AU2012201876A1 - Methods for treating aircraft structures - Google Patents

Abstract

The present invention relates to methods for repairing a structural weakness in an aircraft fuselage, or preventing the advancement of a structural weakness in an aircraft fuselage. Cold spray methods such as supersonic partide deposition have been shown to positively 5 affect structural characteristics of sheet metal and lap joints as used in fuselage construction. 14 100 mm ,1 0000 50mm0 0 125mm 75mm 350 mm 125mm 50mm 0 00 0000o Rivets

Description

M ETHODS FOR TREATING AIRCRAFT STRUCTURES FIELD OF THE INVENTION The present invention relates to the field of aircraft manufacture and repair; and also to 5 preventative maintenance of aircraft. In particular, the invention relates to the prevention and repair of structural weaknesses and environmental degradation in aircraft fuselages and other structures. BACKGROUND TO THE INVENTION 10 The fuselages of many aircraft consist of circumferential frame members, longitudinal stringers, and a thin skin, all made from lightweight aluminium. This construction allows for a balance of flight properties and weight. The sheets of aluminium that make up the skin are connected together as lap joints by generally two to three rows of rivets. The outer skin later is countersunk at each rivet 15 location so the rivet head isflush with the skin, resulting in optimal aerodynamic properties. When the skin is subjected to the stresses of normal operation, particularly in pressurized commercial aircraft, fatigue damage can occur in the metal sheets and especially in high stress locations around fasteners. The problem is exacerbated by the ingress of !0 environmental elements and leads to the joint cracking. -ack growth, if left undetected, can lead to catastrophic failure, as in the case of Aloha Airlines Right 243 in 1988. As the aircraft reached its normal flight altitude of 24,000 feet (7,300 m), a small section on the left side of the roof ruptured. The resulting explosive decompression tore off a large section of the roof, consisting of the entire top half of the aircraft skin extending from just behind the 25 cockpit to the fore-wing area. It was subsequently discovered that the incident was caused by the presence of multiple small cracks which arose as a result of environmental degradation of the joint located aft of the front port side passenger door. This phenomena has subsequently been termed "multi-site damage" (msd) Snoe the Aloha incident, aircraft operators have been directed to regularly check for the 30 presence of cracks and msd in the fuselage skin. In order to identify the presence of cracks before they reach critical lengths, various inspection techniques are utilized. 1 While visual inspection is an important part of the detection process, many naturally occurring cracks in their initiation are simply too small to see or otherwise detect. To assist with the detection of these small and hidden cracks, non-destructive inspection (NDI) methods are used. NDI methods can also be used to detect cracks that exist under paint and 5 detect areas of corrosion between the layers of skin. Some of the more common NDI methods used in aircraft fuselage crack detection are ultrasound and eddy current methods. These methods are not capable of detecting all cracks and are particularly poor in detecting small naturally occurring defects. After the Aloha Airlines Fight 243 accident, all 737's with over 50,000 cycles we required to 10 have their lap joints reinforced with external sheet metal patches. This modification is costly, and takes about 15,000 man hours. Furthermore in April 2011 a fuselage lap joint in a %uthwest Airlines Boeing 737-300 aircraft tore an 18 inch hole in the roof, and led to the grounding of 79 of its older Boeing 15 737 aircraft for inspections [38, 39]. This resulted in the cancelation of almost 700 flights [38, 39]. These inspections, which found cracks in a total of four %uthwest aircraft, [38] led to the USFAA mandating the inspection of 175 737 aircraft that had experienced more than 35,000 cydes. There are more than 931 similar aircraft worldwide. The problem is not confined to 737 and 727 aircraft. On 26th October 2010 an American Airlines 757-200 0 aircraft was forced to land at Miami International Airport due to a sudden decompression arising from cracking in a fuselage joint. This aircraft had experienced less than 23,000 cydes. This led to the discovery of cracking in other 757 aircraft and a January 2011 FAA Airworthiness directive [40] mandating the inspection of all 757-200 and 757-300 aircraft. 25 Environmental degradation and subsequent crack initiation and progression is not just of importance to commercial airlines. Military aircraft, particularly those with advanced age, can also develop environmental degradation and cracking at fastener holes. As the military attempts to keep its fleet flight-worthy for longer periods of time, additional prevention, inspection, and mitigation methods are being developed to prevent both environmental 30 degradation and catastrophic failure. 2 When cracks are discovered, they are typically repaired by the application of external sheet metal patches. Again, this is a costly and time consuming process. Afurther problem is that the application of patches may actually initiate a weakness in the underlying structure. %-ch undetectected and undetectable cracks can compromise the safety of the fuselage/wing 5 skin. These repairs can also locally overstiffen the structure and result in catastrophic failure in the fuselage/wing skin as a result of a crack running from repair to repair. Externally bonded composite bonded repairs have been developed to address this problem. However, these repairs do not prevent the ingress of moisture and hence do not alleviate 10 environmental degradation of the structure. Furthermore, to ensure a durable bond the structure needs to be heated to approximately 120 C Additionally, composite bonded repairs cannot be used in regionswhere there is a tight radius of curvature. It is an aspect of the present invention to overcome or alleviate a problem of the prior art by 15 providing a method for preventing or repairing a structural weakness in an aircraft structure. Afurther aspect of the present invention is to overcome or alleviate a problem of the prior art by providing a method for preventing environmental degradation in an aircraft structure. 0 The discussion of documents, acts, materials, devices, articles and the like is induded in this specification solely for the purpose of providing a context for the present invention. It is not suggested or represented that any or all of these matters formed part of the prior art base or were common general knowledge in the field relevant to the present invention as it existed before the priority date of each provisional daim of this application. 25 SUM M ARYOFTHE INVENTION In a first aspect the present invention provides a method for (i) repairing a structural weakness, and/or (ii) preventing or inhibiting the initiation of a structural weakness, and/or (iii) preventing or inhibiting the progression of a structural weakness in an aircraft structure 30 and/or (iv) preventing the ingress of an environmental element, the method comprising the step of bonding a plurality of parties to the structure, the bonding being effected under 3 conditions allowing the plurality of metallic particles to form a substantially continuous layer. In one embodiment the method is for (i) repairing a structural weakness, and/or (ii) preventing or inhibiting the initiation of a structural weakness, and/or (iii) preventing or 5 inhibiting the progression of a structural weakness in an aircraft structure and (iv) preventing the ingress of an environmental element. In a second aspect the present invention provides an aircraft structure comprising a substantially continuous metallic layer, the layer being deposited on a surface of the structure, the layer being capable of (i) repairing a structural weakness, or (ii) preventing or 10 inhibiting the initiation of a structural weakness, or (iii) preventing or inhibiting the progression of a structural weakness in the aircraft structure, wherein the layer comprises a plurality of metallic particles. In one embodiment, the layer is capable of (i) repairing a structural weakness, and/or (ii) preventing or inhibiting the initiation of a structural weakness, and/or (iii) preventing or 15 inhibiting the progression of a structural weakness in an aircraft structure and (iv) preventing the ingress of an environmental element. In one embodiment of the structure the substantially continuous layer is deposited on the surface of the aircraft structure by a method as described herein. In one embodiment of the method or structure, at least a proportion, or substantially all, of 20 the particles are metallic particles. In one embodiment of the method or structure, the bonding does not involve melting or fusing of the part ides. In one embodiment of the method or structure, the bonding is achieved by a cold spray process, such as supersonic partide deposition. 25 In one embodiment of the method or structure, the substantially continuous layer is at least about 0.05 mm. 4 In one embodiment of the method or structure, the substantially continuous layer has substantially even depth across the application surface. In one embodiment of the method or structure, the aircraft structure is a fuselage component, such as a sheet metal, and may be a lap joint. Where the structure is a lap joint 5 the substantially continuous layer does not extend to cover a junction between the free end of a sheet metal component of the lap joint. In one embodiment, the structural weakness is a crack. BRIEF DESCRIPTION OF THE DRAW INGS 10 RG 1 is a diagram showing the geometry of an edge notched panel, being a 2024T3 test specimen. Material is: Aluminium Alloy 2024T3 Alaad 350 mm x 100 mm x 1.27 mm (0.050"). RG2 is a diagram showing the geometry of the panel of FIG 1 with an SPD doubler. RG3 is a photograph in plan view of a test panel being 2024T3 with PD doubler. 15 RG4 is an IRthermal image showing the stresses is an STD doubler at 11,100 cydes (units in MPa). RG5 is an IRt hermal image showing t he stresses is an SPD doubler at 56,100 cydes (units in MPa). RG6 is a graph showing crack growth histories in S-NT tests. 20 RG 7 is a diagram showing the location of an STD strip repair on a 2024T3 test spedmen. R G8 is a photograph showing cross-section of an SPD strip. RG 9 is an IR thermal image showing the stress field in the skin and the SPD strip (units in MPa). RG 10 is an IRthermal image showing stresses in an STD at 3000 cydes (units in MPa). 25 RG 11 is an IRthermal image showing stresses in an STD at 33,000 cydes (units in MPa). RG 12 is an IRthermal image showing stresses in an SPD at 35,500 cydes (units in MPa). RG 13 is an IRthermal image showing dissipated energy at 33,500 cydes (units in M Pa). RG 14 is a graph showing measured crack length histories with and without an STD patch. RG 15 is a graph showing measured and predicted crack length histories for the -NT 30 sped men with an STD patch. 5 HG 16 is a diagram showing repair configuration: (a) plan view, (b) cross-section along centre line, i.e. x = 0. HG 17 shows two images of (a) Ou (bright) on an Al substrate, from [22], and (b) AJZn onto an Al substrate, from [23]. 5 HG 18 is a schematic diagram demonstrating the stresses used to determine K HG 19 is a diagram of a typical finite element mesh of the cracked structure and the associated 'PM repair. The crack in the base structure (plate) is shown in a different colour to the D and the remainder of the cracked plate. HG 20 is a photograph showing two TD strips on either side of a 20 mm long central crack 10 in a rib stiffened panel. HG 21 is a photograph of a delaminated surface of an TD strip (A), which was 20 mm wide, showing locations where the fractal dimensions were measured. HG 22 is a photograph of a delaminated surface of an TD strip (B), which was 20 mm wide, showing locations where the fractal dimensions were measured. 15 HG 23A is a cross-sectional view of a lap joint having an TD doubler. HG 23B is an enlarged cross-sectional view of the lap joint of HG 24A showing a single rivet. FG23Cis a plan view of the lap joints shown in FGS23A and 23B. HG 24 is a schematic diagram of a fuselage lap joint specimen, without an TD doubler. HG 25 is a photograph of a lap joint detailing the TD application specification, showing the 0 region of application. It will be noted that the TD doubler is applied only up to the edge of the upper fuselage skin, and does not extend onto the lower fuselage skin. Fig 26 is an IRthermal image showing stresses, in MPa, at the critical rows of fasteners in a baseline sped men. Fig 27 isan IRthermal image showing stresses, in MPa, prior to link up. 25 HG 28 is an IR thermal image showing stresses, in MPa, in the joint after approximately 6,500 cydes (Baseline No TD). Fig 29 is an IRt hermal image showing stresses, in MPa, at approximately 29,000 cydes (Baseline No D). HG30 is an IRthermal image showing stresses in the D over the fasteners at 92,000 cycles 30 (Test Panel 1). 6 FIG31 is an IRthermal image showing stresses, in MPa, in the lap joint specimen at 18,000 cydes (Test Panel 2). FIG32 is an IRthermal image showing stresses, in MPa, in the lap joint specimen at 48,000 cydes (Test Panel 2) 5 DETAILED DESCRIPTION OF THE INVENTION Inference throughout this specification to "one embodiment" or "an embodiment" means that a particular feature, structure or characteristic described in connection with the embodiment is induded in at least one embodiment of the present invention. Thus, 10 appearances of the phrases "in one embodiment" or "in an embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment, but may. Furthermore, the particular features, structures or characteristics may be combined in any suitable manner, as would be apparent to one of ordinary skill in the art from this disclosure, in one or more embodiments. 15 Smilarly it should be appreciated that the description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. This !0 method of disclosure, however, is not to be interpreted as reflecting an intention that the aimed invention requires more features than are expressly recited in each daim. Rather, as the following aims reflect, inventive aspects lie in less than all features of a single foregoing disposed embodiment. Thus, the claims following the Detailed Description are hereby expressly incorporated into this Detailed Description, with each daim standing on its 25 own as a separate embodiment of this invention. Furthermore, while some embodiments described herein indude some but not other features included in other embodiments, combinations of features of different embodiments are meant to be within the scope of the invention, and from different 30 embodiments, aswould be understood by those in the art. 7 For example, in the following claims, any of the claimed embodiments can be used in any combination. In the description provided herein, numerous specific details are set forth. However, it is 5 understood that embodiments of the invention may be practiced without these specific details. In other instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description. In the claims below and the description herein, any one of the terms "comprising", 10 "comprised of" or "which comprises" is an open term that means including at least the elements/features that follow, but not exuding others. Thus, the term comprising, when used in the aims, should not be interpreted as being limitative to the means or elements or steps listed thereafter. For example, the scope of the expression a method comprising step A and step B should not be limited to methods consisting only of methods A and B. Any 15 one of the terms "including" or "which includes" or "that includes" as used herein is also an open term that also means induding at least the elements/features that follow the term, but not excluding others. Thus, "including" is synonymous with and means "comprising". In a first aspect the present invention provides a method for (i) repairing a structural 20 weakness, and/or (ii) preventing or inhibiting the initiation of a structural weakness, and/or (iii) preventing or inhibiting the progression of a structural weakness in an aircraft structure, and/or (iv) preventing or inhibiting the ingress of an environmental element, the method comprising the step of bonding a plurality of parties to the structure, the bonding being effected under conditions allowing the plurality of particles to form a substantially 25 continuous layer. In some embodiments of the structure, the method is for (in addition to any one or all of the applications (i), (ii) and (iii) recited supra), (iv) preventing or inhibiting the ingress of an environmental element. Embodiments having all of the applications (i) to (iv) are 30 particularly advantageous given the significant advantages provided in terms of extended life of the aircraft structures. Accordingly, a preferred form of the method provides that the 8 method is for (i) repairing a structural weakness, and/or (ii) preventing or inhibiting the initiation of a structural weakness, and/or (iii) preventing or inhibiting the progression of a structural weakness in an aircraft structure and (iv) preventing the ingress of an environmental element. 5 Applicant has discovered that the deposition of a substantially continuous layer of particles over an area of structural weakness (or potential structural weakness) has a positive effect on the present or future structural integrity of the aircraft structure. It has also been shown that the layer is capable of sealing a substrate against the ingress of environmental 10 elements, which in turn may lead to the development of a structural weakness. The deposition of a substantially continuous layer has been shown to seal joints, including mechanical repairs against the ingress of environmental elements such as water, salts, air, acidified rain and the like. Thus, in some embodiments application of the substantially continuous layer hasa preventative as well as restorative result. 15 As used herein, the term "repair" is not intended to be construed narrowly to mean that the structure must be returned precisely to its original state. It is contemplated that in some embodiments, the structure may be returned to a proportion of its original structural strength, or indeed or a multiple of its original strength. !0 The term "prevent" is not intended to be limited to circumstances where the initiation of a structural weakness is completely prevented. The initiation of the weakness may be delayed in time, or it may manifest as a less severe weakness at initiation. 25 The term "inhibit" is not intended to mean that progression of the structural weakness is completely inhibited. It may just mean that the progression is delayed, or that it progresses to an otherwise less severe weakness. The term "structural weakness" is intended to mean any weakness in the structure (or in 30 any part of the structure where the structure is multi-partite) that alters the ability of the structure to remain integral upon the application of a force. In the context of the present 9 invention a structural weakness indudes a crack, a split, a bend, a deformation, a tear, or damage occasioned by exposure to an environmental element on an aircraft structure in the course of service. It does not include any weakness deliberately inflicted on a structure, nor is the term intended to include any alteration in the geometry of a structure such as may be 5 occasioned on a gearbox component, or an engine component, for example. While the step of bonding may involve the melting and/or fusing of the particles (such as that involved in high velocity or low velocity oxygen fuel thermal spray coatings), preferred embodiments of the method do not involve melting or fusing of the particles. In one 10 embodiment of the method, the step of bonding the plurality of particles is accomplished by bonding directly to the substrate (and also to each other) by the release of kinetic energy from the parties. The parties may acquire the kinetic energy by any means, but the energy is typically acquired by accelerating the particles to a high velocity toward the substrate. Upon impact with the substrate, the parties deform (typically flattening) and 15 form a consolidated structure up to several centimetres thick. Impact of the parties with a solid surface at sufficient velocity causes plastic deformation and bonding with the underlying material without the creation of heat affected zones which are typical of other deposition processes and which are undesirable in many structural 0 applications. Bonding is a result of high strain rate deformation and adiabatic shear instabilities and the bond interface. Specific advantages of this technology indude but are not limited to the following a. TD produces an excellent bond with the substrate 25 b. TD can be used to create almost any required thickness. c. TD produces coatingswith very few defects Reusable for reclamation of eroded surfaces and application of wear resistant coating. (,TD enables the continuing reuse of the base material). d. Can be applied to recover damaged geometry without adversely affecting the 30 substrate (no distortion, heat affected zones or embrittlement). 10 e. Initial trials have shown that it can be used to enhance structural integrity through localized strengthening which may reduce the initiation of cracks or propagation of cracks. f. STD can be used to seal ajoint and a riveted repair against the ingress of 5 environmental elements thereby alleviating the environmental degradation of the structure. g. Sgnificant reduction in occupational health and safety risks associated with a number of current in-servioe applied coatings (e.g. cadmium and hexavalent chromium-containing compounds) 10 Sich embodiments of the method are operative at temperatures below the melting point of the particle used in the method. In some embodiments, the method is operative at a temperature of less than 90%, 80/, 70%, 6 0 / 5 0 0 / 40% 30% 20% 10% 9/q 8% 70% 6% 50% 40% 30% 2%or 1%of the melting point of the partide. 15 One particularly useful method for bonding parties (and particularly metallic particles) to the aircraft structure is a cold spray method. Cold spray methods are known in the art, and are characterized by the application of particles to a substrate at temperatures below the melting point of the particles. As used herein, the term "cold spray" is intended to include any coating process utilizing a high-speed gas jet to accelerate a plurality of particles toward 20 a substrate whereby the parties consolidate on impact. In many cases, the process is conducted at a temperature that is substantially lower than the particle melting point. The term 'cold spray' has been used to describe this process due to the relatively low temperatures of the expanded gas stream that exits the spray nozzle. After exiting the 25 nozzle, the parties are impacted onto a substrate, where the solid parties deform and create a bond with the substrate. Asthe process continues, parties continue to impact and form bonds with the previously consolidated material resulting in a uniform deposit with very little porosity and high bond strength. 30 Snoe bonding of the powder to the substrate, as well as the cohesion of the deposited material, is accomplished in the solid state at low temperatures, the characteristics of the cold sprayed material is advantageous in the context of the present invention. Because 11 partide oxidation as well as deleterious tensile stresses that occur during thermal contraction are minimized, the cold spray process has the ability to produce materials with comparatively superior bond strength to the substrate and greater cohesive strength. 5 The Examples herein demonstrate that fatigue performances of cracked metallic structures having a cold spray applied metallic layer under constant amplitude loading are significantly improved over untreated structures. The experimental data demonstrates that the baseline specimens accrued damage more rapidly and that crack growth was significantly greater than the corresponding treated panels. In the majority of tests cases the SD-treated panels 10 showed little evidence of damage/crack growth. A prediction of the fatigue performance of a treated single edge notch coupon was made using SF values calculated via an approximate analysis and the resultant crack length history is in agreement with experimental data. Weight function solutions for repairs to centre cracked panels have also been developed and validated via three dimensional finite element analysis. 15 In some embodiments of the method the substantially continuous layer is formed by exposing the structure to a high velocity (typically between 300 and 1200 m/s) stream of solid-phase parties, which have been accelerated by a supersonic gas flow, typically nitrogen or helium, at a temperature that can range between about 400 and 900 0 C !0 Cold spray processes are known by alternative names induding supersonic partide Deposition (SD); dynamic spray high velocity powder deposition, kinetic spraying, and Knetic Energy Metallization. 25 The present invention is a significant departure from the accepted uses of cold spray technology. Previously, the method has been used as a coating, much like paint, or to restore geometries in worn parts. The use of the process to result in an aircraft structures having improved structural characteristics is an advance in the art, providing economic and safety advantages. 30 12 In one embodiment, the cold spray process is a low pressure cold spray process comprising injecting the plurality of parties in the diverging section of the spray nozzle from a low pressure gas supply. 5 In low-pressure cold spray, air or nitrogen at relatively low pressure-80-140 psi-is also preheated, up to 550 0 Q then forced through a DeLaval nozzle. At the diverging side of the nozzle, the heated gas is accelerated to about 600 m/s. Fbwder feedstock is introduced downstream in the diverging section and accelerated toward the substrate. As the applicability of cold spray technology expands to new and unique areas of application, there 10 has been an increasing number of commercially available, ready-to-use cold spray systems introduced into the marketplace. In high-pressure cold spray, helium or nitrogen at high pressure, up to 1,000 psi, is preheated--up to 1,000 0 C-and then forced through a converging-diverging DeLaval nozzle. 15 At the nozzle, the expansion of the gas produces the conversion of enthalpy into kinetic energy, which accelerates the gas flow to supersonic regime-1,000 m/s-while reducing its temperature. The powder feedstock is introduced axially into the gas stream, prior to the nozzle throat. The accelerated solid particles impact the substrate with enough kinetic energy to induce mechanical and/or metallurgical bonding. 20 The skilled artisan appreciates that both high pressure and low pressure cold spray processes may be operable in the context of the present invention. Of greater relevance than the pressure per se is the velocity at which the parties are 25 propelled toward the substrate. Pressure is one parameter that will influence velocity, however other factors such as partide size and partide weight will have an effect. Typically, the process is operated such that the particles are propelled at or exceeding a minimum velocity that is sufficient to provide adhesion of particles, and/or provide an acceptable porosity in the resultant coating, and/or provide an acceptable deposition efficiency. 30 13 However, the velocity should not be so high as to damage the substrate, or result in the deflection of significant amounts of particle off the substrate or the building particulate layer. 5 Referred velocities for a given application vary according to the powder type. For some powders a low pressure unit will generate a sufficient velocity to achieve the required adhesion, porosity or deposition efficiency. For aluminium powder (which is a preferred species of partide in the present methods) a low pressure system may achieve the desired outcome if operated at its upper limits. However, it is more typical for a high pressure unit 10 to be used in the present methods. The skilled artisan is enabled to adapt a cold spray method to be operable in the context of the present methods. For example, where a particular hardness is required in the consolidated metal layer relatively simple particle impact models as applied to empirical 15 models for flow stress and hardness may result in useful predications of the hardness resulting from a cold spray metallic powder deposition. In particular, reference is made to the modelling of Giampagne et al (Modelling Smul. Mater. Sai. Eng. 18 (2010) 065011 (8pp)). These authors show that strain hardening of cold sprayed deposits is a result of the flattening of the particles as they impact and bond with the surface. Partide impact velocity 0 is the principle controlled parameter of the cold spray process, where partide velocity and material properties determine partide flattening. A constitutive model often used for high strain rate deformation is that of bhnson and Chok [5]. This model indudes strain hardening, strain rate hardening and thermal softening 25 effects during deformation. A number of variables may be routinely manipulated to achieve a desired outcome for a particular application. While not all variables must necessarily be considered to achieve a desired outcome, some of which may be considered include the following non-limiting 30 parameters: a. Sbst rate material 14 i. Type ii. Condition iii. airface Finish 5 b. FRwder i. Material Type ii. Material Cbndition iii. Sze iv. Shape 10 c. Application Nozzle i. Material Type ii. Nozzle Shape 15 d. Carrier Gas i. Type e. Deposition Parameters i. Gas input pressure ii. Gas expansion temperature 20 iii. Deposition flow rates iv. Deposition transition rates The particles may compromise a single species of partide, or multiple species. The plurality of particles may be metallic parties, polymer parties or composite parties. For aircraft 25 related application the particles are typically metallic particles fabricated from any elementary or alloyed metal, including (but not limited to) aluminium, zinc, tin, copper, nickel, titanium, tantalum, cobalt, iron, niobium, molybdenum, and tungsten. preferably the metallic particles are aluminium particles. The particles are typically utilized in the form of a commercially available powder, generally ranging in size from about 5 to about 100plm. 30 The choice of partide is within the ability of the skilled artisan. Generally, the particle is composed of the same material as the substrate. 15 In one embodiment of the method, the substantially continuous layer is substantially dense and/or substantially impervious to a liquid, induding water, any polar solvent or any nonpolar solvent. An advantage of such layers is that weather is exuded from any 5 underlying surface that would normally be prone to corrosion, this enhancing the operation life of the aircraft. The method may be conducted such that a substantially continuous layer of particles of any depth is created. The skilled artisan will be capable of assessing a minimum required depth 10 for any given structural result required. For example, where the structural weakness is minor or the aircraft structure is not a critical component, a lesser depth may be implemented. Conversely, a greater depth may be indicated where the structure has significant damage, or where the structure has minor damage but is expected to be exposed to high levels of stress during operation. 15 In some embodiments, the method is adapted to deposit a substantially continuous layer having a depth of at least about 0.1 mm, 0.2 mm, 0.3 mm, 0.4 mm, 0.5 mm, 0.6 mm, 0.7 mm, 0.8 mm, 0.9 mm, 1.0 mm, 1.1 mm, 1.2 mm, 1.3 mm, 1.4 mm, 1.5 mm, 1.6 mm, 1.7 mm, 1.8 mm, 1.9 mm. 2.0 mm, 2.1 mm, 2.2 mm, 2.3 mm, 2.4 mm, 2.5 mm, 2.6 mm, 2.7 mm, 2.8 !0 mm, 2.9 mm, 3.0 mm, 3.1 mm, 3.2 mm, 3.3 mm, 3.4 mm, 3.5 mm, 3.6 mm, 3.7 mm, 3.8 mm, 3.9 mm, 4.0 mm, 4.1 mm, 4.2 mm, 4.3 mm, 4.4 mm, 4.5 mm, 4.6 mm, 4.7 mm, 4.8 mm, 4.9 mm, 5.0 mm, 5.5 mm, 6.0 mm, 6.5 mm, 7.0 mm, 7.5 mm, 8.0 mm, 8.5 mm, 9.0 mm, 9.5 mm, 10 mm, 11 mm, 12 mm, 13 mm, 14 mm, 15 mm, 16 mm, 17 mm, 18 mm, 19 mm, 20 mm, 21 mm, 22 mm, 23 mm, 24 mm, 25 mm, 26 mm, 27 mm, 28 mm, 29 mm, 30 mm. Preferably, 25 the substantially continuous layer has a depth of at least about 0.05 mm. At this depth, a layer deposited by 'FD may be capable of achieving one of (i) repairing a structural weakness, (ii) preventing or inhibiting the initiation of a structural weakness, (iii) preventing or inhibiting the progression of a structural weakness in an aircraft structure and (iv) preventing the ingress of an environmental element the aims. For typical applications in 30 aircraft, the substantially continuous layer has a depth of from about 0.2 mm to about 4 mm. Depths of at least about 0.2 mm have greater utility in structural aspects of the invention. 16 In some embodiments, the method is adapted to deposit a substantially continuous layer having a depth of at most about 0.1 mm, 0.2 mm, 0.3 mm, 0.4 mm, 0.5 mm, 0.6 mm, 0.7 mm, 0.8 mm, 0.9 mm, 1.0 mm, 1.1 mm, 1.2 mm, 1.3 mm, 1.4 mm, 1.5 mm, 1.6 mm, 1.7 mm, 5 1.8 mm, 1.9 mm. 2.0 mm, 2.1 mm, 2.2 mm, 2.3 mm, 2.4 mm, 2.5 mm, 2.6 mm, 2.7 mm, 2.8 mm, 2.9 mm, 3.0 mm, 3.1 mm, 3.2 mm, 3.3 mm, 3.4 mm, 3.5 mm, 3.6 mm, 3.7 mm, 3.8 mm, 3.9 mm, 4.0 mm, 4.1 mm, 4.2 mm, 4.3 mm, 4.4 mm, 4.5 mm, 4.6 mm, 4.7 mm, 4.8 mm, 4.9 mm, 5.0 mm, 5.5 mm, 6.0 mm, 6.5 mm, 7.0 mm, 7.5 mm, 8.0 mm, 8.5 mm, 9.0 mm, 9.5 mm, 10 mm, 11 mm, 12 mm, 13 mm, 14 mm, 15 mm, 16 mm, 17 mm, 18 mm, 19 mm, 20 mm, 21 10 mm, 22 mm, 23 mm, 24 mm, 25 mm, 26 mm, 27 mm, 28 mm, 29 mm, or 30 mm. It will be understood that depth of the layer may be adjusted by building up layers of particles in the course of the method. In some embodiments, the substantially continuous layer has substantially even depth across the application surface. 15 In one embodiment of the method, the aircraft structure comprises a single part. The component may be any part of an aircraft for which structural integrity is important, such as a fuselage component. In a preferred embodiment of the method the component isa sheet metal forming the "skin" of the fuselage; an engine cowling, or a flight control such as a 20 wing (including a flap, aileron, spoiler or winglet), tail (including a rudder, elevator or stabilizer), Sieet metal components are thin, relatively flexible and are particularly prone to fatigue and cracking. The cold spray application of aluminium particles to the sheet metal can increase the life of the part, and also repair any structural weakness before it leads to a catastrophic structural failure. 25 In one embodiment, the aircraft structure has two or more components. The present methods have been found to be particularly advantageous for the treatment of lap joints, and particularly riveted lap joints. Typically, such joints comprise an upper skin which overlaps an inner skin, with the overlapping area being secured by rows of fasteners, such as 30 rivets. Often, three rows of rivets are provided across the overlapping area. 17 In one embodiment the aircraft structure is a joint. The joint may comprises one or more fastener holes, such as those found in a lap joint. In that case, the method comprises the step of bonding a plurality of particles on, over or about the one or more fastener holes, the bonding being effected under conditions allowing the plurality of parties to form a 5 substantially continuous layer In some embodiments, the fastener holes indude a fastener. In that case the method comprises the step of bonding a plurality of parties on, over or about the one or more fasteners, the bonding being effected under conditions allowing the plurality of parties to 10 form a substantially continuous layer. Applicant proposes that for structural joints (such as lap joints) and also riveted repairs, environmental elements are capable of entering the skin splice or joint during service. This leads to degradative effects between the two mating surfaces and also around fasteners. In 15 the case of fuselage lap joints the load in the upper skin is transmitted to the lower skin through the rivets. These rivets have countersunk heads resulting in what is termed "knife edges" in the upper skin. The aggressive environment coupled with the high stresses at these knife edges results in crack initiation. It is this crack initiation (which is exacerbated by environmental degradation) that results in crack growth at the fuselage joint. 20 Applicant further proposes that by both sealing the fasteners and reducing stresses in the upper skin, the problem of corrosion initiated cracking at fuselage joints is ameliorated or overcome. In addition the life of the joints is also increased. This may be achieved by the supersonic particle deposition (3PD) of a substantially continuous layer over the riveted 25 region lap joint, riveted repairs, and other structural joints in thin skinned aircraft structure. Turning now to Fig 23A there is shown diagrammatically and in cross-section a lap joint comprised of an upper sheet metal component 1, a lower sheet metal component 2, a rivet hole 4 passing through components I and 2, with a rivet 3. An STD layer 5 has been applied 30 such that the metal parties are bonded to the sheet metal component 1, the rivet 3, and an upper region of the rivet hole 4. It will be noted that edges of the layer 5 are bevelled, 18 and that the layer 5 does not extend to cover the junction between the free end of sheet metal component 1 and the underlying second sheet metal component 2. Fig 23B shows an enlarged view of the lap joint of Fig 23, better showing the countersinking 5 of the rivet 3 within thejoint. A plan view of the lap joint is shown in Fig 23C showing the three rows of rivets. The overlying 'FD layer covers the area defined by 5. The edge of the application area 5 coincides with the edge of the upper sheet metal component i such that the 'FD layer does 10 not extend onto underlying sheet metal component 2, nor does the deposited material enter the interface between sheet metal components i and 2. In one embodiment of the invention the structural weakness is a crack. preferably the crack 15 is initiated at the periphery of an aperture in the aircraft structure. preferably the aperture is a fastener hole. Preferably the fastener hold is a faster hole adapted to receive a faster having a countersunk head. In one embodiment the crack is one formed by a force exerted by the fastener against an edge of a fastener hole. The initiation of the crack may be due to normal fatigue, and may aggravated by the presence of damage as a result of environmental !0 ingress to the location In another aspect the present invention provides an aircraft structure comprising a substantially continuous layer, the layer being deposited on a surface of the structure, the layer being capable of (i) repairing a structural weakness, and/or (ii) preventing or inhibiting 25 the initiation of a structural weakness, and/or (iii) preventing or inhibiting the progression of a structural weakness in the aircraft structure, and/or (iv) preventing the ingress of an environmental element, wherein the layer comprises a plurality of particles. In some embodiments of the structure, the layer is capable (in addition to any one or all of 30 the capabilities (i), (ii) and (iii) recited supra) of (iv) preventing or inhibiting the ingress of an environmental element. Embodiments having all of the capabilities (i) to (iv) are particularly advantageous with regards to the operable life of an aircraft structure. Accordingly, a 19 preferred form of the method provides that the layer is capable of (i) repairing a structural weakness, and/or (ii) preventing or inhibiting the initiation of a structural weakness, and/or (iii) preventing or inhibiting the progression of a structural weakness in an aircraft structure and (iv) preventing the ingress of an environmental element. 5 In some embodiments, the substantially continuous layer has a depth of at least about 0.1 mm, 0.2 mm, 0.3 mm, 0.4 mm, 0.5 mm, 0.6 mm, 0.7 mm, 0.8 mm, 0.9 mm, 1.0 mm, 1.1 mm, 1.2 mm, 1.3 mm, 1.4 mm, 1.5 mm, 1.6 mm, 1.7 mm, 1.8 mm, 1.9 mm. 2.0 mm, 2.1 mm, 2.2 mm, 2.3 mm, 2.4 mm, 2.5 mm, 2.6 mm, 2.7 mm, 2.8 mm, 2.9 mm, 3.0 mm, 3.1 mm, 3.2 mm, 10 3.3 mm, 3.4 mm, 3.5 mm, 3.6 mm, 3.7 mm, 3.8 mm, 3.9 mm, 4.0 mm, 4.1 mm, 4.2 mm, 4.3 mm, 4.4 mm, 4.5 mm, 4.6 mm, 4.7 mm, 4.8 mm, 4.9 mm, 5.0 mm, 5.5 mm, 6.0 mm, 6.5 mm, 7.0 mm, 7.5 mm, 8.0 mm, 8.5 mm, 9.0 mm, 9.5 mm, 10 mm, 11 mm, 12 mm, 13 mm, 14 mm, 15 mm, 16 mm, 17 mm, 18 mm, 19 mm, 20 mm, 21 mm, 22 mm, 23 mm, 24 mm, 25 mm, 26 mm, 27 mm, 28 mm, 29 mm, or 30 mm. 15 In some embodiments, the substantially continuous layer has a depth of at most about 0.1 mm, 0.2 mm, 0.3 mm, 0.4 mm, 0.5 mm, 0.6 mm, 0.7 mm, 0.8 mm, 0.9 mm, 1.0 mm, 1.1 mm, 1.2 mm, 1.3 mm, 1.4 mm, 1.5 mm, 1.6 mm, 1.7 mm, 1.8 mm, 1.9 mm. 2.0 mm, 2.1 mm, 2.2 mm, 2.3 mm, 2.4 mm, 2.5 mm, 2.6 mm, 2.7 mm, 2.8 mm, 2.9 mm, 3.0 mm, 3.1 mm, 3.2 mm, 0 3.3 mm, 3.4 mm, 3.5 mm, 3.6 mm, 3.7 mm, 3.8 mm, 3.9 mm, 4.0 mm, 4.1 mm, 4.2 mm, 4.3 mm, 4.4 mm, 4.5 mm, 4.6 mm, 4.7 mm, 4.8 mm, 4.9 mm, 5.0 mm, 5.5 mm, 6.0 mm, 6.5 mm, 7.0 mm, 7.5 mm, 8.0 mm, 8.5 mm, 9.0 mm, 9.5 mm, 10 mm, 11 mm, 12 mm, 13 mm, 14 mm, 15 mm, 16 mm, 17 mm, 18 mm, 19 mm, 20 mm, 21 mm, 22 mm, 23 mm, 24 mm, 25 mm, 26 mm, 27 mm, 28 mm, 29 mm, or 30 mm. 25 In some embodiments, the substantially continuous layer has substantially even depth across the application surface. In one embodiment, the aircraft structure is a single component, such as a sheet metal. In 30 one embodiment, the aircraft structure has two or more components, such as a lap joint. 20 In one embodiment, the substantially continuous layer is formed by a method as described herein. In one embodiment, the aircraft structure is a joint comprising a first component, and a 5 second component, the first and second components each having an aperture, the first and second components jointed by a fastener extending through the apertures, a substantially continuous layer over or about at least one of the apertures. Preferably, the substantially continuous layer does not extend to cover a junction between 10 the free end of a first sheet metal component of the lap joint and the face of a second sheet metal component of the joint.. This form of the joint is preferred because the SPD doubler does not transfer any load between the upper and lower skin. If the joint were modified in this way, a substantial amount of validation work would be necessary to fully characterise the structural properties of the modified joint. 15 The present invention will now be more fully described by reference to the following non limiting examples. EXAMPLE1: Effect of SPD on the fatigue performance of cracked metallic structures. 20 To study the effect of a supersonic particle deposition (3PD) on the fatigue performance of cracked metallic structures initial tests were performed on a 350 mm long and 1.27 mm thick 2024-T3 dad aluminium alloy dogbone specimen which contained a centrally located 2 mm long edge notch, see Figs. 1-3. These initial tests were performed under constant amplitude loading with amax = 181 MPa and R= amin/Gma = 0.1. (This stress level was chosen 25 since it represents a realistically upper bound on stresses that can be expected in a thin wing skin.) Two specimens were tested, one without a deposited metallic layer (also known as a "SPD doubler"), and one with a 1 mm thick full width doubler, that extended over the working section of the specimen, deposited on either side of the specimen, see Figs. 2 and 3. The doublers were deposited using a 7075 Aluminium Alloy powder with a nominal 30 particle size of between 30 and 40 pm The following deposition parameters were utilized: 21 E.APSD TiME BETWEEN SJAFFREPAPATION AND ODATI NG- 15-20mi ns MAIN GAS PRESJE(Bar): 40 P/FVFRA-PRESJRE (Bar): 38 TBPEPATURE(*C): 400 5 P/ F HOPPER HEATER Active MAIN GAS FLOW (m3/hr): 92-100 P/F GASQUANTITY(m3/h): 6.5 POWDER FEED PATE (RPM): 2.7rpm RLEAS- TEMP(C): 300 PREHEATTEMAP( C): 350 10 INCFRMENT (mm): -0.25mm TPAVERSE PATE (mm/s): 250mm/s STAND OFF (mm): 40mm (where possible) NUMBEROFLAY1R 12- 16 Passesper Patch DEPOSTION TH10(NES 0.005 - 0.012" 15 FOSITTETAMENT: None For the baseline specimen test the crack length was monitored using digital cameras. However, whilst there are numerous non-destructive inspection tools that are commonly used to monitor crack growth in aircraft structures, i.e. ultrasonics, eddy currents, 20 thermography, etc., the present study used Lock-in infra-red thermography to simultaneously monitor the evolution of the stress and the damage states in the 2024-T3 skin and the SPD doublers. (At this point it should be noted that to ensure a uniform emissivity the surface being monitored was sprayed matt black and that thermography was used as a qualitative rather a quantitative measure of the stresses and the fatigue damage. 25 Details on the use of lock-in thermography to measure surface stresses and energy dissipation are given in [13-15].) The baseline specimen, i.e. without a doubler, lasted approximately 35,000 cycles. In contrast the 7075 SPD patched panel test was stopped after approximately 60,000 cycles with little, i.e. no evident, damage in the 7075 SPD or crack 30 growth in the 2024-T3 skin. Rgs. 4 and 5 present infrared pictures of the stress field at 11,100 and 56,100 cycles respectively. These figures show that the stresses in the SPD doubler remained essentially unchanged throughout the test. 22 EXAM PLE2: Sncde edcg notch tension (NET) PD strip tests. To further study the ability of TD doublers to reduce crack growth tests were performed on a single edge notch dogbone specimen, with a geometry as described above and an (initial) 5 1.4 mm long edge notch. In the initial base line test there was no TD and the specimen was tested under constant amplitude loading with a peak stress in the working section of amax = 93.36 MPa and R (amin/amax) = 0.1. This stress level was chosen to represent a typical fuselage skin stress. Cack growth in the 2024-T3 plate was monitored using digital cameras and the resultant crack length versus cycles history is shown in Fig. 6. 10 In the next test, the specimen was first loaded so asto grow a sharp crack. This first phase of the test was stopped at 18,886 cycles when the crack length was approximately 3.2 mm. A 10 mm wide and 1 mm thick STD strip with a nominally (isosceles) triangular crossection, see Figs. 7 and 8, was then installed and the test was continued. The Qack growth in the 15 2024-T3 plate was again monitored using digital cameras whilst the stress field in both the 3 D strip and the 2024-T3 skin and the degradation in the TD strip was monitored using Lock-in infrared thermography. An infrared stress image captured shortly after the restart of the test is shown in Fig. 9. In 20 this figure the picture was captured at a cyclic stress amplitude aG, remote from the centre line of the specimen, of approximately 53 MPa. This was done so as to not overly influence crack growth in the skin. Here it can be seen how the stress field in the 3'D ahead of the crack is contiguous with that in the plate, i.e. the S'D is taking load in t he region ahead of the crack. 25 Hot spots were also noted in the skin outboard of the ends of the SPD strip which establish that the TD strip was indeed pulling load from the skin. This is essential if the process is to enhance the damage tolerance of the skin. The resultant crack growth data is shown in Fig. 6 where it is seen that the use of a 7075 aluminium alloy TD strip has significantly reduced 30 the crack growth rate. 23 A second test was then performed whereby the 'PD strip was applied to a 0.3 mm long initial edm (electrical discharge machine) crack (notch) where the crack was not sharpened (grown) prior to installation of the 'TD strip. In this case the test was stopped after approximately 345,000 cydes since there was no apparent crack growth at t he notch (crack) 5 or damage in the 'D. EXAM PLE3: C-acking in 7050-T7451 SNT tests. It is well known t hat for combat aircraft most of t he fat igue life of t he st ruct ure is consumed in the growth of short cracks [16]. Consequently to evaluate the effect of a 'TD repair on 10 small flaws in aircraft structural components a 3 mm thick S-NT (single edge notch tension) dogbone specimen, was tested with a K = 1.11, with a thin 0.5 mm thick 7075 aluminium alloy 'TD patch on one side. The 7050-T7451 specimen was 350 mm long, 42 mm wide and 3 mm thick and had a 0.69 mm radius semi-circular edge notch on one side. The specimen was tested at a peak stress, in the working section, of 140.0 MPa with R = 0.1. This 15 corresponds to a peak (remote) load of 17.64 kN with R= 0.1 and was chosen to represent the stress, at limit load, in the wing skin of a typical fighter aircraft. A thin 'TD doubler was used so that, in this test program, the damage induced could be evaluated, as the crack opened and closed during fatigue loading, in the interfacial region !0 between the TD and the 7050-T7451. This damage could have been reduced by increasing the thickness of the TD thereby lowering the stresses in the underlying 7050-T7451 and subsequently reducing the opening of the crack. The ability of the TD doubler to pull load from the underlying 7050-T7451 structure is dear from the E-Mode (stress) Lockin thermography picture of the stresses on the specimen side with the TD patch at 3000 25 cydes, see Fig. 10. Although the crack in the 7050-T7451 specimen was not immediately evident an analysis of the infra-red data associated with the left hand side of the specimen shown in this picture, i.e. in the TD directly over the crack, revealed an indication of the crack under the patch. After 33,000 cydes the crack in the 7050-T7451 had grown to a length of approximately 4.2 mm and the resultant stress picture is shown in Fig. 11. At this 30 point there is evidence of delamination damage (disbonding) on the LHSof the TD in the region that lay over the crack. 24 The extent of the damage is illustrated in Fig. 12 which presents a picture of the dissipated energy at 33,500 cydes. (Note that the dissipated energy associated with the crack tip is dearly evident in this figure. This is important because it raises the potential of non destructive inspection of the specimen through the SID doubler.) At 35,500 cydesthe crack 5 had grown to approximately 4.92 mm and the associated stress picture is shown in Fig. 13 where it can be seen that the delamination in the D has grown slightly. It would thus appear that whilst damage to the TD interface can result due to crack opening the onset of damage does not appear to lead to immediate (catastrophic) failure in 10 the TD. As such damage growth in the TD can be induded in the damage tolerance assessment of the TD repair process. Furthermore, given that there was no apparent damage at the ends of the TD the damage in the central region over the crack can be controlled by increasing the thickness of the D in this region thereby reducing the stress in the 7050-T7541 together with the associated crack opening displacement. 15 The test was stopped at 37,000 cydes at which stage the crack was approximately 5.3 mm long. A plot of the measured crack length versus cycles history is presented in Fig. 14 together with test data for the case when there was no TD. Here it will be noted that the TD patch has somewhat reduced the crack growth rate. To further confirm the ability of !0 TD to restore structural integrity and to illustrate the ability to control the onset of delamination damage over the crack a test was subsequently performed on a 1 mm thick 7050-T7451 E-NT specimen, with a 0.8 mm long initial edge crack and an in-plane geometry as per the previous test. This specimen had two 0.5 mm thick D doublers on either side of the specimen. The specimen was subjected to a peak (remote) load of 5.88 kN with R= 0.1 25 which equates to the same remote stress as in the previous test. In this case the test was stopped after 117,000 cydes as there was no apparent crack growth and no apparent degradation in the D. EXAM PLE4: Predictinq crack crowth in the 7050-t7451 sent test. 30 This example is directed to prediction of the crack length history seen in the 7050-T7451 S-NT test outlined in &ction 2.2. Here a 3 mm thick -NT (single edge notch tension) dogbone specimen with a thin 0.5 mm thick 7075 aluminium alloy TD patch on one side 25 was tested. The specimen was 350 mm long, 42 mm wide and 3 mm thick and had a 0.69 mm radius semi-circular edge notch on one side. The specimen was subjected to a peak (remote) load of 17.64 kN with R= 0.1. 5 The stress intensity factor for a through-the-thickness crack of length c emanating from the centre of the notch of radius r isgiven in [17] as: K - f 1 gJ./(rc) (1) 10 Where c is the length of the crack emanating from the notch and a is the stress in the 7050 17451 underneath the SPD. The values of f 1 , g4 and fw taken from [17] are: fi - 1 + 0358<p + 1.425cp 2 - 1.578cp 3 + 2.156t94 (2) (p = 1/(1 + c/r) (3) g4 = Kt(0.36 - 0.32/ f(1 + c/r) (4) f, = 1 + 2.7p2 - 3.5<p4 + 3.8tp6 (5) K= 3.17 (6) 15 Snoe the specimen was tested using hydraulic grips the formulae used for fw was the fixed displacement expression given in [17]. Let us now attempt to use this solution to predict crack growth. Fig. 11-13 revealed that there was (delamination) damage growth in the TPD over the crack. Thus as recommended in [18] for composite repairs to cracked metal skins the problem was analysed by assuming that the resultant stress intensity factor was equal 20 to the solution to the SENT specimen subjected to a stress field ao which corresponds to the stress in the (base) specimen under the TPD in the absence of a crack. The DSTO Cbmbat and Trainer Aircraft Group [19-21] have shown that the growth of small flaws in 7050 17451 conforms to the Generalised Frost-Dugdale crack growth law, viz 25 da/dN C~au -(AK( KKp/G'y)|(1- Km/Ke) (7) 26 where G, y and K, are material constants and a, Kma and AKare crack lengths, the maximum value of the stress intensity factor at cycle N and the range of the stress intensity factor at cycle N respectively. The crack length history was predicted by integrating Eq. (7) using Eqs. (1)-(8) with Pmax = 17.4 kN, R= 0.1 and r = 0.69 mm. In this calculation the values of y, p, C-, 5 yto be as given in [19,21], viz y = 3, p = 0.2, G= 0.50, (y =460 MPa and, for this thickness, Kc -50-65 MPa vm were taken. (In this analysis a value of K. = 60 MPa vm was used. However, for this range of loads and crack lengths the value has a small effect on the crack length predictions.) The resultant predicted crack length history is shown in Fig. 15 where a good agreement is seen between the measured and predicted crack length histories. 10 In this case, as for cracks growing under composite repairs [18] the stress intensity factor can be approximated as the solution to the SENT specimen subjected to the stress field ao which corresponds to the stress in the (base) specimen under the STD in the absence of a crack. One advantage of this approach is that the computed crack length history should be 15 conservative. BAM PLE5: Approximate solutionsfor centre cracked panels repaired usinq SPD. In the previous example the case of a thin (0.5 mm) SD repair to a small flaw in a relatively thick (3 mm) section was considered, and STD delaminated on either side of the crack was 20 noted. In such cases it was reasonable to assume that the dominant effect of the SPD was to merely reduce the net section stress [18]. However, for certification purposes the solution needed for the stress intensity factor associated with an arbitrary length crack where STD patch is not thin. It is also necessary to establish if, for a given crack length, the stress intensity factor range AK is beneath the threshold value AKth as thiswill significantly simplify 25 the certification process. To this end this example will consider an STD repair of thickness tr to a centre cracked panel, thickness tp, with an interfacial region, thickness ti, that has been (potentially) affected by the STD process subjected to a remote stress a as shown in Fig. 16. The SPD process can result in an interfacial region that has been affected by the SPD process [22-24]. For the aluminium alloy powders used in SPD repairs the maximum particle size is 30 approximately 40 pm. Consequently the thickness (ti) of this region is generally very small [22-24] in comparison to the thickness of the underlying plate, i.e. typically less than 0.1 0.15 mm, see Fig. 17. As a result this problem is analogousto that of a bonded repair where 27 the interfacial region mimics the adhesive that joins the repair to the plate. It is known that for small cracks in metal skins repaired using a composite patch the 2D solution for the stress intensity factor is essentially due to the reduction in the stress field under the repair whilst for long cracks the stress intensity factor asymptotes to a limiting stress intensity 5 factor K. as the crack length increases, see [18,25-28]. As such it follows that the 2D solution for the stress intensity factor associated with small cracks repaired using STD is also essentially due to the reduction in the stress field under the 3PD whilst for long cracks repaired using SPD the stress intensity factor should also asymptote to a limiting stress intensity factor K. as the crack length increases. The approximate formulae for this 10 asymptote thusfollowsfrom [26], see pp. 216-218, viz K, = YQL(TOVQ (8) where (o = (-Eptp/(Ept, + Ert,) (9) 15 Y is a geometry factor, = 1 for a large centre cracked panel and OL is a load attraction factor that accounts for the different stiffness of the repaired region. (QL = 1.) The term TEA is given by the expression 71, = Eptpl/f(1± tpEp/Ertr) (10) where 20 3= (t 1 /G, + tt/3G, + tp/3G,)/(t 1 /G + 3tr/8Gr + 3tp/8G)2 (11) Here ti, tp and tr are the thickness' of the interface region where, the SPD has modified the properties of the plate, of the plate, and of the SPD respectively, G and Edenote the shear and Young's modulus and the subscripts i, p and r denote their values for the interfacial 25 bonding region, the plate, and the 3PD repair respectively. (The notation used in this section follows that given in [26] pp. 217-218.) This expression, i.e. Eq. (11), is an extension of the formulae first developed in [25] in that it allows for the interfacial thickness ti to be 28 negligible. This (allowance) is important since for STD repairs the modulii of each region will generally be comparable and the interfacial thickness ti that is affected by the SPD process is expected to be very small. As such the terms in Eq. (11) related to the term ti/G are small in comparison with those terms relating to the PD repair (tr/3$) and the plate (tp/3Gy). 5 Consequently the expression for P can often be approximated as: # = (tr/3Gr + tp/ 3 Gp)/( 3 tr/ 8 G, + 3tp/8Gp) 2 (12) It is expected that, in many instances, the STD powder, used in the repair, and the plate 10 material will have essentially the same modulii, i.e. aluminium plates are expected to be repaired using aluminium alloy powders and steel components are likely to be repaired using steel powders. In such cases G( can be approximated by G so that Eq. (12) reduces to: # = 64G,/27(t, + t,) (13) 15 Having established the asymptotic limit it follows from [28-30] that the functional form of K as a function of the crack length (a) can be approximated as: K = W(a/mA)aov'(7a) (14) where the function W, viz: W(a/(7r,)) = \/[(1 + 2.23a/(7n2))/(1 + 3a /(7i,) + 7(a/(nz))2 (15) 20 describesthe transition from the small crack solution a-+O to the long crack solution a->-1, see [28,30]. Eqs. (14) and (15) reveal that for short cracks the reduction in the stress intensity factor is essentially due to the reduction in the stress in the plate due to the SPD patch, i.e. limitK = aO a (16) 25 a For long cracks K tends to its asymptotic limit K. In Eq. (14) the functional form associated with [30] has been used rather than that given in [28]. 29 EXAMPLE6: SPD repairsto cracks in an arbitrary stressfield This example considers the case of an STD repair to a crack with a total of length 2a subjected to an arbitrary stress field. In this instance the solution for the stress intensity 5 factor Kfollows from the above analogy with a composite repair to a crack in a metal skin under an arbitrary symmetry stress field [29], viz K = W(a/rrA)K (17) 10 where K is the solution to the entre cracked specimen subjected to a stress ao which corresponds to the stress in the (base) specimen under the SPD in the absence of a crack is given by _( = 2v'(a/7T) J ,(x)/(a 2 _ x 2

)

112 dx (18) 15 and cy is the stress in the skin under the TD in the absence of a crack, see Fig. 18. To evaluate the accuracy of this approximation let us consider a 3 mm thick 200 mm 200 mm centre cracked plate repaired using a 3 mm (thick) 200 mm 200 mm TD patch subjected to a remote uniform stress (in the skin) of 100 MPa. To this end three dimensional 0 finite element models were constructed for: 2, 6, 10, 20, 30, and 40 mm long cracks Due to symmetry considerations only one quarter of the structure needed to be modelled. In each case the models had approximately 66,000 three dimensional twenty-one nodded isoparametric brick elements and approximately 300,000 nodes, see Fig. 19. There were eight elements through the thickness of the TD and eight elements through the thickness 25 of the plate. In each case there were ten elements along the crack and the side length of the crack tip elements were approximately 1/100th of the length of the crack. The midside nodes associated with the near tip elements were moved to the quarter points so as to simulate the necessary r- 1 2 singularity. Bending of the STD and the plate was prohibited. Both the aluminium alloy plate and the SPD were assumed to have a Young's modulus E = 30 70,000 MPa and a Poisson's ratio of 0.3. The computed values of the maximum value of the 30 stress intensity factor Kmax are given in Table 1 below along with the associated analytical values, where Eq. (12) was used to compute P, and the quantity. Table 1 Comparison between predicted and computed stress intensity factors, a (mm) Finite element Analytical K, Upper bound Ku, " (MPa yim) (MPa ym) (MPa ,/m) 1 4.88 5.03 5.60 3 6.77 6.82 9.71 5 7.29 7,29 12.53 10 7.56 7,59 17.72 15 7.60 7.65 21.71 20 7.60 7.68 25.07 5 K, =(19) which represents an upper bound on K Here it can be seen that the stress intensity factor associated with STD repairs does indeed asymptote to a constant value and that this 10 asymptote is in good agreement with the analytical approximation, i.e. Eq. (14). EXAM PLE7: Quality control assessment tool. When performing composite repairs to aircraft structural members it is common practice to make travelling specimens that are subsequently used to assess the quality of the repair 15 [11,12]. The challenge is to develop a similar approach for SPD modifications' repairs to aircraft structural components As such this section raises the possibility of using simple specimens that are subsequently fatigue tested and the quality of the bond assessed via the fractal dimension [31] of the resultant fatigue surfaces. In it this context it should be noted that is now known that fracture surfaces can be considered as a fractal set, see Mandelbrot 20 et al. [31]. In this work Mandelbrot et al. [31] wrote: "When a piece of metal is fractured either by tensile or impact loading the facture surface that is formed is rough and irregular. Its shape is affected by the metal's microstructure (such as grains, inclusions, and precipitates where characteristic length is large relative to the atomic scale), as well as by 'macrostructural' influences (such as the size, the shape of the specimen, and the notch 25 from which the fracture begins). However, repeated observation at various magnifications also reveal a variety of additional structures that fall between 'micro' and 'macro' and have 31 not yet been described satisfactorily in a systematic manner. The experiments reported here reveal the existence of broad and dearly distinct zone of intermediate scales in which the fracture is modelled very well by a fractal surface." It is also known [32-34] that, prior to the onset of rapid fracture, fatigue crack surfaces in metals, that are not associated with 5 very small crack lengths, have a fractal box dimension D, as defined in [32], that lies between approximately 1.2 and 1. Thus it may be possible to use this observation to quantify the quality of the SPD process To do this travelling specimens would be fabricated in parallel with the TPD application. These travelling specimens would subsequently be fatigue tested and the associated fractal box dimensions measured. It is hypothesised that if 10 D had a value that was near 1.2, or lower, then you would have a process that produced a fatigue crack surface that was consistent with that associated with fatigue crack growth in the base material, and the process would be acceptable. If it was significantly greater then it is hypothesised that the application process may be deficient. 15 To evaluate this concept the fractal box dimension was measured, the fractal box associated with SPD doublers used on a rib stiffened panel deposited using powders where there was (subsequently) found to be a quality control issue with the powder, i.e. it was found to contain a large proportion of sub 10 micron particles. In this instance the panels had two ten mm wide and 200 mm long TPD doublers located on either side of a centrally 20 located 20 mm long crack, see Fig. 20. As a result of the poor quality powder one end of each of the two STD strips delaminated with the locus of the delaminations lying entirely within the SPD, see Figs. 21 and 22. The fractal box dimensions associated with delamination surfaces on each of the two STD strips, 25 referred to in Figs. 21 and 22 as strips A and B, that delaminated from the structure were measured and the resultant values are given in Tables 2 and 3, below. 32 Table 2 Fractal box dimension (D) associated with the end of strip A. Random area 1 within location Random area 2 within location Loc 1 1.629 1,500 Loc 2 1.409 1.675 Loc 3 1.542 1,684 Loc4 1.416 1.473 Loc 5 1.543 1.530 Loc 6 1.399 1,529 Average 1.49 1,57 Table 3 Fractal box dimension (D) associated with the end of strip B. Random area I within location Random area 2 within location Loc 1 1.673 1.613 Loc 2 1.482 1.521 Loc 3 1.525 1.614 Loc 4 1.551 1.49 Loc 5 1.526 1.516 Loc 6 1.558 1.561 Loc 7 1.578 1.482 Loc 8 1.503 1.593 Loc 9 1.584 1.563 Average 1.553 1.550 5 Here it will be noted that in each case the fractal box dimension D was essentially constant at each of the locations measured on each of the two delaminated strips. Furthermore, the value of the fractal box dimension D was approximately 1.5, see Tables 2 and 3. As such the fractal box dimension D associated with these two poor quality SPD's differed significantly 10 from that associated with macro-scopic fatigue crack growth in metals. Thus whilst a great deal more work is needed to validate the hypothesis that D can be used to quantify the quality of the SPD it looks to be worthy of further evaluation. It is interesting to note that prior to these tests a value of D = 1.5 had only (previously) been found for very small fatigue cracks [32,34]. A more detailed discussion of the role of the fractal dimension D in 15 describing the nature of the crack tip singularity and in characterising fatigue crack growth is given in [20,34-37]. EXAM PLE8: Application of SPD to an aircraft lap joint. Specimens have been prepared to evaluate the application of the SPD on a representative 20 aircraft lap. 33 The spedmen geometry was developed as part of the FAA Aging Aircraft Program, where it was shown to reproduce the crack length history seen in Boeing 727 and 737 fleet data The basic specimen used consists of two 2024-T3 dad aluminium alloy sheets 1.016 mm 0.04 inch) thick, fastened with three rows of BACR15CE-5, 1000 shear head counter-sunk rivets, 5 3.968 mm (5/32 inch) diameter (Fig 24 ). The width of the specimen was chosen to coincide with the typical distance between tear straps of a B-737 aircraft. Snce the amount of out of-plane bending in a typical fuselage joint is an important factor in the fatigue performance of the joint, the amount of local bending in the specimen was made similar to that seen in a typical fuselage joint by testing the specimens bonded back-to-back and separated by a 25 10 mm thick honeycomb core. This test configuration was crucial in ensuring that the specimens reproduced fleet behaviour, see [40, 41]. As in [40, 41] the upper row of rivet holes contained crack initiation sites, induced prior to assembly of the joint by means of an electrical spark erosion technique, on either side of the rivet holes. These initial cracks were (each) nominally 1.25 mm long. This crack length was chosen so that the (initial) defect was 15 obscured by the fastener head and as such was representative of largest possible undetectable flaw size. A 1 mm thick 7075 SPD doubler was deposited over the three rows of fasteners, (Fig 25). The Pbwder was deposited utilising the Knetics 4000 aries GT equipment with the Type !0 33 polycarbonate nozzle. The surface was predeaned utilising 120 Aluminium Oxide grit at 60 psi. The deposition parameters were as follows: ELAPE-D TliME BETWEEN SJRAFCE PFEP AND CDATI NG: 15-20m ins 25 MAIN GASPFESSJFE(Bar): 40 / IFVF99-1 PE5SJRE(Bar): 38 TEMPERATURE (C): 400 P/ F HOPPER HEATER Active MAIN GAS FLOW (m3/hr): 92-100 P/F GASQUANTlY(m3/h): 6.5 30 POWDERIFEED PATE(RPM): 2.7rpm RELFASE TEMP(C): 300 PFREHEATTBMP (C): 350 34 I NCREM ENT (mm): -0.25mm TPAVERSE PATE (mm/s): 250mm/s STAND OFF (mm): 40mm (where possible) NUMBEROF LAM1R 12 - 16 Passes per Patch 5 DEPOS110N TH10(NES 0.005 - 0.012" FOSI TESTAMENT: None EXAM PLE9: Testingc of SPD applied to aircraft lap ioint 10 The specimens were tested under constant amplitude loading, with the maximum and minimum loads as detailed below. PMax (kN): 40 PTAin (kN) 2 15 PMean (kN) 21 Test Frequency (hZ) 5 These loads were determined from operational data obtained for the US DoT MS) Committee review Board for the B-737 aircraft, see [40] for more details, and a stress 20 picture showing the stresses in the baseline specimens is presented in Figure 26 and a stress picture just prior to link up of msd is shown in Figure 27. The fatigue performance of the baseline (no 3PD) specimens is documented in [42]. Here it was found that for specimens without an SPD modification the number of cydes to 25 first link up of cracks from adjacent holes occurs at approximately 30,000 cydes. To illustrate thisand to show the stresses in the baseline joint Figures 28 and 29 present the stresses in a (baseline) joint at approximately 6,500 and 29,000 cydes respectively. For Test Panel 1 the test program revealed that after approximately 110,000 cydes the SPD 30 doubler was still intact. Furthermore, there was no apparent crack growth at any of the fasteners in the lap joint, cracking in the SPD or damage to the bond between the TPD and 35 the skin/fasteners. This is evident from Figure 30 where we show a dose up view of the stresses in three rows of rivets at 92,000 cydes. The test results revealed that the SID doubler significantly reduces the stresses in the joint. 5 This means that the TD seals the fasteners and continues to do so for more than 110,000 cydes. This represents a factor more than 3.6 in the Limit of Viability (LOV) [12] of the joint. Test Panel 2 also achieved 2 times the LOV (60,000 cydes) even though there were pre existing delaminations between the skins (both upper and lower) and the honeycomb core (i.e. loss of panel stability) prior to test . Figures 31 and 32 present the stress distribution in 10 the TD Test Panel 2 at 18,000 cycles and 48,000 cydes respectively. Comparing Figures 26 and 31 a significant reduction in the stresses in the joint is noted. In the upper section of the picture 31, an increase in the stresses in the skin at the edge of the D where load is being attracted up from the skin into the TD. The stress concentrations in the TD over each of the fasteners are also visible. 15 From this Example it can be seen that the TD has remained intact, thereby ensuring that the joint is sealed. Although this study has focused on fuselage lap joints the ability of an ,TD doubler to form a durable bond to both the skin and the fasteners predicts that this approach may well be applicable to other problem areas in an aircraft. !0 36 REFIERENCER [1] Efforts to reduce corrosion on the military equipment and infrastructure of the department of defense, Office of the %cretary of Defense, USA Department of Defense 5 Fbport; June 2007. [2] Karthikeyan, d 2004. (bId spray technology: International status and USA efforts. AM Industries. 1-14. 10 [3] Decker MK Snith MF. Thermal spray and cold spray analysis of density, porosity, and tensile specimens for use with LIGA applications. SAND2000-0339, %ndia National Laboratories; February 2000. [4] %rtwell BD, Kestler PR Legg KO, Assink W, Nardi A, Shell J Validation of HVOFWCCb, 15 WO COQ- and Tribaloy 800 thermal spray coatings as a replacement for hard chrome plating on C-2/E-2/P-3 and 0-130 propeller hub system components, NRL-PP-99-22-FR01, Naval Research Laboratory, Washington; May 2003. [5] Skaki K Cold spray process overview and application trends. Mater Si Forum 20 2004;449-452:1305-8. [6] Karthikeyan, d Development of oxidation resistant coatings on GIop-84 substrates by cold spray process. NASA-CR2007-214706; 2007. [7] Pepi, M. Cold spray technology for repair of magnesium rotorcraft components. NAVAIR corrosion resistant alloy workshop; 8-9, November 2006. 25 [8] Villafuerte d Cirrent and future applications of cold spray technology, Rcent trends in cold spray technology: Fbtential applications for repair of military hardware, NATO RTO-MP AVT-163; 2010. p. 1-14. <ftp://ftp.rta.nato.int>. 30 [9] Matthews N. Supersonic particle deposition (SPD) cutting edge technology for corrosion protection and damaged metallic component recover In: Proceedings 2010 S)E symposium 37 program "Design Engineering in a SRP Environment". RAAF Williams, Melbourne, Australia; 24-25, March 2010. [10] Soltenhoff T. Praxair surface technologies GmbH, Germany, 8th Colloquium, HVOF 5 spraying cold spray, Conference, Erding, Germany; 2009. [11] Baker AA, Rse LRF, bnes R Advances in the bonded composite repair of metallic aircraft structure", vol. I and II, Elsevier Applied Science Publishers; 2002. ISBN 0-08-042699 9. 10 [12] Baker AA bnes R Bonded repair of aircraft structure. The Hague: Martinus Nijhoff Riblishers; 1988. [13] Harwood N, QjmmingsWH. Thermoelastic stress analysis. Bristol: Adam Hilger; 1991. 15 [14] Diaz FA, Yates JR, Patterson EA. Some improvements in the analysis of fatigue cracks using thermoelasticity. Int J Fatigue 2004;26:365-76. [15] bnes PR Rtt S An experimental evaluation of crack face energy dissipation. Int JFatigue 20 2007;28(12):1716-24 (Details on the use of lock-in thermography to measure surface stresses and energy dissipation are given in [13,14].). [16] N.S lyyer, N.E Dowling, Fatigue growth and closure of short cracks, AFWoALTR- 89 3008; June 1989. 25 [17] Newman JC Wu Xfs Venneri S., Li OG. Snall-crack effects in high-strength aluminium alloys, NASA, editor. NASA; 1994. [18] bnes R A scientific evaluation of the approximate 2Dtheories for composite repairs to 30 cracked metallic components, Compos Sruct 87(2) (2009)151-160. 38 [19] bnes R Molent L Citical review of the generalised frost-dugdale approach to crack growth in F/A-18 Hornet structural materials, DSTO-R-0350; March 2010. [20] bnes R Molent L, Pitt S Qack growth from small flaws. Int J Fatigue 2007;29:658 5 1667. [21] Jones R Barter S Chen F. Experimental studies into short crack growth. Eng Fail Anal 2011. doi: 10.1016/j.engfailanal.2011.03.012. 10 [22] Zhao ZB, Gilispie BA, S-nith R Coating deposition by the kinetic spray process. Srf Coat Technol 2006;200:4746-54. [23] Hussain T, McCartney DG, Siipway PH, Zhang D. Bonding mechanisms in cold spraying: the contributions of metallurgical and mechanical components. J Therm Spray Tech 15 2008; 18(3):364-79. [24] Fepi M, Cold spray technology for repair of magnesium rotor craft components. In: proceedings NAVAJRcorrosion resistant alloy workshop, 8-9, November 2006. 20 [25] Ibse LRF. A cracked plate repaired with bonded reinforcements. Int J Fract 1982;18:135-44. [26] bnes R Numerical analysis and design. In: Baker A, Fbse LFT, bnes R editors. Advances in the bonded composite repair of metallic aircraft structure. Bsevier Applied 25 Science Publishers; 2002. ISBN: 0-08-042699-9. [27] bnes R Qack patching: design aspects. In: Baker A, bnes R editors. Bonded lbpair of Aircraft Sructure. The Hague: Martinus Nijhoff Publishers; 1988 [chapter 9]. 30 [28] Wang CH, Fbse. A crack bridging model for bonded plates subjected to tension and bending. Int JSblids lruct 1999; 36:1985-2014. 39 [29] ,bnes R, Qliu WK, Marshall IH. Weight functions for composite repairs to rib stiffened panels. Eng Fail Anal 2004;11(1):49-78. [30] Hart-Smith Li. Recent expansions in the capabilities of Rose's closed-form analyses for 5 bonded crack-patching. In: Baker A, FRse LF, ,bnes PR editors. Bsevier Applied Sience Publishers; 2002. ISBN: 0-08-042699-9 [chapter 8]. [31] Mandelbrot BB, Passoja DE, Paullay Al Fractal character of fracture surfaces of metals. Nature 1984;308:721-2. 10 [32] Bouchaud E &aling properties of cracks. JPhysCbndens Matter 1997;9:4319-44. [33] Mandelbrot BB. Fractal analysis and synthesis of fracture surface roughness and related forms of complexity and disorder. Int J Fract 2006; 138:13-7. 15 [34] Carpinteri A, Paggi M. A unified fractal approach for the interpretation of the anomalous scaling laws in fatigue and comparison with existing models. Int J Fract 2010; 161:41-52. 20 [35] Sagnoli A. lf-similarity and fractals in the Paris range of fatigue crack growth. Mech Mater 2005;37:519-29. [36] Saether E, Ta'asan S. A Hierarchical approach to fracture mechanics, NASA/TM-2004 213499. 25 [37] Carpinteri An, Spagnoli An, Vantadori S5 Viappiani D. Influence of the crack morphology on the fatigue crack growth rate: a continuously-kinked crack model based on fractals. Eng Fact Mech 2008;75(3-4):579-89. 30 [38] http://www.dailymail.co.uk/news/artide-1374574/ Suthwest-Airlines-fully operat ional-plane-cracks-repai red. html 40 [39] http://www.bbc.co.uk/news/world-us-canada-12954335 [40] Jones R and Molent l, Giapter 16, Fpair of Multi-site Damage ,A. Baker, Advances in the Bonded Composite Fpair of Metallic Aircraft Sructure , Edited by L R F. Fbse and 5 Jones R, Esevier Applied :ience Publishers, 2002. ISN 0-08-042699-9 [41] Jones R Cairns K, Baker J., Krishnapillai K And Hinton B. , A study of the effect of CPS on fatigue crack propagation in a representative fuselage lap joint specimen, Engineering Fracture Mechanics, doi: 10.1016/j.engfracmech.2011.11.015. 10 41

Claims (26)

1. A method for (i) repairing a structural weakness, and/or (ii) preventing or inhibiting the initiation of a structural weakness, and/or (iii) preventing or inhibiting the progression of a structural weakness in an aircraft structure and/or (iv) preventing the ingress of an 5 environmental element, the method comprising the step of bonding a plurality of parties to the structure, the bonding being effected under conditions allowing the plurality of metallic particles to form a substantially continuous layer.

2. A method according to claim 1, wherein the method isfor(i) repairing a structural weakness, and/or (ii) preventing or inhibiting the initiation of a structural weakness, and/or 10 (iii) preventing or inhibiting the progression of a structural weakness in an aircraft structure and (iv) preventing the ingress of an environmental element.

3. A method according to claim 1 or claim 2 wherein at least a proportion, or substantially all, of the particles are metallic particles.

4. A method according to any one of claims 1 to 3 wherein the bonding does not 15 involve melting or fusing of the particles.

5. A method according to any one of claims 1 to 4 wherein the bonding is achieved by a cold spray process.

6. A method according to claim 5 wherein the cold spray process is supersonic part ide deposition. 20

7. A method according to any one of claims 1 to 6 wherein the substantially continuous layer is at least about 0.05 mm.

8. A method according to any one of claims 1 to 7 wherein the substantially continuous layer has substantially even depth across the application surface.

9. A method according to any one of claims 1 to 8 wherein the aircraft structure is a 25 fuselage component.

10. A method according to any one of claims 1 to 9 wherein the aircraft structure is a sheet metal. 42

11. A method according to any one of claims 1 to 9 wherein the aircraft structure is a lap joint.

12. A method according to claim 11 wherein the substantially continuous layer does not extend to cover ajunction between the free end of a sheet metal component of the lap 5 joint.

13. A method according to claim any one of aims 1 to 12 wherein the structural weakness is a crack.

14. An aircraft structure comprising a substantially continuous metallic layer, the layer being deposited on a surface of the structure, the layer being capable of (i) repairing a 10 structural weakness, or (ii) preventing or inhibiting the initiation of a structural weakness, or (iii) preventing or inhibiting the progression of a structural weakness in the aircraft structure, wherein the layer comprises a plurality of metallic particles.

15. An aircraft structure according to claim 14, wherein the layer is capable of (i) repairing a structural weakness, and/or (ii) preventing or inhibiting the initiation of a 15 structural weakness, and/or (iii) preventing or inhibiting the progression of a structural weakness in an aircraft structure and (iv) preventing the ingress of an environmental element.

16. An aircraft structure according to claim 14 or daim 15 wherein at least a proportion, or substantially all, of the parties are metallic parties. 20

17. An aircraft structure according to any one of claims 14 to 16 wherein the substantially continuous layer is deposited on the surface of the aircraft structure by a method according to any one of aims 1 to 13.

18. An aircraft structure according to any one of claims 14 to 17 wherein the substantially continuous layer is at least about 0.05 mm. 25

19. An aircraft structure according to any one of aims 14 to 17 wherein the substantially continuous layer has substantially even depth across the application surface. 43

20. An aircraft structure according to any one of claims 14 to 19 wherein the aircraft structure is afuselage component.

21. An aircraft structure according to any one of claims 13 to 18 wherein the aircraft structure is a sheet metal. 5

22. An aircraft structure according to any one of claims 13 to 19 wherein the aircraft st ructure is a lap joint.

23. An aircraft structure according to claim 22 wherein the substantially continuous layer does not extend to cover a junction between the free end of a first sheet metal component of the lap joint and the face of a second sheet metal component of thejoint. 10

24. An aircraft structure according to any one of claims 14 to 23 wherein the structural weakness is a crack.

25. A method according to any one of claims 1 to 13 substantially as hereinbefore described by reference to any one of the Examples.

26. An aircraft structure according to any one of claims 14 to 24 substantially as 15 hereinbefore described by reference to any one of the Examples or Figures. 44