See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/320446584
`
`Thermoforming of HDPE
`
`Conference Paper  in  AIP Conference Proceedings · October 2017
`
`DOI: 10.1063/1.5008069
`
`CITATION
`1
`
`4 authors, including:
`
`G.H. Menary
`Queen's University Belfast
`
`69 PUBLICATIONS   753 CITATIONS   
`
`SEE PROFILE
`
`READS
`2,853
`
`Shiyong Yan
`Queen's University Belfast
`
`20 PUBLICATIONS   107 CITATIONS   
`
`SEE PROFILE
`
`Some of the authors of this publication are also working on these related projects:
`
`PetalRod - A novel method of light-weighting iSBM bottles View project
`
`Thermoforming of Thermoplastic Composites (Experimental & Numerical) View project
`
`All content following this page was uploaded by Shiyong Yan on 14 November 2017.
`
`The user has requested enhancement of the downloaded file.
`
`MacNeil Exhibit 2152
`Yita v. MacNeil IP, IPR2020-01139, Page 1
`
`

`

`Thermoforming of HDPE
`David McKelvey, Gary Menary, Peter Martin, and Shiyong Yan
`
`Citation: AIP Conference Proceedings 1896, 060006 (2017);
`View online: https://doi.org/10.1063/1.5008069
`View Table of Contents: http://aip.scitation.org/toc/apc/1896/1
`Published by the American Institute of Physics
`
`MacNeil Exhibit 2152
`Yita v. MacNeil IP, IPR2020-01139, Page 2
`
`

`

`Thermoforming of HDPE
`
`David McKelvey1, a), Gary Menary1, b), Peter Martin1 and Shiyong Yan1
`
`1 The Queen’s University of Belfast, School of Mechanical and Aerospace, Ashby Building, Belfast, BT9 5AH,
`Northern Ireland
`a)dmckelvey06@qub.ac.uk
`b)g.menary@qub.ac.uk
`
`Abstract. The thermoforming process involves a previously extruded sheet of material being reheated to a softened state
`below the melting temperature and then forced into a mould either by a plug, air pressure or a combination of both.
`Thermoplastics such as polystyrene (PS) and polypropylene (PP) are commonly processed via thermoforming for products
`in the packaging industry. However, high density polyethylene (HDPE) is generally not processed via thermoforming and
`yet HDPE is extensively processed throughout the packaging industry. The aim of this study was to investigate the potential
`of thermoforming HDPE. The objectives were to firstly investigate the mechanical response under comparable loading
`conditions and secondly, to investigate the final mechanical properties post-forming. Obtaining in-process stress-strain
`behavior during thermoforming is extremely challenging if not impossible. To overcome this limitation the processing
`conditions were replicated offline using the QUB biaxial stretcher. Typical processing conditions that the material will
`experience during the process are high strain levels, high strain rates between 0.1-10s-1 and high temperatures in the solid
`phase (1). Dynamic Mechanical Analysis (DMA) was used to investigate the processing range of the HDPE grade used in
`this study, a peak in the tan delta curve was observed just below the peak melting temperature and hence, a forming
`temperature was selected in this range. HPDE was biaxially stretched at 128°C at a strain rate of 4s-1, under equal biaxial
`deformation (EB). The results showed a level of biaxial orientation was induced which was accompanied by an increase in
`the modulus from 606 MPa in the non-stretched sample to 1212MPa in the stretched sample.
`INTRODUCTION
`The thermoforming process involves a preformed sheet of material being reheated to a softened state, deformed
`into a mold via a plug or air pressure, cooled and ejected as shown in Fig.1. The process is ideal for the large scale
`mass production of thin gauge packaging items such as trays, pots and tubs. The reason for this is that during
`processing a level of biaxial orientation is induced, which improves the mechanical properties of the formed part and
`hence, enabling the thickness of the final part to be reduced without compromising the final mechanical properties.
`The softening temperature is a critical aspect of the process as the sample must be readily formable but still maintain
`its structural integrity, as overheating can potentially lead to sagging which results in significant variations in the
`thickness profile of the part. Amorphous polymers such as High Impact Polystyrene (HIPS) can be thermoformed
`relatively easily, due to wide softening temperature range and the relatively simple microstructure. Whereas, semi-
`crystalline polymers such as polypropylene (PP) are significantly harder to process due to the highly temperature-
`dependent crystalline structure. However, due to the enhanced mechanical and chemical properties offered by semi-
`crystalline it is very desirable to process them via thermoforming.
`High density polyethylene (HDPE) like PP is a semi-crystalline polymer which is widely processed within the
`packing industry with 5.5 million tones processing in 2013 (1) however, while PP is widely processed via
`thermoforming HDPE is not. HDPE is typically processed from the molten state via processes such as extrusion blow
`molding, injection and rotational molding. While melt processing HDPE is now standard practice, a few studies have
`highlighted that the mechanical properties can be further enhanced by processing below the melting temperature
`(2)(3). For example, Li et al (2) were able to increase the yield strength of extruded HDPE from 28MPa when extruded
`from the molten state to 181MPa when extruded below the melting temperature, this becomes particularly interesting
`when considering thermoforming HDPE which must be below the melting temperature. However, while this clearly
`
`060006-1
`
`Proceedings of the 20th International ESAFORM Conference on Material Forming
`AIP Conf. Proc. 1896, 060006-1–060006-6; https://doi.org/10.1063/1.5008069
`Published by AIP Publishing. 978-0-7354-1580-5/$30.00
`
`MacNeil Exhibit 2152
`Yita v. MacNeil IP, IPR2020-01139, Page 3
`
`

`

`highlights the potential there is currently a lack of literature published on the response of HDPE under comparable
`processing conditions i.e. high temperature, high-rate (0.1-10s-1) equal biaxial deformation (4). The understanding of
`the materials response it vital when it comes to setting up the process. The QUB biaxial stretcher has previously been
`used to apply representative thermoforming processing conditions for HIPS, PP and aPET (5,6), where a key outcome
`from this work was an understanding of the temperature and strain-rate dependence of the materials.
`The aim of this study was to investigate the potential of thermoforming HDPE, by applying representative
`processing conditions offline. The objectives were to firstly investigate the mechanical response under comparable
`loading conditions and secondly, to investigate the final mechanical properties post-forming.
`
`Preformed
`sheet
`
`Heated
`sheet
`
`Plug Stretches
`Sheet
`
`Part cooled and
`ejected
`
`Sheet
`
`Plug
`
`Mold
`
`FIGURE 1. Thermoforming process.
`
`EXPERIMENTAL
`
`Material
`HDPE samples with dimensions 76x76x2mm were injection molded using an Arburg 320S Allrounder 500-350
`machine. The injection temperature was 235°C and the injection pressure was 85MPa, samples were held in the mold
`for 15seconds before ejection and were allowed to cool to room temperature. The HDPE resin used in this study had
`a Mn of 28623 Daltons, Mw of 151174 Daltons and Mz of 850210 Daltons. A Perkin Elmer DSC6 was used to analyze
`10mg samples of HDPE, cut from the injection-molded sheet. The sample was sealed in an aluminum pan and heated
`at a controlled rate along with an empty reference aluminum pan. The reference sample was subtracted from the final
`results to account for the heating of the pan. Both samples were heated to 180°C, to ensure the sample was completely
`melted. The endothermic melting peak obtained via DSC is shown in Fig.2a along with the baseline used to determine
`the degree of crystallinity. The crystallinity was determined by dividing the area under the endotherm by the enthalpy
`of fusion for PE, which was taken as 293J/g (7). This was repeated three times and an average was taken for the
`degree of crystallinity and peak crystalline melting temperature, which were 64% and 132°C respectively, for a heating
`rate of 10°C per minute.
`
`DMA
`Dynamic Mechanical Analysis (DMA) was conducted using a Triton Tritec 2000 DMA. Specimens of dimensions
`25 x 7.75 x 1.85mm were loaded in dual cantilever configuration with a span length of 15mm. Temperature sweeps
`at constant frequency of 1Hz and displacement of 0.025mm were conducted between 35°C and 135°C at a rate of
`1°C/min. The tan delta curve is shown in Fig.2b as a function of temperature.
`
`060006-2
`
`MacNeil Exhibit 2152
`Yita v. MacNeil IP, IPR2020-01139, Page 4
`
`

`

`0.50
`
`0.40
`
`0.30
`
`0.20
`
`0.10
`
`0.00
`
`tan delta
`
`35
`
`55
`
`115
`
`135
`
`95
`75
`Temperature (°C)
`(b)
`FIGURE 2. (a) DSC melting endotherm with baseline used for crystallinity calculations. (b) tan delta curve obtained from DMA
`
`(a)
`
`Biaxial Stretching
`
`Biaxial stretching experiments were carried out on the QUB biaxial stretcher, which is capable of stretching square
`samples at rates up to 32s-1 and temperatures up to 200°C i.e. representative thermoforming processing conditions.
`The sample was gripped around its perimeter by pneumatic clamps and heating on both sides via two convection
`heaters, as shown in Fig.3. During the experiments, the air was heated to the stretching temperature and then the
`sample was held for the specified soaking time of 4 minutes, the heater was then turned off and the sample was
`stretched immediately. The temperature of the air close to the sample was controlled via a thermocouple, in a closed
`loop system. A more in depth description of the QUB stretcher is provided in (5)and (6).
`
`(b)
`(a)
`FIGURE 3. (a) QUB biaxial stretching machine. (b) QUB biaxial stretching machine sample holder and instrumentation
`
`060006-3
`
`MacNeil Exhibit 2152
`Yita v. MacNeil IP, IPR2020-01139, Page 5
`
`

`

`Tensile testing
`Dogbone specimens were cut from biaxially stretched HDPE sheets, in both the MD and TD in accordance with
`ISO 527-2:2012(E) – Type 1BA. The modulus was determined from nominal strain range of 0.0005 to 0.0025 and an
`average value was determined from 4 repeats. The accurate determination of the Young’s Modulus from tensile tests
`requires an accurate measurement of the strain within the sample. Digital Image Correlation (DIC) was used to
`accurately determine the strain in the center of the done bone specimen, using a randomly applied pattern as shown in
`Fig.4a. It was verified that the average strain measured over the gauge length of the specimen corresponded to the
`strain determined from the crosshead displacement at a strain-rate of 1mm/min, as shown in Fig.4b. Therefore, the
`crosshead displacement was used to determine the strain throughout all the experiments.
`
`instron
`DIC
`
`0.035
`
`0.03
`
`0.025
`
`0.02
`
`0.015
`
`0.01
`
`0.005
`
`0
`
`nominal strain
`
`60
`40
`time(sec)
`(a) (b)
`FIGURE 4. (a) Tensile sample with random pattern used for DIC. (b) Comparison of nominal strain measured by
`DIC and Instron crosshead.
`
`80
`
`100
`
`0
`
`20
`
`RESULTS AND DISCUSSION
`
`Biaxial Stretching
`The DMA temperature sweep from 30°C to 135°C indicated a single tan delta peak at 133°C, this corresponded to
`the peak crystalline melting temperature determined via DSC, as HDPE is already in the rubbery state at room
`temperature. The tan delta peak is significant as it indicates the materials ability to dissipate energy and hence,
`indicates the processing temperature range. From Fig.2b it is clear that the gradient of the tan delta curve changes
`significantly at 125°C, indicating the onset of the peak. Therefore, the temperature range from 126°C to 133°C was
`investigated. It was found that for temperatures below 126°C samples could not be stretched at strain rates above 1s-
`1, as the sample was too stiff. Whereas, at temperatures above 130°C the samples were too soft and could not be
`successfully stretched. Therefore, sheets were biaxially stretched at 4s-1 in the temperature range 126-130°C. The
`stress-strain response for each temperature under EB deformation at 4s-1 is shown in Fig5. A clear temperature
`dependence was observed with stress decreasing with increasing temperature, with the most pronounced difference
`between 126°C and 127°C. Furthermore, over the strain levels applied no clear strain hardening was observed. The
`data in Fig.5 highlights the significant temperature dependence of the grade of HDPE used in this study, with 1°C
`having a significant impact on the stress response.
`
`060006-4
`
`MacNeil Exhibit 2152
`Yita v. MacNeil IP, IPR2020-01139, Page 6
`
`

`

`126°C TD
`126°C MD
`127°C TD
`127°C MD
`128°C TD
`128°C MD
`129°C TD
`129°C MD
`130°C TD
`130°C MD
`
`0.5
`
`0123456
`
`True Stress (MPa)
`
`0
`
`1
`Nominal Strain
`FIGURE 5. EB stress-strain data obtained on the QUB biaxial stretcher
`
`1.5
`
`2
`
`Tensile Testing – Modulus
`The key driving factor behind thermoforming HDPE below the melt temperature is the potential to enhance the
`Young’ Modulus and as shown in Fig.6 the modulus was increased by approximately 100% in the MD and 200% in
`the TD. It was also evident that the initial injection molded sheet had a level of anisotropy, with a modulus of 606MPa
`in the MD and 391MPa in the TD. The same anisotropy was observed within the biaxial stress-strain data as shown
`in Figure XX with a stiffer response in the MD direction. The anisotropy was an inherent aspect from the injection
`molding process, where flow-induced orientation leads to a preferred orientation and hence, a stiffer response.
`Interestingly, there was minimal variation between the MD and TD modulus values obtained, with and average
`difference of 64MPa between the MD and TD for each stretching temperature. Furthermore, for the temperature range
`considered there was no real distinguishable difference between the average modulus for each case with an
`approximate modulus of 1200MPa. It is acknowledged that the temperature range considered was particularly small
`but the processing range limited this. Furthermore, it was observed from the biaxial stress-strain data that while the
`temperature range was small it had a significant impact on the response. Therefore, the results indicate the modulus
`of the biaxially stretched sheet is dependent on the strain level applied which is in agreement with the findings of Hiss
`et al (8), who found the deformation under tensile loading is strain controlled.
`
`CONCLUSIONS
`Injection molded HDPE sheets were formed under industrially relevant processing conditions and based on the
`results the following conclusions can be made;
`1. A small temperature processing window of 126-130°C exists, to process the grade HDPE used, for the high-
`rate EB deformation applied.
`2. The stress-strain response was clearly temperature dependent over the temperature range investigated,
`indicating high temperature sensitivity with 1°C having an impact on the response.
`3. The Young’s Modulus was increased by approximately 100% and 200% in the MD and TD.
`4. The processing temperature range applied had minimal impact on the Young’s Modulus.
`
`060006-5
`
`MacNeil Exhibit 2152
`Yita v. MacNeil IP, IPR2020-01139, Page 7
`
`

`

`RT MD
`
`1400
`
`1200
`
`1000
`
`800
`
`600
`
`400
`
`200
`
`0
`
`Modulus (MPa)
`
` 130°C
`MD
`
` 130°C
`TD
`
` 129°C
` 129°C
` 128°C
` 128°C
` 127°C
` 127°C
` 126°C
`RT TD 126°C
`MD
`TD
`MD
`TD
`MD
`TD
`MD
`TD
`FIGURE 6. Young's Modulus values determined across the processing range
`
`REFERENCES
`Plastics Europe [Internet]. [cited 2016 Jul 18]. Available from: http://www.plasticseurope.org
`Li Y, Chen J, Yuan Y, Shen K, Guo J. Study on Properties and Structure of Near Melt Point Extruded High-
`Density Polyethylene. J Macromol Sci Part B. 2008;47(3):426–33.
`Zheng H, Wang B, Zheng G, Wang Z, Dai K, Liu C, et al. Study on Crystallization Kinetics of Partially
`Melting Polyethylene Aiming To Improve Mechanical Properties. 2014;
`Bourgin P, Cormeau I, Saint-Matin T. A first step towards the modelling of the thermoforming of plastic
`sheets. J Mater Process Tech. 1995;54(1–4):1–11.
`Menary GH, Tan CW, Harkin-Jones EMA, Armstrong CG, Martin PJ. Biaxial Deformation and
`Experimental Study of PET at Conditions Applicable to Stretch Blow Molding. Polym Eng Sci.
`2012;52:671–88.
`Martin PJ, Tan CW, Tshai KY, McCool R, Menary G, Armstrong CG, et al. Biaxial characterisation of
`materials for thermoforming and blow moulding. Plast Rubber Compos. 2005;34(5–6):276–82.
`Wunderlich B. Crystal structure morphology, defects. Macromol Phys. 1973;1(Academic Press).
`Hiss R, Hobeika S, Lynn C, Strobl G. Network stretching, slip processes, and fragmentation of crystallites
`during uniaxial drawing of polyethylene and related copolymers. A comparative study. Macromolecules.
`1999;32(13):4390–403.
`
`1.
`2.
`
`3.
`
`4.
`
`5.
`
`6.
`
`7.
`8.
`
`
`
`View publication statsView publication stats
`
`060006-6
`
`MacNeil Exhibit 2152
`Yita v. MacNeil IP, IPR2020-01139, Page 8
`
`