Nuclear Instruments and Methods in Physics Research B 323 (2014) 71–74

Contents lists available at ScienceDirect

Nuclear Instruments and Methods in Physics Research B

journal homepage: www.elsevier .com/locate /n imb

Depth profiles of the Doppler-broadening S parameter for polymersobtained with two measuring patterns: The role of accumulated charges

0168-583X/$ - see front matter � 2014 Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.nimb.2013.12.019

⇑ Corresponding author. Tel.: +86 10 88235896; fax: +86 10 88233186.E-mail address: wangboy@ihep.ac.cn (B.Y. Wang).

J. Yang a,b, P. Zhang a, E.Y. Lu a, X.Z. Cao a, R.S. Yu a, B.Y. Wang a,⇑a Key Laboratory of Nuclear Radiation and Nuclear Energy Technology, Institute of High Energy Physics, Chinese Academy of Sciences, No. 19 Yuquan Lu, Beijing 100049, Chinab University of Chinese Academy of Sciences, Beijing 100039, China

a r t i c l e i n f o a b s t r a c t

Article history:Received 28 November 2013Received in revised form 15 December 2013Available online 27 January 2014

Keywords:PositronPolymersDepth profileAccumulated charges

Depth profiles of Doppler broadening S parameter for oxygen containing polymer polycarbonate (PC),fluoropolymer poly (tetrafluoroethylene) (PTFE) and chlorine containing polymer polyvinylchloride-unplasticized (UPVC) were obtained with two measuring patterns, i.e. energy increase pattern and energydecrease pattern. The two curves can’t coincide with each other for that a trough appeared between 1 and5 keV in the curve obtained with energy decrease pattern. It was found that charges induced by highenergy incident positrons greatly influenced the annihilation of low energy incident positrons, whilecharges induced by low energy incident positrons showed little influence on the annihilation of highenergy incident positrons. With energy increase measuring pattern, charges induced by low energy inci-dent positrons showed little influence on the annihilation of later incident high energy positrons, thus themeasurement can give the depth profile of S parameter in polymer as it was.

� 2014 Elsevier B.V. All rights reserved.

1. Introduction

Doppler broadening energy spectroscopy on a variable-energypositron beam has been widely applied to study damage and defectdepth-profiling of polymers [1–4]. By changing the incident energyof positrons, one can obtain information on the depth of damagelayer [5,6], and interface layer [7], which are important to macro-scopic properties of polymers.

When positron annihilation Doppler broadening energy spec-troscopy on slow positron beam is applied, usually the depthprofile of sample is obtained using the following measuringpattern, first low energy positrons, and then high energy positronsare injected into sample [8,9] (energy increase pattern). In fact,there is the other measuring pattern, first high energy positronsare used, and then the accelerating voltage is decreased so thatlow energy positrons could inject into sample (energy decreasepattern). With both measuring patterns, usually, it takes more than12 h to obtain a depth profile of S parameter. It is well known thatin normal positron annihilation measurements, prolonged positronirradiation in polymer can cause the drop of o-Ps intensity(so called self-irradiation effect) as a result of the build-up of aninternal electric field [10–12], since the accumulated charges can’tbe dissipated in the insulating polymers. While in slow positronbeam measurement, where the incident positron energy is muchlower compared to fast positron directly emitting from 22Na

source, self-irradiation effect is seldom considered. However, evenin slow positron beam measurement, long time positron irradia-tion can cause charges accumulation in polymer, which may influ-ence the characteristics of positron annihilation. In this study,depth profiles of the Doppler-broadening S parameter for some po-lar and nonpolar polymers were obtained with the two measuringpatterns. We try to find the measuring pattern which is less influ-enced by accumulated charges, thus better reflecting the real char-acteristics of polymer sample.

2. Experimental details

The polymer samples studied here were purchased fromGoodfellow (Cambridge, UK), including polycarbonate (PC), poly(tetrafluoroethylene) (PTFE), polypropylene (PP), polyvinylchlo-ride-unplasticized (UPVC). The silicon slice was single crystal. Pos-itron annihilation Doppler broadening measurements wereconducted with a magnetically guided variable-energy positronbeam (0–20 keV) at IHEP, Beijing. The beam intensity was about105 positrons/s. Positrons are accelerated to specific energy by neg-ative high voltage within the distance of 1 cm from the sample.When no accelerating voltage was applied, positron energy wasabout 0.18 keV. With energy increase pattern, the acceleratingvoltage was increased from 0 to 10 (or 20) kV. With energydecrease pattern, the accelerating voltage was decreased from 10(or 20) to 0 kV. (For convenience, the positron energy below refersonly the accelerating energy, not including the initial energy, e. g.the energy of a positron accelerated by 10 kV voltage was denoted

0.46

0.48

0.50

S pa

ram

eter

a1 a2PC

72 J. Yang et al. / Nuclear Instruments and Methods in Physics Research B 323 (2014) 71–74

as 10 keV.) The Doppler broadening energy spectra (DBES) wererecorded using a high-purity Ge detector. The obtained DBES spec-tra were expressed in S parameter, which was defined as a ratio ofintegrated counts in the central parts (510.2–511.8 keV) to thetotal counts of the 511 keV annihilation peak after proper back-ground subtraction. To obtain depth profiles of S parameter for apolymer, two identical films were prepared. One piece of filmwas measured with energy increase pattern, and the other wasmeasured with energy decrease pattern.

0 5 100.42

0.44

Incident positron energy (KeV)

(a)

0 5 10

0.44

0.45

0.46

0.47

S pa

ram

eter

Incident positron energy (KeV)

b1 b2

UPVC

(b)

0 5 100.42

0.43

0.44

0.45

0.46

S pa

ram

ter

Incident positron energy (keV)

c1 c2

PTFE

(c)Fig. 1. Depth profiles of S parameter for PC, UPVC, and PTFE polymers; black curveswere measured with energy increase pattern, red curves were measured withenergy decrease pattern. (For interpretation of the references to colour in this figurelegend, the reader is referred to the web version of this article.)

3. Results and discussion

Fig. 1 shows the depth profiles of Doppler broadening S param-eter for oxygen containing polymer PC, fluoropolymer PTFE andchlorine containing polymer UPVC obtained with the two measur-ing patterns. With energy increase pattern, PC, PTFE and UPVC givedifferent depth profiles of S parameter. For PC, S parameter in-creases with increasing incident positron energy until 4 keV, andthen reaches leveling-off at higher energies. For PTFE, S parameterdecreases from polymer surface to bulk. And for UPVC, S parameteris independent of incident positron energy and remains nearly con-stant. As discussed in our previous paper [13], the increase of Sparameter for PC is owing to depth-dependent positronium forma-tion, the decrease of S parameter for PTFE is due to the influence ofhighly electronegative atoms on positron annihilation characteris-tics, and nearly constant S parameter for UPVC is the result ofequivalent effect of depth-dependent positronium formation andhighly electronegative atoms. However, depth profile obtainedwith energy decrease pattern is very different from the one ob-tained with energy increase pattern. For all the three samples,though S parameter after 5 keV can accord with the data obtainedwith energy increase pattern, a trough forms in the energy rangefrom 1 to 5 keV.

To examine whether the trough is due to some problems of ourslow positron beam, depth profiles of S parameter for silicon sliceare measured with the two measuring patterns, as in Fig. 2. It isclear that the curve measured with energy increase pattern andthe one obtained with energy decrease pattern can accord witheach other within the error range in the whole energy range. Thereis no doubt that the beam is on the normal operation condition.

Silicon slice is conductive, while PC, PTFE and UPVC films areinsulating. Besides, from positron lifetime experiments using stan-dard 22Na fast positron sources it is known that self-irradiationeffects in polymers may induce electrical charging of the samples[14]. It is inferred then that the notable difference observed inpolymers may relate to accumulated charges during Dopplerbroadening energy spectroscopy measurement.

With both measuring patterns, charges will accumulate. It isnecessary to pick out the one which is less influenced by accumu-lated charges. As the two measuring patterns give different depthprofiles of S parameter, the better one may be found by studythe interaction of the two measuring patterns. Four identical PTFEfilms are prepared. First, a piece of film is measured with energyincrease pattern, and another one is measured with energydecrease pattern. Then the rest two films are continuously mea-sured with both measuring patterns (energy increase pattern andenergy decrease pattern) but in different sequence. The resultsare shown in Figs. 1(c) and Fig. 3. In Fig. 1(c), curve c1 and c2 rep-resents the depth profile of S parameter independently obtainedwith energy increase pattern and energy decrease pattern, respec-tively. Fig. 3(a) shows the result of a PTFE film first measured withenergy increase pattern (curve e1) and then measured with energydecrease pattern (curve e2), Fig. 3(b) shows the result of anotherPTFE film first measured with energy decrease pattern (curve f1)and then measured with energy increase pattern (curve f2). It is

reasonable that curve e1 and c1, f1 and c2 are nearly the same,since these two curves are measured on the same condition. It isquite interesting that curve e2 and curve c2 are quite similar, withthe only difference being a little shallower trough in curve e2. Incontrast, curve f2 shows notable difference from curve c1. It isinferred that charges accumulated during the measurementwith energy decrease pattern show much greater influence ondepth profile of S parameter.

0 10 20

0.44

0.46

0.48S

para

met

er

Incidence positron energy (keV)

d1

d2Si

Fig. 2. Depth profiles of S parameter for silicon slice obtained with two measuringpatterns; black curve was measured with energy increase pattern, red curve wasmeasured with energy decrease pattern. (For interpretation of the references tocolour in this figure legend, the reader is referred to the web version of this article.)

0 5 100.42

(a)

(b)

0.43

0.44

0.45

0.46

S pa

ram

ter

Incident positron energy (keV)

e1 e2

PTFE

0 5 10

0.42

0.43

0.44

0.45

S pa

ram

ter

Incident positron energy (keV)

f1 f2

PTFE

Fig. 3. Depth profiles of S parameter for PTFE; (a) The result of a PTFE film firstmeasured with energy increase pattern (curve e1) and then measured with energydecrease pattern (curve e2), (b) the result of another PTFE film first measured withenergy decrease pattern (curve f1) and then measured with energy increase pattern(curve f2).

0 10 20

0.42

(a)

(b)

0.44

0.46

0.48

S pa

ram

eter

Positron irradiation time (h)

0 keV

6 keV h2

3 keV g2

10 keV

irradiation measurementPP

0 5 10 15

0.42

0.44

0.46

0.48

S pa

ram

eter

Irradiation time (h)

g1 g2 h1 h2

PP

after 10keV irradiation

3 keV

6 keVafter 0 keV irradiation

Fig. 4. Influence between high energy incident positrons and low energy incidentpositrons in polymer. (a) The variation of S parameter with irradiation time,magenta red curve shows the result of a pp film first irradiated by 10 keV positronsfor 12 h, and then irradiation effect of 3 keV positrons was obtained, navy bluecurve shows the result of another pp film first irradiated by 0 keV positrons for 12 h,and then irradiation effect of 6 keV positrons was obtained. The processes on theleft of the dashed line were to generate space charges, while the processes on theright were the measurement. (b) Comparison between the S–t curve for 3 keVpositrons (curve g1) and 6 keV positrons (curve h1) obtained directly and the S–tcurve for 3 keV positrons (curve g2) and 6 keV positrons (curve h2) obtained afterthe irradiation processes described in (a). (For interpretation of the references tocolour in this figure legend, the reader is referred to the web version of this article.)

J. Yang et al. / Nuclear Instruments and Methods in Physics Research B 323 (2014) 71–74 73

The difference between curves measured with the two measur-ing patterns is related to the influence range of space charges.When positrons annihilate with electrons in an insulating polymer,

positive charges will be accumulated inside the polymer, thereforean electric field will be built up during the measurement [15,16]. Inslow positron beam measurement, the mean implantation depth ofthe positron can be calculated by an empirical formula [17].Z(E)=(400/q)E1.6, where Z is expressed in angstrom, q is the densityin g/cm3, E is the incident energy in keV. According to this formula,positrons with energy E0 can penetrate into the depth Z(E0) andannihilate with an electron there, positive charges then accumu-late. Consequently, the area from surface to the depth Z(E0) is themainly influence area of space charges induced by positrons withenergy E0. When positrons with energy lower than E0 come, posi-tronium formation is severely suppressed since their annihilationsites are within the influence range of space charges accumulatedbefore. When positrons with energy higher than E0 come, thoughtheir injection paths may be influenced, their annihilation sitesare beyond the influence range of space charges. In such a way,space charges induced by low energy incident positrons willshow little influence on the annihilation of high energy incidentpositrons.

To study the influence between high energy incident positronsand low energy incident positrons, PP films are measured, asindicated in Fig. 4. One piece of film is first irradiated by 10 keV

0 5 100.42

0.44

0.46

0.48

0.50S

para

met

er

Incident positron energy (KeV)

i k

PP

Fig. 5. Depth profiles of S parameter for PP films; curve k represents the result of aPP film measured with energy increase pattern after being irradiated by 5 keVpositrons for 12 h, curve i represents the result of another PP film measured withenergy decrease pattern.

74 J. Yang et al. / Nuclear Instruments and Methods in Physics Research B 323 (2014) 71–74

positrons for 12 h, and then irradiation effect of 3 keV positrons ismeasured (curve g2). Another film is first irradiated by 0 keV pos-itrons for 12 h, and then irradiation effect of 6 keV positrons ismeasured (curve h2). For comparison, two extra films are mea-sured with 3 keV positrons (curve g1) and 6 keV positrons (curveh1) for 12 h respectively. It is obvious that 12 hours’ irradiationby 3 keV, 6 keV or 10 keV positrons doesn’t generate observablechanges in S parameter. In Fig. 4(b), curve h1 and h2 can basicallyaccord with each other, indicating that the irradiation by lowenergy incident positrons (0 keV) doesn’t have much influence onthe annihilation of high energy incident positrons (6 keV). Whilethere is a big gap between curve g1 and g2, suggesting that theirradiation by high energy incident positrons (10 keV) has greatinfluence on the annihilation of low energy incident positrons(3 keV).

Fig. 5 shows the depth profile of S parameter for PP filmobtained with energy increase pattern (curve k) after irradiationby 5 keV incident positrons for 12 h. The depth profile of S param-eter obtained with energy decrease pattern (curve i) is also shownin Fig. 5 for comparison. It is clear that curve k is quite similar tocurve i, which indicates that with energy decrease measuring pat-tern, the trough between 1 and 5 keV appears due to influence ofcharges induced by high energy incident positron (similar toprolonged irradiation by 5 keV incident positrons).

Now it is confirmed that with energy decrease measuring pat-tern, a trough between 1 and 5 keV appears due to influence ofcharges induced by high energy incident positron. With energydecrease measuring pattern, it is high energy positrons that areinjected into polymer first, charges induced by high energyincident positrons and the electric filed built thereafter willseverely suppress positronium formation of low energy incidentpositrons. While with energy increase measuring pattern, itis low energy positrons that are injected into polymer first,accumulated charges show little influence on the annihilation oflater incident high energy positrons. As a result, energy increase

measuring pattern can give the depth profile of S parameter inpolymer as it is.

4. Conclusion

Depth profiles of Doppler broadening S parameter for oxygencontaining polymer PC, fluoropolymer PTFE and chlorine contain-ing polymer UPVC were obtained with two measuring patterns,i.e. energy increase and energy decrease measuring pattern. Thetwo curves measured with different patterns can’t coincide witheach other for that a trough appeared between 1 and 5 keV inthe curve obtained with energy decrease measuring pattern. Itwas found that space charges induced by high energy incident pos-itrons greatly influenced the annihilation of low energy incidentpositrons, while space charges induced by low energy incident pos-itrons showed little influence on the annihilation of high energyincident positrons. With energy decrease pattern, it was high en-ergy positrons that were injected into polymer first, charges in-duced by high energy incident positrons and the electric filedbuilt thereafter would severely suppress positronium formationof low energy incident positrons. While with energy increase pat-tern, it was low energy positrons that were injected into polymerfirst, charges accumulated before showed little influence on theannihilation of high energy incident positrons. As a result, energyincrease pattern can give the depth profile of S parameter in poly-mer as it was. Our study revealed that the depth profile of S param-eter in polymer should measure with energy increase pattern.

Acknowledgements

This work is supported by NSFC under Grant Nos. 11175191,91226103.

References

[1] Y. Kobayashi, I. Kojima, S. Hishita, T. Suzuki, E. Asari, M. Kitajima, Phys. Rev. B52 (1995) 823.

[2] K. Ito, Y. Kobayashi, A. Nanasawa, Appl. Phys. Lett. 82 (2003) 654.[3] Y.C. Jean, H. Cao, G. Dai, R. Suzuki, T. Ohdaira, Y. Kobayashi, K. Hirata, Appl.

Surf. Sci. 116 (1997) 251.[4] R.S. Yu, X.P. Hao, Y.Y. Ma, P. Wang, X.B. Qin, Z. Zhang, B.Y. Wang, L. Wei, T.

Suzuki, Y. Ito, V.P. Shantarovich, Appl. Surf. Sci. 255 (2008) 205.[5] K. Hirata, Y. Kobayashi, S. Hishita, Y. Saitoh, Nucl. Instr. Meth. B 164 (2000) 471.[6] K. Hirata, Y. Kobayashi, Y. Saitoh, S. Hishita, Nucl. Instr. Meth. B 171 (2000) 236.[7] Y.C. Jean, J. Zhang, H. Chen, Y. Li, G. Liu, Spectrochim. Acta A 61 (2005) 1683.[8] S. Kuna, P. Coleman, A. Nejim, F. Cristiano, P. Hemment, Semicond. Sci. Technol.

13 (1998) 394.[9] K. Hirata, Y. Kobayashi, Nucl. Instr. Meth. B 136 (1998) 763.

[10] D. Cangialosi, H. Schut, M. Wübbenhorst, J. van Turnhout, A. van Veen, Radiat.Phys. Chem. 68 (2003) 507.

[11] Z. Peng, Y. Itoh, S. Li, S. Wang, Phys. Status Solidi 155 (1996) 299.[12] T. Suzuki, T. Miura, Y. Oki, M. Numajiri, K. Kondo, Y. Ito, Radiat. Phys. Chem. 45

(1995) 657.[13] J. Yang, P. Zhang, G.D. Cheng, D.X. Li, H.B. Wu, Z.X. Li, X.Z. Cao, Q.J. Jia, R.S. Yu,

B.Y. Wang, Appl. Surf. Sci. 280 (2013) 109.[14] A. Alba García, Radiat. Phys. Chem. 68 (2003) 515.[15] C.Z. Qi, S. Zhang, Y. Wu, H. Li, G. Wang, G. Huai, T. Wang, J. Ma, J. Polym. Sci.

Polym. Phys. 37 (1999) 2476.[16] X.S. Li, M.C. Boyce, J. Polym. Sci. Polym. Phys. 31 (1993) 869.[17] P. Schultz, K. Lynn, Rev. Mod. Phys. 60 (1988) 701.