CA2186848A1

CA2186848A1 - Lubricated metal workpiece and method

Info

Publication number: CA2186848A1
Application number: CA002186848A
Authority: CA
Inventors: Alan Robert Carr; Peter Geoffrey Sheasby; William Francis Marwick
Original assignee: Individual
Current assignee: Rio Tinto Alcan International Ltd
Priority date: 1994-04-12
Filing date: 1995-04-12
Publication date: 1995-10-19
Also published as: CN1044003C; KR970702352A; AU682599B2; KR100388008B1; AU2218595A; DE69503773D1; BR9507319A; MX9604675A; EP0755427B1; ATE169055T1; JPH09511542A; EP0755427A1; WO1995027767A1; CN1149311A; ES2120742T3; DE69503773T2

Abstract

A lubricant is provided which has a hardness in the range 0.1 - 10 N/mm at all temperatures in the range 15 - 40 .degree.C. The lubricant consists of at least one full ester of a glycol with a fatty acid, e.g. ethylene glycol dilaurate, optionally mixed with a minor amount of a fatty acid such as stearic acid, and/or a minor amount of a partial ester of a glycol with a fatty acid. The lubricant is useful in the production of press-formed components, and particularly in techniques for converting aluminium sheet to adhesively bonded aluminium structures.

Description

WO 95127767 P~
218fi848 - 1 -LI~BRICATED N~T~T~ 'lY~ D J~El~OD
, 5 This invention relates to lubricated metal workpieces, particularly of steel and aluminium, used in the production of press-formed components, and in particular to a method of using such workpieces to make structures of shaped components.
There is current interest in techniques for producing adhesively bonded structures of shaped aluminium cmnrnnl~ntq for use in the automotive industry. Such a technique is described for example in EPA 127343. The technique of converting a coil of aluminium metal sheet into a structure of shaped components f or use in the automotive industry may typically involve the following steps:-- The metal surface is pre-treated to provide a strongly bonded layer thereon which acts as a base for subsequently applied adhesive.
- A lubricant is applied to the treated metal coil. The coil may then be stored or transported, with the lubricant serving to protect the treated metal surface, and is cut up into pieces ready 2 5 f or press - f orming .
- The pieces of metal sheet are press-formed into ~ ^nt,~ of desired shape. This and subsequent operations are all performed on an automobile product ion l ine .

3 - Adhesive is applied to selected areas of the shaped components, without first removing the lubricant .
- The I ,~nn~ntc are assembled into the shape of the desired structure, and may be spot welded or otherwise fixed to hold the structure together until the adhesive is cured.

W0 9~127767 21 8 ~ ~ ~ 8 I ~
_ ~ _ - The a&esive is cured at elevated ;~
temperature .
- The metal surfaces of the structure are -subjected to an aqueous ;~lk~linP cleaner which removes 5the lubricant. ~-- The structure is painted.
Alternatively, the presg-formed c~r~mPntq may be secured together to form the structure by mechanical means, e.g. by rivets or spot-welds, either 10in ~ n to or instead of adhesive bonding.
A lubricant f or use in such a technique needs to fulfil several requirements:
a) The lubricant must, obviously, have suitable lubricating properties for the press-forming operation.
5b) The lubricant should be solid at likely metal storage temperatures in order to prevent stacked sheets from sticking together. Furthermore, a film of lubricant that is liquid or sticky is prone to smear and to pick up dust and dirt.
20c) Since it is not practicable in a production line to remove lubricant prior to application of adhesive, the lubricant needs to be compatible with an adhesive if one i8 to be used.
d) After the a&esive has been applied and 25cured, the lubricant must be readily removable by an aqueous ~lk~l;nP cleaner of the type convPn~ n~lly used to prepare metal surfaces for r~;ntin~
The lubricants of EPA 227360 are designed to be useful, not only for the technique described above, 3but also for other forming and shaping operations perf ormed on a variety of metals .
In one aspect, EPA 227360 provides a lubricating composition for press forming consisting of a lubricant dissolved or dispersed in a volatile liquid 35medium, wherein the lubricant comprises at least one ester of a polyhydric alcohol having two or three ~ Wo 95/27767 ~ 1 8 6 8 ~ 8 p I .,~
hydroxyl groups of which one or two are esterified with a long chain carboxylic acid and has a melting point above ambient temperature but low enough to permit removal f rom a metal surf ace by an as~ueous A 1 kA~ l; n,A
5 cleaner.
EPA 227360 mentions that mixtures of esters may be used and may be advantageous; and that the lubricant may contain a minor proportion up to 50~ of one or more other lubricating compounds such as long-10 chain carboxylic acids. The lubricants exemplified are: diethylene glycol monostearate in solution in xylene; and diethylene glycol distearate in solution in xylene.
Although the lubricants described in BPA
227360 are generally successful at meeting re~uirements c) and d), they are sometimes less successful at meeting re~{uirements a) and b) . It is sur~risingly found that lubricants of this kind are ineffective, so far as aluminium forming operations are cnArArn~d, at 20 temperatures above their liyuidus. For good aluminium lubricating properties, in ester lubricants of this kind, it appears necessary that some, , An~ be present in the solid state, so that the lubricant is solid or at least mushy or viscous, at the forming 25 ten~perature which may be as high as 35C or 40C or even higher.
It might appear a simple matter to solve this isolated problem by using a different ester with a higher melting point. A difficulty with this strategy 3 is that higher melting esters tend to be relatively hard at low and ambient temperatures, to the extent that they readily spall and flake off metal surface to which they are applied. Metal forming at 15 or 20C
cannot satisfactorily be performed under conditions 35 where the lubricant flakes off the metal workpiece.
For use in various parts of the world, there is a need 21868~8 for a single lubricant system which meets both these high- and low-temperature criteria. It is an object of this invention to meet that need. '` '-In one aspect the invention provides lubricated metal, wherein a surface of the metal '-carries a film of a lubricant which a) consists essentially of at least one full ester of a di- or poly-hydroxy compound with a C8 - C18 saturated carboxylic acid optionally in admixture with a minor amount of at least one long-chain carboxylic acid and/or a minor amount of at least one partial ester of a di- or poly-hydroxy compound with a C8 - C1 saturated carboxylic acid, and b) has a hardness in the range 0 . 2 - lO N/mm at all temperatures in the range 15 - 30C, preferably in the range 0.1 - lO N/mm at all temperatures in the range 15 - 3 5 C .
In another aspect, the present invention provides a method of making a structure of shaped aluminium ~ ~ AntS starting from lubricated ~ m;n;llm metal sheet as defined, comprising the steps:
- forming pieces of the sheet into components, - bringing the ,~ _ -ntR together in the shape of the desired structure, - and securing the ~l AntA together by -'hAn; ~-Al and/or adhesive means .
Hardness of the lubricant is measured by a techni~ue whereby a block of the uncoated lubricant is 3 equilibrated at a given temperature and is penetrated by a steel needle. ~he test procedure used eRsAnt;~lly involves driving a pointed 12 mm diameter needle into the lubricant at a 5peed of 20 mm/minute, achieved with the use of materials testing machine such as an 35 Instron, and recording the load as a function of the needle penetration into the lubricant. Separate tests _ _ . , _ . , = ~ . _ . , . .. . . _ _ _ _ _ . _ ~ Wo 9~l27767 2 1 8 6 8 ~ o are conducted at various temperatures to derive the full curves. The hardness value ~uoted i9 then found as the slope of the graph of penetration load versus penetration distance.
Although forming, e.g. press-forming, of metal sheet is generally performed at temperatures in the range 15 - 30C or 35C, in some tropical l~^,r~ti. nA
tt",~eld~ures in the press may rise to 40C or even 45C. It is therefore preferred that lubricant films of this invention have specif ied hardness values at temperatures within the range 15 - 40C, in the case of particularly preferred lubricants, within the range 15 - 45C.
If the lubricant film is too hard, it is likely to be brittle and have poor frictional characteristics during forming e.g. press-forming.
If the lubricant film is too soft, then again the lubricating characteristics are inferior.
Preferably, the lubricant film has a hardness in the range 0.1 - 5 N/mm at all temperatures within the range specified at which forming e.g. press-forming is likely to take place in different parts of the world. It is surprising that the hardness of the lubricant film has useful predictive value for its lul~ricating characteristics.
The major c, ^^t of the lubricant film is a Eull ester of a polyhydric alcohol with a long-chain carboxylic acid. Dihydric or trihydric alcohols having 2 - 6 carbon atoms are suitable, for example ethylene 3 glycol, propylene glycol, diethylene glycol and glycerol. The long chain carboxylic acid is preferably a saturated straight-chain monocarboxylic acid having from 12 to 18 carbon atoms in the chain, such as lauric, palmitic or stearic acid. Mixtures of esters may be used and may be advantageous. Particularly preferred esters are ethylene glycol di-laurate (~GD~) Wo gSl27767 21 8~8 4 8 - 6 ~ c ~- ~
and propylene glycol di-6tearate.
The full ester or esters may be used optionally in ~' Yt~lre wit~ a minor ,amount of at least one long-chain carboxylic acid, preferably a saturated S straight-chain monocarboxylic acid having from 14 - 20 -carbon atoms in the chain. This optional minor ^nt iS present in an amount of less than 509~ most usually 5 - 2096, by weight on the weight of the mixture. A particularly preferred fatty acid is 10 stearic acid.
The full ester or esters may be used preferably in A~' Ytllre with a minor amount of at least one partial ester of a polyhydric alcohol with a long-chain carboxylic acid. Dihydric or trihydric alcohols 15 are suitable, f or example ethylene glycol, propylene glycol, diethylene glycol and glycerol. The long chaincarboxylic acid is preferably a saturated straight-chain monocarboxylic acid having from 12-18 carbon atoms in the chain, such as lauric, palmitic or stearic 20 acid. Mixtures of partial esters may be used and may be advantageous. Particularly preferred partial esters are ethylene glycol -1 ~lrate IEGMT~ and propylene glycol monostearate.
The full esters should be present in the 25 lubricant mixture at a rrnrPntration of 40 - lO0 wt~.
Preferably the lubricant consists of 50 - 85 wt~ of a full ester such as EGDL, 10 - 30 wt~ of a partial ester such as EGML and up to 20~ e.g. 5 - 20 wt~ of a fatty acid such as stearic acid. Lubricant composition can 3 drift during storage, resulting in a somewhat differellt composition on the lubricated metal surf ace, and these f igures ref er to the lubricant when f reshly made .
Proportions herein are determined by analysis e.g. by standard technir~ues involving gas chromatography, mass 35 spectrometry and IR spectrometry; they do not n~.r~-cs~rily correspond closely to manufacturers ' stated _ _ _ _ ,, , , , _ .. . . . . _ . ... .

Wo 95l2776,7 2 1 8 6 8 4 8 P~l,. . 7 proportions in commercially available materials.
To the best of our belief, there is no single ester of commercial purity which meets the above- stated hardness requirement. Suitable l1~hrirAntq may be 5 achieved in one or both of two ways . The f irst is by blending two or more, ~n~ ~ together . The second is by using purer material.
These esters are not easy to purify. But to the best of our knowledge and belief, EGDL has a 10 melting point of about 50C; and EGML has a melting point in the range 23 - 25C; and mixtures of the two have melting ranges intermediate these two f igures .
Full esters and partial esters such as EGDL
and EGML can have impurities arising from two main 15 Sources:-a) The nominal fatty acid, e.g. lauric acid, isin fact a mixture of saturated long-chain monocarboxylic acids, typically cnntAin;n~ more than 30~1; of acids other than the nominated one. An effect 20 of these cnnr~m; nAting acid3 is to depress the melting point of the ester.
b~ The ester is derived from a fatty acid mi~:ture which cnn~;nA ethylenically unsaturated acids. Such impurities make the lubricant less 25 adhesive-compatible and less easy to remove from the metal surface, and are therefore preferably absent or present in amounts below 5~ by weight.
As noted above, the lubricating characteristics of the lubricant film on lubricated 3 metal according to this invention fall off at both excessively high and excessively low temperature. We have developed a test, which is described below, for measuring lubricating characteristics in terms of a frictional coefficient (mu). This frictional 35 coefficient is preferably below about 0.1 at all temperature9 within the range of interest, that is to Wo 95/2~76~ 2 1 8 6 8 ~ 8 P~ IAID~ S~

say 15C up to 30C or 35C or 40C or 45~C. As noted above, the hardness of the lubric~t film at any temperature is predictive of its frictional coPf f; ripnt, Depending on its intended use, the lubricant may need to be compatible with subsequently applied adhesive. In general, the esters described herein are compatible as a result of being either absorbed or displaced by subsequently applied adhesive without grossly impairing the adhesive bond strength obtainable. By contrast, resinous lubricants and metal soap lubricants are generally not adhesive compatible in this sense.
The lubricant has a melting point above ambient temperature, preferably of at least 30C, more preferably at least 40C. This ensure6 that the lubricant is present as a solid f ilm on the metal substrate, which avoids problems with 6mearing and blocking during coiling, decoiling, slitting and cutting. The use of such a lubricant avoids ~l~ntAminAtion of the metal surface with a possible adhesive-incompatible oil or contæm;nAnt and prevents local build up of lubricant to an undesirably thick Iayer .
2~ The lubricant melts at a temperature low enough to permit its removal from a metal surface by an aqueous AlkAl ;nP cleaner, such as is used in automotive production lines to prepare metal parts for F~;ntin~
The highest practicable temperature for aqueous 30 alkaline cleaners in such circumstances is about 70C.
Lubricants melting below 70C and preferably below 65C
can thus be removed by aqueous A 1 kA l; n~ cleaners .
Lubricants melting above 70C may or may not be removable depending on whether they have chemical 35 groups, e.g. hydroxyl groups, which can react with the alkali to assist removal from the metal surface. Thus _ _ _ _ _ . . .. . . . . . . .. . .. . _ _ ~ _ Wo gs/27767 21 8 6~ ~ 8 g fo~- example, a commercially available wax having a melting point of 85C and an acid number of 135 to 155 by DIN 53402, was found not to be removable by aqueous ;~kAl inf~ cleaners. A lubricant is deemed removable by aqueous ~ l k:~ 1; n~ cleaners if it can be removed by treatment for 2 minutes at 70C with a 15~
by weight aqueous solution of Ridolene 160 (a silicate-based proprietary cleaner marketed by I . C . I . plc . ) A further aspect of this invention involves applying the lubricant to the metal in the absence of any volatile solvent or diluent. This avoids the need to evaporate volatile liquid from the lubricant film, and avoids t~e need to include any surface active agent in the lubricant. It is found that the molten lubricants have satisfactory viscosity for spraying or for application by roller coat. To ensure rapid soli-~; ficatio~ of the lubricant film, the metal may be pre-cooled. To ensure good adhesion of a uniform f ilm, the metal may be pre -heated .
Alternatively the lubricant may be dissolved in a volatile solvent for application to the metal.
Indeed, very thin films can only be applied from solution. The use of solution permits control of lubricant film thickness to within less than 0.5 g/m2.
The lubricant may be applied to steel or other metals, but is likely to be pr;n~ y used on aluminium, which term is used to cover the pure metal and alloys in which Al is the maj or, ~~ ^n~ . A metal surface may carry a strongly-bonded inorganic and/or 3 organic pretreatment or primer layer, on the top of which the lubricant is present. Such non-metallic layers are well known, and may be provided for example as chemical conversion coatings or deposited coatings of the no-rinse type, based on chromium, titanium or 35 zirconium; or may be an anodic oxide layer (on Al or Ti) or a siloxane layer. The metal may be in sheet WO 95/27767 2 ~ ~ 6 8 ~ 8 P~ ~ 5~

form. The rate of application of lubricant will depend on the intended use, but may typically be ~in the range of 0.1 - 10 g/m2, e.g. 0.25 - 8 g/m', particularly 1 -4 g/m', for aluminium coil to be formed into 5 adhesively bonded structures.
Reference is directed to the AC ~ ying drawings in which:
- Figure 1 is a schematic view of a strip-draw apparatus used for testing lubricated metal;
- Figure 2 is a perspective view of a modif ied strip-draw apparatus;
- Figure 3 is a graph of hardness against temperature for several lubricants.
- Figures 4 and 6 are Bar Charts showing frictional coefficients of two lubricants at different temperatures and different rates of application.
Figure 4 is for lubricant 2. Figure 6 is for lubricant 1.
- Figure 5 is a bar chart showing lubricant residues after different bakes followed by cleaning.
~ purpose built strip-draw rig was designed and constructed with reference to ASTM 4173-82 for testing sheet metal forming lubricants. The apparatus is shown in Figure5 1 and 2. The die set shown in Figure 1 was designed to simulate material flowing between pressurised binder surfaces ~ ntA;nin~ a draw bead a ~ L ~ t . The die set of Figure 2 was designed to simulate flow between parallel binder surfaces 50 as to allow conventional frictional values to be obtained.
3 Referring to Figures 1 and 2, one die 10 of each tool set is mounted on a load cell 12. The other die 14 of the tool set is mounted on a hydraulic cylinder 16. Elat strips 18, hydraulically pressurised between the two dies, can then be pulled through a particular tool set while the clamp load is measured.
The draw load is also measured using a second load cell ., .. , _ . _ . ... _ _ _ .. , , . , .. ,, _ _ _ _ _ . .

~ W09S12776'7 21 8 ~8~ 8 Z0 mounted between a testing machine gripping jaw 22 and a cross head 24. Thus, when used in conjunction with the f lat parallel platen set of Figure 2, a conventional~ frictional value is obtained.
The strip draw rig is designed to be mounted on either a press s;r-llAtQr or a standard te~sile testing f rame, depending on the variables under investigation .
Lubricated strips of material, 50 mm wide, were placed between the two f aces of the f lat tool set of Figure 2 and hydraulically pressurised to a particular load. The strips were then drawn through the die set of Figure 1 for a distance of approximately 250 mm, the draw and clamp forces being recorded as a function of time/displacement of the drawn strip.
Results presented in the form of a graph (draw force/2) versus clamp load have a slope eciual to the conv,onti~AnAl friction coefficient.
E:XANPI,E 1 A lubricant f ormulation according to the invention had the composition, in wt5c:-61~f ethylene glycol dilaurate (EGDL).
19% ethylene glycol monolaurate (EGML) 11% stearic acid 9~ other ester species The identity of the ~ Ant ,A, was tl_t~Arml nAd by standard gas chromatography/mass spectrometry 3 techni~ues. This formulation is hereinafter called lubricant Another lubricant f ormulation according to the invention had the composition, in wt~:-70% ethylene glycol dilaurate 21. 596 ethylene glycol monolaurate 8 . 5% other ester species W09sl27767 21 8fi8 ~ 8 .~

The identity of these components also was determined by ætandard gas chromatography/ma;ss spectrometry techniques. The fn l~t;nn is hereinaf ter called lubricant 3 .
A formulation called lubricant 2 was made up for comparison. Lubricant 2 nnnt~;nq commercially available EGML 90~ and 5tearic acid 10~. This lubricant falls outside the scope of the present invention, and is included for comparison purposes 1 0 onlY.
This commercially supplied ethylene glycol monolaurate has been analysed by us and f ound to contain seven different acids in proportions as follows: caprylic (C8) 3.9~; capric (C10) 5.89f;
lauric (C12) 339~; myristic (C14) 16.89~;
palmitic (C16~ 11.996; oleic and stearic (C18) 28~.
Lubricants 1, 2 and 3 were applied by spraying on to aluminium alloy sheets which had been preheated to 50C. By this means, uniform films could be applied at controlled thickness. The hardness of the lubricants was measured (by the method described above) and the results are recorded in Figure 3.
Lubricants 1 and 2 were further tested in the strip draw rig illustrated in Figures 1 and 2. In each case, tests were performed at different t~ ~ dLL-r ~s in the range 0 - 50C; and at five different rates of lubricant application ranging from 1 - 6 g/m~. The results of these tests are shown in Figure 4 (for lubricant 2) and Figure 6 (for lubricant 1 batch 2, see 3 below)-~MPLE 2 Lubricants 1 and 3 from Example 1 were evaluated. Lubricant 2 from Example 1 was used for comparative purposes.

wo95l27767 ~ 8 E~erimental P~ ~cel.. e The experimental ~ork described below was carried out on 1. 6 mm gauge 5754 material.
5 2.1 Application of Lubricant to AlllTn;n;llm Sheets The procedure for lubricant application consisted of pre-heating a reservoir of the new lubricant to 70C, and applying this onto sheets using air-assisted airiess spray nozzles. Lubricant was 10 applied to sheets which were held at both room temperature (20C), and preheated to 60C. These sheets were then placed in stacks. In the case of the pre-heated material, the sheets were placed in a stack when the lubricant had solidified.

2 . 2 Adhe6ive Co:npati~ility The standard test method for adhesive COlllpatibility i8 to assemble standard lap shear j oints wi th a 10 mm overlap, using lubricated 1. 6 mm 20 pretreated coupons and a standard adhesive. A string of six such joints are then exposed to c ' ;n~d stress/humidity testing under a constant load. The time to failure of the first three joints in a set of six joints is then noted. Individual lap shear joints 25 are also exposed to salt spray for given periods of time, and then tested for static strength retention.
Tests were carried out on j oints manuf actured with the luoricant 1 on their surfaces prior to bonding. Two lubricant weight levels were evaluated, 3 namely 2 . 0 g/m' and 5 . 5 g/m' .
2.3 Lubriclnt Softening as a Functio~ of Temperature The Wax Penetration Test, was used to determine the softening response as a function of 35 temperature. The test procedure used essentially in~olves driving a pointed 12 mm diameter needle into WO 95l27767 ~18 6 8 ~ 8 r~ .s/c. ~

the lubricant at a speed of 20 mm/minute, àchieved with the use of materials testinq machine sùch as an Instron, and recording the load as a function of the needle penetration into the lubrican~. Separate tests are conducted at various temperatures to derive the full curves. The hardness value quoted is then found as the slope of the graph of penetration load versus penetration distance.
2.4 Strip Draw Evaluation Lubricated sheets were produced with 3 g/m~
of different lubricants via the pre-heated blank route, as indicated in section 2.1. These 3heets were guillotined into strips 50 mm wide and then drawn through the strip draw rig, using the described procedure, to allow friction values to be determined at temperatures of 10, 20, 30, 40 and 50C.
2 . 5 Pre6~ For:ning Evaluation o~ the Lubric~mt 1 Press forming testa were carried out on a press simulator to evaluate the lubricant. Two distinct trials were used, namely:
(a) Square pan depth to failure.
(b) Pressed dome height to failure.
The above trials were carried out under standard conditions on a 275 mm square tooling without the draw bead sections.
Sheets of AA5754-0 were pressed with 3 g/m' of both lubricants 1 and 2 to allow the comparative 3 performance to be assessed.
2 . 6 Simulation of Po~6ible Thermal Cycle6 of Pre-Lubricated Stacks In order to simulate possible thermal cycles 35 which may be experienced by pre-lubricated material, lubricated stacks were produced by applying the _ _ _ _ _ . . .. _ ... . , , .. . _ ~ W0951277fi7 ?~ ~l 8~8 A ~

lubricant to pre-heated blanks, as described in section 2.1.
Four stacks were produced l-nnt~;n;n~ some thirty sheets, each 500 x 500 mm, with a nominal 3 g/m2 5 lubricant weight. These stacks were heated to four different temperatures in an oven at, 30, 35, 40 and 45C respectively. After removal from the oven, each stack was le~t to cool with a centrally applied weight of 18.1 kg. After destacking, coupons were removed 10 from a number of adjacent sheets to quantify any lubricant transfer observed.
2 . 7 Cleaning, Oven E:vaporation and ~
Two distinct tests were carried out in this 15 section, namely different oven bakes followed by a cleaning stage, including no bake, and a typical bonded stL-ucture route with the adhesive cure cycle included.
For the first series of tests, pretreated strips of aluminium were lubricated with lubricant 2 20 (3 4 g/m2 ) and lubricant 1 (3 . 8 g/m2 ), and given the following treatment:
a ) 2 0 minutes at 17 0 C
b) 20 minutes at 180C
c) 20 minutes at 190C
d) 20 minutes at 200C
e ) No oven bake .
All strips were then cleaned in stirred 20 g/litre solutions of Chemkleen C~t165 at a temperature of 60C for 3 minutes. After drying, 3 organic ~ ~rt~m;n~tion on the strips was measured, as carbon, by analysis at 600OC.
For the second series of tests, clean sheets of aluminium were coated with lubricant 1 and lubricant 2 at a coating weight of approximately 4.5 g/m'. The 35 sheets were then subjected to a cumulative oven-bake and alkali-clean cycle. This consisted of:

WO 95l27767 218 6 8 4 8 r~ r ~

a) 10 mins at 145C
b) 20 mins at 190C ; - ' c) 20 mins at 190C
d) 30 secs alkali clean, (stirred 2.59~ w/w Ridolene 336 at 60C).
Final coat weights were measured after the cleaning stage using gravimetric det~inat;on~
3. RESOT~TS

3.1 Applic~tlon of the Lubricant to the Sheet~
Satisfactory results were obtained by spraying the lubricant 1 onto sheets held at both room temperature, approximately 20C, and sheets pre-heated to 60C. The lubricant solidified upon contact with the sheets held at ambient temperature. However, the latter condition allowed the lubricant to remain liquid on the sheets f or a short time period .
The lubricant itself passed through the spray nozzles without any additional problems to those encountered with the lubricant 2.
3.2 Adhe~ive Co~pati~ility The results of stress-humidity and salt spray testing on joints produced with lubricant 1 on their surfaces are presented in Table 1. This Table shows a good strength retention after 20 weeks salt spray, and a testing duration in exces~ of 100 days during stress/humidity with a 5 MPa applied stress.

3.3 Lubrica~t Softening a~ a Punctiorl of TenLperature The results of the Wax Penetration Testing carried out on lubricants 1, 2 and 3 are presented in Figure 3.
Two batches of lubricant 1 were made on separate occasions. Batch 1 is shown by filled _ _ _ _ _ _ _ _ _ , . , . . . _ . .. . , .. , ..... , . . . . . _ _ o 95/27767 ~ 8 q j oined by a solid line . Batch 2 is shown by shaded squares joined by a dotted line. Both materials fall within the scope of the invention, as does lubricant 3, shown by stars.
Lubricant 2 is shown for comparison. The hardness was relatively low at all temperatures.
3.4 Strip Draw Evaluation Table 2 shows the comparative performance of the lubricants l, 2 and 3 over the measured temperature range for a given lubricant weight of 3 g/m'. These f igures show an improved perf ormance of the lubricant batch 1 at temperatures of 30, 40 and 50C. They also indicate a similar performance at 20C.

3 . 5 Pre~s Forming E~aluation of the Lubricant 1 The results of the press forming trial8 are sho~Am in Table 3. This Table indicates that both lubricants give a similar performance during 8tretch 20 forming, but an improved performance is obtained during square pan forming wi~h the lubricant 1. The values quoted are the average of five tests in each case. The tests themselves were carried out in ambient conditions of around 22-24C.

3 . 6 Simulation of Pos~ible Thennal Cycles of Pre-Lubricated Stacks Stacks of lubricated sheets, having lubricant on their surf ace, which were heated to 3 0, 3 5 and 30 40C and subsequently cooled with a centrally applied weight showed no evidence of de-stacking ~roblems or lubricant transfer between adjacent sheets. The corr~qp~n-l;n~ stack heated to 45C was more difficult to separate, showing clear evidence of a "patchy~
35 appearance, and slight lubricant transfer between adj acent sheets . With the lubricant 2, a similar _ _ _ . . . .

W095/27767 ~1 8~848 .

effect was seen at a temperature of 35C.
3.7 C'le~;n~, Oven ~vaporzltion and R~
The results of the different oven bakes 5 followed by a cleaning stage, including no bake followed by a cleaning stage, are given in Figure 5.
This figure shows that high oven bakes contribute to the surface cl~nl ;n~s.
The results of the oven evaporation trials 10 show that lubricant l evaporates almost totally compared to lubricant 2, 0.03 g/m2 and 0.35 g/m' residue respectively. Final coat weights were also measured after the final alkali clean using gravimetric determination. The results of both lubricants fell to 5 between -O.Q1 and -0.03 g/m~ indicating that the cleaning stage is removing all of the residues.
4 DIS~- S
20 4.1 Applicatio~ of the Lubricant to the Sheets No difficulties were experienced with the spray applicatio~ of lubricant l. Adherence to the metal surface was improved by pre-heating the sheets, although a satis~actory appearance was obtained with 25 room temperature metal.
4 . 2 A&esive Compatibility The results of the-stress/humidity testing show that joints manufactured with lubricant 1 at a 3 level of 2 g/m2 are still on test, with l90 days achieved to date for all stress levels tested.
Salt spray data after 20 weeks exposure shows P~r~Pl 1 ~n~ strength retention at both lubricant weight levels, out -perf ormlng the lubricant 2 .

~w09sl27767 2186848 ~- r~

4.3 Lubricant Softeni~g as a Function of ~remperature Results of the Wax Penetration te8t, presented in Figure 3, show the hardness i~ Luvl ' over the temperature range ~ m; nPd . Earlier work had 5 suggested that the hardness value 6hould be m~; nti~ i nPd between 0.1 and 1. 0 N/mm ûn the vertical logarithmic axis. Significantly ~Y~ee~;n~ the higher hardness values at lower temperatures will produce a wax which is brittle and has iimited value in terms of press die 10 forming. At the higher temperatures, hardness values less than 0 . 1 N/mm correspûnd to the melting range of the wax. Figure 3 shows that the hardness of lubricant 1 falls to 0.1 N/mm around 41-42C, corresponding to its melting range. This is approximately 10-12C higher than lubricant 2. Thus the upper melting range has been significantly increased without producing a brittle wax at the lower temperature range. ~ubricant 3 has a hardness that is strikingly ind ~ .deL,t of temperature.

4.4 Strip Draw Evaluation In terms of frictional performance versus temperature for a lubricant weight of 3 . o g/m2, Table 2, the lubricant 1 demonstrated a signif icant 25 il~l~JLUV~ t in the frictional coefficient at 30, 40 and 50C, whilst having a similar performance at 20C.
4 . 5 Pres~ Forming Evaluation of the Lubricant 1 The results of the press forming evaluation 3 indicate that the lubricant 1 formulation has an equal performance during stretch forming, but a somewhat better performance during square pan pressing.
this should give an advantage to c- ~ U~ S such as door rings and door inner pressings, where deep corner 35 features are required having small corner sweep radii.
Thus the performance; ~uv, ^nt will be beneficial.
_, . . .. ,, _ _ _ _ _ . , , WO 95/27767 '~1 868q 8 4 . 6 Simulation of Po~sible Thermal Cycle,~ of Pre-,ubricated Stack~ , Results of thermal cycleæ applied to pre- .
lubricated stacks have shown no evidence of lubricant 5transfer at 40C, and only slight evidence of lubricant -transfer when the stack was heated to 45C and then cooled . Thus, problems of lubricant transf er with the lubricant l will now become apparent at temperatures between 40 and 45C. This performance is much better 10than the lubricant 2 where no evidence of transf er was vislble when the stack was heated to 30C, but evidence was visible when the stack was heated to 35C. Hence a 10C temperatur- i _LiV. ~ in terms of h=inr~1 ;n~
perf ormance has been achieved .

4 . 7 Cleaning, Oven Evaporation and ~
Trials to assess the cleaning of the surf ace after either various oven bakes or no bake, Figure 5, have shown that ~ n solution does clean the 20surfaces reasonably effectively. High oven bake conditions, 190C and 200C, definitely contribute to surface ~-1eAn1 ini~Sq in the case of lubricant l, which evaporates very cleanly from the aluminium surfaces.
Trial8 to determine the relative evaporation 25and ,~ ;n;ng residues of lubricant l versus lubricant 2 have shown that the lubricant l evaporates almost completely prior to the cleaning 6tage. Further, lower levels of carbon residuals are obtained.
3o ~ W0 9512M67 218 6 8 ~ 8 Tal~le 1: Stress/~Iunidity and Salt 5pray Dat~ for the Lubricant 1 .. S
Lubri~ant Stress/};umidity Salt Spray Re5ults Weight g/m' Days to Failure ~MPa~
3 MPa 4 MPa 5 MPa O wks 8 wks 20 wks 102.0 lso~ lso+ lso+ 27.8 2s.s (92~) 25.1 (9o~) s.s lso+ lso+ 137 28.4 2s.2 (89~) 23.6 (83s) 15 Table 2 Friction Value at temperature C
Lubricant 10 20 30 40 50 0 . 017 0 . 012 0 . 046 0 . 069 2 0 . 020 0 . 036 0 . 096 0 . 107 30 . 092 0 . 020 0 . 043 nd nd nd - not ~et~r~n~ nf~d Tal~le 3: ~ffect of Lu~ricant on Press Fn~-h; 1; ty L~hricant Coat Weight Ss~uare Pan Dome ~Ieight (g/m2 ) Depth mm mm 23.0 59.1 51.5 35 1 3.0 74.3 52.0 WO 9S127767 ~1 8 68 ~ 8 ~Y~MPLE 3 Propylene-1, Z-glycol distearate (PG12DS) was made as f ollows:
,.
Material~:
Stearic Acid MW 284.47 mp 69 - 70C
Propane-1,2-diol: MW 76.09 bp 188.2C
Xylene bp 137 - 140C

~quip:ne~ t:
Heated 5 l reaction vessel fitted with th~ -ter, Dean and Stark, and reflux condenser.
Procedure:
Stearic acid (1500 g) and propane-1, 2-diol (200 g) were placed in the reaction vessel.
Approximately 1. 5 l of xylene was added, and the mixture stirred.
The vessel was heated to boiling and reaction continued under reflux for apprn~ tPly 10 hours (bp_135C) .
Azeotropic li~luid gathered in the Dean and Stark was removed at f requent intervals . A tDtal of 25 apprnYir~tely 100 ml being collected. Two additions of 10 ml of propane-1, 2-diol were made during the course of reaction to c~ ,-nC~te for 1088 of diol in the azeotrope .
Once the boiling point of the mixture had 3 risen to 137C, xylene was removed by distill~t;on.
Distillation was stopped as the boiling point reached 142 C .
The reaction product was cooled, placed in an open flat tray and r-;nt~in,od at 75C in an explosion 35 proof forced air oven for a further two hours to remove any residual solvent (rotary évaporation would normally . _ _ _ _ . . ... . _ , . ,, .. , _ . .

~ wo s5/27767 2 t 8 6 8 ~ 8 r~ Y

be a preferred method).
Blends of this product with different proportions of commercial purity EGML ~see Example 1) were made. The hardness of these luoricant 5 forr-~11At;nn~ at varioug temperatures were determined.
Ester A B C D E

Hardness N/mm 11C 3.8 1.5 1.5 1.1 0.95 15 20C 2.5 1.8 1.4 0.85 0.55 35C 0.51 0.75 0.47 0.33 liquid Lubricants A, B, C and D but not E are in 20 accordance with the invention.

3o

Claims

-24-

1. Lubricant which consists essentially of (in wt %) pure ethylene glycol dilaurate 50 - 85 pure ethylene glycol monolaurate 10 - 30 pure stearic acid up to 20 other glycol, ester and carboxylic acid species up to 20 as determined by analysis, which lubricant has a hardness in the range 0.1 - 10 N/mm at all temperatures in the range 15 - 35°C.

2. Lubricant as claimed in claim 1, wherein the lubricant has a hardness in the range 0.1-5 N/mm at all temperatures within the range specified.

3. Lubricant as claimed in claim 1 or claim 2, wherein the lubricant has a hardness within the range specified at all temperatures in the range 15 - 40°C.

4. Lubricant as claimed in any one of claims 1 to 3, wherein the stearic acid content is 5-20% by weight.

5. Lubricated metal, wherein a surface of the metal carries a film of the lubricant of any one of claims 1 to 4.

6. Lubricated metal as claimed in claim 5, wherein the metal is sheet.

7. Lubricated metal as claimed in claim 5 or claim 6, wherein the metal is aluminium.

8. Lubricated metal as claimed in claim 7, wherein the aluminium carries a strongly-bonded - artificial inorganic surface layer on top of which the film of lubricant is present.

9. A method of making a structure of shaped aluminium components starting from lubricated aluminium metal sheet according to any one of claims 6 to 8, comprising the steps:
- forming pieces of the sheet into components, - bringing the components together in the shape of the desired structure, - and securing the components together by mechanical and/or adhesive means.

10. A method as claimed in claim 9, comprising the steps:
- forming pieces of the sheet into components, - applying adhesive to the components, - bringing the components together in the shape of the desired structure, and - curing the adhesive.

11. A method as claimed in claim 9 or claim 10, comprising the additional steps of subjecting the structure to the action of an aqueous alkaline cleaner, and thereafter painting the structure.