The Use of Electrostatic Techniques for The

8/9/2019 The Use of Electrostatic Techniques for The

http://slidepdf.com/reader/full/the-use-of-electrostatic-techniques-for-the 1/8

Resources, Conservation and Recycling 44 (2005) 91–98

The use of electrostatic techniques for theidentification and sorting of waste

packaging materials

G.L. Hearn∗, J.R. Ballard

Electrical Power Engineering Group, School of Electronics and Computer Science,

University of Southampton, Highfield, Southampton SO17 1BJ, UK

Received 12 December 2002; accepted 19 August 2004

Abstract

Techniques have been developed which enable items of domestic waste packaging to be identified

and sorted into material groups prior to recycling. Plastic packaging items can be separated from

items of similar size and shape constructed from cardboard, wood or paper. This is achieved by

applying a controlled level of electrostatic charge and comparing the rate of charge relaxation using

a non-contacting electric field meter. Plastic packaging items such as bottles, food tubs and trays aresubsequently streamed into their polymer groups using a series of tribo-electric probes. The probes

are brought into contact with the packaging materials to be identified and comprise a rotating head

or drum, which generates an electrostatic charge due to the friction between the head and the item

to be identified. Since the magnitude and polarity of the generated charge depends on the polymers

comprising both the head material and the packaging material, a degree of identification can be

obtained.

Using this technique has resulted in: (a) the successful separation of plastics from non-plastics and

(b) plastics separated into polypropylene (PP), polyethylene terephthalate (PET)/polystyrene (PS),

polyvinyl chloride (PVC) and high-density polyethylene (HDPE) streams. The techniqueideally lends

itself to incorporation on an automated recycling line.

© 2004 Elsevier B.V. All rights reserved.

Keywords: Electrostatics; Plastic; Packaging; Sorting

∗ Corresponding author. Tel.: +44 2380 594995; fax: +44 2380 593015.

E-mail address: [email protected] (G.L. Hearn).

0921-3449/$ – see front matter © 2004 Elsevier B.V. All rights reserved.

doi:10.1016/j.resconrec.2004.08.001



92 G.L. Hearn, J.R. Ballard / Resources, Conservation and Recycling 44 (2005) 91–98

1. Introduction

Data from the UK Department of Trade and Industry indicates that on average eachhousehold in the United Kingdom throws away 1.6 kg of dense plastic waste, each week, in

the form of bottles, containers andfood trays (DTI, 2002). This amounted to a total of around

15,000 tonnes during 2001 (Recoup, 2002). According to the results of recent research

(Hampshire County Council, 2001), common household plastic waste items comprise the

following polymers.

• Polyethylene terephthalate (PET/PETE)—squash and carbonated drink bottles, alcoholic

drinkbottles andsome food trays.For recyclingpurposes this polymer is usually separated

into three groups; clear (or with a slight blue tint), green and other mixed colours. The

clear group is the most desirable as it is closest to the natural PET state. Currently, the

green group is produced in sufficient quantity to be reused while the mixed group is

recycled but only in small amounts.

• High-density polyethylene (HDPE)—milk bottles, oil containers, bleaches, cleaning

products and some bags. HDPE can be sub divided into two groups, natural and coloured.

Natural is used mainly for milk bottles.

• Polyvinyl chloride (PVC)—some fruit cordial and spring water bottles and some food

trays.• Polypropylene (PP)—PP is used for spread containers, such as butter and margarine, and

most bottle caps.• Polystyrene (PS)—trays and drinks dispenser cups; expanded polystyrene is used for

meat trays.

HDPE and PET are by far the most commonly encountered household waste plastics(Hooper et al., 2001).

In order to reduce the amount of material going to landfill a degree of recycling is already

being applied to household waste collection. One particularly important development is the

use of the material recovery facility (MRF) or super MRF (ETSU, 1998). This applies a

degree of sorting to general household waste in a semi-automated way. With the exception

of two MRF’s in the UK, plastic packaging items, including bottles are currently sorted by

hand, whichis both laboriousand inefficient.The useof an automatedsystem forrecognising

the various plastics is an obvious requirement.

Theconcept of using theelectrostatic properties of thematerialas a basis foridentification

was investigated by Hearn et al. (1997). Natural materials such as wood,cardboard andpaper

have an intrinsic electrical conductivity, several orders of magnitude greater than virtuallyall polymers (Davies, 1967; Hearn, 1997). Thus, the migration of electrical charge across

the surface of a cardboard box is much greater than that of a plastic carton. This difference

in the rate of charge movement can therefore, be used as a basis of discriminating between

plastic and non-plastic packaging items.

The development of the ‘Tribopen’ (Hearn et al., 1996) illustrates how another electrical

property that of tribo-electrification can also be used in order to distinguish between polymer

types. Tribo-electrification occurs when two unlike materials are brought together and then

separated. Upon separation, one of the materials acquires a positive charge while the other



G.L. Hearn, J.R. Ballard / Resources, Conservation and Recycling 44 (2005) 91–98 93

acquires a negative charge. Rubbing the two materials together, as opposed to simple contact

or separation, tends to enhance the triboelectric effect. The generation of electrical charge

in this way is particularly suited to plastics since they are highly electrically insulating bynature. As a consequence, any charge that is generated is retained on the surface of the

interacting materials and can be measured. Plastics may be arranged in a table according to

whether they charge preferentially positive or negative when rubbed against other materials.

The table is known as a tribo-electric series and is analogous to the electrochemical series

for metals (Davies, 1969). The series is usually displayed with the most positively charging

materials at the topand most negatively chargingmaterials at the bottom. It follows therefore

that if two materials are picked from this series and rubbed together the higher of the two

in the series will charge positively and the other negatively.

Measurement of tribo-electric charge is relatively easy and can be used to distinguish

between two plastics. This can be achieved by rubbing each of the two plastics with a third

‘reference’ material chosen from a position in the series between the two plastics of interest.

The electrostatic charge generated on this reference material will be negative for one plastic

and positive for the other.

The attraction of the two techniques described above is their inherent simplicity and

suitability for incorporation on an automated sorting line.

2. Experimental study

For this work, the use of the charge relaxation technique to differentiate between plastics

and non-plastics was investigated under controlled laboratory conditions of humidity and

temperature. Representative items of packaging were chosen including the following: card-

board crisp/snack tubes, PET bottles, HDPE milk bottles, PP-coated cardboard juice cartonsand assorted plastic food trays. Each of the samples in turn were placed on a grounded,

electrically conductive rubber substrate and charged by means of a +20 kV dc ioniser bar

passed over the top of each sample at a constant height above the substrate (simulating the

action of a conveyor belt). Immediately after charging, the electrostatic surface potential

on the item under test was monitored using an electric field meter. Fig. 1 gives a com-

parison of the charge relaxation characteristics of the 5 packaging materials obtained at

a temperature of 20 ◦C and a relative humidity of 30%. At time T = 0 in Fig. 1, the ionic

charging source is removed and replaced with an electric field meter measuring the surface

potential on the packaging sample in kilovolts. As time elapses, charge migration occurs

over the surface of the sample to the grounded substrate and the monitored potential be-

gins to decay. It can be seen from Fig. 1 that there are significant differences between therelaxation rates of the plastic and non-plastic samples. The presence of a PP coating on

cardboard, which is likely to be less than 20m in thickness, has some effect over plain

cardboard, however, there is a significant difference between the curves for the non-plastic

and plastic items. Comparing electrostatic potentials after 10 s of charge relaxation enabled

plastic and non-plastic items to be easily discriminated. There is also a difference between

the PET and HDPE bottles with the electrostatic potential on the HDPE decaying more

rapidly. The curves in Fig. 1 were found to be representative of the above packaging items

generally and there was very little variation observed between different test samples of the




Fig. 1. Electrostatic charge relaxation rates for different packaging materials.

same material. The presence of residual liquid within the packaging items appeared to have

little effect.

In order to segregate different plastics from each other using tribo-electrification it was

necessary to identify reference materials which when brought into frictional contact with the

various waste plastics would charge to the appropriate polarity. A large number of trials were

undertaken in which potential reference materials were manually rubbed against a variety

of plastic packaging. The resultant magnitude and polarity of electrostatic charge developedwere recorded using an electric field meter. These trials were undertaken in a controlled

environment of 20 ◦C and 30% R.H. as before. Many potential reference materials that were

tested were found to be unsuitable because they did not register a polarity shift. Two suitable

materials were found; polyvinylidene fluoride (PVDF), which acquired a negative charge

when rubbed against polypropylene, but charged positively when rubbed against the other

plastics of interest and polybutylene terephthalate (PBT) which charged negatively against

PET and PS and positively against HDPE and PVC.

Since plastic packaging waste arrives at the recycling point in a variety of forms it was

necessary to identify a practical, controlled method of bringing the reference material and

waste items into contact. Furthermore, the presence of labels on many of the bottles gave

confused readings in terms of charge generation. The tribo-electric system designed to solvethese issues is shown in Fig. 2. It takes the form of a probe with a rotating cylindrical head,

which is brought into contact with the waste plastic to be identified. The head is constructed

of the reference tribo-electric material (PVDF or PBT) driven by a small electric motor at

around 900 Hz. Electrostatic charge generated on the cylindrical head in contact with the

waste plastic is conducted away by an array or ‘comb’ of fine metallic filaments on the

opposite side of the cylinder. This generates a small positive or negative electrical current,

which can be detected and processed accordingly. The use of a probe with a relatively small

head of 25 mm diameter makes this a suitable technique for items of variable geometry.




Fig. 2. Tribo-electric probe showing rotating cylindrical head (white PBT) and filament array.




Fig. 3. Schematic of proposed mixed waste sorting line.

It also addresses the problem of labels since the probe can be brought into contact with

sections of the packaging, which are not covered by the label. The application proposed for

an automated recycling line would have the probe suspended above a conveyor and brought

down to contact the items as they pass beneath.

A schematic of a mixed household waste sorting line is given in Fig. 3. Non-plastics, such

as cardboard cartons, are removed at the beginning of the line using the charge relaxation

technique. Plastic items remaining on the line are then brought into contact with the PVDF

tribo-electric probe. Polypropylene materials such as margarine tubs register a negativecharge on the PVDF and are removed from the line. All other plastics register a positive

charge on the probe and are allowed to pass to the next sorting point. At this point, a similar

probe is used this time with the head composed of PBT. This head charges negative when

rubbed against PET and PS and these two materials are ejected from the line at this point.

HDPE and PVC packaging items generate a positive charge on the probe and pass to the

next sorting point. At this point in the line the vast majority of the material will consist of

HDPE in the form of milk bottles, detergent bottles, etc. There may also be some PVC food

trays and squash bottles. PVC and HDPE can be separated relatively simply by using clear




object recognition. This technique uses an optical wavelength light beam to sort clear items

from opaque. This splits PVC and HDPE since PVC packaging is always clear and HDPE

is never present in packaging materials as a transparent plastic.

3. Discussion and conclusions

Two sorting techniques have been developed which use the electrostatic properties of

materialsto produce separate material streams forthe purposes of recycling. Trials have been

undertaken using typical common items of waste packaging giving encouraging results.

The techniques described in this paper are ideally suited to an automated recycling

line and could be considered for incorporation into currently operating super MRF and

other semi-automated recycling facilities. Early results indicate reliable operation under a

range of environmental conditions, however, the effects on sorting efficiency of extremes of

surface contamination, moisture, temperature and humidity have yet to be quantified. It is

recognised that the presence of high levels of surface contamination on the waste items to be

streamed may cause problems particularly for the tribo-electric sensor probe. Preliminary

examination of material from an MRF, however, suggests that these materials are generally

not heavily contaminated with anything other than moisture. The presence of surface water

does significantly influence both charge generation and tribo-electrification. MRF’s and

plants where these techniques are likely to be applied lend themselves to the application of

driers or air curtains which could be installed upstream of the electrostatic sorting area. It is

also recognised that this technique may not be appropriate for all polymer types but can be

used in conjunction with other techniques. Such techniques may include an optical sensor

to separate PVC from HDPE and Fourier transform infrared spectroscopy (Hearn, 2003) to

sort the stream containing PET/PETE and PS.Difficulties may be encountered with certain packaging geometries and the presence of

labels and coatings. On all of these issues the use of the small tribo-electric probe is an

advantage as it can be directed at an area of packaging most likely to constitute exposed

polymer. For example, most plastic bottles have labels, which do not extend fully along

their length. It is envisaged that an automated line would use optical sensors to detect the

approach of the bottle and initiate the tribo-electric probe to make contact with only the

first few centimetres of the bottle above or below the label. Coatings, which are applied

to the full surface of the packaging, may confuse the tribo-electric probe. Indications are

however that such coatings applied to the outside of packaging materials are not common.

Furthermore, their presence may not necessarily cause incorrect streaming of the plastic

since the polarity of electrostatic charge generated by the coating may be similar to the basepolymer. It is recognised, however that further analysis is required in this case.

4. Future work

Work is continuing at the University of Southampton under the sponsorship of the Onyx

Environmental Trust. The next phase of work is to incorporate the techniques described

in this paper into a pilot scale recycling line for further analysis. The objectives of this




future work are: (a) to develop a system which is capable of handling the diverse range

of packaging geometries currently found in domestic waste, (b) to assess, on a pilot scale

conveyor system, the resilience of these techniques (in particular the tribo-electric probe)and(c) to gain information relating to the accuracy andreproducibility of theaforementioned

techniques, in a practical and quasi-industrial situation.

Acknowledgements

The authors wish to express their gratitude to the Onyx Environmental trust for their

financial support of this project. The authors would also like to extend their thanks to the

Ford Motor Company for their funding and support of initial work on the development of

electrostatic techniques for the recycling industry. Particular thanks goes to Professor John

Amner, Ford of Europe’s Automotive Recycling Specialist, who has provided a great deal of

time and effort with the University of Southampton investigating novel methods of plastics

identification.A final word of thanks is extended to Mr. P.E.R. Mucci of Powertile Ltd. and

formerly a member of the academic staff of the School of Engineering at the University of

Southampton for his advice on mechanical engineering issues associated with this project.

References

Davies DK. Charge generation on dielectric surfaces. Brit J Appl Phys (J Phys D) 1969;2:1533–7 (Series 2).

Davies DK. The generation and dissipation of static charge on dielectrics in a vacuum. In: Proceedings of the

Static Electrification Conference, Institute of Physics, London; 1967. p. 29–36.

Department of Trade and Industry (DTI), UK Environmental ResourcesManagement.PlasticsRecycling Factsheet,

2002. Crown Copyright URN 02/1231.

ETSU for UK Department of Trade and Industry, 1998. An Introduction to Household Waste Management.

Hampshire County Council (UK) Waste Management Team, 2001. Project Integra statistics.

Hearn GL, Mucci PER, Eyres A. Plastics Identification. European Patent EP 0 680 606 B1. Proprietors: Ford

Motor Company, 1997.

Hearn GL. May the force be with you. IEE Review, Inst. Elec. Eng., London, 1997.

Hearn GL, Mucci PER, Eyers A, Amner J. The triboelectric pen: an electrostatic method for the identification of

plastics in recycling. In: Proceedings of the IEEE IAS 31st Annual Meeting; 1996.

Hearn GL. Plastics Identification for Recycling (Shanks First Fund). Wolfson Electrostatics report 304B/GLH for

EB Nationwide, 2003.

Hooper R, Harder MK, Potter AKN. Profit from Plastic. In: Proceedings of the I Mech E Conf. Trans VI Int. Conf.

on Eng. for Profit from Waste, London; 2001.

Recoup. Affordable Plastic Bottle Recycling? Analysis and review of collection, sorting, reprocessing, end-market

issues and economics, 2002. http://www.recoup.org.

http://www.recoup.org/

The Use of Electrostatic Techniques for The

Documents