Towards recyclable insulation materials for high voltage cables

Towards recyclable insulation materials for high voltage cables I. L. Hosier, A. S. Vaughan and S. G. Swingler January 2011 Thanks to EPSRC (Supergen AMPer. ES Project) for financial support

Outline of presentation • Introduction • Experimental • Results - Ethylene based systems • Results - Propylene based systems • Conclusions • Towards the future 2

Introduction • Paper/oil cables were the first type to be used commercially but oil leaks are a serious environmental issue • Nowadays these have been largely superseded by XLPE cable systems but these also have drawbacks – At end of life XLPE cannot be easily recycled, although several re-use schemes have been proposed – Disposal and environmental issues are getting more and more important in every area of life • So, we need a recyclable alternative to XLPE for the next generation of extruded high voltage cables 3

Possible alternatives to XLPE • Both PE and PP are already widely recycled through existing waste channels – ideal replacements. . . ? • But what is required from a cable material? – Good flexibility at low temperatures (~-20 o. C) – Good stability at high temperatures (~105 o. C or better) – Sufficiently high melting point (>105 o. C) – Low dielectric loss, high breakdown strength – Easy and cheap to manufacture – The same (or better!) than XLPE

Analytical process • Identify suitable recyclable starting materials – Various ethylene and propylene based systems • Make samples and perform tests – Melting point – Morphology – Thermo mechanical stability – Breakdown strength • Optimise their performance if required by combining the properties of more than one material (blending) and through the control of crystallisation conditions

Experimental

Materials and sample preparation • Materials – Ethylene systems – LDPE, HDPE, EVA 09, EVA 20, EVA 33, EVA 40 (EVA systems offer tree retardancy) – Propylene systems – i. PP, s. PP, PE 02, PE 12, PE 40, PB 12 • Blending – Solution blends using xylene as solvent • Samples (melt pressed) – 60 μm disks for ASTM D 149 breakdown tests, ~2 mm thick plaques for thermo-mechanical/morphology. Isothermal crystallisation or quenched

Characterisation process • Thermal - DSC on 2 -5 mg samples (Perkin Elmer DSC-7) • Morphology - Microtome followed by permanganic etching, gold coating then SEM (Cambridge Instruments Stereoscan 360 operated at 20 k. V, SE mode) • Mechanical - Tensile testing to 100% strain at room temperature (Instron 4301) • Thermo-mechanical - Modulus from -20 to ~150 o. C (Rheometrics RSA II) • Electrical – ASTM D 149 ramp to breakdown test (50 Hz AC, 50 Vs-1, opposing 6. 25 mm ball bearings)

Ethylene based systems

Thermal properties Single components Only LDPE and HDPE have high enough melting points for use in a cable system, EVA’s have much lower melting points Blends (20% HDPE) Blending HDPE into LDPE or EVA systems increases the final melting points of the composite material!

Morphology Single components LDPE to EVA 40 - the crystalline texture becomes increasingly disrupted by the presence of VA groups Blends (20% HDPE) Increasing VA content leads to increased phase separation between the HDPE and EVA components

Mechanical Single components HDPE is brittle (necks), LDPE offers more ideal (rubbery) behaviour whereas increasing the VA content offers softer materials Blends (20% HDPE) Adding HDPE increases the modulus but without making the composite system brittle – they offer us a range of mechanical properties

Thermo-mechanical Single components HDPE offers us good performance at high temperatures, LDPE is OK up to ~60 o. C whereas EVA materials would be of limited use in distribution cables Blends (20% HDPE) Adding HDPE improves the high temperature performance, Blend A (LDPE based) offers us properties comparable to XLPE

Electrical Single components HDPE offers the best breakdown strength, LDPE offers electrical properties identical to XLPE whereas the EVA systems are electrically weak Blends (20% HDPE) Adding HDPE improves the electrical breakdown performance, Blend A offers us improved performance compared to LDPE or XLPE

Conclusions • LDPE offers comparable electrical properties to XLPE but is unsuitable for use above ~60 o. C • HDPE has good high temperature performance, good breakdown strength but is mechanically brittle • Various EVA co-polymers show reduced breakdown strength and reduced high temperature performance • Adding HDPE to EVA or LDPE improves the breakdown strength and the high temperature performance • Blend A (20 % HDPE in LDPE) offers a recyclable system with comparable properties to XLPE

Propylene based systems

Thermal properties • All of the materials melt above ~120 o. C, s. PP has the lowest melting point and i. PP the highest • Addition of an ethylene or butene co-unit generally reduces the melting point • Some systems can behave strangely (PE 12) • Might permit cables to operate at higher temperatures!

Morphology i. PP and s. PP Both materials show spherulitic textures Copolymers Unspherulitic PE 02 (left), leads onto an amorphous texture in PE 40. PE 12 shows a two phase structure (right) consistent with DSC

Mechanical properties • Both i. PP and PE 12 exhibit brittle fracture – not good! • The remaining materials neck except PE 40 which is rubbery • Despite the advantage of having high melting temperatures and excellent high temperature stability, all the propylene materials (except PE 40) are too stiff at low temperatures

Electrical properties • Whilst being the best mechanically, PE 40 is the worst electrically! • When crystallised isothermally (right) i. PP and PE 12 also show very poor breakdown performance • In conclusion NONE of the propylene materials offer the right combination of mechanical and electrical properties – BLENDS

Selected blends I (Thermal) • Out of the materials s. PP and PE 40 offer the best (but still not ideal) mechanical properties • Can we add some i. PP to overcome their poor electrical performance without introducing unwanted brittleness? • Melting behaviour indicates that both systems can still operate to 120 o. C (better than XLPE) Designation Composition Blend 1 20 % i. PP in s. PP Blend 2 50 % i. PP in s. PP Blend 3 20 % i. PP in PE 40 Blend 4 50 % i. PP in PE 40

Selected blends II (Morphology) Blends 1 and 2 (s. PP) Crystals of i. PP in an s. PP “matrix”, organise into spherulites in blend 2 (50 % i. PP content) Blends 3 and 4 (PE 40) Crystals of i. PP in a largely amorphous PE 40 “matrix” (organise into spherulites in blend 4)

Selected blends III (Mechanical) • The s. PP based blends 1 and 2 exhibit undesirable necking and are too stiff at low temperatures for a cable system • Blends 3 and 4 show a more desirable (rubbery) behaviour • All systems provide high temperature integrity to ~120 o. C but blends 3 and 4 provide mechanical properties close to XLPE

Selected blends IV (Electrical) • Under both quenching and isothermal conditions blends 1 and 2 exhibit dielectric strength better than XLPE! • However, the current work indicates that blend 3 (20 % i. PP in PE 40) offers the best balance of electrical and mechanical properties • Increased operating range to 120 o. C is possible!

Conclusions • Unfortunately none of the propylene based materials in isolation offered good breakdown performance combined with suitable mechanical properties for a cable – As in the case of HDPE/LDPE blends, combining a rubbery (PE 40) system with a stiff (i. PP) system combines the rubbery nature of PE 40 with the good breakdown performance afforded by the i. PP – Care must be taken with this system to avoid slow crystallisation which can lead to a more brittle system • Advantages of such a blend include better breakdown strength than XLPE and an improved operating temperature range – thinner insulation for the same power!

Conclusions

Conclusions • We have looked at various recyclable ethylene and propylene based systems – In isolation none of the single component systems provides the required balance of properties for a cable – Blending is a powerful tool for properties optimisation allowing the advantages of different materials to be combined effectively to yield a better composite system – We have identified suitable ethylene and propylene based blend systems having optimised properties for future recyclable cable systems

Towards the future • There has been considerable interest by various cable manufacturers to commercialising these ideas – TSB project at Southampton/Surrey is looking at “minicables” (both PE and PP based) – focussing on nonisothermal crystallisation, dielectric and mechanical properties optimisation – Papers in press – C. D. Green, A. S. Vaughan et. Al. – Successfully making the transition from “laboratory samples” to real extruded cable systems • The future is indeed bright and we expect implementation of practical recyclable cable systems within the next decade!

References • An investigation of the potential of polypropylene and its blends for use in recyclable high voltage insulation systems, I. L. Hosier, A. S. Vaughan and S. G. Swingler, , Submitted to J. Mat. Sci. (December 2010). • An investigation of the potential of ethylene vinyl acetate/polyethylene blends for use in recyclable high voltage cable insulation systems, I. L. Hosier, A. S. Vaughan and S. G. Swingler, J. Mat. Sci. , 45, 10, pp. 2747 -2459, 2010. • Propylene based systems for high voltage cable insulation applications, I. L. Hosier, L. Cozzarini, A. S. Vaughan and S. G. Swingler, J. Phys. : Conf. ser. , vol. 183, 012015, 2009. • Morphology, thermal, mechanical and electrical properties of propylenebased materials for cable applications, I. L. Hosier, S. Reaud, A. S. Vaughan and S. G. Swingler, In Conf. Rec. of the 2008 International Symposium on Electrical Insulation, pp. 502 -505, 2008. • Effect of polyethylene on morphology and dielectric breakdown in EVA blends, I. L. Hosier, A. S. Vaughan and W. Tseng, In Proc. 2007 International Conference on Solid Dielectrics, pp. 184 – 187, 2007.