trend of industrial design is tending to upgrade grease requirements, and to give enough payoff value to unifonn performance to justify the higher cost of uni- FIG. I - Effect of soap composition on apparent form raw materials. . (Moses and Pudd ington data) A base grease, for example, may be satis- factory when the ratio of stearate to oleate is, say, 50:50. Aluminum oleate, however, is not a satisfactory grease component, and saturated acids will be pre- ferred when aluminum is the meml base. Two - base grease fonnulations may have similar character- istics, despite wide variations in fatty acid polyun- saturant content or chain-length distribution. But two lithium-base greases will respond vety critically to fatty-acid differences of the same order. In general, metals with a narrower tolerance to variations fatty- acid length and structure are being favored in high- quality greases.

Effect of Petroleum Oil Variations

Variation in petroleum oil composition, as it affects grease performance, is too broad and basic a matter for review here. Again, an intricate inter relationship with the soap phase exists. Physical characterisacs of the oil, such as viscosity, are important to the strength and length of the soap fibers formed. In fact, the same soap may form an entirely different fiber saucture as oil viscosity changes. Also, the structure of the fibrous soap network is affected by the chemical composition (cyclic vs. chain hydro- carbons) of the oil. C. J. Boner writes, "Until lubri- cating oils of more definite chemical composition are FIG. 2 - Structure of a lithiuwcalcium stearate -. available, it will be difficult to draw rigid conclusions type grease as shown by the electron as to the effect on saucture. . .". (2) This, of course, microscope (10,000~) adds to ow difficulty in relating the oil-retention and service of existing greases to their fatty acid com- Srandard grades of tallows and animal greases position, except where petroleum oils of identical are defined only in terms of solidifying point (titer), composition are assured. free fatty acids, M.I.U.*, and color, with no control over the particular combination of fatty acid radicals Variation in Fats except that arising from the simila; tendency of fats of particular animals to hold to the same proportions. Animal fats are estimated to constitute over 80 There is also lack of control over water-soluble per cent of the fatty materials used in the production impurities such as proteins and gelatinous matter. of lubricating greases. In this class, the biggest Even for the buyer of animal fats who limits his single component will be inedible tallow, totalling 40 purchases to a single source to minimize these varia- million pounds or more per year. As a traditional tions, the effect of seasonal and geographical factors material to which the art of the greasemaker has long and the time and temperature changes in movementof accustomed itself, tallow makes upin familiarity what stock to market will still exist. it lacks in uniformity. Particularly in the conventional lime-base and sodium-base greases, it can be "doc- tored" to produce greases of structural uniformity and Will Greater Uniformity Pay? reasonably good performance. But as one supplier was recently quoted: "All my business life I've sup- It would be idle to contend that a grease manu- plied fats to your industry and wondered how you facturer, having built a paying business on his skill could possibly make uniform products. Your specifi- in coping with such variable raw materials, should cations could be met by hundreds of combinations of now abandon tallow and grease "across the board" beef, pork, and sheep fars." (3) for something higher-priced. At the same time, the *Moisture, Insolubles, Unsaponifiable -- a "catch-all" classification of the valueless content of the fat. exploration of the full possibilities of uniform raw fatty acids may require wider use of corrosion-resis- materials on performance is now receiving wider rant tanks and equipment. (5) Moreover, the point has study. One tendency has been to seek some inorganic been raised that the claim of greater uniformity for or synthetic thickener which is inherendy uniform. fatty acids has not always been substantiated in But it is proving even more rewarding to explore practice: Acids meeting the conventional test speci- soaps and soap source materials freed from "un- fications as to titer, , iodine knowns .' ' value and the lik,e may still produce differences in performance. There are technical as well as economic limita- tions to he degree that natural fats as such, parti- During World War 11, for example, regulations for cularly those of by-product origin, can be supplied glycerine recovery forced a shift to fatty acids on with any assurance of chemical uniformity. But these the part of grease manuhcturers. Many of these limitations are much less severe when glycerides are r r emergency" mixed acids had no more uniformity hydrolized to fatty acids and glycerol. Distillation, than the "bottom-of-the-bauel" type of fat available solvent separation, fractional crysmllization, and at that time. Experience with these mixed fatty acids other processes such as hydrogenation, are more ef- can not be applied to the purified, specialty types fectively applied to the fatty acids than to glycerides available today. to alter or adjust their composition and to eliminate impurities. Source materials of different origin permit Of course, since uniformity and reproducibility enrichment to arrive at a uniform composition of the are obtained at some extra production cost, this desired chain-length distribution, degree of saturation, brings us to a question of which particular fatty acid and - if necessary - molecular anangement. characteristics are worth paying for. Are we measur- ing and specifying the right things to produce a better Even more strictly standardized fatty acids (or product, in terms of actual grease service? their pre-formed soaps) can be expected in the future. Their commercial development is technically possible. What price - composition analysis? As the value of uniformity becomes more widely held, and thus improves the economics of their manufacture, As shown in Table 1, seven conventional tests the grease industry will have a much wider range of used to characterize fatty materials have gained ac- products to choose from to meet specific needs. ceptance, and are reflected in purchasing specifica- tions of the grease industry. Some of these, taken Fatty Acids vs. Glycerides together, establish the average composition of the material. For example; under certain conditions, the Aside from factors of uniformity and controllabi- composition of stearic-palmitic mixtures can be lity of composition, purified fatty acids have been calculated from titer (6). cited as being preferable to raw fats for the following reasons. (4) It may be asked, should not these specifications 1. Greater ease of saponification. be supplemented or replaced by a breakdown such as shown in Table 2? From the standpoint of research 2. More complete saponification. and product development, such composition analysis 3. Greater yield of anhydmus soap (per 100 lbs. has justification despite the fact that analytically it of component). is a complex task, calling for special experience and 4. In the case of calcium soaps, a more trans- expensive equipment (see Note A). Most suppliers of lucent product. fatty acids, when- cooperating in a new product de- velopment with grease producers, are in a position to 5. More controllable stabilization against oxida- supply such data. As yet, however, the economic justi- tion. fication does not exist for composition analysis as a 6. More controllable length of soap fiber (by regu- routine specification. lating the titer and unsaturated content). 7. Directcontrol of glycerol content in the grease. This can best be understood if we realize that some 30 or more man-hours of analytical time are in- Against these advantages, the question of higher volved in fractional distillation and spectophotometric costs must be measured. The storage and handling of work to set up and run an analysis such as shown in Table 2. Any simplification of the test, by combining ing upon: (a) the fatty-ecid chain-length distribuaon, certain groups of acids, at once reintroduces the (b) the acid-to-base ratio (mono-, di- and tri-aluminum chance that significant effects of minor constituents stearate percentage), (c) the particle size, (dl any will be unrevealed. impurities, such as sodium sulphate, moisture, etc .

Many generalizations are found in the litername Aluminum base soaps are compounded almost as to the effect of fatty acid chain length on grease entirely with fully samrated acids, so that little characteristics. Some (9) have held that only C,, and tendency to oxidize, or other instability,' will arise C,, Acids led to optimum results. Others state that from a properly selected soap component. increasing saturation of the fatty acid portion raises the thermal stability of the grease, as well as the One study favors a di-stearatepalmitate eueectic dropping point. But the fact remains that specific over either stearate or palmitate alone. Another has formulations violating almost all such generalizations reported that hydrogenated marine acids, having some have been found desirable for one purpose or another. C, and C, chain-length acids, appear to contribute to smoothness and gel strength of aluminum stearate One investigator, for example, cautions against greases. Different lubricating oils apparently up the the use of fats or fatty acids with an iodme value scales toward different aluminum soap specifications. above 20. Others have set the limit as high as 70. Hydrogenated materials, particularly those derived from fish oils containing about 75% of C, and C, Acids, are favored for lutxicants by some producers, but not others. Table I1 - Composition Analysis Of A) TYPICAL ANIMAL FATTY ACIDS Many more grease performance tests, made with A 0 no variable except the fatty-acid chain-length com- Myristic ...... 2%...... 2% position, would appear to be necessary before signi- Palmitic ...... 23% ...... 27% licunt differences can be clearly separated from in- Stearic ...... 19% ...... ,.... 17% significant differences. It may then prove to be pos- Oleic ...... 47% ...... 48% sible to concentrate the analytical work on areas and Linoleic ...... 8% ...... 6% degrees of uniformity vital to good grease. Linolenic ...... 1%...... - In he meantime, experimentll lots of fully- 0) TYPICAL HYDROGENATED FISH ACIDS characterized fatty acids are being made available to grease matlufacmrets conscious of this long-range Myristic ...... 4%...... 7% goal, as well as to manufacturers of pre-formed soaps, Palmitic ...... 32% ...... 31% for study and development wok. Stear ic...... ,..22%...... 23% Arachidic ...... 23%...... 21% Use of Pre-forrned,Soaps Behenic ...... 13% ...... 13% Oleic ...... 6%...... 3% Technical as well as historical reasons have Palmitoleic ...... 2% favored separate preparation of metal soaps of alu- minum and lithium. Similarly, metallic soaps of lead, C) TYPICAL DOUBLE-PRESSED zinc and other metals used as additives in mixed base greases are generally pre-formed. Myristic ...... 2%...... - Palmitic ...... 51%...... 48% Producers of metallic soap almost invariably use Stearic ...... 41% ...... 46% fatty acids, and have been in the forefront in seeking ...... 6%...... 6% uniformity of fatty acids as a means of producing soaps of reproducible quality in the grease. They D) TYPICAL HYDROGENATED ANIMAL FATTY ACIDS may employ direct reaction (fusion) or precipitation Myristic ...... 1% ...... 2% from a solublc soap solution. Palmitic ...... 2% ...... 31% Stearic ...... 69% ...... '...... 66% Aluminum stearate, as a typical example, is Olac...... 1% ...... 1% reported (10) to alter grease characterisacs depend- - 1 Table f - Fat and Fatty Acid Specifications NORMALLY DETERMINED BY ANALYSIS Analytical Method Fats Fatty Acids

1. Acid Value - (Free Fatty Acids)...... AOCS - Ca-4-25 L-1-55 2. Ash ...... AOCS - Ca-11-46 L-3a-55 3. Color ...... ASTM-NP A D-155-45T 4. Iodine Value ...... AOCS - Cd-1-25 L-8a-55 5. Saponification Value ...... AOCS - Cd-3-25 L-7a-55 6. Titer ...... AOCS - CC-12-41 L-6a-55 7. Unsaponi fiable...... AOCS - Ca-6a-40 L-40-55 8. M.I.U. (fats only)...... AOCS - (Includes Ca-46)

The use of pre-formed lithium soaps for produc- acids. Saponification temperatures may be lower. tion of lithium greases, whether produced by a separ- Fatty acid-lime soaps may be formed in a much ate supplier or by the grease manufacturer himself, greater pmportion of oil than are fat-lime soaps. pua even more stringent demands on the composition Thus, a wide variety of methods of combinmg the and uniformity of the fattyacid employed. ingredients may be employed. At least 5 such methods have recently been described, all offering benefits According to one study, a from in saponification or hydration as compared to the stearic acid was preferred with a paraffin-base mineral use of fat. (1 2) oil, but for naphthenic oil, a soap is more stable(l1). Another investigator favors a mixture When lithium grease is prepared without the use of fatty acids containing 7.5% C,,; 12.1% C,,; 23.5% of pre-formed soap in the open kettle procedye, the C,;; 54.8% C,,and 2.1% greater than C,,. flexibility of the use of itty acids is again empha- sized. Fatty acids may be dispersed in 10 to 50 per The use of 12hydrorystearic acid, both alone cent of the total oil before lithium solution and in combination with highly purified stearic acid, is added. Thereafter, heating to some 400° F. provides has also been the subject of detailed performance for dispersion, following by cooling to form a gel and tests in a lithium soap for grease usage. working of the gel to form the finished grease.

Grease Production With Fatty Acids In the case of mled-base lubricating greases, the soap may be fonned in situ with fatty acids and a In general, use of fatty acids rather than gly- portion of the mineral oil, combined with the mired cerides will simplify and increase the rate of saponi- metallic bases. Or, one or more pre-formed soaps, fication, but the degree of chis advantage again will may be dispersed in the oil, or a second soap may be depend on the metal hydroxide or alkali employed. added following saponification of the fatty acids with one of the metallic bases. In each case, a more con- A sodium soap base, produced by the open kettle mehod using caustic soda solution, will follow the trollable rate of saponification, as well as of glycerol same general cycle with either fats or fatty acids. and water content, is attained. The time for saponification may be 1%hours, but a "running up" time of 4 to 5 hours in the kettle to The reasoning that fats are to be preferred to mill in the oil and develop the structure will be fatty acids because of the desired effects of glycerol utilized. in certain greases, is not complete. Some producers have found it advantageous to add glycerol as such - Calcium base greases show more definite time again on the basis that more flexibility and control and steam-saving advantages when made from fatty justifies the higher first cost. Fatty Acid Soap vs. NonSoap Gels which they are made, due perhaps m the catalytic As previously mentioned, advances in grease- effect of certain ht impurities on the oil, or vice making are not being confined to the field of those versa. The inorganic thickeners neither increase nor using fatty-acid soaps. Silica gel, and new organic decrease oil stability. However, if we select purified of silica, are effective colloidial agents for fatty acids from the same "specialty" view-point as thickening oil. Others include organophilic bentonite some of the new thickeners, particularly with respect and colloidial attapulgite, which are modified clays. to unit cost, we open up the economic possibility of Phthalocyanine blue is an effective grease-thickening high stability with fattyacid soaps as well, particu- pigment with wide-range temperature stability. Aroma- larly because the effect of anti+xidants will apply to tic thioureas, used with , open up another both the oil and soap components. new class of lubricating agents. The same "quality-mindedness" that is attract- Currently, the estimate for total use of non-soap ing attention to non-soap gels can, if applied to up- thickeners is about 5 per cent of the market(l3). They grade fatty-acid soaps, produce even greater gains for have several unusual advantages from the standpoint the buyer of greases. of temperature and oxidative stability. While they have been advanced as the ultimate answer to the Applica~ionService Expanded search for an "all-purpose grease", their acceptance Expanded use of purified fatty acids has been is limited on several counts. The silica gel and variously estimated at from 30 to 45 per cent of the bcmnite types depend on water-proof surface coat- total fatty materials used Unfortunately, accurate ings of the particles, and these have temperature statistics on this point are unavailable to reveal the limits in the 250-300~F. range, according to recent extent of the trend. But throughout the fattyacid in- observations (14). dustry, the past ten years have seen a wide expansion From the standpoint of cost, effectiveness of in laboratory and development work. Severalcompanies additives, compatibility, and acceptance, the non- have enlarged their research staffs two or more times soap thickeners also have some serious limitations. in this period. The trend to develop and supply fatty Because they absorb but do not dissolve in the oil, acids matched to the need of the lubricating grease they have limited "oiliness" (15). Ignoring cost, it is producer, or the producer of preformed soaps for such possible to make a case that they are superior to use, is certain to continue. What is needed to assure fats, but not necessarily to fatty acids, from the benefit to all concerned is a closer tie-in between stability standpoint. Soap greases have been reputed fatty acid possibilities and lubricating abilities in to be less oxidation stable than the soap or oil from tomorrow's improved line of greases.

REFERENCES

(I) C. J. Boner - The Manufacture and Aplication of (8) Earle, F. R. and Milner, R. T. -- Journal A.O.C.S. Lubricating Grease (1954) Reinhold Publishing (Oil & Soap) Vol. 17, page 106 (1940). Company, New York. page 32(Contains bibliography). (9) Boner, C. J. (See Ref. 1) - page 34. (2) Ibid - page 37. (10) Ibid - page 274. (3) Ibid - page 128. (11) Ibid - page 434. (4) Ibid - page 140. (12) Ibid - page 194. (5)How to Handle Patty Acids - Published by Fatty Acid Division - A.A.S.G.P.,, New York. (13) Carmichael, E. S. - Lubricating Greases, Lectures of Shut Course on Inedible Fats - August 15-20,

(6) Peters, R. M. and Clark, W. C. '- Journal of the 1754 (Publishd as November 1754 issue, Journal A.O.C.S., Vol. 27, page 201 - 1750. A.O.C.S.).

(7) Whitkamp, A. W. and Brunstrum, L. C. - Journal of (14) Boner, C. J. (See Ref. 1) - page 680. the A.O.C.S. (Oil & Soap) Vol. 18, page 47 (1941) .,.Also: Norris, F. A. and Terry, D. E. - Oil & Soap, (15) Ehrlich, M. - Proceedings, 1955 Conv., A.A.S.G.P., Vol. 22, page 41 (1745). New York. Note A - Procedures for Composition Analysis of Fatty Acids

The procedure for determining fatty-acid chain- bond cannot absorb in the ultraviolet region and length by a combination of distillation and spectro- accordingly ,it is determined by difference. Linoleic phommmic methods has been outlined in detail and linolenic acids also do not absorb in the ulua- (7). It calls for a fractionating column with a mini- violet region but they can be made to do so by mum of 15 to20 theoretical plates and nearly perfect cmverting them to their conjugated forms. Thus, adiabatic conditions. Podbelniak columns are fre- these acids are heated with alkali under specified quently used. Normally, the methyl esters are conditions to shift the double bonds from an un- distilled. The distillation is most effective in de- conjugated to a conjugated position. Thereafter, the termination of shorter chain saturated acids, such conjugated will absorb at 233 mu as capric, myristic, palmitic. Molecular distillation whereas the conjugated linolenic acid will absorb is also sometimes applied, particularly for separat- at 268 mu. A teaaenoic acid, which is present and ing stearols and other impurities and for the frac- which has been conjugated will absorb at 350 mu. tionation of unsaturated acids. Because of these absotption characteristics, the transmission is determined at the appropriate Fractional crystallization as an analytical wave length and &is transmission is a direct method to separate saturated from unsamrated measure of the quantities of each of the conjugated acids is frequently employed (8). The individual polyunsaturated acids present. An appropriate unsaturated acids in a given fraction .are then blank is also run. Oleic Acid, as indicated above, determined by s pecuophotome~ic procedure. A is calculated by difference. recent summary of the procedure appears in Oil, Paint and Chemical Review (July 14, 195): If some coojugated acids are present initially in the acid mixture, the absoktion of these may be "The three most common unsaturated acids in determined directly. Subsequently, the conjuption the Co fraction are oleic, linoleic and linolenic procedure is carried out in order to determine the acids and hese have one, two and three double tom1 quantity of each of the polyunsaturated bonds respectively. Oleic acid with its one double acids." Prepared by FATTY ACID DIVISION ASSOCIATION OF AMERICAN SOAP & GLYCERINE PRODUCERS, INC.

Reprinted for