Method for treating aluminium surfaces

Description

[0001] This invention relates to the use of hydrogen peroxide in the surface treatment of aluminium, mainly though not exclusively to prepare aluminium for applications where it is to be bonded using a thermosetting polymer adhesive. It has long been known that the use of such etching mixtures causes the surface to develop a filamented porous structure, somewhat similar in thickness and morphology to that produced by A.C. anodising in hot phosphoric acid. This structure is very suited to adhesive bonding; the adhesive penetrates deep into the pores of the oxide structure and hardens within it, producing a firm anchoring of the polymer which is manifested in a high tensile strength of the resulting bond. It is known that the combination of hydrogen peroxide and mineral acid imparts to the metal just such a porous surface and furthermore results in a similar high tensile strength of an adhesive joint when bonded with a single part epoxy adhesive. For example, Bijlmer disclosed in a paper at Interfinish 76: Proc. 9th World Congress on Metal Finishing (Amsterdam 1976) results that related the peel strength of adhesively bonded aluminium specimens to the micro-roughness of the metal surface. Similarly, Japanese Kokai 53/97037 (78/97037) to Asahi Glass Co. Ltd. discloses the use of peroxide pretreatment to give strongly adhering fluoropolymer coatings on aluminium. Japanese Patent 52/86937 and Kokai 53/132035 contain similar disclosures.

[0002] The background to the invention is, as outlined above, the increasing use of aluminium in applications where it is structurally bonded with adhesive; that is to say, applications where the adhesive joint is exposed to significant tensile and shear loads and where in many cases the joint also has to be durable despite exposure to adverse conditions of temperature and humidity. The earliest such application was the use of the metal in aircraft construction, and it was recognised from an early date that some anodised surface finishes were particularly suitable as a preparation for adhesive bonding. However anodized finishes have been regarded by the aircraft industry as not suitable for coil-application to aluminium, and rapid coil-applicable pretreatments with good durability are very desirable, particularly if they do not contain chromium.

[0003] Similarly, oxidizing treatments using peroxide have been carried out slowly; for example, Bijlmer (see above) used treatment times of 10 and 30 minutes. There is a need for treatments of this kind which are sufficiently rapid to be carried out on aluminium coil.

[0004] A clear distinction needs to be drawn between peroxide pretreatments designed to form an oxide layer on aluminium, and peroxide cleaning treatments. Peroxide cleaning mixtures, typified by UK Patent GB 2200136 to Nihon Parkerizing Co. Ltd., are frequently used on aluminium and copper. An oxide-dissolving agent such as fluoride is always included in these formulations, and the object of the treatment is to remove the naturally occurring oxide layer and not form anything in its place. Such peroxide cleaning mixtures give metal surfaces which show poor adhesive bonding.

[0005] In a report on the bondability of peroxide-­cleaned metals, Venables and co-workers confirmed that peroxide pre-treating produced strong initial bonds, but expressed doubts that durable bonds would be produced from aluminium because of the known instability of aluminium oxides in moist environments. (Ditchek, Breen & Venables, Martin Marietta labs report MML-TR80-17C, April 1980).

[0006] The need for an adhesive bond that is durable in adverse environments has led to the development of no-rinse coatings for aluminium surfaces. As described in US patent 3706603, such coatings may comprise a solution of hexavalent and trivalent chromium compounds mixed with sub-micron particulate silica. A commercial no-rinse treatment of this kind is available under the trademark Accomet C. Recommended pretreatments involve an acid or alkaline rinse to remove the naturally occurring oxide layer from the aluminium surface, so as to present a clean bare metal surface for application of the no-rinse coating.

[0007] EPA 34040 teaches applying to an aluminium metal surface a solution containing peroxide and a metal salt in order to form a conversion coating containing the metal (in chemically combined form) on the aluminium surface. A fluoride is preferably included to prevent aluminium oxide formation To achieve uniform conversion coatings, the peroxide concentration is preferably kept below 20 gl⁻¹.

[0008] CH 540350 describes a chemical treatment to form an aluminium oxide coating on aluminium metal. The treating solution preferably contains fluoride. A peroxide solution may be used; or alternatively an alkaline solution containing a heavy metal salt for incorporation in the coating. Treatment may take 20 - 30 min.

[0009] In one aspect, the invention provides a method which comprises applying to an aluminium surface a fluid composition comprising a per-compound under conditions to form an artificially applied oxide layer and applying an inorganic coating on top of the oxide layer.

[0010] In another aspect the invention provides a method which comprises applying to an aluminium surface a composition comprising at least 30 gl⁻¹ of a per-­compound under conditions to form an artificially applied oxide layer on the surface characterised in that the fluid composition contains an effective concentration of a metal ion which accelerates formation of the artificially applied oxide layer.

[0011] The term aluminium is used herein to include the pure metal and its alloys. The per-compound can be an organic peroxide such as a peracid such as peracetic acid, but is preferably hydrogen peroxide. Salts such as KHSO₅ can be used, but are believed to be effectively an expensive source of H₂O₂. Because of the naturally occurring oxide layer on its surface, these per-compounds are not by themselves capable of forming an artificially applied oxide layer, but need to be used in conjunction with an acid or alkali which removes the naturally occurring layer and enables the per-compound to react with aluminium to generate the desired artificially applied oxide layer without significantly re-dissolving it. The use of mineral acids such as sulphuric acid is preferred for this purpose as giving mechanically stronger oxide films and thus stronger adhesive joints, but the use of alkali such as sodium hydroxide is possible where the aluminium substrate does not contain significant concentrations of reactive alloying constituents such as lithium or magnesium.

[0012] Suitable metal ions comprise transition metal ions including particularly Cu. Care may need to be taken to ensure that the per-compound is sufficiently stable in the presence of required concentrations of the chosen metal ion. It is believed that these metal ions form bimetal cells with the aluminium metal substrate which hasten formation of the artificial oxide layer. The metal ion should be used in a concentration sufficient to accelerate the overall process, which includes removal of naturally occurring oxide and formation of an artificially applied oxide layer on the surface of the aluminium. The metal ion concentration should not be so great as to rapidly decompose the hydrogen peroxide or other per-compound. It is also important that the metal ion concentration be not so high that the metal plates out and is deposited on the aluminium surface in a form which accelerates corrosion. In practice, it is not difficult to choose metal ion concentrations which meet these criteria. Fluid compositions containing from 0.05% to 5%, particularly from 0.1% to 1.0%, of a soluble metal salt should provide suitable metal ion concentrations.

[0013] The fluid composition may contain an amine in a concentration effective to accelerate the formation of the artificially applied oxide layer. It is believed that the amine may increase the action of the copper salt, perhaps by forming a cupramine complex. Ammonia is suitable if its volatility can be controlled. Preferred amines are dimethylethanolamine and triethanolamine. Preferred concentrations are 0.5 - 5% particularly 1 - 3% by weight. The use of amines is beneficial when less reactive Al alloys, e.g. those of the 6000 series such as 6009 and 6111 are being treated.

[0014] Peroxide treatment conditions may be as described in the literature referred to above. Hydrogen peroxide concentration should preferably be high, consistent with stability, to minimize treatment time. We have used 8% w/w H₂O₂ successfully. Higher concentrations are not readily available and do not appear to provide any advantage. H₂O₂ concentrations below 3% (w/w) can be made to work under suitable conditions, but tend to give flatter less profiled oxide films. We have found standard 6% w/w H₂O₂ solution to be satisfactory.

[0015] Acid (or alkali) concentration needs to be sufficient to dissolve the naturally occurring oxide film, but not so high as to dissolve the artificially applied oxide layer; depending on conditions, up to 20% by weight sulphuric acid may be suitable. Some H₂O₂ stabilisers contain sulphuric acid (see below), and this needs to be taken into account when considering overall acid concentration. We have performed successful experiments using only the acid present in the H₂O₂ stabilizer, that is to say only 0.6% H₂SO₄ overall. Acid concentration over 10% may increase rate of oxide dissolution at higher temperatures. Preferred overall acid concentration are 1 - 10% by weight.

[0016] Treatment temperature may be ambient or elevated consistent with the stability of the per-­compound. When hydrogen peroxide is used, preferred temperatures are in the range 50 to 90^oC. It is generally thought that hydrogen peroxide should be used at temperatures of 50 - 65^oC to avoid rapid decomposition. However hydrogen peroxide can be used at higher temperatures of 65 - 90^oC in the presence of stabilizer. Temperatures of 75 - 85^oC are preferred. Use of these higher temperatures can permit shorter treatment times, or can give rise to higher bond strengths at equivalent treatment times.

[0017] Treatment times of 4s to 10 min are preferred, with shorter time appropriate to higher temperatures and more concentrated solution, especially when using spray application. There is an optimum treatment time, often in the range 4 - 60s. For continuous coil coating, 4 - 30s may be preferred, whereas batch operation may require 15 - 60s. Excessively long treatment times may cause the resulting profiled oxide layer to start to deteriorate. These treatment times make possible continuous treatment of aluminium coil, sheet or extrusion.

[0018] The fluid composition can be applied to the metal surface by dipping, roll-coating or otherwise spreading. But particularly for coil treatment, a preferred method of application is by spraying. Spraying accelerates the reaction between the composition and the metal. Pre-cleaning of the aluminium metal surface is possible, but is expected to be necessary only if the surface carries a relatively thick natural oxide film, as may be the case where the aluminium coil has been annealed. It is an important advantage of the invention that the step of pre-­cleaning the aluminium metal surface can be avoided, not only at high treatment temperatures of 65 - 90^oC, but also at conventional peroxide treatment temperatures.

[0019] After treatment, the aluminium metal surface needs to be rinsed. Surprisingly, rinsing temperature has some effect on performance. Rinsing temperatures above ambient, e.g. from 50 - 90^oC as for the treatment, are preferred.

[0020] A stabilizer may be added to the mixture to prolong its lifetime; saturated solution of a proprietary stabilizer sold by Interox Limited under the trademark Stabtabs has been found useful. An activator may be used where rapid development of a highly braided and filamented oxide surface structure is important; an example of a useful activator is 1% (w/v) sodium thiosulphate.

[0021] Another effective stabilizer is an alcohol or glycol such as propylene glycol. The proprietary stabilizer B222, sold by Interox Limited, is believed to be a mixture of a glycol with an acid. The proprietary stabilisers B33, B104 and B222 permit the use of hydrogen peroxide solutions at temperatures above 65^oC as high as 90^oC.

[0022] Fluorides should not be present in the mixture; they cause dissolution of the structure as it is formed.

[0023] The result of this treatment is an artificially applied oxide layer which is typically from 100 to 1000 Angstroms thick. By suitable choice of acid (or alkali) concentration and treatment temperature, this applied oxide layer may be arranged to have a profiled surface with fine oxide protrusions or whiskers. Thus, mechanical interlocking by whisker reinforcement of an adhesive appears to play a role in enhancing adhesive bonding. Scanning electron microscope examination of the profiled oxide layer indicates porosity on a scale typically of 50 to 100 nm.

[0024] The artificially applied oxide layer forms an excellent base for subsequently applied organic coatings such as paint, lacquer or adhesive. In order to obtain adhesive bonds having improved durability under damp, corrosive or other adverse conditions, it may be preferred to apply a further coating on top of the oxide layer in order to inhibit or prevent hydration of the oxide layer.

[0025] A preferred method is to apply an inorganic coating on top of the artificially applied oxide layer. This inorganic coating is preferably a no-rinse coating, and preferably one containing inorganic particles, which may have been pre-formed or formed in situ. The inorganic coating is preferably so thin that the profiled surface topography of the artificially applied oxide layer is substantially maintained. It is believed that the artificially applied oxide layer provides improved initial adhesion for such subsequently applied organic films by mechanical interlocking; and that the inorganic coatings applied according to this step insure that the initial excellent adhesion properties are not reduced on prolonged exposure to humid or corrosive environments.

[0026] The no-rinse coating may comprise a hydrous metal oxide sol of the kind described in EPA 358338. Because the artificially applied oxide layer has the desired profile, there is no need to include a passenger powder in the sol in order to change the surface topography, although such a powder can be added if desired. One such powder is fumed silica Aerosil R202 from Degussa

[0027] A dissolved adhesion promoter may also be applied, either in the no rinse composition or separately before or after the sol. When the composition is intended for a no-rinse treatment, the constituent should preferably be substantially non-­toxic. The constituent promotes adhesion, for example by providing suitable links to the underlying oxide layer and to the overlaying organic layer, or by inhibiting corrosion at the organic coating/oxide layer/metal interfaces. It is believed that adhesive bond strength falls on exposure to water or more aggressive agents because of corrosion or hydration at these interfaces. Inhibition of this corrosion helps to retain adhesive bond strength.

[0028] The adhesion promoter may comprise phosphate or phosphonate. Phosphate esters are known to bond well onto aluminium surfaces and to be able to inhibit corrosion. In addition to the phosphoric acids and inorganic phosphates, there are a number of organic phosphorus-containing compounds which may be used, examples being amino-phosphates for example nitrilotris (methylene) phosphonic acid (NTMP) or other nitrilo-­ substituted phosphonic acids or phosphate esters such as bis-(nonyl phenyl ethylene oxide) phosphate.

[0029] The adhesion promoter may comprise one or more organosilanes, for example glycidoxypropyltri­methoxy silane or aminopropyltriethoxy silane. The composition may contain one or more of these or other classes of dissolved adhesion-promoting and/or corrosion-inhibiting constituents, including molybdates, zirco-aluminates, organo-metallic trivalent chromium compounds and hexavalent chromium compounds.

[0030] The use, following peroxide treatment, of chromium-based no-rinse coatings such as Accomet C is possible but not particularly preferred. When the underlying oxide coating is profiled, it is advantageous to use a no-rinse coating based on a very fine particle size material which enters the pores of the oxide layer such that the profiled surface thereof shows through. The oxide layer formed by peroxide treatment also appears advantageous when used with other no-rinse coatings, e.g. non-toxic coatings not containing chromium, which may otherwise not adhere reliably to the underlying aluminium metal.

[0031] Another method involves applying a compound to the oxide layer which after the action of heat and/or moisture decomposes to form an inorganic coating. Preferred examples of this type of coating are titanate esters and chelates which may be subjected to the action of heat and moisture to form titanium dioxide coatings. Commercially available materials of this kind are supplied by Tioxide International Limited under the trademarks Tilcom PI2 and Tilcom AT31 and Tilcom PBT. These are titanate esters and chelates which decompose under the action of heat and moisture to give hard and stable coatings of titanium dioxide. Useful coatings are likely to be in the region of 0.01 to 0.1 microns thick on top of the artificially applied oxide layer.

[0032] The no-rinse composition may be applied to the metal surface (carrying an artificial applied oxide layer with a profiled surface) by any convenient technique, such as spin coating, immersion, flow or roller coating, or by spraying. For aluminium coil, roller coating is likely to be an attractive option. The formulation may need to be adjusted to provide a convenient viscosity for application by the desired method. After application, the coating on the metal surface is dried, but rinsing is not normally necessary. Drying temperatures may typically be up to 200^oC.

[0033] The metal surface with the artificially applied oxide layer preferably carries the coating in the range of from 0.005 to 0.5 gm⁻², preferably between 0.01 and 0.1 gm⁻²

[0034] The invention envisages also carrying out the two steps simultaneously, by applying to the aluminium metal surface a fluid mixture including a per-compound and ingredients for an inorganic (e.g. no-rinse) coating, so that there is formed on the metal surface an oxide layer with an overlying inorganic coating.

[0035] The invention envisages as an additional method step the application to the protective coating of an organic coating such as paint, lacquer, varnish or adhesive. There is increasing interest in the use of adhesively bonded aluminium components as structures for motor vehicles. One example of a commercially available adhesive suitable for this application is Permabond ESP105.

[0036] The following examples illustrate the invention.

[0037] Two no-rinse coating compositions are used in the examples, namely Accomet C, a proprietary chrome-­containing pretreatment sold by Albright & Wilson Ltd., and the chrome-free pretreatment JT10 whose composition is given below.
JT10 3% (w/w) phosphoric acid: 190g
fumed alumina (Degussa): 11g
fumed silica Aerosil 380 (Degussa): 33g
Water: 80g

[0038] The components are mixed in a Silverson or other high shear stirrer, and dispersed to form a stable and uniform dispersion.

Example 1

[0039] Sheets of 0.7mm 5251-HO aluminium were treated for 60s at 60^oC in a mixture containing: Hydrogen peroxide (w/w) 6%
Sulphuric acid (w/w) 15%
referred to below as solution (I), and also in solution (I) with a saturated solution of "Stabtabs" (solution (II)) and in solution (I) with 10g/litre sodium thiosulphate (solution (III)). For purposes of comparison, a 2% solution of Ridoline 124/120E was also used (solution IV). This is a commercial (ICI) sulphuric acid/HF/wetting agent mixture which could be described as a more conventional acid cleaning medium.

[0040] All the treated sheets were coated with approximately 100mg/m2 JT10 using a roll-coater, and the coatings were dried at 150^oC. The thus-pretreated sheets were then cut to form 20mm x 100mm coupons, bent to form L-shaped adherends and bonded with a standard heat-cured single-part structural epoxy adhesive to give T-shaped joints with a 60mm long bondline. These were peeled at 5mm/min on an Instron 1115 tensile tester and the steady-state peel load was recorded during the peel event.

[0041] Results were as follows:

Solution	Peel load in Dry air (N)	Peel load in Water spray (N)
I	77,78,80	65
II	72,74,75	57
III	70,77,80	70
IV	36,39,42	--

[0042] The thickness of the porous structure was examined by SEM and was found to be as follows:

I 370 Angstroms

II 480 Angstroms

III 560 Angstroms

Example 2

[0043] Solution 1 was used at various times and temperatures in conjunction with coatings of Accomet C which were coated onto the treated metal specimens at a constant coatweight of 15mg/m². This time only the dry peel loads were measured.

Results:
Etch condition	Peel load (N)
40oC. 30 seconds	25,27,24
40oC. 120 seconds	45,44,48
60oC. 30 seconds	58,54,55
60oC. 120 seconds	75,69,73

[0044] It is clear that initial strengths of the adhesive joints are very high indeed. But treatment times of more than 30 seconds are required to achieve the highest bond strength.

Example 3

[0045] Coil pretreatment processing rates require metal contact times of not more than about 20 seconds if the line is not to be of uneconomical length. Sheets of 0.7 mm 5251-HO aluminium were treated for 20 seconds at 60^oC in a mixture containing:
hydrogen peroxide (w/w) 6%
sulphuric acid (w/w) 15%
referred to below as solution (I), and also in solution (1) with an addition of 1% (w/w) copper sulphate pentahydrate (solution V). The thus-pretreated sheets were then cut to form 20 mm x 100 mm coupons, bent to form L shaped adherends and bonded with a standard heat-cure single-part structural epoxy adhesive to give T-shaped joints with a 60 mm long bond line. These were peeled at 5 mm/min on an Instron 1115 tensile tester and the steady-state peel load was recorded during the peel event. Results were as follows:

Solution	Slow Peel Strength (N)
I	46, 53, 53 (average 51)
V	50, 57, 66 (average 58)

[0046] The addition of copper ions to the peroxide composition represents a significant improvement in view of the fact that the range of slow peel strengths in this test is from 20 N (completely featureless morphology) to 82 N (completely cohesive failure locus within the adhesive; no interfacial failure.)

Example 4

[0047] The experimental conditions used here were the same as in Example 3, except that different stabilisers were present in the peroxide compositions, and a different adhesive, a standard heat-cure single-­part structural epoxy adhesive was used.

[0048] Results were as follows:

Etching Mixture	Stabiliser	Slow Peel Strength (N)
8% H₂SO₄ + 6% H₂O₂,	3% B33	97
8% H₂SO₄ + 6% H₂O₂, 1% CuSO₄	3% B33	144
8% H₂SO₄ + 6% H₂O₂,	3% B104	100
8% H₂SO₄ + 6% H₂O₂, 1% CuSO₄	3% B104	148
8% H₂SO₄ + 6% H₂O₂,	3% B222	97
8% H₂SO₄ + 6% H₂O₂, 1% CuSO₄	3% B222	135

Example 5

[0049] This example shows the use of two different coatings applied on top of the artificially produced oxide layer in order to improve the storage stability in a warm and humid environment.

[0050] Sheets of 0.7mm 5251-HO aluminium were treated for 60 seconds at 60C in a mixture containing hydrogen peroxide 6% (w/w) and sulphuric acid 15% (w/w), rinsed in deionised water and dried for 3 minutes at 100C, and were then roller-coated at a dry-­coat thickness of 0.01 to 0.1 microns with the following coatings:

1. a 1% (w/w) solution of an alkanolamine titanate chelate AT31 in water, dried at 300C for 1 minute.

2. a mixture containing a 3% (w/w) dispersion of fumed silica Aerosil R202, 20% (w/w) dispersion of glycidoxypropyltrimethoxysilane, 20% (w/w) solution of ethoxyethanol, all in deionised water, dried at 200^oC for 3 minutes.

[0051] AT31 is obtained from Tioxide International Ltd., Cleveland, U.K., while Aerosil R202 is obtained from Degussa Ltd., Wilmslow, Cheshire, U.K.

[0052] The resulting samples were cut into coupons, and these coupons were artificially aged in a humidity cabinet at 25C, 98% R.H. for periods of time before being assembled into peel joints with a structural single-part epoxy adhesive, cured at 190C, and subjected to slow peel testing under the conditions of Example 3. Results were as follows:

	0 weeks	2 weeks		4 weeks
	Peel Strength (N)	% Retention		% Retention
uncoated oxide:	81	73	90	70	86
oxide + 1:	99	79	80	--	--
oxide + 2:	107	98	91	110	103

Example 6

[0053] This example shows the use of inorganic coatings applied on top of the artificially applied oxide layer in order to improve its storage stability. Experimental conditions were as in Example 5. However, after the peroxide treatment had resulted in an artificially applied oxide layer on the aluminium specimens, coatings of two titanate esters or chelates were applied. These were obtained from Tioxide International Limited, Cleveland, U.K., and are sold under the trademarks Tilcom PI2 and Tilcom PBT. The applied coatings were decomposed under the action of heat and moisture to give hard and stable coatings of titanium dioxide, generally in the region of 0.01 to 0.1 microns thick. The resulting samples were subjected to slow strain rate testing, again under the conditions of Example 3, with the following results. Tilcom PBT is a polybutyl titanate. Tilcom PI2 is an ethoxy isopropoxy titanium bisacetylacetonate.

Coating	Initial slow peel strength (N)	% retention after 2 weeks storage
Tilcom PI2	77	79
Tilcom PBT	67	86

Example 7

Treatment Temperature

[0054] The treatment solution contained 8% H₂SO₄; 6% H₂O₂; 1% CuSO₄.5H₂O; 3% stabiliser. Sheets of 0.7 mm 5251 - HO aluminium were treated for 20s at various temperatures. Joints were formed as described in Example 3 and were subjected to slow peel strength testing under wet conditions (which is a more severe test than under dry conditions). The results were:

	Wet Slow Peel Strength (N)
Stabiliser	Treatment Temp.
	40oC	60oC	80oC
B33	42	70	129
B104	56	78	130

[0055] Slow peel strength is an indirect measure of the surface topography of the artificially applied oxide layer. Results are better at higher temperatures.

Example 8

Salt Spray Tests

[0056] The treatment solution contained 4% H₂SO₄; 6% H₂O₂; 3% propylene glycol. Sheets of 1.6 mm 5754 Al alloy were acid cleaned and treated at 80^oC for 20s. Lap shear joints were formed and held at 43^oC in 5% neutral salt spray. Joint strength after exposure was:-

Weeks	0	8	20
Joint Strength (MPa)	29.9	25.4	24.9

Example 9

Amines

[0057] The treatment solution contained 4.6% H₂SO₄; 6% H₂O₂; 2.4% propylene glycol; 0.1% CuSO₄.5H₂O; 2% amine. Sheets of 1mm 6009 alloy were treated, without any pre-cleaning step, for 30s at 80^oC. Joints were formed and tested as described in Example 7. Results were:-

Amine	Wet Peel Strength (average) N
None	83
Dimethylethanolamine	109
Triethanolamine	98
Morpholine	90

Example 10

Rinsing Temperature

[0058] The treatment solution contained 4% H₂SO₄; 6% H₂O₂; 3% B222 stabiliser. Sheets of 1mm 5754 H40 Al alloy were subjected, without pre-cleaning, to treatment for 20s at 80^oC. Treated sheets were rinsed with deionised water at various temperatures and dried (at 100^oC or 180^oC; the drying temperature is not critical). Joints were made and tested as described in earlier examples. Results were:

Rinse Temp (oC)	Dry Peel Strength (average) N
20	77
50	77
80	92

Example 11

[0059] Two treatment solutions were used in this experiment:
A: 4.6% H₂SO₄; 6% H₂O₂; 2.4% propylene glycol.
B: 4.6% H₂SO₄; 6% H₂O₂; 2.4% propylene glycol; 0.1% CuSO₄.5H₂O.

[0060] Sheets of 1mm 5754 Al alloy were treated for 30s at various temperatures and then rinsed with deionised water at the treatment temperature. In some cases, the sheet was acid cleaned for 60s prior to treatment. Joints were made up and tested as described in Example 7. Results were:-

Treatment Solution	Acid Cleaned Y/N	Temperature (oC)	Wet Peel Strength (average) (N)
A	Y	60	42
A	N	60	51
A	N	80	53
A	Y	80	49
B	N	60	111
B	N	80	120
B	Y	80	50

Example 12

Inorganic Coatings

[0061] Sheets of 1.6 mm 5754-HO Al alloy were subjected to two alternative treatments:-

(1) Acid cleaned for 60s (Ridolene 124/120E) followed by treatment with Accomet C under recommended conditions. This provides a standard for comparison.

(2) Treatment at 80^oC for 30s using a treatment solution of composition 4% H₂SO₄, 6% H₂O₂, 0.1% CuSO₄.5H₂O, 3% B222, rinsing at 80^oC, then application of a sol according to EPA 358338 containing 0.3% zirconia and 1.5% glycidoxypropyl trimethoxysilane.

[0062] Then storage of the pretreated metal as in Example 5 for eight weeks gave the following strengths after using the stored metal to prepare 20 mm x 10 mm overlap single lap joints:

	0 weeks	8 weeks
(1)	28.8	26.1 MPa
(2)	29.0	28.1 MPa

[0063] This demonstrates that the coated peroxide pretreatment possesses a good stability, in contrast to the known instability of uncoated peroxide pretreatments as pointed out by Venables in the prior art.

[0064] When the above pretreated metal was used to make lap joints without prior storage, the joints retained their strength in salt-spray exposure (5% NaCl) to the following extent:

	0 weeks	8 weeks	20 weeks
(1)	28.8	23.8	21.2
(2)	29.0	23.8	20.7

[0065] This demonstrates the durability of these in the tests used in Example 8.

Claims

1. A method which comprises applying to an aluminium surface a fluid composition comprising at least 30 gl⁻¹ of a per-compound under conditions to form an artificially applied oxide layer on the surface, characterised in that the fluid composition contains an effective concentration of a metal ion which accelerates formation of the artificially applied oxide layer.

2. A method as claimed in claim 1, which includes the additional step of applying an inorganic coating on top of the oxide layer.

3. A method which comprises applying to an aluminium surface a fluid composition comprising a per-compound under conditions to form an artificially applied oxide layer and applying an inorganic coating on top of the oxide layer.

4. A method which comprises applying to an aluminium surface a fluid composition comprising a per-compound under conditions to form an artificially applied oxide layer on the surface, wherein the fluid composition is used at a temperature of 65 to 90^oC.

5. A method as claimed in claim 4, wherein the fluid composition is used at a temperature of 75 to 85^oC.

6. A method as claimed in any one of claims 1 to 5, wherein the fluid composition is applied to an aluminium surface which has not been subjected to pre-­cleaning.

7. A method as claimed in any one of claims 3 to 6, wherein the fluid composition contains an effective concentration of a metal ion which accelerates formation of the artificially applied oxide layer.

8. A method as claimed in any one of claims 1, 2 or 4 to 7, wherein the metal ion is Cu⁺⁺.

9. A method as claimed in any one of claims 1 to 8, wherein the per-compound is hydrogen peroxide applied at a concentration of at least 30 gl⁻¹ in solution in a mineral acid.

10. A method as claimed in any one of claims 2 to 9, wherein the artificially applied oxide layer has a profiled surface and the inorganic coating is so thin that the profiled surface shows through.

11. A method as claimed in any one of claims 1 to 10, wherein the fluid composition includes an amine in a concentration effective to accelerate formation of the artificially applied oxide layer.

12. A method as claimed in any one of claims 1 to 11, including the additional step of applying an adhesive to the artificially applied oxide layer or the overlying inorganic coating.

13. A method as claimed in Claim 10, wherein the inorganic coating is formed by application of a hydrous metal oxide sol.