Final Report Summary - ADMACOM (Advanced manufacturing routes for metal/Composite components for Aerospace)
Executive Summary:
The aim of ADMACOM (Advanced manufacturing routes for metal/Composite components for aerospace) was to develop innovative manufacturing technologies based on advanced design of interfaces and of joining materials for aerospace components.
Several innovative joining technologies and materials for Ceramic Matrix Composites (CMC) and bulk ceramics to metals for aerospace components have been developed within ADMACOM.
The objective of ADMACOM is to have industrials designers, material scientists and engineers to work together to find the best novel, reliable and affordable manufacturing technology for joining dissimilar materials.
The ADMACOM activity focussed on novel manufacturing technologies and joining materials, coupled with surface engineering to maximize interface strength for the joints, and to model/design physical-chemical reactions at the interface; surface engineering means also selective oxidation of the composite surface, laser structuring and/or mechanical machining of the composite/metal surface. The main investigated joining techniques were adhesive bonding, active soldering, « in situ» composite processing, joining with SPS by direct bonding or with interlayer.
Thermodynamic prediction of high temperature interactions and interface formation in dissimilar materials, advanced wettability studies supported by computer simulations has been used to select joining materials.
ADMACOM gathers three industrial partners, academia and research institutions with proven world class expertise in production and joining of CMC. In particular some CMC have been produced by two industrial partners of this project (MTA and AIRBUS Group). The industrial partner Nanoforce exploited the innovative manufacturing technologies.
The main achievements are listed hereafter:
- AGI successfully manufactured some representative demonstrators for satellite applications ( SiC/Invar Tooling Hole Boss, Si3N4/Si3N4 and Si3N4/Invar, laser structured and joined with an epoxy adhesive)
- Laser pre-treatment: laser irradiated SiC surface zones were analysed and the studies led to understand the laser structuring behavior and the phenomena of observed microstructural features.
- A joining procedure was optimized to obtain Cf/SiC - Cf/SiC and SiCf/SiC - SiCf/SiC demonstrators for MT Aerospace joined with Mo-Si composite through a new pressureless process.
- Cf/SiC composites have been successfully joined to Ti6Al4V alloy with Cusil/ABA using a pressureless brazing process.
- Nanoforce sucessfully joined SiCf/SiC to SiCf/SiC and Cf/SiC to Cf/SiC composites with Ti3SiC2 ternary carbide for MT Aerospace demonstrators using optimised Spark Plasma Sintering process.
- Fraunhofer IFAM used Ta and Co foils instead of the Ta/Co powder mixture and obtained sound joints by using pressure-assisted Spark Plasma Sintering Process.
- CNR designed an ad hoc transient liquid phase bonding procedures through Al-Ti interlayers to join SiC.
- Recommended practice draft for determining the torsional shear stress of joined dissimilar materials was submitted to ESIS (Empa, POLITO).
Project Context and Objectives:
A wide number of new composite materials with exceptional properties, such as high thermo-mechanical and thermo-chemical resistance and low density compared to traditional materials (i.e. metals and polymers), are available in the EU market. Their use is still limited to very few applications because of the problem of joining them reliably.
Just to cite a few possibilities: fibre reinforced Ceramic Matrix Composites (CMC, e.g. SiC/SiC, C/SiC, C/C or Al2O3/mullite) could be used right now to replace existing aerospace components. This will contribute to an increasing demand for novel components with improved properties, such as light weight, for reducing fuel consumption and CO2 emission. To enable this, the joining of CMC to metals needs to be easier, more rapid, and possibly permit inspection.
ADMACOM (Advanced manufacturing routes for metal/Composite components for aerospace) developed several innovative technologies and materials to join Ceramic Matrix Composites (CMC) and bulk ceramics to metals for aerospace components.
In ADMACOM, industrials designers, material scientists and engineers worked together to find the best novel, reliable and affordable manufacturing technology for joining dissimilar materials.
The ADMACOM activity focused on novel manufacturing technologies and joining materials, coupled with surface engineering to maximize interface strength for the joints, and to model/design physical-chemical reactions at the interface; surface engineering means also selective oxidation of the composite surface, laser structuring and/or mechanical machining of the composite/metal surface.
Thermodynamic prediction of high temperature interactions and interface formation in dissimilar materials, advanced wettability studies supported by computer simulations were used to select joining materials.
A suitable joint design was developed through an extensive work on the design and model of the joined components, taking into account the requirements and specifications defined by the industrial partners (AIRBUS Group and MTA). The assembly of demonstrators and an accurate cost/benefit analysis were carried out in order to exploit the project results.
The experimental activity of the last 18 months of the project focused on the validation of the model of interface by the manufacturing of successful joints.
The surface engineering was optimized both on the metallic and the ceramic surfaces to be joined: for SiC and Si3N4 materials, the laser nano-structuring technique - to improve the joint performance - was extensively studied. The effect of laser irradiation on the surface morphology and chemical composition of SiC and Si3N4 and its influence on ceramic/joining material interface were investigated. Concerning CMCs, a surface engineering based on the Selective Thermal Removal (STR) of SiC fibers from SiCf/SiC composites was used to obtain a brush-like surface.
The following joining materials were proposed, discussed within the consortium and tested during the period M18-M36:
- adhesive HYSOL EA9321, active soldering using Sn-Ag-Ti solders (for Airbus Group Innovations).
- In-situ Mo-Si composite joints; Ta-based brazing alloys (Ta-Ni and Ta-Co); Ti, Ti6A4lV or Ti3SiC2 foils or slurry with Spark Plasma Sintering (SPS) (for MTAerospace).
The main investigated joining techniques during the period M19-M36 were:
adhesive bonding, active soldering, « in situ» composite processing, joining with SPS by direct bonding or with interlayer.
All these joining materials and techniques were used to prepare joined samples for testing. According to industrial specifications, experimental work focuses on the manufacture of joints on flat components.
In order to define the most promising novel joining materials and techniques and to optimize the joining conditions a detailed microstructural characterization of all joined materials and interfaces was carried out. Concurrently, a mechanical validation of each joined components was supported by a careful selection of available mechanical tests as well as environmental tests (i.e. cryogenic cycles) to reproduce in service conditions. Strong collaboration within the partners led to substantial work on the standardization of selected mechanical tests.
Concerning the activities carried out to maximize ADMACOM impact on EU economy, a exploitation plan for the results, as well as a dissemination plan were defined and implemented by all the ADMACOM partners.
The main achievements are listed hereafter:
- AGI successfully manufactured the following representative demonstrators for satellite applications:
• SiC/Invar Tooling Hole Boss (laser target fixation, design on basis of NIRSpec aerospace experiences), joined by active soldering at Fraunhofer IFAM, for representative testing at Airbus Defence and Space facilities;
• Si3N4/Si3N4 and Si3N4/Invar, laser structured and joined with epoxy Hysol adhesive at AGI, mechanically tested at POLITO at room temperature and after environmental cycling (5 cycles, 50K). The laser pre-treatment of Si3N4 and Invar
- Laser pre-treatment: laser irradiated SiC surface zones were analysed (POLITO, AGI, Nanoforce) and the studies led to understanding the laser structuring behavior and the phenomena of observed microstructural features, such as graphite layer formation.
- A joining procedure was optimized to obtain Cf/SiC - Cf/SiC and SiCf/SiC - SiCf/SiC demonstrators for MT Aerospace joined with Mo-Si composite through a new pressureless process.
- Cf/SiC composites have been successfully joined to Ti6Al4V alloy with Cusil/ABA using a pressureless brazing process.
- Nanoforce sucessfully joined SiCf/SiC to SiCf/SiC and Cf/SiC to Cf/SiC composites with Ti3SiC2 ternary carbide for MT Aerospace demonstrators using optimised Spark Plasma Sintering process.
- Fraunhofer IFAM used Ta and Co foils instead of the Ta/Co powder mixture and obtained sound joints using pressure-assisted Spark Plasma Sintering Process.
- CNR designed an ad hoc transient liquid phase bonding procedures through Al-Ti interlayers to join SiC.
- Recommended practice draft for determining the torsional shear stress of joined dissimilar materials was submitted to ESIS (Empa, POLITO).
Project Results:
WP1 Definition of specifications and requirements for joined components, manufacture of composites
The objectives of this WP are to implement the activity according to industrial requirements on the components to be joined and the manufacturing and supplying the materials to be joined (composites, alloys) by the industrial partners in a quantity agreed as necessary for the completion of the project.
The industrial partner AGI provided the specification and requirements for joined components in space applications (satellites) according to industrial requirements. Furthermore, the representative ceramic / metal demonstrators as well as corresponding bonding partner materials and joining techniques were selected, defined, specified and designed:
• SiC / Invar Tooling Hole Boss (laser target fixation, design on basis of NIRSpec aerospace experiences), joined by active soldering, for representative mechanical validation
• Si3N4/Si3N4 and Si3N4/Invar, joined with epoxy Hysol adhesive, for SLO mechanical testing
The industrial partners provided the amount of materials to be supplied to the other partners; material type and surface quality as well as amount of material samples to be provided was continuously discussed with project partners.
WP2 Design and modelling of joined components
The objectives of WP2 are: to design and model the joined component(s) as agreed in WP1, including demonstrator(s); cost/benefit analysis and definition of designers/engineers cooperation best practices, i.e. a close collaboration between designers and engineers in order to guarantee the experimental feasibility of the designed components
The experimental work was supported by combined thermal-mechanical simulations of bonds with structured interfaces (FRAUNHOFER Bremen). Stress evaluation was done using Mises-stresses for the solder and the maximum principal stress for the ceramic (supposing mode-1 cracking failure type for the ceramic). Stress relaxation was only allowed in the solder by means of plastic yield. The simulations showed that the stress situation near the solder-ceramic interface is dominated by the residual thermal stresses. Structuring the surface of the ceramic leads to increasing of tensile as well as compressive stresses in the ceramic. Maximum stresses in the ceramic were found near the bottom of the holes. Tensile and compressive peak stresses were partly in close vicinity. Maximum tensile stresses in the ceramic seem to be caused by the restrained shrinkage of the solder deep in the hole. In comparison the maximum stresses are caused by triangular stress superposition between neighbored solder-filled holes. Further simulations showed that a reduction of the structuring depth can lead to a significant reduction of the residual thermal stresses in both the solder as the ceramic – while still offering a better interlocking of the bonding partners.
Laser structuring of SiC surface: deeper studies of laser influenced zones were done (POLITO, AGI, Nanoforce). More detailed analysis of the laser irradiated SiC surface zones were performed in order to better understand the laser structuring behavior and the phenomena of observed microstructural features. The understanding is of high importance because of the joint low mechanical performance. Based on literature research and experimental analysis results, a hypothesis for the mechanism of the graphite layer formation at the SiO2/SiC interface was proposed. Details were summarized in the common paper “Effect of laser pulsed irradiation on the SiC surface” for the ADMACOM special issue on the Int. Jour. of App. Ceram. Tech. (submitted).
Design of demonstrator on basis of NIRSpec aerospace experiences (AGI): The demonstrator design and demonstrator set-up for testing and validation of the developed joining technology on technological but representative sample level (in WP7) as well as the fabrication of the INVAR Tooling Hole Boss parts and SiC parts according to the technical drawings were completed.
Cost/benefit analysis
• Add-on activities on cost/benefit analysis (MTA, AGI): some add-on activities were established, knowing more precisely the process steps of the main relevant joint methods. Since all methods were running in parallel at several partners and institutes, a final cost-benefit analysis and evaluation could not been performed after M12. Therefore, MTA and AGI performed analysis best after the final selection of joining processes with most relevant success and definition of pre-prototypes. The main cost/benefit analysis results are summarized in the additional requested Deliverable D2.4.
• The AGI cost / benefit analysis was done in context with reproducibility, bonding performance, risk mitigation and savings in effort. The laser structuring for surface pre-treatment of e.g. Si3N4 brings benefit in terms of
o Improved bonding performance which was demonstrated (for both similar & dissimilar at RT & cryo-cycling: strength, failure mode, homogeneity)
o Factor human being: the process brings
▪ high potential to substitute manual process
▪ highly reliable in one shot
▪ high reproducibility
▪ high degree of automation
▪ high quality
▪ reduced risks
o Substitution of critical chemicals for pre-treatment (REACH)
o Substitution of lapping as a surface pre-treatment process 75% savings in efforts (and by this cost)
o Surface pre-treatment: use of laser nano-structuring brings to a 99,8% saving in the overall environmental burden of the joining process (see LCIA) → “green” process
• Main benefits of ADMACOM joining techniques for MTA can be summarized as following:
o General:
▪ Weight reduction
▪ Higher flexibility in design layout
▪ Less efforts in machining of parts and joints (screws, pins, bolts..)
▪ Less space needed
▪ Risk reduction
o Active Soldering
▪ - Compared to adhesives no release of humidity during operation in orbit
▪ - Wetting of difficult-to-wet surfaces
▪ - Thermal conductive joints
▪ - Low joining temperature
o Partial Transient Liquid Phase Bonding:
▪ - A (transient) liquid phase assures intimate contact between surfaces
▪ - Joining temperature in principle lower than service temperature (high melting phases formed by reaction and diffusion)
▪ - Reduction of thermal stresses arising upon cooling
• Nanoforce also performed cost-benefit analysis of the joining process developed on site within Admacom project by comparing with the current state of the art (joining processes with Ti foils or MAX phase powder). The newly developed approach consists of joining of CVD-SiC and CMC materials with the pre-sintered Ti3SiC2 MAX phase foils via solid state diffusion bonding using SPS at temperatures as low as 1250°C. It was concluded that the joining process has the potential to both be cheaper (due to lower temperature and time) and produce high quality joints (higher shear strength). The main disadvantage of this technique is the additional processing step of preparing the joining foil. This extra step might be compensated by the potential to remove the surface preparation step
T2.3 Designers/engineers cooperation best practice
• MTA and AGI have defined the demonstrator design, during the reporting period M19-M36 the prototype realization has been developed by MTA, AGI, POLITO, Nanoforce, FhG IFAM in an excellent designers/engineers cooperation best practice way. So, on-going activities were ensured by effective and very close collaboration among designers from industry and materials scientists from institutes (All).
• The effective collaboration among component designers and materials scientists was also focused on mechanical test and standardization. One of the objectives was to “translate” data available now at AGI into pure shear data by starting a round robin test based on torsion test, asymmetric four point test and AGI single lap test. The final goal was to provide AGI the possibility of “translating” apparent shear strength measured up to now to pure shear, and to provide designers with suitable results for modelling of new components.
• Several telephone conferences and as well as meetings industry / institutes were hold. Furthermore, bilateral meetings have been organized, also with researches’ visits to industrial premises and to other partner facilities.
WP3 Surface engineering, modelling and physical-chemical design of interfaces
The objective of this WP was to acquire new knowledge of the physico-chemical phenomena required to model and design the composite interface in order to maximize its strength and the overall performances of both adhesive and high temperature joints.
Surface engineering to obtain the brush-like joints (POLITO)
POLITO performed several experiments to reproduce the “brush-like” surfaces of SiC/SiC samples: the composites have been treated at 1450°C for 2h in order to selectively remove part of the fibers at the surface. The goal is to increase the surface area and to maximize the interface strength and the overall performances of joints. The study of the effectiveness of the surface engineering treatment on SiC/SiC has been evaluated by AFM analysis at POLITO; the measurements have been performed both on the surface engineered and on as-received SiC/SiC; in any case, it is very difficult for these analysis to be successful because the maximum z range of the AFM is about 5 µm and the SiC/SiC show high porosity and roughness on the cross section.
Evaluation of effectiveness of surface engineering by 3D profilometry and high-temperature wetting tests (CNR)
In order to assess the effectiveness of the surface engineering performed at POLITO on SiC/SiC composites to obtain the “brush-like” joints, confocal profilometry was implemented for the 3D acquisition of the surfaces of SiC/SiC composites and to measure their characteristic features. The average depth of the fibers removal was measured to be 2.4 µm with an increase of the total surface area ranging from 89% to 129%.
Wetting tests were conducted using the brazing alloy AgCuTi (Ag: 57.7 Cu: 36.8 Ti: 5.5 at%) at 850°C. For the sake of comparison, the tests were conducted with several materials: ultra-pure SiC, SiC/SiC as cut, SiC/SiC polished and SiC/SiC thermally treated. For the materials not thermally treated, contact angles range from 10° to 20° with similar wetting kinetics. For the SiC/SiC composites thermally treated the final contact angles were measured to be slightly higher (28°) with kinetics relatively slower. This was attributed to the fact that the surfaces after the thermal treatment are in fact oxidized and in principle not wettable. The Ti in the alloy reacted with the oxide layer allowing the wetting. The micrographs taken on cross-sectioned samples showed that the alloy perfectly adhered to the SiC/SiC surfaces filling all the holes and gaps introduced by the thermal treatment. In this way, the adhesion between the composite and the brazing alloy increased coherently to the increase of surface area measured by profilometry. This constitutes the most interesting and important result of this activity demonstrating the promising effect of surface engineering to produce brush-like features and their effectiveness to increase the joint quality.
Wetting experiments for Co-Ta/SiC and Ni-Ta/SiC systems (CNR)
The wetting and interfacial behaviour of liquid Ta alloys (Co-Ta 12 and 50 at%, and Ni-Ta 14 and 38 at%) in contact with SiC surfaces (CVD-SiC, C/SiC, SiCf/SiC) has been determined by sessile drop experiments. The Ta alloys were selected by Fraunhofer as potential brazing media for SiC based composites. The influence of process parameters (atmosphere, alloy preparation, surface roughness) has been tested and discussed. For all the compositions and experimental conditions, good wetting was observed with fast spreading. The interfacial behaviour is determined by the competition between the typical interfacial phenomena of the pure elements; specifically, either the dissolution of the ceramic phase by Ni (or Co) or the formation of a new interfacial layer, TaC, by reaction between Ta and SiC. The competition and the prevalence of one or the other of the two phenomena depend on the relative amount of the single element in the alloy. The main results are: - all the systems showed very good wetting (i.e. θ < < 90°); - whatever the composition, interfacial Ta carbide was found; for low contents of Ta, dissolution of the SiC substrate was found, similar to what is observed for the well known systems Ni-Si-C and Co-Si-C; - a high content of Ta led to the suppression of dissolution of the ceramic phase and to the formation of a continuous layer of TaC at the metal/ceramic interface; this layer stops diffusion and enhances wetting making the use of this alloy for brazing feasible; - using high vacuum rather than a protective Ar/5% H2 atmosphere resulted in a faster wetting as a consequence of an enhanced cleaning effect of the surface; - the comparison between pre-melted alloys and pellets of compacted powders (that one would likely use in a joining process) showed no relevant differences in both wetting kinetics and microstructure.
Basic wetting data of Ta alloys on SiC were still not present in the scientific literature, therefore their importance and usefulness go beyond the framework of the ADMACOM project. These results were used by Fraunhofer to choose the most proper brazing alloy compositions and process parameters necessary to obtain reliable joints (see WP4).
Thermodynamic modeling through CALPHAD calculations (CNR)
In order to obtain thorough understanding of the interaction phenomena occurring during the contact of liquid Co-Ta and Ni-Ta alloys on SiC materials observed during wetting tests (see D3.1 and D3.2) the thermodynamic evaluation of SiC/Ta-Ni and SiC/Ta-Co interfaces was performed through CALPHAD calculations. A deep study of the binary and ternary phase diagrams, between the elements of the ceramic substrate (C, Si) and of the alloys (Co, Ni, Ta), has been performed in order to predict and evaluate the different potential intermediate ternary phases formed at the interface zone. A database input file, containing the thermodynamic parameters of the pure elements and of the intermediate compounds, fundamental to obtain the ternary and the quaternary systems (C-Ni-Si-Ta, C-Co-Si-Ta), has been produced. Thanks to the CALPHAD modeling, it was possible to better interpret the experimental results regarding the high temperature reactivity between Co-Ta alloys (12 and 50 at% Ta) and silicon carbide. Even if the studied systems are dynamic, a comparison with the equilibrium phase diagrams is a useful way to understand the interactions occurring during the brazing processes such as high temperature isothermal interactions and phase transformations during the subsequent cooling process.
High temperature wetting test in SPS (Nanoforce)
A new advanced technique to assess the wetting behaviour of metal/ceramic systems at very high temperatures was developed in Nanoforce Technology ltd. The advantage of this technique is a possibility to measure in-situ a contact angle between alloys with high melting points and ceramic substrates using Spark Plasma Sintering (SPS) up to 2300°C. The wetting results of Ta-based alloys on the CVD SiC were compared with the ones obtained by CNR using their well-controlled wetting apparatus. A good match in terms of the wetting and spreading kinetics, contact angles, reaction and solubility of the SiC etc., was obtained, confirming a good potential of the new technique. Using this newly developed wetting technique, the influence of the heating rate on the wetting behaviour was investigated using Ti6Al4V alloy on the polished CVD β-SiC surface. The results showed a significantly better wetting kinetics and spreading behaviour of the alloy when the faster (500°C/min) heating rate was used rather than the slower one (100°C/min). In addition, the slower heating slightly lowered melting temperature of the alloy because there is more time for the phases with a low melting point to form due to the reaction with SiC. The results showed that the wetting of an alloy can be modified by the heating rate and it might be necessary to apply a higher heating rate in order to obtain a good bonding while joining.
Design of transient liquid phase bonding procedures: wetting of SiC by AlTi alloys (CNR)
In order to determine the feasibility of using Al-Ti interlayers for the brazing of SiC based ceramics through the transient liquid phase bonding technique, the wetting and interfacial phenomena related to the contact between SiC and Al-Ti alloys were studied. Three compositions were selected: Al 3at% Ti, Al-25at%Ti and Al-46at%Ti and tested by the sessile drop method at 1500°C on pure SiC substrates. All the systems exhibited very low contact angles (<< 90°C) that in principle would assure an intimate contact between the materials to be joined. From the interfacial chemistry point of view, the best results were obtained with the Al-25 at% Ti alloy: the formation of a continuous interlayer of Ti3Si1 xAlxC2 at the metal-ceramic interface was observed and this phase perfectly adhered to the SiC substrate. These results proved the possibility of using Al-Ti interlayers for the joining of SiC-based materials: the Al-Ti liquids assure the contact between the adjoining materials and the (transient) liquid phase evolves by reaction and diffusion, solidifies at the testing temperature and forms a Ti3Si1 xAlxC2 MAX phase at the interface that has a re-melting temperature higher than the process itself. The outcome of this activity is the manufacturing, at the laboratory scale, of sound SiC-SiC joints (see WP4).
Investigations of the joined interfaces using HR-SEM equipped with FIB (Nanoforce)
Nanoforce investigated the interface between Boostec SiC and the Sn-Ag-Ti-RE solder (joined and supplied by Fraunhofer) using High Resolution Scanning Electron Microscopy (HR-SEM) equipped with Focused Ion Beam (FIB) and Transmission Electron Microscopy (TEM). Thin lamellas were prepared using FIB and the area of interest was subsequently observed using TEM. The analysis revealed that there was a sharp interface between the SiC and the solder, suggesting good, apparently strong bonding. However, some small well-separated pores (< 1μm) were observed all along the interface. This may suggest that the solder did not fill small surface pores in some areas while it strongly bonded to the matrix in others. On the other hand, no obvious reaction layer was found by both SEM and TEM analysis of the interface.
WP4 Novel joining materials and techniques
T4.1 Novel joining materials and surfaces (Task Leader Fraunhofer)
This task was focused on :
− Experimental application of novel joining materials and techniques selected in WP 1
− Preparation of joint samples for testing
− Optimization of joints
− Production of demonstrator parts
The experimental activity can be summarized as follows:
− Manufacturing of joints using modified surfaces, according to results of WP3; the modification of the surfaces has been carried out by mechanical machining, i. e 1) laser structuring of selected ceramics and metal alloys by Airbus Group Innovations ; the modified materials are BOOSTEC®SiC and Invar alloy 2) selective fibre removal of the composite matrix (“brush joint”) performed by POLITO and optimization of the process parameters
− Surface modification by chemical reaction (CNR);
− Study of novel pressure-less, flexible and low cost joining approaches by three different joining materials for CMC composites (calcia-alumina glass-ceramic, metallic brazes, in-situ formed Mo-Si based composite) (POLITO, EMPA).
− POLITO performed a study on the HYSOL adhesive (epoxy paste) to characterize its behaviour and the curing parameters; the joining process was optimized and several joints have been performed using BOOSTEC®SiC and Invar both laser structured and non laser structured
− Fraunhofer investigated two new joining technologies for SiC-based materials:
o High temperature brazing with Ta-Ni and Ta-Co alloys
o Active soldering using Sn-Ag-Ti-based solders
− CNR continued optimization of a Partial Transient Liquid Phase Bonding by Al-Ti interlayers
− For mechanical testing purposes EMPA produced SiC-SiC bending and torsion test samples by active brazing using Incusil ABA
Significant results:
• Surfaces of SiC/SiC to be joined with different joining materials have been manufactured; the surface modification method has been developed by POLITO to obtain “brush-like” for SiC/SiC composites
• Joining of surface engineered SiCf/SiC composites by Ag-Cu-Ti alloy (POLITO, CNR); Micro-sized brush-like structure increased the adhesion area between the composite and the brazing alloy and this enhances the mechanical strength of the joint due to the mechanical interlocking.
• A laser surface structuring technology for nano-scaled modification of ceramic materials such as ceramics and CMCs was investigated by Airbus Group. Laser parameter studies on SiC and Si3N4 were conducted by Airbus Group to investigate influence on microstructure, chemical composition & mechanical joint performance; Mechanical test results do not confirm the effectiveness of laser structuring on SiC surfaces (“weak intermediate layer“). Mechanical test results DO confirm the effectiveness of laser structuring on Si3N4 surfaces.
• The joint shear strength for laser-structured BOOSTEC®SiC - BOOSTEC®SiC samples joined by the adhesive Hysol® EA 9321 was measured by single lap-shear test. Promising shear strengths of up to 43 MPa were obtained.
• HYSOL joined BOOSTEC®SiC to Invar, both laser structured and not laser structured have been manufactured and compared
• The joining process by HYSOL and the joining material itself have been deeply investigated in terms of reduction of porosity during curing and ways to improve the thermos-mechanical resistance of the joints
• For the joining of Keraman® SiC/SiC and C/SiC samples POLITO investigated two joining approaches: the joining materials are CA glass-ceramic and a Mo-Si based filler, both used successfully.
• A literature review (limited to the last five years) of the most promising joining processes applicable to selected materials (SiC, CfSiC, SiSiC, Invar and Ti6Al4V) has been prepared by EMPA and provided to all consortium;
• For mechanical testing purposes EMPA produced SiC-SiC bending and torsion test samples by active brazing using Incusil ABA.
• CNR worked on the transient liquid phase bonding technique by means of Al-Ti interlayers. These processes are meant to be effective for C/SiC and SiC/SiC composites for high temperature applications. Joined samples were produced using Al-Ti alloys assemblies (interlayers and bulk pieces flowing by capillarity) showing the potential of the method and the formation of new phases with melting point higher than the process temperature.
• A high temperature brazing approach with Ta-Ni and Ta-Co alloys was tested by Fraunhofer to join SiC/SiC composites for high temperature applications.
• Concerning the high temperature brazing with Ta-Ni and Ta-Co alloys, the main results are:
o Promising wetting behavior of selected Ta-Ni and Ta-Co alloys on SiC-surfaces (CNR)
o Joining using powder mixtures and SPS technology is feasible
o Formation of reaction zones for both Ta-Ni and Ta-Co alloy system
o Control of the SiC decomposition reactions (reaction zones) within the Ta-Co/SiC system by choosing a layered starting stack set-up using metal foils (SiC/Ta/Co/Ta/SiC)
o The joining parameters for monolithic SiC were successfully transferred to SiC/SiC composite joints (Fraunhofer).
• Active soldering for the joining of BOOSTEC®SiC - BOOSTEC®SiC for low temperature and cryo applications has been conducted by Fraunhofer.
• Concerning the active soldering using Sn-Ag-Ti-based solders, the main results are:
o Active soldering is considered to be an attractive joining process for SiC components for low temperature applications
o Active soldering of monolithic SiC and SiC/Invar joints for low temperature applications was successfully conducted. Parameter studies for both material combinations were performed. Joints with apparent shear strength of about 30 MPa were obtained for SiC/SiC and SiC/Invar joints (Fraunhofer). The influence of the sourrounding atmosphere during joining was investigated. Air, Argon and Nitrogen were tested. The best results were achieved for air atmosphere. Demonstrator production for WP7 was conducted.
o The active soldering was also successfully applied to join SiC to TiAl6V4.
T4.2 SPS (Task Leader NanoForce)
New joining materials based on powder mixtures are being prepared and used in combination with fast pressure assisted technologies (Spark Plasma Sintering, (Fraunhofer, NanoForce) .
Significant results:
• The SPS at Nanoforce has been used to join both monolithic CVD silicon carbide (as a representative of the CVD SiC coating on CMC materials) as well as the SiC-SiC and the C-SiC composites. The max phase - ternary carbide Ti3SiC2 - was used as the joining material.
• Four point bending strength results of joined monolithic CVD SiC showed that the strength of the joint was significantly higher when a pre-sintered Ti3SiC2 foil rather than a slurry made of Ti3SiC2 powder was used. The foil was pre-sintered using SPS at 1300°C and then polished down to around 50-100 μm. The samples were joined at relatively low temperatures 1300-1350°C and using an external pressure of 60 MPa in vaccuum.
• The bending strength of the monolithic CVD SiC joined with Ti3SiC2 was independent on the final thickness of the joint. Since the Ti3SiC2 foil did not shrink or react with the SiC, the final thickness of the joint was predetermined by the initial thickness of the foil.
• The Ti3SiC2 foil showed some ductility during the joining process as it conformed to the rough surface of the CMC materials. It also penetrated into the cracks in the CVD coating of CMC materials (formed during the processing of the composites); effectively healing the CMC composites.
• Joining of the monolithic CVD SiC with Ti6Al4V foil was also performed using SPS at Nanoforce. The foil with different thicknesses was used as the joining filler and joining process was carried out at temperature of 1650°C and an external pressure of 60MPa. The foil shrunk during the joining and showed a significant infiltration into the SiC material, confirming a high reactivity of this metal alloy with SiC surface.
• Bending strength of the monolithic CVD SiC joined with Ti6Al4V exhibited a similar value as non-joined monolithic reference material, proving a high strength of the joint.
• Nanoforce performed additional investigation of the cross section of the joints. Although only coated CMC samples have priority for Admacom demonstators, two uncoated CMCs were used as reference samples for the sake of comparison. Therefore, two uncoated Cf/SiC were joined together with a Ti3SiC2 pre-sintered foil using SPS at 1300°C and 50 MPa for 5 minutes.
• The SEM analysis revealed the formation of SiC reaction layer in the joining filler when the Ti3SiC2 interlayer was in contact with carbon fibres.
• The infiltration of the joining filler into the inter-spaces of carbon fibres was limited because of the limited number of free interspaces of Cf near the interface and due to the SiC reaction layer that acted as a diffusion barrier.
• Unlike for the coated CMCs (diffusion bonding only), this time the joints were obtained by the combination of solid- state reaction and diffusion bonding when the Ti3SiC2 pre-sintered foil was in contact with Cf and SiC, respectively. However, the mechanical tests showed that the apparent shear strength of the joined components was lower when compared to the coated CMC ones.
WP5 Micro-structural characterization of joints
The micro-structural characterization of all selected joined materials and interfaces has been done by using different techniques. The detailed investigation of materials allowed to define the most promising novel joining materials and techniques and to optimize the joining conditions. The activity within this task aimed at a complete microstructural characterization of all joined materials and interfaces developed in WP2, WP3 and WP4. It regarded also the investigation of the fracture surfaces after mechanical characterization according to WP6. To accomplish this task, the most suitable characterization techniques have been identified, shared among partners and extensively used. In summary, the following results were achieved in WP5:
✓ A full list of techniques for the micro/nanostructural characterization of joined components has been identified and shared among Partners.
✓ All the partners involved in WP5, CNR, POLITO, Fraunhofer and NanoForce performed standard and conventional characterization methods (e.g. optical and electron microscopy, EDS, XRD, etc.) on their own equipments routinely.
✓ Advanced techniques (e.g. nano tomography, TEM) were performed on selected samples when necessary and made available to Partners.
✓ Samples were exchanged among partners in order to perform tests and microstructural analyses using the most appropriate techniques.
✓ The results obtained within WP 5 were tremendously important for the entire ADMACOM project and essential to evaluate results and processes within other work packages. Without the analyzing techniques used inside this work package it would have been impossible to assess the quality of produced joints and interfaces.
✓ Measurement results like images, chemical compositions or phase assignments contribute to new knowledge when disclosed in publications.
✓ A few techniques, not planned in Annex I of the DoW, have been adopted (e.g. HR-SEM equipped with focused ion beam, non-contact profilometry) in order to improve the micro/nanocharacterization of joined interfaces. These methods and the related results constitute a further added value within the whole ADMACOM project.
In the following some examples of the most interesting and outstanding results obtained are provided:
Non-contact 3D optical profilometer (CNR): a multimodal non-contact 3D optical profilometer combining the confocal, interferometric and focus variation techniques has been acquired by CNR in the mainframe of the ADMACOM Project. Compared to other techniques (e.g. AFM) confocal profilometry allows the acquisition of relatively large surfaces at an adequate z scale in order to observe and measure the characteristic features of a surface. Thus, the peculiarities offered by the instrument have been employed, in particular, to analyze the surface characteristics of SiC/SiC composites with reference to the selective fiber removal to obtain the “brush-like” surfaces (surface engineering in WP3). The average depth of SiC fibers removal was measured by confocal profilometry to be 2.4 µm. For the surface engineered samples, the surface area increased of 89% and 129% for measurements performed on areas with fibers // or ⊥ to the surface respectively demonstrating the effectiveness of the fiber removal.
Microindentation (Nanoforce): in order to confirm that the joining interlayer remained intact and was not damaged during joining, microhardness of the interlayer was measured and compared to the microhardness of the initial Ti3SiC2 foil that was used as a joining filler. The results showed that the hardness in both cases was at the same level, confirming that neither decomposition nor reaction occurred between the interlayer and the base materials (CVD-SiC, CMCs).
Microindentation (CNR): microhardness measurements were performed by CNR on joined samples in the framework of WP6. The information acquired had not only a mechanical meaning but were useful from the microstructural examination point of view. In fact, the chemical nature of the phases observed and identified by SEM-EDS can be confirmed by microhardness values measured on cross-sectioned samples.
MicroCT analysis (Nanoforce, POLITO): this technique was used to investigate the small-scale demonstrator that consisted of the coated CMCs joined with the Ti3SiC2 interlayer. The analysis confirmed that defect-free interface was obtained and the Ti3SiC2 interlayer showed conformal behaviour as it bent and nicely copied the rough surface of the CMCs without breaking or cracking.
SEM-FIB and TEM (Nanoforce): the intact interfaces between the monolithic Boostec SiC and solders were investigated by SEM-FIB without being influence by cutting, polishing, etc. Thin lamellas prepared by FIB were then investigated by TEM. This helped to reveal that there was no additional reaction layer between the SiC and the solder. It also revealed the presence of a number of small well-separated pores along the interface that were not found by conventional SEM analysis.
Cryotest (POLITO): Si3N4/Si3N4 and Si3N4/INVAR joints (joining material: HYSOL EA9321) have been tested in order to reproduce the in service conditions and to study the evolution of microstructures during thermal cycles down to cryogenic temepratures. During each thermal cycle, the samples were cooled from room temperature down to 50 K then kept at 50 K for one hour. Finally, the cryocooler was switched off and the samples warmed up to room temperature. After the first thermal cycle, the samples were dismounted for a preliminary optical check. Then, the samples were mounted again and submitted to four consecutive thermal cycles. At the end, they have been dismounted and the joints have been tested mechanically.
WP6 Thermo-mechanical tests
Determination of the coefficient of thermal expansion (CTE)
CNR, Polito and Fraunhofer measured and collected the CTE on selected materials to be joined, on joining materials and obtained joints.
Microhardness measurements of Ti3SiC2
In order to confirm that the Ti3SiC2 joining filler remained intact after joining, Nanoforce performed microindentation to determine the hardness of both the initial pre-sintered Ti3SiC2 foil and the Ti3SiC2 joining interlayer. Neither decomposition of Ti3SiC2 nor reaction between the interlayer and matrix materials occurred during joining.
Influence of the laser treatment on the strength of SiC Boostec 100
Samples of monolithic SiC (type SiC 100, BOOSTEC) were delivered by Airbus Group to Empa to study the influence of the laser treatment on the strength. The chosen method was co-axial Ring on Ring test, based on DIN-EN 1288-5. A fractographic study was performed after testing to identify the fracture origin. The ring on ring measurements confirmed that there is no effect of the laser treatment on the strength of the SiC samples.
Single Lap Offset Internal Round Robin
The round robin test within the ADMACOM partners for single-lap offset test (SLO) was launched to determine the reproducibility of the test method and so to confirm and compare the experimental results with previous ones obtained by the industrial partner, Airbus. 4 partners (EMPA, POLITO, IFAM and Airbus) tested 10 samples each. The samples were manufactured and joined by EMPA. The shearing tool was designed by Airbus. The joining material was the same adhesive selected by Airbus for SiC joints, Hysol 9321.The material for the RR was structural steel S235.
The main conclusions of the Round Robin are:
• The Round Robin of SLO was successfully performed by 4 partners of the Admacom project without any outlier.
• All the samples of RR have the same type of failure mechanism, adhesive fracture at hysol/steel interface.
• The inter-laboratory SLO measurements were reliable within the intrinsic deviation of this method, ±3 MPa.
• SLO was chosen as comparative tool.
Apparent shear strength of SiC-SiC and SiC-Invar joints produced by active soldering
Fraunhofer worked on mechanical testing of SLO samples of the following material combinations produced by active soldering:
- SiC-SiC joints, in this combination it was studied, the influence of the thickness fillers, joining temperature and atmosphere on the mechanical properties. Best results with average values of 30.4 ± 6 MPa were achieved for samples joined under air atmosphere at 280°C-300°C and 50 µm filler thickness.
- SiC-Invar joints, in this combination it was studied, the influence of the joining thickness and joining temperature on the mechanical properties. Best results with average values 28 ± 4 MPa for samples joined under air atmosphere at 280°C and 25 µm joining thickness.
Mechanical characterization of SiC joined with Mo-Si based materials
SiC-SiC joints have been manufactured at POLITO using Mo-Si based material as joining material; details on the joining process are reported in WP4. Two different tests at RT have been carried out:
• Torsion test
• SLO test
4 joints have been tested by torsion test (geometry THG-4) and the average shear strength is 27 MPa (s.d. ±4 MPa). These results show very low values of mechanical strength for this kind of joint; it could be explained by possible misalignment of the tested samples, due to significant difficulties in their preparation. Concerning SLO tests, only two samples have been tested and the preliminary results show very high values of apparent shear strength.
Mechanical characterization of CMCs (C/SiC and SiC/SiC) joined with MoSi2-Si
POLITO tested by single lap test CMC joined with MoSi2-Si, the average value of the apparent shear value obtained was ~7 MPa, and the main problem during the measurements was the delamination of the CMC.
Mechanical characterization of C/SiC to Ti64 joints
C/SiC to Ti64 joints have been manufactured and characterized at POLITO. For the mechanical characterization of C/SiC to Ti64 joints manufactured using the brazing alloy TiB-590 the single-lap tests was used, but for all the tested samples the composites delamination occurred. Other tests were carried out by SLO: the average value of the apparent shear test is 43 MPa; also in this case, a critical issue is the delamination of the composite during the test.
SLO tests on joined laser structured SiC and Si3N4
Polito performed SLO tests on joined laser structured SiC and Si3N4 samples provided by AGI. All the joints were manufactured using HYSOL EA9321. The tested samples have been manufactured using different laser structuring parameters (LS), different adhesive thicknesses and different lapping conditions on facing surfaces before laser structuring process. The apparent shear strength of LS SiC is lower than that measured on lapped SiC. The apparent shear strength in the Si3N4 is higher for the laser structured joined samples if compared with the lapped reference sample without laser structuring. The mechanical test results confirm the effectiveness of laser structuring on the Si3N4 surface. Details on this activity are reported in D6.2.
Mechanical characterization of SiC/SiC samples before and after surface engineering treatment ( 3 point bending strength)
SiC/SiC samples have been subjected to 3-point bending tests at Polito before and after surface engineering treatment, i.e a selective oxidation process that involves a treatment at 1450°C for 2 hours in Ar atmosphere. The average value of the flexural strength of as received sample is 576 MPa, while heat treated samples showed a decrease of the mechanical strength to 387 MPa.
Torsional shear stress test at cryogenic conditions at elevated temperature on Hysol joined samples.
Empa adapted a torsional shear stress setup to perform measurements from cryogenic temperatures, -190°C up to 1200°C. Two adhesives used in aerospace industry were characterized between -190°C to 150°C. The torsional shear strength of the glue drops by 40% at 50°C with respect to its RT value. Cryogenic temperatures do not affect negatively the torsion shear strength.
Development of a shear stress fatigue test.
Empa developed a fatigue shear stress method based on the torsional shear stress test. The setup produces a peripherally uniform stress distribution on the joining area, the load can be adjusted and the samples can be cycled over 20,000 times per day. Fatigue tests on samples joined with hysol with a load of 90% have been performed, on average the samples failed after 60 cycles.
Single-lap offset tests on SiC joined by brazing alloy in collaboration with CNR
POLITO tested by SLO, SiC joined with TiAl by CNR – IENI( joining area 7 or 8 mm x 5 mm). The joining process consists of partial transient liquid phase bonding by Al-Ti interlayers (details of the bonding process is described in D4.2).
The average value of the apparent shear strength is more than 200 MPa; these high values of mechanical strength indicate good joints and demonstrate the usefulness of this joining method.
Simulations: effect of the meniscus geometry on the shear strength
Experimental shear tests were supported by macro-simulations with the aim to evaluate the influence of residual thermally induced stresses on the stress distributions in the chosen torsional shear test-setup. The basic model system consisted of 2 steel parts soldered with S-Bond 220-1. Soldering temperature was 250°C, solder thickness 0.2 mm, contact radius r = 4 mm, maximum torsional moment 250 Nmm.
The simulation showed that the superposition of the thermal and mechanical loads leads to reduced shear stress maxima in the solder in comparison to the mechanical load case alone. That means that the influence of the thermal history may lead to an underestimating of the real stresses resp. to an overestimation of the shear strength of the solder. On the other hand, the thermal stresses are highest near to the solder free surface which means that relaxation and micro-damaging can occur in this area before the mechanical test, leading to underestimating of the shear strength of the solder.
Simulations: effect of the geometry close to the joining area
The simulations of residual stresses originating from the soldering process and the mechanical stresses during shear test were extended in order to evaluate the influence of the specimen shape close to the soldering layer. Specifically, the cross-section geometry of the steel specimens was changed from straight (truncated cone) to circular with a radius of r = 4 mm. All other aspects of the model remained the same: it consisted of 2 steel parts soldered with S-Bond 220-1. For the simulations straight solder meniscus geometry was assumed.
As expected, the geometry variation affected the shear stress in the solder only in the vicinity of the triple contact lines steel-solder-atmosphere. A smooth geometrical transition of solder and specimen is advantageous, though the overall effect is comparatively small. In comparison to the simulations done before (variation of solder meniscus and contact angle) it can be concluded that for the chosen configuration, meniscus shape and contact angle have a higher influence on the stress situation in the solder than the specimen shape.
SLO and Torsional shear strength measurements with different roughness values
With the purpose to check the sensitivity of the different shear strength testing methods, Empa prepared samples of steel S235 with different surface roughness, the specimens were ground with different SiC papers, and the roughness was measured. The results show a clear increase of the shear strength with the roughness of the samples, this tendency is clearer in the measurements performed with the torsional setup than with the SLO setup.
Mechanical joints characterization at elevated temperatures of SiC Incusil ABA brazed.
The behavior of SiC ceramic joints brazed with commercially available Incusil ABA (Ag-32.25Cu-12.5In-1.25Ti in wt.%) was characterized by Empa with respect to the mechanical performance at temperatures up to 550°C using four-point bending and torsional shear tests. In four-point bending tests we interestingly observed a strong retention in bending strength up to 300°C with only a slight 10% decrease. With the torsional shear apparatus it was not possible to characterize the joint accurately below 450°C as the samples always broke in the ceramic part. This is due to a combination of the accumulation of very high localized stresses at contact points where the shear load is transferred and the low KIc of the material, initiating fracture in the ceramic. What is remarkable is the high values of the shear strength obtained, which are higher than the bending strength at the same temperatures, indicating both that a more complex stress system exists during bending strength tests and the adverse effects of the tensile stresses.
Torsional shear strength measurements of Steel S235 glued with Hysol 9321 with different radius
In all the cases the measured shear strength is higher than the one measured in the SLO RR for the same materials, 41±3 MPa reported in the D6.2. The values are especially high in the cases of specimens with an outer diameter of 5 mm but also the deviations of the measurements are higher. For large outer diameter, 14 and 20 mm, the measured shear strength values remain for all the inner diameters around 50 MPa. When the inner diameter approaches the size of the outer diameter, the deviation of the measurements increases, probably due to the difficulty to control the glue thickness, lower thickness can increase the values of the shear strength.
Simulations: Effect of the radius on the residual stresses
Macro-simulations were performed with the aim to evaluate the influence of the specimen shape close to the joining layer and the influence of the radius size on the residual stresses in the chosen torsional shear test-setup. The basic model system consisted of 2 steel parts glued with Hysol 9321, a glue thickness of 0.2 mm, joining area of 5 mm diameter, mesh size of 0.1 mm and a maximum torsional moment of 250 Nmm. Two geometries were compared Empa geometry with chamfered angle of 45, and a geometry with a chamfered radius of r = 4 mm. The size radius dependence simulation was performed with the Empa geometry.
As in the previous simulations, the geometry variation affected the shear stress in the vicinity of the triple contact lines steel-glue-atmosphere. A smooth geometrical transition of glue and specimen is advantageous. The results show that the larger the radius of the specimens, the less shear stresses in the vicinity of the triple contact point are observed.
Comparison of different configurations to measure the joint strength (shear and apparent shear).
Empa and Polito compared different test methods to measure the joint strength (shear and apparent shear) of different combinations of joining materials and substrates. The aim of this activity is to validate the torsional shear stress test and collect information for the torsional shear stress round robin.
Recommended practice for determining the torsional shear stress of joined dissimilar materials.
Empa in collaboration with Polito submitted a first draft of the recommended practice to the European Structural Integrity Society (ESIS) a non-profit organization. The purpose of the document is dissemination of the practice as basis of the planned Round Robin on torsional shear stress test.
WP7 Manufacture of demonstrators, repair, maintenance and non-destructive tests
The aim of this WP is to validate what done in the previous ones, and to start technology transfer to the industrial partners. It is focused on:
(i) construction of the demonstrators with the novel manufacturing technology,
(ii) validation of the structural behavior and ease of implementation in industrial environment. Proof of concept in terms of at least one demonstrator per joining based manufacturing technology, demonstrating here the scalability towards industrial needs and pilot implementations in industrial settings;
Life Cycle Impact Assessment of the demonstrators
In the last period of the project, all involved partners manufactured the final demonstrators and pre-prototypes for both industrial partners, AGI and MTA. This is significant, because effectiveness of all new promising joining techniques developed within WP4 was verified by the construction of the final pre-prototypes and demonstrators. It was confirmed that all selected joining techniques were appropriate for manufacturing the following final demonstrators and pre-prototypes. This is considered as the promising final output of the project, all according to DoW. However, the joining of curved surfaces for MTA demonstrators did not lead to the desirable results and more process optimisation is still required.
• SiCf/SiC to SiCf/SiC: flat to flat surfaces, 25 mm x 10 mm, were joined with MoSi2-Si composite by the presureless technique (POLITO)
• Cf/SiC to Cf/SiC: flat to flat surfaces, 10 mm x 10 mm, were joined with MoSi2-Si composite by the presureless technique (POLITO)
• Cf/SiC to TiAl6V4: flat to flat surfaces, 30 mm x 30 mm, were joined with Cusil/ABA by the pressureless technique (POLITO)
• Cf/SiC to Cf/SiC: flat to flat surfaces, 10 x 10 mm, were joined with Ti3SiC2 using SPS process (Nanoforce)
• SiCf/SiC to SiCf/SiC: flat to flat surfaces, 10 x 10 mm, were joined with Ti3SiC2 using SPS process (Nanoforce)
• On basis of EMA rod a tubular shaped demonstrator has been manufactured and machined and on basis of the really complex shaped bodyflap, T- and angle-shaped parts (>5 parts) have been manufactured as well as finally one (1) representative component for the box design of the flap; a combination of edges and angles, dimension ca. 350x200x200 mm.
• Boostec SiC to Boostec SiC, flat to flat lapped surfaces, 25 mm x 25 mm x 5 mm, were joined with Sn91Ag4Ti4 using active soldering (FRAUNHOFER)
• Boostec SiC to Invar, lapped surface of SiC was joined to the laser-modified surface of Invar with Sn91Ag4Ti4 using active soldering (FRAUNHOFER, AGI). This constitutes the final demonstrator, which is a satellite part where a Tooling Hole Boss (Invar alloy) is joint to a Boostec SiC plate.
• Si3N4 to Si3N4 and Si3N4 to Invar, flat to flat lapped (reference) and laser structured surfaces, 25 mm x 25 mm x 10 mm, were joined with adhesive HYSOL® EA 9321 (AGI)
Life Cycle Impact Assesment
Life Cycle Impact Assessment (LCIA). This step involves the evaluation of environmental impacts associated with the system outputs as well as the consumption of resources related to the inputs. In this phase there is the changeover from the objective data, collected during the inventory phase, to the judgment of environmental hazard.
The goal of this study is to analyse the environmental burden of two innovative joining processes for advanced materials used in the aerospace sector.
In particular, the assessment covers:
• An active brazing process to join a silicon carbide based composite material (C/SiC) and a titanium alloy (Ti6Al4V). An Active Braze Alloy (CuABA) is used and the process runs at a temperature around 900°C.
• An adhesive bonding process to join silicon nitride (Si3N4) and an INVAR alloy, by means of an epoxy adhesive at room temperature (for 8 days).
In both cases, a surface treatment is required before the actual joining of the two materials. In the case of the active brazing process, a selective oxidation is performed on the C/SiC while the Ti6Al4V alloy is machined. As far as the adhesive bonding is concerned, on both materials (Si3N4 and INVAR alloy) two alternative surface treatments are used: a mechanical lapping and a laser surface nanostructuring.
The processes under study are currently at lab/pilot scale. Therefore, the related impacts are directly comparable to optimized industrial processes.
Most of the data were directly measured and collected by the partners running the processes. Where primary data were not available, secondary data were used, referring to the ecoinvent database .
The present assessment was performed using the SimaPro 8.1 software (PRé Consultants), that integrates the ecoinvent database. The Impact Categories chosen are the most representative for industrial processes and among the easiest to be communicated and understood also by a non-technical audience. In particular, they are the Cumulative Energy Demand (CED) and the Global Warming Potential (GWP, also referred as Carbon Footprint).
The Life Cycle Inventory analysis involves the creation of a model of each process under study. In particular, two cases are considered:
• Case 1: it concerns the active brazing process used to join C/SiC and the Ti6Al4V alloy.
• Case 2: it concerns the adhesive bonding process used to join Si3N4 and the INVAR alloy. This case study is further divided in two alternative scenarios, according to the surface treatment used before the actual joining of the two materials:
o Case 2a: the surface treatment used on both the Si3N4 and the INVAR alloy is Lapping;
o Case 2b: the surface treatment used on both the Si3N4 and the INVAR alloy is Laser Nanostructuring.
LCIA conclusions
From the analysis of the results analysis, a common outcome is that the energy consumption of each process represents the major source of the environmental impact. This implies that a higher degree of sustainability is straightforwardly related to a lower energy consumption. Therefore, it is evident that high temperature processes (Case 1) lead to a higher impact with respect to those performed at room temperature (Case 2).
Another leitmotiv emerged from the results is the fact that surface pre-treatment plays a major role in the joining processes analysed. In fact, in Case 1, the selective oxidation of the C/SiC produce the 66,67% of the impacts. In Case 2, this consideration is even more radical since a change in the laser pre-treatment brings to a 99.8% saving in the overall environmental burden of the joining process.
The environmental impact of the active brazing is moderate, but not low in general terms. To give a daily life reference, the CO2 emissions produced by the joining of 1 cm2 of materials correspond to the CO2 emissions produced by a person travelling for about 50 km by car. The same reasoning applied to the adhesive bonding, with Lapping pre-treatment, leads to an equivalence of 1.7 km by car, while, with Laser Nanostructuring, the process impacts as much as 2.88 m travelled by car .
It is important to underline that the functional unit does not consider changes in the quality of the joining process. A combination of the environmental results and the mechanical performances will lead to a more comprehensive and consistent comparison among processes. In particular, if the performance of a joining process, it can be evaluated if the same mechanical performance of the aerospace component can be reached with a lower extension of the joins. In this respect, the results of this analysis are intended to be ready to be used as inputs in such evaluation.
Potential Impact:
The ADMACOM impact can be summarized as follows:
1. Development of new and exploitable manufacturing technologies based on joining of hybrid components made of ceramic matrix composites (C/SiC) joined to metals (Ti alloy) and ceramics (SiC and Si3N4) to metals (Invar). The impact can be seen in enabling new design solutions and obtaining light-weight structural demonstrators for aerospace applications with improved reliability and in-service performance, in term of weight reduction, fuel consumption, pollution reduction and high temperature resistance.
2. Improved competitiveness of European industries via more favourable cost/benefit solutions due to the use of CMC and advanced ceramics joined to existing and reliable traditional materials (metal alloys).
3. The Lyfe Cycle Impact Assessment results showed that the laser-pretreatment of ceramic and metal surfaces, before adhesive bonding, corresponds to 99.8% carbon footprint reduction of the joining process in comparison with the reference surface treatment (lapped surfaces).
4. Acquisition of new skills in the European workforce resulting also in indirect socio-economic benefits: this has been done through two Advanced Training Modules on mechanical characterization of joined components.
5. Significant effort have been done for standardization of mechanical tests (shear strength) of joined components, in cooperation with EU and national standard and measurements agencies. It must be underlined the high impact that a EU standardization of mechanical test procedures on joined components to support reliable design and foster the development of new industrial components where different materials are joined together.
Finally, the potential areas and markets of application of ADMACOM will be in the field of high performance aerospace systems, including satellites. A reliable CMC or ceramic to metal joined component is an emerging market for lightweight technologies. The joining of these advanced ceramics and CMC to other dissimilar materials can be undertaken using mechanical fasteners, but such an approach is not a cost effective and a technologically reliable solution. The new joining technologies investigated in ADMACOM can offer benefits with regards to reduced mass, reduced radar signature, reduced costs through simpler construction, improved fatigue performance and a more uniform stress distribution.
The output of ADMACOM will be considered for industrial partners applications well beyond those described in ADMACOM, in particular advanced ceramics/metal and CMC/metal joints could impact on the production of:
• space industry/Re-entry vehicles/Capsules/Landers, namely thermal protection systems/panels, heat shields, insulation, nose caps, corner edges, structural and moveable parts like flaps, rudders, leading edges, fixation of sensors/instrumentation.
• Aerospace civil-military/Propulsion Systems, namely turbines and turbine environment, jet engines, housings, nose caps, leading edges, fins, exhaust nozzles, combustion chambers or parts for combustion chambers, nozzles for small thrusters, and fixation of sensors/instrumentation.
• Power generation/turbines/Renewable industry/Marine, namely pump and bearing environment, insulation, environmental industry, chemical industry, renewable industry (power generation, wind, water and sun).
OUTREACH ACTIVITIES:
- ADMACOM participated in the 10th European Researchers’ Night, 25 September 2015, Turin and in the 11th European Researchers’ Night, 30 September 2016, Turin (POLITO and AGI).
- The ADMACOM newsletters were periodically published in the ADMACOM website.
- FRAUNHOFER and AGI prepared a presentation of the “Active soldering of ceramic-ceramic and ceramic-metal compounds” developed within the ADMACOM project. This informative material was published and disseminated through the ADMACOM website.
- The ADMACOM Industrial workshop was held on September 14th - 15th, 2016 in Brussels. The aim of the ADMACOM workshop was to present innovative manufacturing technologies based on advanced design of interfaces and joining materials for aerospace components http://www.admacomproject.eu/article/admacom-workshop-september-14th-15th-2016.
- POLITO participated in the METAL MORPHOSIS project seminar 24th February 2016 at the BWI - Belgian Welding Institute, Ghent and attended and exhibited at Composites Innovation 2016 (Sheffield, UK 22-23 June 2016).
A very intensive dissemination activity was done and planned during the last 18 months: project website, newsletter, conferences, workshops. The project website contains also the list of conferences at which ADMACOM partners disseminated scientific results. Advanced training modules have been organized on demand, on the request of the involved companies.
List of Websites:
http://www.admacomproject.eu/
The aim of ADMACOM (Advanced manufacturing routes for metal/Composite components for aerospace) was to develop innovative manufacturing technologies based on advanced design of interfaces and of joining materials for aerospace components.
Several innovative joining technologies and materials for Ceramic Matrix Composites (CMC) and bulk ceramics to metals for aerospace components have been developed within ADMACOM.
The objective of ADMACOM is to have industrials designers, material scientists and engineers to work together to find the best novel, reliable and affordable manufacturing technology for joining dissimilar materials.
The ADMACOM activity focussed on novel manufacturing technologies and joining materials, coupled with surface engineering to maximize interface strength for the joints, and to model/design physical-chemical reactions at the interface; surface engineering means also selective oxidation of the composite surface, laser structuring and/or mechanical machining of the composite/metal surface. The main investigated joining techniques were adhesive bonding, active soldering, « in situ» composite processing, joining with SPS by direct bonding or with interlayer.
Thermodynamic prediction of high temperature interactions and interface formation in dissimilar materials, advanced wettability studies supported by computer simulations has been used to select joining materials.
ADMACOM gathers three industrial partners, academia and research institutions with proven world class expertise in production and joining of CMC. In particular some CMC have been produced by two industrial partners of this project (MTA and AIRBUS Group). The industrial partner Nanoforce exploited the innovative manufacturing technologies.
The main achievements are listed hereafter:
- AGI successfully manufactured some representative demonstrators for satellite applications ( SiC/Invar Tooling Hole Boss, Si3N4/Si3N4 and Si3N4/Invar, laser structured and joined with an epoxy adhesive)
- Laser pre-treatment: laser irradiated SiC surface zones were analysed and the studies led to understand the laser structuring behavior and the phenomena of observed microstructural features.
- A joining procedure was optimized to obtain Cf/SiC - Cf/SiC and SiCf/SiC - SiCf/SiC demonstrators for MT Aerospace joined with Mo-Si composite through a new pressureless process.
- Cf/SiC composites have been successfully joined to Ti6Al4V alloy with Cusil/ABA using a pressureless brazing process.
- Nanoforce sucessfully joined SiCf/SiC to SiCf/SiC and Cf/SiC to Cf/SiC composites with Ti3SiC2 ternary carbide for MT Aerospace demonstrators using optimised Spark Plasma Sintering process.
- Fraunhofer IFAM used Ta and Co foils instead of the Ta/Co powder mixture and obtained sound joints by using pressure-assisted Spark Plasma Sintering Process.
- CNR designed an ad hoc transient liquid phase bonding procedures through Al-Ti interlayers to join SiC.
- Recommended practice draft for determining the torsional shear stress of joined dissimilar materials was submitted to ESIS (Empa, POLITO).
Project Context and Objectives:
A wide number of new composite materials with exceptional properties, such as high thermo-mechanical and thermo-chemical resistance and low density compared to traditional materials (i.e. metals and polymers), are available in the EU market. Their use is still limited to very few applications because of the problem of joining them reliably.
Just to cite a few possibilities: fibre reinforced Ceramic Matrix Composites (CMC, e.g. SiC/SiC, C/SiC, C/C or Al2O3/mullite) could be used right now to replace existing aerospace components. This will contribute to an increasing demand for novel components with improved properties, such as light weight, for reducing fuel consumption and CO2 emission. To enable this, the joining of CMC to metals needs to be easier, more rapid, and possibly permit inspection.
ADMACOM (Advanced manufacturing routes for metal/Composite components for aerospace) developed several innovative technologies and materials to join Ceramic Matrix Composites (CMC) and bulk ceramics to metals for aerospace components.
In ADMACOM, industrials designers, material scientists and engineers worked together to find the best novel, reliable and affordable manufacturing technology for joining dissimilar materials.
The ADMACOM activity focused on novel manufacturing technologies and joining materials, coupled with surface engineering to maximize interface strength for the joints, and to model/design physical-chemical reactions at the interface; surface engineering means also selective oxidation of the composite surface, laser structuring and/or mechanical machining of the composite/metal surface.
Thermodynamic prediction of high temperature interactions and interface formation in dissimilar materials, advanced wettability studies supported by computer simulations were used to select joining materials.
A suitable joint design was developed through an extensive work on the design and model of the joined components, taking into account the requirements and specifications defined by the industrial partners (AIRBUS Group and MTA). The assembly of demonstrators and an accurate cost/benefit analysis were carried out in order to exploit the project results.
The experimental activity of the last 18 months of the project focused on the validation of the model of interface by the manufacturing of successful joints.
The surface engineering was optimized both on the metallic and the ceramic surfaces to be joined: for SiC and Si3N4 materials, the laser nano-structuring technique - to improve the joint performance - was extensively studied. The effect of laser irradiation on the surface morphology and chemical composition of SiC and Si3N4 and its influence on ceramic/joining material interface were investigated. Concerning CMCs, a surface engineering based on the Selective Thermal Removal (STR) of SiC fibers from SiCf/SiC composites was used to obtain a brush-like surface.
The following joining materials were proposed, discussed within the consortium and tested during the period M18-M36:
- adhesive HYSOL EA9321, active soldering using Sn-Ag-Ti solders (for Airbus Group Innovations).
- In-situ Mo-Si composite joints; Ta-based brazing alloys (Ta-Ni and Ta-Co); Ti, Ti6A4lV or Ti3SiC2 foils or slurry with Spark Plasma Sintering (SPS) (for MTAerospace).
The main investigated joining techniques during the period M19-M36 were:
adhesive bonding, active soldering, « in situ» composite processing, joining with SPS by direct bonding or with interlayer.
All these joining materials and techniques were used to prepare joined samples for testing. According to industrial specifications, experimental work focuses on the manufacture of joints on flat components.
In order to define the most promising novel joining materials and techniques and to optimize the joining conditions a detailed microstructural characterization of all joined materials and interfaces was carried out. Concurrently, a mechanical validation of each joined components was supported by a careful selection of available mechanical tests as well as environmental tests (i.e. cryogenic cycles) to reproduce in service conditions. Strong collaboration within the partners led to substantial work on the standardization of selected mechanical tests.
Concerning the activities carried out to maximize ADMACOM impact on EU economy, a exploitation plan for the results, as well as a dissemination plan were defined and implemented by all the ADMACOM partners.
The main achievements are listed hereafter:
- AGI successfully manufactured the following representative demonstrators for satellite applications:
• SiC/Invar Tooling Hole Boss (laser target fixation, design on basis of NIRSpec aerospace experiences), joined by active soldering at Fraunhofer IFAM, for representative testing at Airbus Defence and Space facilities;
• Si3N4/Si3N4 and Si3N4/Invar, laser structured and joined with epoxy Hysol adhesive at AGI, mechanically tested at POLITO at room temperature and after environmental cycling (5 cycles, 50K). The laser pre-treatment of Si3N4 and Invar
- Laser pre-treatment: laser irradiated SiC surface zones were analysed (POLITO, AGI, Nanoforce) and the studies led to understanding the laser structuring behavior and the phenomena of observed microstructural features, such as graphite layer formation.
- A joining procedure was optimized to obtain Cf/SiC - Cf/SiC and SiCf/SiC - SiCf/SiC demonstrators for MT Aerospace joined with Mo-Si composite through a new pressureless process.
- Cf/SiC composites have been successfully joined to Ti6Al4V alloy with Cusil/ABA using a pressureless brazing process.
- Nanoforce sucessfully joined SiCf/SiC to SiCf/SiC and Cf/SiC to Cf/SiC composites with Ti3SiC2 ternary carbide for MT Aerospace demonstrators using optimised Spark Plasma Sintering process.
- Fraunhofer IFAM used Ta and Co foils instead of the Ta/Co powder mixture and obtained sound joints using pressure-assisted Spark Plasma Sintering Process.
- CNR designed an ad hoc transient liquid phase bonding procedures through Al-Ti interlayers to join SiC.
- Recommended practice draft for determining the torsional shear stress of joined dissimilar materials was submitted to ESIS (Empa, POLITO).
Project Results:
WP1 Definition of specifications and requirements for joined components, manufacture of composites
The objectives of this WP are to implement the activity according to industrial requirements on the components to be joined and the manufacturing and supplying the materials to be joined (composites, alloys) by the industrial partners in a quantity agreed as necessary for the completion of the project.
The industrial partner AGI provided the specification and requirements for joined components in space applications (satellites) according to industrial requirements. Furthermore, the representative ceramic / metal demonstrators as well as corresponding bonding partner materials and joining techniques were selected, defined, specified and designed:
• SiC / Invar Tooling Hole Boss (laser target fixation, design on basis of NIRSpec aerospace experiences), joined by active soldering, for representative mechanical validation
• Si3N4/Si3N4 and Si3N4/Invar, joined with epoxy Hysol adhesive, for SLO mechanical testing
The industrial partners provided the amount of materials to be supplied to the other partners; material type and surface quality as well as amount of material samples to be provided was continuously discussed with project partners.
WP2 Design and modelling of joined components
The objectives of WP2 are: to design and model the joined component(s) as agreed in WP1, including demonstrator(s); cost/benefit analysis and definition of designers/engineers cooperation best practices, i.e. a close collaboration between designers and engineers in order to guarantee the experimental feasibility of the designed components
The experimental work was supported by combined thermal-mechanical simulations of bonds with structured interfaces (FRAUNHOFER Bremen). Stress evaluation was done using Mises-stresses for the solder and the maximum principal stress for the ceramic (supposing mode-1 cracking failure type for the ceramic). Stress relaxation was only allowed in the solder by means of plastic yield. The simulations showed that the stress situation near the solder-ceramic interface is dominated by the residual thermal stresses. Structuring the surface of the ceramic leads to increasing of tensile as well as compressive stresses in the ceramic. Maximum stresses in the ceramic were found near the bottom of the holes. Tensile and compressive peak stresses were partly in close vicinity. Maximum tensile stresses in the ceramic seem to be caused by the restrained shrinkage of the solder deep in the hole. In comparison the maximum stresses are caused by triangular stress superposition between neighbored solder-filled holes. Further simulations showed that a reduction of the structuring depth can lead to a significant reduction of the residual thermal stresses in both the solder as the ceramic – while still offering a better interlocking of the bonding partners.
Laser structuring of SiC surface: deeper studies of laser influenced zones were done (POLITO, AGI, Nanoforce). More detailed analysis of the laser irradiated SiC surface zones were performed in order to better understand the laser structuring behavior and the phenomena of observed microstructural features. The understanding is of high importance because of the joint low mechanical performance. Based on literature research and experimental analysis results, a hypothesis for the mechanism of the graphite layer formation at the SiO2/SiC interface was proposed. Details were summarized in the common paper “Effect of laser pulsed irradiation on the SiC surface” for the ADMACOM special issue on the Int. Jour. of App. Ceram. Tech. (submitted).
Design of demonstrator on basis of NIRSpec aerospace experiences (AGI): The demonstrator design and demonstrator set-up for testing and validation of the developed joining technology on technological but representative sample level (in WP7) as well as the fabrication of the INVAR Tooling Hole Boss parts and SiC parts according to the technical drawings were completed.
Cost/benefit analysis
• Add-on activities on cost/benefit analysis (MTA, AGI): some add-on activities were established, knowing more precisely the process steps of the main relevant joint methods. Since all methods were running in parallel at several partners and institutes, a final cost-benefit analysis and evaluation could not been performed after M12. Therefore, MTA and AGI performed analysis best after the final selection of joining processes with most relevant success and definition of pre-prototypes. The main cost/benefit analysis results are summarized in the additional requested Deliverable D2.4.
• The AGI cost / benefit analysis was done in context with reproducibility, bonding performance, risk mitigation and savings in effort. The laser structuring for surface pre-treatment of e.g. Si3N4 brings benefit in terms of
o Improved bonding performance which was demonstrated (for both similar & dissimilar at RT & cryo-cycling: strength, failure mode, homogeneity)
o Factor human being: the process brings
▪ high potential to substitute manual process
▪ highly reliable in one shot
▪ high reproducibility
▪ high degree of automation
▪ high quality
▪ reduced risks
o Substitution of critical chemicals for pre-treatment (REACH)
o Substitution of lapping as a surface pre-treatment process 75% savings in efforts (and by this cost)
o Surface pre-treatment: use of laser nano-structuring brings to a 99,8% saving in the overall environmental burden of the joining process (see LCIA) → “green” process
• Main benefits of ADMACOM joining techniques for MTA can be summarized as following:
o General:
▪ Weight reduction
▪ Higher flexibility in design layout
▪ Less efforts in machining of parts and joints (screws, pins, bolts..)
▪ Less space needed
▪ Risk reduction
o Active Soldering
▪ - Compared to adhesives no release of humidity during operation in orbit
▪ - Wetting of difficult-to-wet surfaces
▪ - Thermal conductive joints
▪ - Low joining temperature
o Partial Transient Liquid Phase Bonding:
▪ - A (transient) liquid phase assures intimate contact between surfaces
▪ - Joining temperature in principle lower than service temperature (high melting phases formed by reaction and diffusion)
▪ - Reduction of thermal stresses arising upon cooling
• Nanoforce also performed cost-benefit analysis of the joining process developed on site within Admacom project by comparing with the current state of the art (joining processes with Ti foils or MAX phase powder). The newly developed approach consists of joining of CVD-SiC and CMC materials with the pre-sintered Ti3SiC2 MAX phase foils via solid state diffusion bonding using SPS at temperatures as low as 1250°C. It was concluded that the joining process has the potential to both be cheaper (due to lower temperature and time) and produce high quality joints (higher shear strength). The main disadvantage of this technique is the additional processing step of preparing the joining foil. This extra step might be compensated by the potential to remove the surface preparation step
T2.3 Designers/engineers cooperation best practice
• MTA and AGI have defined the demonstrator design, during the reporting period M19-M36 the prototype realization has been developed by MTA, AGI, POLITO, Nanoforce, FhG IFAM in an excellent designers/engineers cooperation best practice way. So, on-going activities were ensured by effective and very close collaboration among designers from industry and materials scientists from institutes (All).
• The effective collaboration among component designers and materials scientists was also focused on mechanical test and standardization. One of the objectives was to “translate” data available now at AGI into pure shear data by starting a round robin test based on torsion test, asymmetric four point test and AGI single lap test. The final goal was to provide AGI the possibility of “translating” apparent shear strength measured up to now to pure shear, and to provide designers with suitable results for modelling of new components.
• Several telephone conferences and as well as meetings industry / institutes were hold. Furthermore, bilateral meetings have been organized, also with researches’ visits to industrial premises and to other partner facilities.
WP3 Surface engineering, modelling and physical-chemical design of interfaces
The objective of this WP was to acquire new knowledge of the physico-chemical phenomena required to model and design the composite interface in order to maximize its strength and the overall performances of both adhesive and high temperature joints.
Surface engineering to obtain the brush-like joints (POLITO)
POLITO performed several experiments to reproduce the “brush-like” surfaces of SiC/SiC samples: the composites have been treated at 1450°C for 2h in order to selectively remove part of the fibers at the surface. The goal is to increase the surface area and to maximize the interface strength and the overall performances of joints. The study of the effectiveness of the surface engineering treatment on SiC/SiC has been evaluated by AFM analysis at POLITO; the measurements have been performed both on the surface engineered and on as-received SiC/SiC; in any case, it is very difficult for these analysis to be successful because the maximum z range of the AFM is about 5 µm and the SiC/SiC show high porosity and roughness on the cross section.
Evaluation of effectiveness of surface engineering by 3D profilometry and high-temperature wetting tests (CNR)
In order to assess the effectiveness of the surface engineering performed at POLITO on SiC/SiC composites to obtain the “brush-like” joints, confocal profilometry was implemented for the 3D acquisition of the surfaces of SiC/SiC composites and to measure their characteristic features. The average depth of the fibers removal was measured to be 2.4 µm with an increase of the total surface area ranging from 89% to 129%.
Wetting tests were conducted using the brazing alloy AgCuTi (Ag: 57.7 Cu: 36.8 Ti: 5.5 at%) at 850°C. For the sake of comparison, the tests were conducted with several materials: ultra-pure SiC, SiC/SiC as cut, SiC/SiC polished and SiC/SiC thermally treated. For the materials not thermally treated, contact angles range from 10° to 20° with similar wetting kinetics. For the SiC/SiC composites thermally treated the final contact angles were measured to be slightly higher (28°) with kinetics relatively slower. This was attributed to the fact that the surfaces after the thermal treatment are in fact oxidized and in principle not wettable. The Ti in the alloy reacted with the oxide layer allowing the wetting. The micrographs taken on cross-sectioned samples showed that the alloy perfectly adhered to the SiC/SiC surfaces filling all the holes and gaps introduced by the thermal treatment. In this way, the adhesion between the composite and the brazing alloy increased coherently to the increase of surface area measured by profilometry. This constitutes the most interesting and important result of this activity demonstrating the promising effect of surface engineering to produce brush-like features and their effectiveness to increase the joint quality.
Wetting experiments for Co-Ta/SiC and Ni-Ta/SiC systems (CNR)
The wetting and interfacial behaviour of liquid Ta alloys (Co-Ta 12 and 50 at%, and Ni-Ta 14 and 38 at%) in contact with SiC surfaces (CVD-SiC, C/SiC, SiCf/SiC) has been determined by sessile drop experiments. The Ta alloys were selected by Fraunhofer as potential brazing media for SiC based composites. The influence of process parameters (atmosphere, alloy preparation, surface roughness) has been tested and discussed. For all the compositions and experimental conditions, good wetting was observed with fast spreading. The interfacial behaviour is determined by the competition between the typical interfacial phenomena of the pure elements; specifically, either the dissolution of the ceramic phase by Ni (or Co) or the formation of a new interfacial layer, TaC, by reaction between Ta and SiC. The competition and the prevalence of one or the other of the two phenomena depend on the relative amount of the single element in the alloy. The main results are: - all the systems showed very good wetting (i.e. θ < < 90°); - whatever the composition, interfacial Ta carbide was found; for low contents of Ta, dissolution of the SiC substrate was found, similar to what is observed for the well known systems Ni-Si-C and Co-Si-C; - a high content of Ta led to the suppression of dissolution of the ceramic phase and to the formation of a continuous layer of TaC at the metal/ceramic interface; this layer stops diffusion and enhances wetting making the use of this alloy for brazing feasible; - using high vacuum rather than a protective Ar/5% H2 atmosphere resulted in a faster wetting as a consequence of an enhanced cleaning effect of the surface; - the comparison between pre-melted alloys and pellets of compacted powders (that one would likely use in a joining process) showed no relevant differences in both wetting kinetics and microstructure.
Basic wetting data of Ta alloys on SiC were still not present in the scientific literature, therefore their importance and usefulness go beyond the framework of the ADMACOM project. These results were used by Fraunhofer to choose the most proper brazing alloy compositions and process parameters necessary to obtain reliable joints (see WP4).
Thermodynamic modeling through CALPHAD calculations (CNR)
In order to obtain thorough understanding of the interaction phenomena occurring during the contact of liquid Co-Ta and Ni-Ta alloys on SiC materials observed during wetting tests (see D3.1 and D3.2) the thermodynamic evaluation of SiC/Ta-Ni and SiC/Ta-Co interfaces was performed through CALPHAD calculations. A deep study of the binary and ternary phase diagrams, between the elements of the ceramic substrate (C, Si) and of the alloys (Co, Ni, Ta), has been performed in order to predict and evaluate the different potential intermediate ternary phases formed at the interface zone. A database input file, containing the thermodynamic parameters of the pure elements and of the intermediate compounds, fundamental to obtain the ternary and the quaternary systems (C-Ni-Si-Ta, C-Co-Si-Ta), has been produced. Thanks to the CALPHAD modeling, it was possible to better interpret the experimental results regarding the high temperature reactivity between Co-Ta alloys (12 and 50 at% Ta) and silicon carbide. Even if the studied systems are dynamic, a comparison with the equilibrium phase diagrams is a useful way to understand the interactions occurring during the brazing processes such as high temperature isothermal interactions and phase transformations during the subsequent cooling process.
High temperature wetting test in SPS (Nanoforce)
A new advanced technique to assess the wetting behaviour of metal/ceramic systems at very high temperatures was developed in Nanoforce Technology ltd. The advantage of this technique is a possibility to measure in-situ a contact angle between alloys with high melting points and ceramic substrates using Spark Plasma Sintering (SPS) up to 2300°C. The wetting results of Ta-based alloys on the CVD SiC were compared with the ones obtained by CNR using their well-controlled wetting apparatus. A good match in terms of the wetting and spreading kinetics, contact angles, reaction and solubility of the SiC etc., was obtained, confirming a good potential of the new technique. Using this newly developed wetting technique, the influence of the heating rate on the wetting behaviour was investigated using Ti6Al4V alloy on the polished CVD β-SiC surface. The results showed a significantly better wetting kinetics and spreading behaviour of the alloy when the faster (500°C/min) heating rate was used rather than the slower one (100°C/min). In addition, the slower heating slightly lowered melting temperature of the alloy because there is more time for the phases with a low melting point to form due to the reaction with SiC. The results showed that the wetting of an alloy can be modified by the heating rate and it might be necessary to apply a higher heating rate in order to obtain a good bonding while joining.
Design of transient liquid phase bonding procedures: wetting of SiC by AlTi alloys (CNR)
In order to determine the feasibility of using Al-Ti interlayers for the brazing of SiC based ceramics through the transient liquid phase bonding technique, the wetting and interfacial phenomena related to the contact between SiC and Al-Ti alloys were studied. Three compositions were selected: Al 3at% Ti, Al-25at%Ti and Al-46at%Ti and tested by the sessile drop method at 1500°C on pure SiC substrates. All the systems exhibited very low contact angles (<< 90°C) that in principle would assure an intimate contact between the materials to be joined. From the interfacial chemistry point of view, the best results were obtained with the Al-25 at% Ti alloy: the formation of a continuous interlayer of Ti3Si1 xAlxC2 at the metal-ceramic interface was observed and this phase perfectly adhered to the SiC substrate. These results proved the possibility of using Al-Ti interlayers for the joining of SiC-based materials: the Al-Ti liquids assure the contact between the adjoining materials and the (transient) liquid phase evolves by reaction and diffusion, solidifies at the testing temperature and forms a Ti3Si1 xAlxC2 MAX phase at the interface that has a re-melting temperature higher than the process itself. The outcome of this activity is the manufacturing, at the laboratory scale, of sound SiC-SiC joints (see WP4).
Investigations of the joined interfaces using HR-SEM equipped with FIB (Nanoforce)
Nanoforce investigated the interface between Boostec SiC and the Sn-Ag-Ti-RE solder (joined and supplied by Fraunhofer) using High Resolution Scanning Electron Microscopy (HR-SEM) equipped with Focused Ion Beam (FIB) and Transmission Electron Microscopy (TEM). Thin lamellas were prepared using FIB and the area of interest was subsequently observed using TEM. The analysis revealed that there was a sharp interface between the SiC and the solder, suggesting good, apparently strong bonding. However, some small well-separated pores (< 1μm) were observed all along the interface. This may suggest that the solder did not fill small surface pores in some areas while it strongly bonded to the matrix in others. On the other hand, no obvious reaction layer was found by both SEM and TEM analysis of the interface.
WP4 Novel joining materials and techniques
T4.1 Novel joining materials and surfaces (Task Leader Fraunhofer)
This task was focused on :
− Experimental application of novel joining materials and techniques selected in WP 1
− Preparation of joint samples for testing
− Optimization of joints
− Production of demonstrator parts
The experimental activity can be summarized as follows:
− Manufacturing of joints using modified surfaces, according to results of WP3; the modification of the surfaces has been carried out by mechanical machining, i. e 1) laser structuring of selected ceramics and metal alloys by Airbus Group Innovations ; the modified materials are BOOSTEC®SiC and Invar alloy 2) selective fibre removal of the composite matrix (“brush joint”) performed by POLITO and optimization of the process parameters
− Surface modification by chemical reaction (CNR);
− Study of novel pressure-less, flexible and low cost joining approaches by three different joining materials for CMC composites (calcia-alumina glass-ceramic, metallic brazes, in-situ formed Mo-Si based composite) (POLITO, EMPA).
− POLITO performed a study on the HYSOL adhesive (epoxy paste) to characterize its behaviour and the curing parameters; the joining process was optimized and several joints have been performed using BOOSTEC®SiC and Invar both laser structured and non laser structured
− Fraunhofer investigated two new joining technologies for SiC-based materials:
o High temperature brazing with Ta-Ni and Ta-Co alloys
o Active soldering using Sn-Ag-Ti-based solders
− CNR continued optimization of a Partial Transient Liquid Phase Bonding by Al-Ti interlayers
− For mechanical testing purposes EMPA produced SiC-SiC bending and torsion test samples by active brazing using Incusil ABA
Significant results:
• Surfaces of SiC/SiC to be joined with different joining materials have been manufactured; the surface modification method has been developed by POLITO to obtain “brush-like” for SiC/SiC composites
• Joining of surface engineered SiCf/SiC composites by Ag-Cu-Ti alloy (POLITO, CNR); Micro-sized brush-like structure increased the adhesion area between the composite and the brazing alloy and this enhances the mechanical strength of the joint due to the mechanical interlocking.
• A laser surface structuring technology for nano-scaled modification of ceramic materials such as ceramics and CMCs was investigated by Airbus Group. Laser parameter studies on SiC and Si3N4 were conducted by Airbus Group to investigate influence on microstructure, chemical composition & mechanical joint performance; Mechanical test results do not confirm the effectiveness of laser structuring on SiC surfaces (“weak intermediate layer“). Mechanical test results DO confirm the effectiveness of laser structuring on Si3N4 surfaces.
• The joint shear strength for laser-structured BOOSTEC®SiC - BOOSTEC®SiC samples joined by the adhesive Hysol® EA 9321 was measured by single lap-shear test. Promising shear strengths of up to 43 MPa were obtained.
• HYSOL joined BOOSTEC®SiC to Invar, both laser structured and not laser structured have been manufactured and compared
• The joining process by HYSOL and the joining material itself have been deeply investigated in terms of reduction of porosity during curing and ways to improve the thermos-mechanical resistance of the joints
• For the joining of Keraman® SiC/SiC and C/SiC samples POLITO investigated two joining approaches: the joining materials are CA glass-ceramic and a Mo-Si based filler, both used successfully.
• A literature review (limited to the last five years) of the most promising joining processes applicable to selected materials (SiC, CfSiC, SiSiC, Invar and Ti6Al4V) has been prepared by EMPA and provided to all consortium;
• For mechanical testing purposes EMPA produced SiC-SiC bending and torsion test samples by active brazing using Incusil ABA.
• CNR worked on the transient liquid phase bonding technique by means of Al-Ti interlayers. These processes are meant to be effective for C/SiC and SiC/SiC composites for high temperature applications. Joined samples were produced using Al-Ti alloys assemblies (interlayers and bulk pieces flowing by capillarity) showing the potential of the method and the formation of new phases with melting point higher than the process temperature.
• A high temperature brazing approach with Ta-Ni and Ta-Co alloys was tested by Fraunhofer to join SiC/SiC composites for high temperature applications.
• Concerning the high temperature brazing with Ta-Ni and Ta-Co alloys, the main results are:
o Promising wetting behavior of selected Ta-Ni and Ta-Co alloys on SiC-surfaces (CNR)
o Joining using powder mixtures and SPS technology is feasible
o Formation of reaction zones for both Ta-Ni and Ta-Co alloy system
o Control of the SiC decomposition reactions (reaction zones) within the Ta-Co/SiC system by choosing a layered starting stack set-up using metal foils (SiC/Ta/Co/Ta/SiC)
o The joining parameters for monolithic SiC were successfully transferred to SiC/SiC composite joints (Fraunhofer).
• Active soldering for the joining of BOOSTEC®SiC - BOOSTEC®SiC for low temperature and cryo applications has been conducted by Fraunhofer.
• Concerning the active soldering using Sn-Ag-Ti-based solders, the main results are:
o Active soldering is considered to be an attractive joining process for SiC components for low temperature applications
o Active soldering of monolithic SiC and SiC/Invar joints for low temperature applications was successfully conducted. Parameter studies for both material combinations were performed. Joints with apparent shear strength of about 30 MPa were obtained for SiC/SiC and SiC/Invar joints (Fraunhofer). The influence of the sourrounding atmosphere during joining was investigated. Air, Argon and Nitrogen were tested. The best results were achieved for air atmosphere. Demonstrator production for WP7 was conducted.
o The active soldering was also successfully applied to join SiC to TiAl6V4.
T4.2 SPS (Task Leader NanoForce)
New joining materials based on powder mixtures are being prepared and used in combination with fast pressure assisted technologies (Spark Plasma Sintering, (Fraunhofer, NanoForce) .
Significant results:
• The SPS at Nanoforce has been used to join both monolithic CVD silicon carbide (as a representative of the CVD SiC coating on CMC materials) as well as the SiC-SiC and the C-SiC composites. The max phase - ternary carbide Ti3SiC2 - was used as the joining material.
• Four point bending strength results of joined monolithic CVD SiC showed that the strength of the joint was significantly higher when a pre-sintered Ti3SiC2 foil rather than a slurry made of Ti3SiC2 powder was used. The foil was pre-sintered using SPS at 1300°C and then polished down to around 50-100 μm. The samples were joined at relatively low temperatures 1300-1350°C and using an external pressure of 60 MPa in vaccuum.
• The bending strength of the monolithic CVD SiC joined with Ti3SiC2 was independent on the final thickness of the joint. Since the Ti3SiC2 foil did not shrink or react with the SiC, the final thickness of the joint was predetermined by the initial thickness of the foil.
• The Ti3SiC2 foil showed some ductility during the joining process as it conformed to the rough surface of the CMC materials. It also penetrated into the cracks in the CVD coating of CMC materials (formed during the processing of the composites); effectively healing the CMC composites.
• Joining of the monolithic CVD SiC with Ti6Al4V foil was also performed using SPS at Nanoforce. The foil with different thicknesses was used as the joining filler and joining process was carried out at temperature of 1650°C and an external pressure of 60MPa. The foil shrunk during the joining and showed a significant infiltration into the SiC material, confirming a high reactivity of this metal alloy with SiC surface.
• Bending strength of the monolithic CVD SiC joined with Ti6Al4V exhibited a similar value as non-joined monolithic reference material, proving a high strength of the joint.
• Nanoforce performed additional investigation of the cross section of the joints. Although only coated CMC samples have priority for Admacom demonstators, two uncoated CMCs were used as reference samples for the sake of comparison. Therefore, two uncoated Cf/SiC were joined together with a Ti3SiC2 pre-sintered foil using SPS at 1300°C and 50 MPa for 5 minutes.
• The SEM analysis revealed the formation of SiC reaction layer in the joining filler when the Ti3SiC2 interlayer was in contact with carbon fibres.
• The infiltration of the joining filler into the inter-spaces of carbon fibres was limited because of the limited number of free interspaces of Cf near the interface and due to the SiC reaction layer that acted as a diffusion barrier.
• Unlike for the coated CMCs (diffusion bonding only), this time the joints were obtained by the combination of solid- state reaction and diffusion bonding when the Ti3SiC2 pre-sintered foil was in contact with Cf and SiC, respectively. However, the mechanical tests showed that the apparent shear strength of the joined components was lower when compared to the coated CMC ones.
WP5 Micro-structural characterization of joints
The micro-structural characterization of all selected joined materials and interfaces has been done by using different techniques. The detailed investigation of materials allowed to define the most promising novel joining materials and techniques and to optimize the joining conditions. The activity within this task aimed at a complete microstructural characterization of all joined materials and interfaces developed in WP2, WP3 and WP4. It regarded also the investigation of the fracture surfaces after mechanical characterization according to WP6. To accomplish this task, the most suitable characterization techniques have been identified, shared among partners and extensively used. In summary, the following results were achieved in WP5:
✓ A full list of techniques for the micro/nanostructural characterization of joined components has been identified and shared among Partners.
✓ All the partners involved in WP5, CNR, POLITO, Fraunhofer and NanoForce performed standard and conventional characterization methods (e.g. optical and electron microscopy, EDS, XRD, etc.) on their own equipments routinely.
✓ Advanced techniques (e.g. nano tomography, TEM) were performed on selected samples when necessary and made available to Partners.
✓ Samples were exchanged among partners in order to perform tests and microstructural analyses using the most appropriate techniques.
✓ The results obtained within WP 5 were tremendously important for the entire ADMACOM project and essential to evaluate results and processes within other work packages. Without the analyzing techniques used inside this work package it would have been impossible to assess the quality of produced joints and interfaces.
✓ Measurement results like images, chemical compositions or phase assignments contribute to new knowledge when disclosed in publications.
✓ A few techniques, not planned in Annex I of the DoW, have been adopted (e.g. HR-SEM equipped with focused ion beam, non-contact profilometry) in order to improve the micro/nanocharacterization of joined interfaces. These methods and the related results constitute a further added value within the whole ADMACOM project.
In the following some examples of the most interesting and outstanding results obtained are provided:
Non-contact 3D optical profilometer (CNR): a multimodal non-contact 3D optical profilometer combining the confocal, interferometric and focus variation techniques has been acquired by CNR in the mainframe of the ADMACOM Project. Compared to other techniques (e.g. AFM) confocal profilometry allows the acquisition of relatively large surfaces at an adequate z scale in order to observe and measure the characteristic features of a surface. Thus, the peculiarities offered by the instrument have been employed, in particular, to analyze the surface characteristics of SiC/SiC composites with reference to the selective fiber removal to obtain the “brush-like” surfaces (surface engineering in WP3). The average depth of SiC fibers removal was measured by confocal profilometry to be 2.4 µm. For the surface engineered samples, the surface area increased of 89% and 129% for measurements performed on areas with fibers // or ⊥ to the surface respectively demonstrating the effectiveness of the fiber removal.
Microindentation (Nanoforce): in order to confirm that the joining interlayer remained intact and was not damaged during joining, microhardness of the interlayer was measured and compared to the microhardness of the initial Ti3SiC2 foil that was used as a joining filler. The results showed that the hardness in both cases was at the same level, confirming that neither decomposition nor reaction occurred between the interlayer and the base materials (CVD-SiC, CMCs).
Microindentation (CNR): microhardness measurements were performed by CNR on joined samples in the framework of WP6. The information acquired had not only a mechanical meaning but were useful from the microstructural examination point of view. In fact, the chemical nature of the phases observed and identified by SEM-EDS can be confirmed by microhardness values measured on cross-sectioned samples.
MicroCT analysis (Nanoforce, POLITO): this technique was used to investigate the small-scale demonstrator that consisted of the coated CMCs joined with the Ti3SiC2 interlayer. The analysis confirmed that defect-free interface was obtained and the Ti3SiC2 interlayer showed conformal behaviour as it bent and nicely copied the rough surface of the CMCs without breaking or cracking.
SEM-FIB and TEM (Nanoforce): the intact interfaces between the monolithic Boostec SiC and solders were investigated by SEM-FIB without being influence by cutting, polishing, etc. Thin lamellas prepared by FIB were then investigated by TEM. This helped to reveal that there was no additional reaction layer between the SiC and the solder. It also revealed the presence of a number of small well-separated pores along the interface that were not found by conventional SEM analysis.
Cryotest (POLITO): Si3N4/Si3N4 and Si3N4/INVAR joints (joining material: HYSOL EA9321) have been tested in order to reproduce the in service conditions and to study the evolution of microstructures during thermal cycles down to cryogenic temepratures. During each thermal cycle, the samples were cooled from room temperature down to 50 K then kept at 50 K for one hour. Finally, the cryocooler was switched off and the samples warmed up to room temperature. After the first thermal cycle, the samples were dismounted for a preliminary optical check. Then, the samples were mounted again and submitted to four consecutive thermal cycles. At the end, they have been dismounted and the joints have been tested mechanically.
WP6 Thermo-mechanical tests
Determination of the coefficient of thermal expansion (CTE)
CNR, Polito and Fraunhofer measured and collected the CTE on selected materials to be joined, on joining materials and obtained joints.
Microhardness measurements of Ti3SiC2
In order to confirm that the Ti3SiC2 joining filler remained intact after joining, Nanoforce performed microindentation to determine the hardness of both the initial pre-sintered Ti3SiC2 foil and the Ti3SiC2 joining interlayer. Neither decomposition of Ti3SiC2 nor reaction between the interlayer and matrix materials occurred during joining.
Influence of the laser treatment on the strength of SiC Boostec 100
Samples of monolithic SiC (type SiC 100, BOOSTEC) were delivered by Airbus Group to Empa to study the influence of the laser treatment on the strength. The chosen method was co-axial Ring on Ring test, based on DIN-EN 1288-5. A fractographic study was performed after testing to identify the fracture origin. The ring on ring measurements confirmed that there is no effect of the laser treatment on the strength of the SiC samples.
Single Lap Offset Internal Round Robin
The round robin test within the ADMACOM partners for single-lap offset test (SLO) was launched to determine the reproducibility of the test method and so to confirm and compare the experimental results with previous ones obtained by the industrial partner, Airbus. 4 partners (EMPA, POLITO, IFAM and Airbus) tested 10 samples each. The samples were manufactured and joined by EMPA. The shearing tool was designed by Airbus. The joining material was the same adhesive selected by Airbus for SiC joints, Hysol 9321.The material for the RR was structural steel S235.
The main conclusions of the Round Robin are:
• The Round Robin of SLO was successfully performed by 4 partners of the Admacom project without any outlier.
• All the samples of RR have the same type of failure mechanism, adhesive fracture at hysol/steel interface.
• The inter-laboratory SLO measurements were reliable within the intrinsic deviation of this method, ±3 MPa.
• SLO was chosen as comparative tool.
Apparent shear strength of SiC-SiC and SiC-Invar joints produced by active soldering
Fraunhofer worked on mechanical testing of SLO samples of the following material combinations produced by active soldering:
- SiC-SiC joints, in this combination it was studied, the influence of the thickness fillers, joining temperature and atmosphere on the mechanical properties. Best results with average values of 30.4 ± 6 MPa were achieved for samples joined under air atmosphere at 280°C-300°C and 50 µm filler thickness.
- SiC-Invar joints, in this combination it was studied, the influence of the joining thickness and joining temperature on the mechanical properties. Best results with average values 28 ± 4 MPa for samples joined under air atmosphere at 280°C and 25 µm joining thickness.
Mechanical characterization of SiC joined with Mo-Si based materials
SiC-SiC joints have been manufactured at POLITO using Mo-Si based material as joining material; details on the joining process are reported in WP4. Two different tests at RT have been carried out:
• Torsion test
• SLO test
4 joints have been tested by torsion test (geometry THG-4) and the average shear strength is 27 MPa (s.d. ±4 MPa). These results show very low values of mechanical strength for this kind of joint; it could be explained by possible misalignment of the tested samples, due to significant difficulties in their preparation. Concerning SLO tests, only two samples have been tested and the preliminary results show very high values of apparent shear strength.
Mechanical characterization of CMCs (C/SiC and SiC/SiC) joined with MoSi2-Si
POLITO tested by single lap test CMC joined with MoSi2-Si, the average value of the apparent shear value obtained was ~7 MPa, and the main problem during the measurements was the delamination of the CMC.
Mechanical characterization of C/SiC to Ti64 joints
C/SiC to Ti64 joints have been manufactured and characterized at POLITO. For the mechanical characterization of C/SiC to Ti64 joints manufactured using the brazing alloy TiB-590 the single-lap tests was used, but for all the tested samples the composites delamination occurred. Other tests were carried out by SLO: the average value of the apparent shear test is 43 MPa; also in this case, a critical issue is the delamination of the composite during the test.
SLO tests on joined laser structured SiC and Si3N4
Polito performed SLO tests on joined laser structured SiC and Si3N4 samples provided by AGI. All the joints were manufactured using HYSOL EA9321. The tested samples have been manufactured using different laser structuring parameters (LS), different adhesive thicknesses and different lapping conditions on facing surfaces before laser structuring process. The apparent shear strength of LS SiC is lower than that measured on lapped SiC. The apparent shear strength in the Si3N4 is higher for the laser structured joined samples if compared with the lapped reference sample without laser structuring. The mechanical test results confirm the effectiveness of laser structuring on the Si3N4 surface. Details on this activity are reported in D6.2.
Mechanical characterization of SiC/SiC samples before and after surface engineering treatment ( 3 point bending strength)
SiC/SiC samples have been subjected to 3-point bending tests at Polito before and after surface engineering treatment, i.e a selective oxidation process that involves a treatment at 1450°C for 2 hours in Ar atmosphere. The average value of the flexural strength of as received sample is 576 MPa, while heat treated samples showed a decrease of the mechanical strength to 387 MPa.
Torsional shear stress test at cryogenic conditions at elevated temperature on Hysol joined samples.
Empa adapted a torsional shear stress setup to perform measurements from cryogenic temperatures, -190°C up to 1200°C. Two adhesives used in aerospace industry were characterized between -190°C to 150°C. The torsional shear strength of the glue drops by 40% at 50°C with respect to its RT value. Cryogenic temperatures do not affect negatively the torsion shear strength.
Development of a shear stress fatigue test.
Empa developed a fatigue shear stress method based on the torsional shear stress test. The setup produces a peripherally uniform stress distribution on the joining area, the load can be adjusted and the samples can be cycled over 20,000 times per day. Fatigue tests on samples joined with hysol with a load of 90% have been performed, on average the samples failed after 60 cycles.
Single-lap offset tests on SiC joined by brazing alloy in collaboration with CNR
POLITO tested by SLO, SiC joined with TiAl by CNR – IENI( joining area 7 or 8 mm x 5 mm). The joining process consists of partial transient liquid phase bonding by Al-Ti interlayers (details of the bonding process is described in D4.2).
The average value of the apparent shear strength is more than 200 MPa; these high values of mechanical strength indicate good joints and demonstrate the usefulness of this joining method.
Simulations: effect of the meniscus geometry on the shear strength
Experimental shear tests were supported by macro-simulations with the aim to evaluate the influence of residual thermally induced stresses on the stress distributions in the chosen torsional shear test-setup. The basic model system consisted of 2 steel parts soldered with S-Bond 220-1. Soldering temperature was 250°C, solder thickness 0.2 mm, contact radius r = 4 mm, maximum torsional moment 250 Nmm.
The simulation showed that the superposition of the thermal and mechanical loads leads to reduced shear stress maxima in the solder in comparison to the mechanical load case alone. That means that the influence of the thermal history may lead to an underestimating of the real stresses resp. to an overestimation of the shear strength of the solder. On the other hand, the thermal stresses are highest near to the solder free surface which means that relaxation and micro-damaging can occur in this area before the mechanical test, leading to underestimating of the shear strength of the solder.
Simulations: effect of the geometry close to the joining area
The simulations of residual stresses originating from the soldering process and the mechanical stresses during shear test were extended in order to evaluate the influence of the specimen shape close to the soldering layer. Specifically, the cross-section geometry of the steel specimens was changed from straight (truncated cone) to circular with a radius of r = 4 mm. All other aspects of the model remained the same: it consisted of 2 steel parts soldered with S-Bond 220-1. For the simulations straight solder meniscus geometry was assumed.
As expected, the geometry variation affected the shear stress in the solder only in the vicinity of the triple contact lines steel-solder-atmosphere. A smooth geometrical transition of solder and specimen is advantageous, though the overall effect is comparatively small. In comparison to the simulations done before (variation of solder meniscus and contact angle) it can be concluded that for the chosen configuration, meniscus shape and contact angle have a higher influence on the stress situation in the solder than the specimen shape.
SLO and Torsional shear strength measurements with different roughness values
With the purpose to check the sensitivity of the different shear strength testing methods, Empa prepared samples of steel S235 with different surface roughness, the specimens were ground with different SiC papers, and the roughness was measured. The results show a clear increase of the shear strength with the roughness of the samples, this tendency is clearer in the measurements performed with the torsional setup than with the SLO setup.
Mechanical joints characterization at elevated temperatures of SiC Incusil ABA brazed.
The behavior of SiC ceramic joints brazed with commercially available Incusil ABA (Ag-32.25Cu-12.5In-1.25Ti in wt.%) was characterized by Empa with respect to the mechanical performance at temperatures up to 550°C using four-point bending and torsional shear tests. In four-point bending tests we interestingly observed a strong retention in bending strength up to 300°C with only a slight 10% decrease. With the torsional shear apparatus it was not possible to characterize the joint accurately below 450°C as the samples always broke in the ceramic part. This is due to a combination of the accumulation of very high localized stresses at contact points where the shear load is transferred and the low KIc of the material, initiating fracture in the ceramic. What is remarkable is the high values of the shear strength obtained, which are higher than the bending strength at the same temperatures, indicating both that a more complex stress system exists during bending strength tests and the adverse effects of the tensile stresses.
Torsional shear strength measurements of Steel S235 glued with Hysol 9321 with different radius
In all the cases the measured shear strength is higher than the one measured in the SLO RR for the same materials, 41±3 MPa reported in the D6.2. The values are especially high in the cases of specimens with an outer diameter of 5 mm but also the deviations of the measurements are higher. For large outer diameter, 14 and 20 mm, the measured shear strength values remain for all the inner diameters around 50 MPa. When the inner diameter approaches the size of the outer diameter, the deviation of the measurements increases, probably due to the difficulty to control the glue thickness, lower thickness can increase the values of the shear strength.
Simulations: Effect of the radius on the residual stresses
Macro-simulations were performed with the aim to evaluate the influence of the specimen shape close to the joining layer and the influence of the radius size on the residual stresses in the chosen torsional shear test-setup. The basic model system consisted of 2 steel parts glued with Hysol 9321, a glue thickness of 0.2 mm, joining area of 5 mm diameter, mesh size of 0.1 mm and a maximum torsional moment of 250 Nmm. Two geometries were compared Empa geometry with chamfered angle of 45, and a geometry with a chamfered radius of r = 4 mm. The size radius dependence simulation was performed with the Empa geometry.
As in the previous simulations, the geometry variation affected the shear stress in the vicinity of the triple contact lines steel-glue-atmosphere. A smooth geometrical transition of glue and specimen is advantageous. The results show that the larger the radius of the specimens, the less shear stresses in the vicinity of the triple contact point are observed.
Comparison of different configurations to measure the joint strength (shear and apparent shear).
Empa and Polito compared different test methods to measure the joint strength (shear and apparent shear) of different combinations of joining materials and substrates. The aim of this activity is to validate the torsional shear stress test and collect information for the torsional shear stress round robin.
Recommended practice for determining the torsional shear stress of joined dissimilar materials.
Empa in collaboration with Polito submitted a first draft of the recommended practice to the European Structural Integrity Society (ESIS) a non-profit organization. The purpose of the document is dissemination of the practice as basis of the planned Round Robin on torsional shear stress test.
WP7 Manufacture of demonstrators, repair, maintenance and non-destructive tests
The aim of this WP is to validate what done in the previous ones, and to start technology transfer to the industrial partners. It is focused on:
(i) construction of the demonstrators with the novel manufacturing technology,
(ii) validation of the structural behavior and ease of implementation in industrial environment. Proof of concept in terms of at least one demonstrator per joining based manufacturing technology, demonstrating here the scalability towards industrial needs and pilot implementations in industrial settings;
Life Cycle Impact Assessment of the demonstrators
In the last period of the project, all involved partners manufactured the final demonstrators and pre-prototypes for both industrial partners, AGI and MTA. This is significant, because effectiveness of all new promising joining techniques developed within WP4 was verified by the construction of the final pre-prototypes and demonstrators. It was confirmed that all selected joining techniques were appropriate for manufacturing the following final demonstrators and pre-prototypes. This is considered as the promising final output of the project, all according to DoW. However, the joining of curved surfaces for MTA demonstrators did not lead to the desirable results and more process optimisation is still required.
• SiCf/SiC to SiCf/SiC: flat to flat surfaces, 25 mm x 10 mm, were joined with MoSi2-Si composite by the presureless technique (POLITO)
• Cf/SiC to Cf/SiC: flat to flat surfaces, 10 mm x 10 mm, were joined with MoSi2-Si composite by the presureless technique (POLITO)
• Cf/SiC to TiAl6V4: flat to flat surfaces, 30 mm x 30 mm, were joined with Cusil/ABA by the pressureless technique (POLITO)
• Cf/SiC to Cf/SiC: flat to flat surfaces, 10 x 10 mm, were joined with Ti3SiC2 using SPS process (Nanoforce)
• SiCf/SiC to SiCf/SiC: flat to flat surfaces, 10 x 10 mm, were joined with Ti3SiC2 using SPS process (Nanoforce)
• On basis of EMA rod a tubular shaped demonstrator has been manufactured and machined and on basis of the really complex shaped bodyflap, T- and angle-shaped parts (>5 parts) have been manufactured as well as finally one (1) representative component for the box design of the flap; a combination of edges and angles, dimension ca. 350x200x200 mm.
• Boostec SiC to Boostec SiC, flat to flat lapped surfaces, 25 mm x 25 mm x 5 mm, were joined with Sn91Ag4Ti4 using active soldering (FRAUNHOFER)
• Boostec SiC to Invar, lapped surface of SiC was joined to the laser-modified surface of Invar with Sn91Ag4Ti4 using active soldering (FRAUNHOFER, AGI). This constitutes the final demonstrator, which is a satellite part where a Tooling Hole Boss (Invar alloy) is joint to a Boostec SiC plate.
• Si3N4 to Si3N4 and Si3N4 to Invar, flat to flat lapped (reference) and laser structured surfaces, 25 mm x 25 mm x 10 mm, were joined with adhesive HYSOL® EA 9321 (AGI)
Life Cycle Impact Assesment
Life Cycle Impact Assessment (LCIA). This step involves the evaluation of environmental impacts associated with the system outputs as well as the consumption of resources related to the inputs. In this phase there is the changeover from the objective data, collected during the inventory phase, to the judgment of environmental hazard.
The goal of this study is to analyse the environmental burden of two innovative joining processes for advanced materials used in the aerospace sector.
In particular, the assessment covers:
• An active brazing process to join a silicon carbide based composite material (C/SiC) and a titanium alloy (Ti6Al4V). An Active Braze Alloy (CuABA) is used and the process runs at a temperature around 900°C.
• An adhesive bonding process to join silicon nitride (Si3N4) and an INVAR alloy, by means of an epoxy adhesive at room temperature (for 8 days).
In both cases, a surface treatment is required before the actual joining of the two materials. In the case of the active brazing process, a selective oxidation is performed on the C/SiC while the Ti6Al4V alloy is machined. As far as the adhesive bonding is concerned, on both materials (Si3N4 and INVAR alloy) two alternative surface treatments are used: a mechanical lapping and a laser surface nanostructuring.
The processes under study are currently at lab/pilot scale. Therefore, the related impacts are directly comparable to optimized industrial processes.
Most of the data were directly measured and collected by the partners running the processes. Where primary data were not available, secondary data were used, referring to the ecoinvent database .
The present assessment was performed using the SimaPro 8.1 software (PRé Consultants), that integrates the ecoinvent database. The Impact Categories chosen are the most representative for industrial processes and among the easiest to be communicated and understood also by a non-technical audience. In particular, they are the Cumulative Energy Demand (CED) and the Global Warming Potential (GWP, also referred as Carbon Footprint).
The Life Cycle Inventory analysis involves the creation of a model of each process under study. In particular, two cases are considered:
• Case 1: it concerns the active brazing process used to join C/SiC and the Ti6Al4V alloy.
• Case 2: it concerns the adhesive bonding process used to join Si3N4 and the INVAR alloy. This case study is further divided in two alternative scenarios, according to the surface treatment used before the actual joining of the two materials:
o Case 2a: the surface treatment used on both the Si3N4 and the INVAR alloy is Lapping;
o Case 2b: the surface treatment used on both the Si3N4 and the INVAR alloy is Laser Nanostructuring.
LCIA conclusions
From the analysis of the results analysis, a common outcome is that the energy consumption of each process represents the major source of the environmental impact. This implies that a higher degree of sustainability is straightforwardly related to a lower energy consumption. Therefore, it is evident that high temperature processes (Case 1) lead to a higher impact with respect to those performed at room temperature (Case 2).
Another leitmotiv emerged from the results is the fact that surface pre-treatment plays a major role in the joining processes analysed. In fact, in Case 1, the selective oxidation of the C/SiC produce the 66,67% of the impacts. In Case 2, this consideration is even more radical since a change in the laser pre-treatment brings to a 99.8% saving in the overall environmental burden of the joining process.
The environmental impact of the active brazing is moderate, but not low in general terms. To give a daily life reference, the CO2 emissions produced by the joining of 1 cm2 of materials correspond to the CO2 emissions produced by a person travelling for about 50 km by car. The same reasoning applied to the adhesive bonding, with Lapping pre-treatment, leads to an equivalence of 1.7 km by car, while, with Laser Nanostructuring, the process impacts as much as 2.88 m travelled by car .
It is important to underline that the functional unit does not consider changes in the quality of the joining process. A combination of the environmental results and the mechanical performances will lead to a more comprehensive and consistent comparison among processes. In particular, if the performance of a joining process, it can be evaluated if the same mechanical performance of the aerospace component can be reached with a lower extension of the joins. In this respect, the results of this analysis are intended to be ready to be used as inputs in such evaluation.
Potential Impact:
The ADMACOM impact can be summarized as follows:
1. Development of new and exploitable manufacturing technologies based on joining of hybrid components made of ceramic matrix composites (C/SiC) joined to metals (Ti alloy) and ceramics (SiC and Si3N4) to metals (Invar). The impact can be seen in enabling new design solutions and obtaining light-weight structural demonstrators for aerospace applications with improved reliability and in-service performance, in term of weight reduction, fuel consumption, pollution reduction and high temperature resistance.
2. Improved competitiveness of European industries via more favourable cost/benefit solutions due to the use of CMC and advanced ceramics joined to existing and reliable traditional materials (metal alloys).
3. The Lyfe Cycle Impact Assessment results showed that the laser-pretreatment of ceramic and metal surfaces, before adhesive bonding, corresponds to 99.8% carbon footprint reduction of the joining process in comparison with the reference surface treatment (lapped surfaces).
4. Acquisition of new skills in the European workforce resulting also in indirect socio-economic benefits: this has been done through two Advanced Training Modules on mechanical characterization of joined components.
5. Significant effort have been done for standardization of mechanical tests (shear strength) of joined components, in cooperation with EU and national standard and measurements agencies. It must be underlined the high impact that a EU standardization of mechanical test procedures on joined components to support reliable design and foster the development of new industrial components where different materials are joined together.
Finally, the potential areas and markets of application of ADMACOM will be in the field of high performance aerospace systems, including satellites. A reliable CMC or ceramic to metal joined component is an emerging market for lightweight technologies. The joining of these advanced ceramics and CMC to other dissimilar materials can be undertaken using mechanical fasteners, but such an approach is not a cost effective and a technologically reliable solution. The new joining technologies investigated in ADMACOM can offer benefits with regards to reduced mass, reduced radar signature, reduced costs through simpler construction, improved fatigue performance and a more uniform stress distribution.
The output of ADMACOM will be considered for industrial partners applications well beyond those described in ADMACOM, in particular advanced ceramics/metal and CMC/metal joints could impact on the production of:
• space industry/Re-entry vehicles/Capsules/Landers, namely thermal protection systems/panels, heat shields, insulation, nose caps, corner edges, structural and moveable parts like flaps, rudders, leading edges, fixation of sensors/instrumentation.
• Aerospace civil-military/Propulsion Systems, namely turbines and turbine environment, jet engines, housings, nose caps, leading edges, fins, exhaust nozzles, combustion chambers or parts for combustion chambers, nozzles for small thrusters, and fixation of sensors/instrumentation.
• Power generation/turbines/Renewable industry/Marine, namely pump and bearing environment, insulation, environmental industry, chemical industry, renewable industry (power generation, wind, water and sun).
OUTREACH ACTIVITIES:
- ADMACOM participated in the 10th European Researchers’ Night, 25 September 2015, Turin and in the 11th European Researchers’ Night, 30 September 2016, Turin (POLITO and AGI).
- The ADMACOM newsletters were periodically published in the ADMACOM website.
- FRAUNHOFER and AGI prepared a presentation of the “Active soldering of ceramic-ceramic and ceramic-metal compounds” developed within the ADMACOM project. This informative material was published and disseminated through the ADMACOM website.
- The ADMACOM Industrial workshop was held on September 14th - 15th, 2016 in Brussels. The aim of the ADMACOM workshop was to present innovative manufacturing technologies based on advanced design of interfaces and joining materials for aerospace components http://www.admacomproject.eu/article/admacom-workshop-september-14th-15th-2016.
- POLITO participated in the METAL MORPHOSIS project seminar 24th February 2016 at the BWI - Belgian Welding Institute, Ghent and attended and exhibited at Composites Innovation 2016 (Sheffield, UK 22-23 June 2016).
A very intensive dissemination activity was done and planned during the last 18 months: project website, newsletter, conferences, workshops. The project website contains also the list of conferences at which ADMACOM partners disseminated scientific results. Advanced training modules have been organized on demand, on the request of the involved companies.
List of Websites:
http://www.admacomproject.eu/