Nederlandse Organisatie voor toegepast-natuurwetenschappelijk onderzoek /Netherlands Organisation TNO Industrial Technology for Applied Scientific Research Centrum voor Lichtgewicht Constructies TUD-TNO / Centre of Lightweight Structures TNO report TUD-TNO Kluyverweg 3 2629 HS DELFT Experimental Resin Flow Investigation for Vacuum Netherlands Infusion www.clc.tao.nl T +3115278 1778 Contract order number F61775-02-WE044 F +3115 278 72 99 Date 13 January 2003 Author(s) M. Labordus, R.C. Verhoef Copy no I No. of copies 3 Number of pages 25 Number of appendices - Sponsor European Office of Aerospace Research and Development, Air Force Office of Scientific Research Laboratory Project name Experimental Resin Flow Investigation for Vacuum Infusion Project number 007.62465.01.01 All rights reserved. No part of this publication may be reproduced and/or published by print, photoprint, microfilm or any other means without the previous written consent of TNO. In case this report was drafted on instructions, the rights and obligations of contracting parties are subject to either the Standard Conditions for Research Instructions given to TNO, or the relevant agreement concluded between the contracting parties. Submitting the report for inspection to parties who have a direct interest is permitted. C 2002 TNO Copy Best Available 070 20030610 AQFo31-o?-/43c)- REPORT DOCUMENTATION PAGE Form Approved OMB No. 0704-0188 Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing the burden, to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS. 1. REPORT DATE (DD-MM-YYYY) 2. REPORT TYPE 3. DATES COVERED (From - To) 16-01-2003 Final Report 6 July 2002 -06-Dec-02 4. TITLE AND SUBTITLE Sa. CONTRACT NUMBER F61775-02-WE044 Vacuum infusion of low-cost aerospace composites 5b. GRANT NUMBER 5c. PROGRAM ELEMENT NUMBER 6. AUTHOR(S) 5d. PROJECT NUMBER Mr. Maarten Labordus 5d. TASK NUMBER 5e. WORK UNIT NUMBER 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION TNO Industrie REPORT NUMBER Kluyverweg 3 Delft 2629 HS N/A The Netherlands 9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR'S ACRONYM(S) EOARD PSC 802 BOX 14 11. SPONSOR/MONITOR'S REPORT NUMBER(S) FPO 09499-0014 SPC 02-4044 12. DISTRIBUTION/AVAILABILITY STATEMENT Approved for public release; distribution is unlimited. 13. SUPPLEMENTARY NOTES 14. ABSTRACT This report results from a contract tasking TNO Industrie as follows: A mold system will be designed in which the hydrostatic pressure can be measured in the resin during the vacuum infusion process. Pressure transducers will be mounted in the surface of a flat aluminum plate, and data will be digitally recorded. Simple compression tests on dry and impregnated laminate will provide data to relate the pressure gradient (out of plane) to the local laminate thickness. From a combination of the pressure data and the compression data, the correction factor on Darcy's Law can empirically be determined. Outputs of the research will be: 1. Empirical quantification of pressure gradient during vacuum infusion and Resin Transfer Molding (RTM); 2. Correction of Darcy's Law;, 3. Practical test to measure compressibility of reinforcements and translation of data into above-mentioned correction term. 15. SUBJECT TERMS EOARD, Materials, Composites, Polymers, Materials Process Design 16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF 18, NUMBER 19a. NAME OF RESPONSIBLE PERSON ABSTRACT OF PAGES Charles H. Ward, Lt Col, USAF a. REPORT b. ABSTRACT c. THIS PAGE UL UNCLAS UNCLAS UNCLAS 25 19b. TELEPHONE NUMBER (Include area code) +44 (0)20 7514 3154 Standard Form 298 (Rev. 8/98) B est AIvailab e Prescribed by ANSIStd. Z39-18 TNQ report 12/25 Declarations This material is based upon work supported by the European Office of Aerospace Research and Development, Air Force Office of Scientific Research Laboratory, under Contract No. F61775-02-WE044 Any opinions, findings, and conclusions expressed in this material are those of the authors and do not necessarily reflect the views of the European Office of Aerospace Research and Development, Air Force of Scientific Research, Air Force Research Laboratory The Contractor, TNO Industrie, hereby declares that, to the best of its knowledge and belief, the technical data delivered herewith under Contract Number No. F61775-02- WE044 is complete, accurate, and complies with all requirements of the contract. I certify that there were no subject inventions to declare as defined in FAR 52.227-13, during the performance of this contract. Date: Name and Title of Authorised Official: Best Available Copy TNO report 1 3/25 Contents 1 Introduction 4 2 Introduction to resin flow 6 3 Review of vacuum infusion research 8 3.1 Compaction of the reinforcement 8 3.2 Permeability 10 3.3 Resin flow analysis 11 4 Subjects for experimental investigation 13 4.1 Comparison vacuum infusion with RTM 13 4.2 Pressure gradient in laminate during vacuum infusion 13 4.3 Compaction of the reinforcement 13 4.4 Effect of flow enhancement material on flow and pressure gradient 14 5 Description of test set-up and procedure 15 5.1 Test procedure 15 5.2 Test Equipment specifications 17 6 Results 18 7 Conclusions and recommendations 23 8 References 25 Best Available Copy TNO report 1 4 /25 1 Introduction Resin Transfer Moulding (RTM) processes are characterised by a fibre preform laid-up in a closed mould system and a pressure difference to drive the resin in the mould cavity. Traditional RTM, which uses high pressures at the resin inlet to inject the resin, has been successfully applied in the aerospace industry and in the automotive industry. RTM can result in high quality composite parts, however there are many drawbacks, such as high costs for tooling and the inflexibility of the process with respect to (minor) design changes. The vacuum infusion process, a low-cost RTM variant, was developed especially for the boat building industry. Instead of two solid mould halves and high pressures, one single solid mould halve is used, combined with an airtight flexible film and vacuum pressures to force the resin to flow. This vacuum infusion process was originally introduced as a sound alternative for open-mould processes like spray-up and hand lay- up. Besides the intended environmental and health benefits, a considerable quality improvement was noticed. Specifically, the fibre volume content was increased and the void content decreased. This led to the assumption that the vacuum infusion process could be developed into a low-cost alternative for both the RTM and the prepregging process. Figure 1 illustrates the vacuum infusion principle and the basic equipment. The dry reinforcement is placed on a mould and is subsequently covered with a flexible foil, which is sealed at the edges leaving room for the resin inlet and outlet. During infusion the resin flows from inlet, through the fibre reinforcement and subsequently the outlet. vacuum seal foil dry reinforcement resin inlet resin outlet mould Figure 1: Schematic overview of vacuum infusion process An important aspect of both RTM and vacuum infusion is the injection strategy. An injection strategy should be developed for each product in order to establish that the part may be filled within a reasonable cycle time (e.g. within the geltime of the resin) and that all risk of air enclosures is eliminated. Resin flow simulation software tools have been available for many years and have proven to be a useful tool to validate injection strategies. The parameters involved can be related by Darcy's Law, which serves as the basis for all resin flow simulation software. However, the underlying models used in these software tools are all based on resin flow in a stiff mould system (so on the RTM process) whereas for the vacuum infusion process the resin flow is influenced by the flexible mould halve. Best Available Cop'." TNO report I 5/25 In order to get a better understanding of the resin flow during the vacuum infusion process, an experimental research program was initiated, during which vacuum infusion tests were performed on different types of reinforcement materials and different lay- ups. The test set-up was designed to provide a complete overview of the resin flow parameters. This reports presents background information on the vacuum infusion process, interesting focus points for experimental research and the development of a complete test set-up. The results from the tests are presented and related to existing vacuum infusion theories. Best Available Copy TNO report I 6/25 2 Introduction to resin flow The mathematical analysis of resin flow through a fibre preform is based on the theory of (viscous) flow through porous media, which was formulated by Darcy. Darcy's Law, combined with the continuity equation of two-dimensional incompressible flow and the appropriate boundary conditions form the basis of resin flow simulation. Applied to a one-dimensional flow, Darcy's law is: Q = KA (1) P Ax Where Q = volumetric flow rate [m3/s] p the fluid viscosity [Pa-s] A = the cross section area [mr2]; Ap/lx = the pressure gradient in the flowing resin [Pa/m] K = the reinforcement permeability [m2] From Darcy's law we can conclude the following about the basic behaviour of the resin flow: "* The larger the pressure difference, the quicker the part is filled; "* The lower the resin viscosity, the quicker the part is filled; "* The higher the permeability, the quicker the part is filled. The first two parameters, pressure difference over the mould and resin viscosity, may be considered constant during infusion of the product and can be approached similarly to the ordinary RTM process. However, it is obvious that for vacuum infusion the maximum achievable pressure difference is about 1000 mbar. Related to this relatively small pressure difference (compared to ordinary RTM), low viscosity resin systems have been developed to enhance the flow rate and subsequently reduce the fill time. The permeability is a geometrical property of the reinforcement, which describes the ease of flow through the material. The permeability can be determined from experiments, but can also be predicted by the Kozeny-Carman model. The model relates the permeability to the porosity, but in case of fibrous media, the fibre volume fraction is used. 1 0(-Vfs (2) K C.S2 V 2 Where K = permeability [m2] C = Kozeny constant S = specific surface area [m2/m3] Vf= fibre volume fraction TNO report I 7/25 With vacuum infusion, a solid mould halve is combined with a flexible film to close the mould. Due to the vacuum applied to force the resin through the reinforcement, the reinforcement itself is also compressed. Due to this compression, the fibre volume fraction is increased and thus the permeability of the reinforcement is decreased. A decrease in permeability in its turn alters the pressure gradient and subsequently the resin flow. Simulation software does usually not take into account the influence of the compression of the reinforcement and its related effects. Neglecting these effects introduces errors in the resin flow simulation. For critical production processes (complicated 3D products or infusion with short resin geltimes) this can lead to incorrect judgements. TNO report 1 8/25 3 Review of vacuum infusion research The vacuum infusion process is a promising manufacturing concept for multiple applications. This has been recognised by the boat building industry, automotive industry and recently by the aerospace industry. This has resulted in several patented manufacturing processes based on similar concepts. The first patent dates back to 1950 with the production method called the Marco Process. A complete overview of the historical development of the vacuum infusion process is presented by C.D. Williams et al. [1]. A useful and extensive general overview of the vacuum infusion process (cost comparison, process applications, basic resin flow theory, injection strategy) is provided by Hoebergen and Holmberg [2]. More fundamental research on the vacuum infusion process focuses on the one thing that distinguishes vacuum infusion from the other RTM processes: the flexible mould halve. This flexible mould halve influences the production process by permitting compaction of the reinforcement and, thus, causing non-constant permeability. These influences on the production process also affect the product itself. In comparison with solid moulding RTM, the thickness of the product is not simply given by the dimensions of the mould cavity; it is also dependent on the compressibility and relaxation of the reinforcement under pressure. 3.1 Compaction of the reinforcement Experiments have been performed on dry fabric to understand the compaction and relaxation behaviour. Many different models have emerged from these experiments. An example of experimental work on this subject is given by Robitaille et al. [3]. The representation of the compaction and relaxation of the fabric is based on the following relations: Vf =A .pB (3) -P = I_-C.t(IID) (4) P 0 Where Vf = fibre volume fraction P = compaction pressure [N/n2] Po = initially applied pressure [N/m2] t = time [s] A = fibre volume fraction for a pressure equal to 1 Pa. B = compaction stiffening index (B<I) C = pressure decay after 1 sec. D = relaxation index The compaction of the reinforcement can also be expressed in the form of the thickness or the porosity of the laminate. The results of the experiments presented by Robitaille are all expressed in terms of the indices A through D. Comparison of the results for different testing parameters led to some general trends. For example: as the number of layers increases, the initial fibre volume fraction (A) increases while the stiffening Best Av.a IbI .... TNO report 1 9/25 index (B) decreases for both random mats and woven rovings. Subsequent research [4] also focused on the compaction and relaxation of reinforcement material in fluid- saturated condition. In general, the most pronounced effects on the compaction and relaxation of fibrous reinforcements are found to be, respectively, the number of cycles and the compaction rate. The lubrication of the reinforcement seems to have the greatest influence on the time-dependent relaxation. The aforementioned experimental work was performed on a specially designed test set- up using compacting pressures up to 1.0 MPa and compaction speeds up to 1 m/n/mn. These high pressures and high speeds are only approachable when using a solid mould RTM process. However, the relations described above are also found to be viable for the lower pressure range (0-0.1 MPa), which is used for the vacuum infusion process. Power law relations are used in a number of vacuum infusion research programmes to model the compaction of the reinforcement material, e.g. Hammami et al. [5] and Ragondet et al. [6]. An extensive description of the compaction behaviour of a number of reinforcement materials under vacuum infusion conditions is given by Hammami [7]. In comparison to the experiments described above, the compaction pressure does not exceed 1 bar and the compaction rate is much lower (max. 2 mm/min). Figure 2 gives an example of compaction test data for a specific fibrous reinforcement material. The graph shows the typical exponential relation between the applied pressure and the fibre volume fraction, which reaches its maximum around 0.55. Also, differences are noticeable for the dry or wet situation and different compaction rates (which vary from 0.05 mm/nmin to 2.0 mm/min). 0 ....... L-900 E11-O.O5mm/min-dry i 0.10] - -- LO-E 1-2mmlmin-dry " /I .I;- - 0. L0O9E11-0.O5mn/min-wet . o - ...- .L900-Eli-2mmblmn-wet [. .. 0.06 ' .2 0 .04 LD . . .. . ..... 3 -0.02-2 - 0.0 0.35 0.4 0.45 0.5 0.55 0.6 Fibre volume fraction Figure 2: Reinforcement compaction test data (taken from [7]) for a 884gr/m2 0°/900 fabric from Devold Hammarni describes compaction results for various reinforcement materials, and also for preforms with flow enhancement layers. Each material with or without flow enhancement layers has its own distinct behaviour. However, in the presence of a flow enhancement layer the compaction behaviour seems to be dictated by this layer (NB. in the experiments only Rovicore and Multimat were evaluated). In accordance with the