Atmospheric dielectric barrier discharge and its application to surface modification of blood-filtering nonwoven fabrics

Indian Journal of Fibre & Textile Research Vol. 3 1 , September 2006, pp. 432-438

Atmospheric dielectric barrier discharge and its application to surface modification of blood-filtering nonwoven fabrics

Tang Xiaoliang" & Qill Gao

Plasma and Surface Research Center, College of Science, Dong Hua University, Shanghai 20005 1 , China and

Feng Xianping

Department of Physics, University of Puerto Rico, San Juan, P. R. 0093 1 -3343, USA Received I I March 2005; revised received and accepted 28 June 2005

Electricity parameters of discharge current and discharge power measured by the oscilloscope have been analyzed by using improved dielectric barrier discharge equipment. After carefully controlled discharge voltage, current, power and gap between the electrodes, an i mproved quasi-stable ntmospheric pressure dielectric bnrrier dischnrge pbsmn source is achieved. This plnsma source hns been used to modify the surfnce of melt-blown polybutylene terephthnlnte (PBT) nonwovens. It is observed thnt both the wettnblity and permention of trented melt-blown PET nonwovens nre greatly improved. The result is of great i mportance to dielectric bnrrier discharge nt ntmospheric pressure nnd its npplication to surfnce modificntion of textile mnterials.

Keywords: Dielectric barrier dischnrge, Melt-blown polybutylene terephthnlnte nonwovens, Nonwoven fnbric, Plnsmn dingnostics, Surface modification

IPC Code: Int. CI.8 A6 1 F2/00, D06M IO/OO 1 Introduction

Dielectric barrier discharge (DBD) is characterized by the presence of at least one i nsulating layer in contact with the discharge between two planar or cylindrical electrodes connected to an AC power supply (Fig. 1 ). The main advantage of this type of electrical discharge is that the nonequilibrium plasma conditions in atmospheric pressure gases can be maintained in an economic and reliable way. This has led to a number of i mportant applications i ncluding industrial ozone generation, plasma-chemical vapor deposition, pollution control, excitation of CO2 lasers, excimer lamps, and most recently the large area flat plasma display panels. I -7 Recently, extensive work in materials' surface modi fication has been done successfllll y. 8- 10

Nonwoven fabrics are found to be very i mportant in our daily lives, whether used alone or as components i n clothing, home furnishings, health care, engineering, i ndustrial and consumer goods.

Nonwoven fabrics are flat structures mainly defined as sheets or webs made by bonding and entangli ng

"To whom all the correspondence should be nddressed.

E-mnil: xltnng @dhu.edu.cn

fibres or filaments by mechani cal, thermal or chemical means. I I It is known that the blood is one of the most i mportant and essential materials used to maintain normal human activities. B ut in clinical transfusion, it is the existence of WBC that makes transfusion dangerous and risky, because WBC i n blood will transfer viruses during the transfusion. It is very necessary to transfuse blood with little, even no WBC. The nonwovens made from melt-blown process are the third excellent material used for blood filtration because of their advantages of peculiar

I

^.,,--. High voltage

, ^{____ electrode}

(n ^..^.. i· ^.

.

^. J ^...

. · t .

^h::··J -... Dielectric Discharge gap barrier

1.:-: ^{.... : ....}• • v ••.•• .• · . • • :

':·· ··"··r

^{.. ''' · ··:}

^;,v, y ^J

-____ Ground electrode

Fig. 1 -Typicnl bnrrier-dischnrge configurations

structure and performance, such as thi n fibre, small effective aperture, big surface square ratio, high i nterstice rate, well-proportioned interstice distribution, low price, etc. 12 Therefore, melt-blown polybulylene terephthalate (PBT) nonwoven fabrics are probably regarded as one of the best materials to remove WBC from blood. But a problem is that the wettabil ity of this material is poor, resulti ng i n not effectively filtering WBC in the permeation of blood.

In order to solve thi s problem, the technology of surface modification based on low-temperature plasma (L TP) is used to grant material surface new performances including wettabili ty, whereas bulk properties remain unchanged.

In a plasma source, there are sufficient amounts of energetic particles as well as the same amount of negative and positive charged particles. Bombing and etching the surface of materials will significantly modify the physical and chemical properties of the surface. In 1 994, Cao and Zhou 13 reported that the surface modification of flexible polyvinyl chloride (PYC) fi lms for medical uses is carried out by plasma treatment and polymerization. Four years later, Cheng et ^{af. 14} modified the surface of polyvinyl alcohol (PY A) by L TP. In 2000, by using electrical discharges, surface-modified polyester (PET) fabrics were obtained by loan et ^{af. ls} During early 1 999, extensive works in the field of surface-modified PBT melt-blown nonwoven were done and a great progress was made by Ke et 01. 16-1 9 at Dong Hua University, Shanghai , China. However, extensive studies were carried out on low gas pressures condition, which requires a vacuum system, thereby limiting their range of applications.

In order to avoid the expensive experi ments, a renewed interest was shown i n generating high­

pressure non-equi librium plasma, particularly at atmospheric pressures. This has been driven by the advantage offered by these plasmas in comparison to low-pressure plasmas. In this study, atmospheric pressure plasma discharge has been used to effectively modify the PBT material .

2 Materials and Methods 2.1 Melt-blown PBT Nonwovens

The melt-blown PBT nonwovens were prepared for the experiments. These materials have average fibre diameter of 2.382 Jl.m, base weight of 85 g/m2, average fabric thickness of 0.536 mm, and average aperture of 1 0 Jl.m. Based on the better understanding

of the polar groups on the surface of filter materials, which can adhere selectively leukocytes, and uniform filtration surface, the method of surface modification by atmospheric DBD plasma is adopted to i ncrease the Critical Wetting Surface Tension (CWST) and to i mprove the filtration coefficient ^(AD). To meet the requirements of both the filter material structure and surface performance, over 75 mN/m CWST and 1 0 cm- I ^Ao are required.20

2.2 Atmospheric DBD Plasma Source

The experimental system used to study DBD at atmospheric pressure i s shown i n Fig. 2. The DBD plasma is produced in an organi c glass discharge chamber mainly consists of two plane-parallel copper electrodes (250 mmx55 mm) i n which i nsulating oil was forced to circulate by two i ndependent pumps to provide active cooling. The upper electrode is covered with a dielectric barrier 0 .5 mm quartz glass). The working gases at atmospheric pressure flow through the varied discharge gap, which can be i ncreased up to several centi metres; however, to ensure stable plasma operation the gap width was limited to 5 mm.

A gauge controls the flow rate. A sinusoidal type of voltage with several tens kY is applied to the electrodes, and the frequency of power supply is varied i n the range from 10 kHz to 20 kHz. The electrical measurements on the plasma were carried out on a Tektronics TDS2000 Digital Storage

5

Fig. 2-Experimental setup of spectroscopy diagnoslics 01

dielectric barrier discharge [ I-Main chamber, 2-Electrodes, 3- Dielectric, 4-DB D power, 5-0scilloscope, 6-PC. 7- Spectrograph, 8-Lens system, 9-Vacuum pump, HI-Flow meter, and l l-Gas bottle and 12-Current-sampling resistance (R= ^{1 0} ohm), and 13-Series stabilization resistance]

434 INDIAN J. FIBRE TEXT. RES., S EPTEMBER 2006

Oscilloscope. Plasma spectroscopic emissions were recorded by using a BP-300 grating spectrograph, and all signals were directly sent to computer for data processing i mmediately during the experiment. In this study, all the experimental data and emission spectroscopic results were obtained when the materials were kept in between the electrodes.

3 Results and Discussion

The experimental determination of the power dissipated i n DBDs has often proved to be difficult because i n reali ty the power is consumed in a large number of short-lived microdischarges,z ' Therefore, the method for obtaining the plasma power involves the use of discharge Lissajous figure, when plotting transported electrical charge q through the discharge as a function of the applied periodical voltage V (Fig. 3). Thi s can be ach ieved i n a simple way by placing a measuring capacitor C.n (selected based on the maximum i nput range of the measuring i nstrument) in the current line of DBD as shown i n Fig. 4. The voltage across Cm is proportional to the charge. It can be rigorously shown that the area of a closed loop of the applied voltage vs. charge always represents the energy consumed during one period.

Figure 5 shows that the DBD can be represented as a series connection of two capacitors Cg and Cd, where Cg is the capacitance of the discharge gap and Cd the capacitance of the adjacent dielectric barrier.

In the PC-based measuring system, a microcomputer with a high-speed analog-to-digital (AID) conversion board is employed to acquire data of the charge q(t) and voltage Vet) waveforms, and these two signals are then processed to obtain the power dissipated in the DBD. The algorithm for the power calculation i s described briefly here.22

For deriving the numerical algorithm of the power calculation in the DBD, a recall of power defi nition is necessary. If l(t) is the current flow through the DBD and Vet) i s the voltage applied to it, power is obtained by the following relationship: ^{p = -}_T

1 ^f'o+TI2 ,o-TI2

V (t)I (t)dt . . . ( 1 ) where T i s the period; and P, the mean power during one period.

Since the current flow through the measuring capacitor Cm can be expressed by

I (t) ⁼ dq(t)

dt ' . . . (2)

· . . . ^{.. . .}

. .^.. ._{· ·} .^...._{. .} .^{.. . . .}_{, .} ^...._{. .}. .^..^{.. . . .}_{.. .} ^.._. ^...... _. ^....^{. . ..} . .. . .^. . ^..

· . . ^{. . . .} .

. . .^{. :} .^{. . . :} . . .. : . . . . : .... : .^{. ..... .. : .}.^{.. : " ..}

� � � � �

^:

: .'d' :

· .l/"�" '. . ^{- . . . .}

: -r. ... ; ^. : ^{: . :}

. . . _· ^{. .} _{. .} . ^. . ^. _. . . . ^{. . . . . .} _. . . .^.... _.. . ^{. . . . .} _. ^{. . .} . ^. . _. ^{. . . . . . .} _. ^{. . . ..} _. ^{.. .}

· . . ^{. .. .} . . .

· . . . ^{.. . .} . .

.^..^.._{· ·} .^{. . . .}_{. .}... . . . ._..^{' .}. . . .. .. ^_ .. _{.. -}. ..^{. _} . ^....^.. . . . ...^{. . ..} .. .. . . . .. .

••• i •••. .•• ⁱ ••••. " XVMo ^e

Fig. 3-Corresponding q-V Lissajous figure

c

em

Fig. 4-Experimental setup for voltage and charge transfer measurements

Fig. S-Diagram showing the equivalent circuit for DBD the power is obtained by

1 ^{f'o+T 12}

p ⁼ -_T

lo-TI2

V (t)dq(t) where q i s the charge released.

. . . (3)

It should be noted that Eq. (3) could be resolved by examining the i nstantaneous plot of Vet) and q(t) and integrating the closed q- V Lissajous figure. Thus, the energy consumed during one cycle of the DBD is equal to the area of the q-V trace. This means that the

• AI flow: o.4m'JH 2.e

80 • NfAr flow: O.6m'JH ^T

... ",-Ar Bow: O.8m'lH ^{• 2.4 • • •}

• ...

70 T 02-Ar flow: O.8m'JH _..

�5 ^{2.2 .. •}

T �

� 60

^{T T • .. • • T}

�

^{. . •}

f

^{2.0 A}

0 .'

':; 50 ^T • AI. ^•

l 1.e

^{• .. •}

I�

^{• • 1 ... � 0 1.e • t • • •}

{ 1.4

.. • •

..

• .. " ...

30 • ... •

... ...

• 1.2

20

0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 H O.S ^{1.0 1.5' 2.0 2.5 3.0 3.5 4.0 4.5}

Discharge ClIp, ^mm

Fig. 6-Discharge power and discharge power density of different discharge gap

q- V trace method is based on a display of Vet) and q(t). If Vet) is connected to the x-axis of the oscilloscope and q(t) to the y-axis, the corresponding q-V Lissajous figure can be obtained as shown i n Fig. 3. By means o f suitable calibration o f the vertical and horizontal scales, the area of the displayed oscillogram on the screen can be calculated to obtain discharge power by the DBD.

The PC-based measuring system calculates the discharge power of the DBD by performing the integration in Eq. (3) numerically. The analog signals of the charge q(t) and voltage Vet) are sampled simultaneously and converted into digital values qk and Vk respectively. Thus, Eq. (3) can be approximated by

11 1

p "",

/I

-(qk+1 - qk )(Vk+1 ⁺ Vk )

k=1 2 . . . (4)

where Il is the number of samples i n one cycle; and J, the fundamental frequency of the applied voltage.

Argon (Ar) and gas mixtures (NrAr, HrAr and Or Ar) were used as working gases i n the experiment.

Figure 6 shows the relationship of discharge power, power density and discharge gap distance i n atmospheric DBD. By using a kind o f above­

mentioned working gas arbitrarily, the discharge power rises on i ncreasing the discharge gap, whereas the discharge power density decreases. In practice, the plasma condition of low power and high power density is suitable for surface modification of textile materials. Therefore, short gap distance i n plasma treatments of DBD is better selected.

Air I 0.22

�v/

^{0.20 0.21} _0.19

0.18 0.17

.'-'.-�" _" ?-.. 2 ^0.16 '\ _{_---a-lit} / / ^0.15

0.14

AI - air

" 2

I 0.13

0.12

3 F-. ^{0 . 1 1}

.... .1

... _.. ,""

o

� .. -"'� A{ 0. 1 0

0.09

2 3 4 0

Discharge Gap, ^mn

..... - ...

2 3 4

Fig. 7-Relationship between discharge current and discharge gap [Resistance: I-lOOQ, 2-1000Q, and 3-1 5000Q]

Figure 7 shows the relationship of plasma discharge current and discharge gap distance i n DBD equipment at atmospheric pressure. The frequency of power supply is 1 8kHz. A series of resistances l OOQ, l OOOQ and 1 5000Q has been used for change of discharge current while power supply voltage remains unchanged during experiments. The results under two discharge gap conditions are given below:

Gap distance d ^{> -}2mm

While the discharge gap i ncreases, the plasma discharge current rises, whereas plasma discharge will be far away from glow discharge and gradually grow

436 INDIAN 1. FIBRE TEXT. RES., SEPTEMBER 2006

sparser and sparser filamentary. Plasma energy is gathered and gigantic in energy channels, resulting in punched clothing material whose structure i s destroyed i n the process o f surface modifi cation.

Finally, the filamentary discharge d isappears. I t is believed that at this period of d ischarge the line i nduced current dominates for the measured current.

Gap distance d ^< -2mm

While the discharge gap shrink, the plasma current also rises. The plasma discharge grows stronger and stronger while its homogeneity becomes better and better. The measured current is mainly related to plasma discharge current.

Since plasma discharge shows very typical performance at gap of 2mm, this typical gap distance of 2mm is defined as a critical discharge gap (CDG).23 Figure 8 shows emission spectroscopy of air and Ar-air plasma at atmospheric pressure. In general, the analysis of plasma spectroscopy at atmospheric pressure is more difficult than that at pure gas filled case. At low pressure, a simple spectral structure should be observed in plasma emission spectroscopy.

In the case of the atmospheric pressure, the recorded spectroscopic emission is much complicated due to many elements plasma emissions as shown i n Fig. 8.

Based on l iterature24.26, part of Ar plasma lines has been identified for approxi mately quantitative calculation of plasma parameters, such as electron temperature, i n order to control the process of material modification.

Assuming homogeneous and optically thi n plasma, the relative transition probabilities of two different

lines (indicated by suffixes 1 and 2) may be determined using the following expression 27.28:

. . . (5)

where A i s the tranSItIOn probability; I, the total i ntensity; A, the wavelength;

E,

the excitation energy;

k,

the Boltzmann constant; and g, the degeneracy of the upper level. Equation (5) i mplies that the upper levels involved are i n partial Local Thermodynamic Equilibrium (LTE) ; for our plasma thi s condition i s easily satisfied.28

The natural logarithm of Eq. (5) can be written as the following expression:

. . . . (6)

Considering some influence factors, such as self­

absorption, signal-to-noise ration and mutual i nterference of emission spectroscopy (Fig. 8), three argon spectral lines (wavelength: 4 1 5 .6nm, 427.22nm, 727.29 nm) were chosen. The g, A and

E

have already been reported i n the l iterature.24.26.30,3 1 In Table 1 , I i s the experimental i ntensity o f the spectral line except spectral response. Figure 9 shows a straight line of least-squares fit about

Elk

and In(IA.IgA). According to Eg. (6), the slope of the straight line is j ust electron temperature. The measurement method of electron temperature is practical and helpful to obtain other related parameters so that the process of material modification will be controlled.

5000 ,---,

4000 Air

3000

u�

2000 :: 1000

r:

�. _{350 400 450 500 550 600 650 700 750 8 00}

:.:; 2500 e

v

� S

^{. ..u -au}

<;::; 2000 H . ^. 3 I

..:: ^{- '"} . ..u gOll \1X 1.1.1 �m ; 1

�

.. ..

1 5 00 1 1 III III f;; ^...

1 000

� ..

¹

l)) �J���'��fi-.-.",

^c..

ⁱ

500 ^'" �

0 � A

350 400 450 500 550 600 650 700 750 800

\V;l'.-eIeu!:!th. Hili

Fig. 8-Emission spectroscopy of DBD plasma at atmospheric pressure

Table I -Data of spectral lines

1.8

1.7

:.: 1.B '0

::.

^{1 .5}

-I.Ll 1.4 1.3

A n m 4 1 5.6 427.22 727.29

Spectral term

3P6 � I Ss 3P7 � IS4 2P2 � I S4

• •

g

5 3 3

1 .2 L...o---..L--'---'-�-'-�-'---o-J�--'-_-L-...

o 2 3 � ^{5 8 7 8}

In (! /... .1 gil )

Fig. 9-Relationship between Elk and I n (/ AlgA)

'"

-0-

0

·c

Po.. ^.,

.5 t:: CIl

�

26 24 22 20

18

1 6 1 4 1 2 1 .5

4

2.0 2.5 3.0 3.5 4.0 4.5

Discharge Gap, om

Fig. 1 0-Treatment effects of discharge gap

O UO��10---2�O--�30�-4�O��W��60�

Lay period, days

Fig. I I -Treatment effects of lay period

1 O.6s-1 A

1 .49 0.87 2.01

After plasma treatment, the PBT sample was floated on the surface of 5% potassium bichromate solution. Through the measured wetting period of the sample, the wettability and permeation were indicated. The shorter the wetting period, the better i s the wettability.

For Fig. 1 0, the samples were pretreated for 3 min by air plasma with different discharge gaps and then kept in dry box for 14 days. For Fig. 1 1 , with 3 mm

E In/AlgA ^Elk

eV a.u. 1 0sK

14.53 894 3.9 1 1 .685

14.74 920 5.0 1 1 .683

1 3.32 1 63 2.98 1 .544

Table 2-Wettabi lity with treatment period by different working gas

Treatment period, min 3 6 1 0

Wetting period, s

Air 1 4.27 14.68 15.02

Argon < I < I < 1 < I

Fig. 1 2-SEM photographs o f PBT fibre before and after the DDD plasma treatment [(a) u ntreated, (b) after 30 s Ar-air DBD treatment, and (c) after 1 80 s air DBD treatment]

438 INDIAN J. FIBRE TEXT. RES., SEPTEMBER 2006

discharge gap the samples were also pretreated for 3 min by air plasma but kept i n dry box for different days.

Treated by different working gas (air or argon), the wettability of the samples is shown in Table 2. For pretreatment with air, when treatment period t is 1 min, the samples could not be permeated. But for samples pretreated with argon, short treatment period (t ^< 1 min) produces better results.

Figure 1 2 shows the SEM of the PBT fibre before and after the DBD plasma treatment. It can be seen that the surface is etched and the surface roughness i s increased both after 30 s Ar-air DBD treatment and after 1 80 s air DBD treatment. Compared with air DBD treatment, the Ar-air DBD treatment seems to be more quick and effective, which indirectly proves that the Ar-air i s more homogeneous than the air DBD.

4 Conclusions

The variation i n plasma discharge power and current following the change i n discharge gaps indicates the existence of critical gap d istance. When the gap between electrodes is less than that the critical gap, a quasi-stable atmospheric pressure dielectric barrier discharge plasma sowce can be achieved after carefully controlled discharge voltage and current.

Based on this plasma source treatment, both the wettabli ty and permeation of treated PBT melt-blown nonwovens are greatly improved to conform to medical use.

Acknowledgement

The authors are thankful to the Chinese Education Ministry for sponsoring the Science and Technological Research Emphasis Project (No.

03077), the "Dawn" Program of Shanghai Municipal Education Commission (No. 02SG28) and Dong Hua University.

References

1 Kogelschatz U, IEEE Trans Plasma Sci, 30 (2002) 1 400.

2 Kogelschatz U, Pure Appl Chem, 62 ( 1 990) 1 667.

3 Kogelschatz U, Proceedings, 1 0th International Conference on Gas Discharges and Their Applications (Swansea, UK),

1 992, 972.

4 Eliasson B & Kogelschatz U, IEEE Trans Plasma Sci, 1 9 ( 1 99 1 ) 309.

5 Kogelschatz U, Eliasson B & Egli W, J Phys (IV), 7 ( 1 997) 47.

6 Kogelschatz U, Eliasson B, & Egli W, Pure Appl Chem, 7 1 ( 1 999) 1 8 1 9.

7 Kogelschatz U & Salge J, High pressure plasmas: Dielectric­

barrier and corona discharges. Properties and technical applications, Low Temperature Plasma Physics, edited by R Hippler, S Pfau, M Schmidt & K H Schoenbach (Wiley­

VCH, Verlag, Berlin), 200 1 , 33 1 .

8 Yang C & Qiu G, J Dong Hua Univ, 27 (200 1 ) 9 1 .

9 Yang C, Development of New A tmospheric-Pressure Plasma Reactor and Modification to Blood-Filtering Material, Master thesis, Dong Hua University, Shanghai, China,

200 1 .

1 0 Kan C W, Chan K & Yuen C W M , Illdian J Fibre Text Res, 30 (2005) 60.

1 1 Mahmud R & Ramkumar S S, Man-made Text India, 44

(200 1 ) 34 1 .

1 2 Sun X M , Study on Graft Modificatioll of Nonwovells for Leukocytes Filtration by UV Irradiation, Master thesis, Dong Hua University, Shanghai, China, 2002.

1 3 Cao W M & Zhou K L, Chin J Synth Chem, 2 ( 1994) 57.

1 4 Cheng S H, Ning Z Y & Ge S B , PolYIll Mater Sci Ellg, 1 4 ( 1 998) 96.

1 5 loan I N , Simona D & Jonathan C, Text Res J, 70 (2000) 1 . 1 6 Ke Q F & He F M , J China Text Un iv, 25 ( 1999) 73.

1 7 Li Y & Ke Q F, J China Text Univ, 25 ( 1 999) 1 0 1 . 1 8 Ke Q F & L i Y , J China Text Univ, 2 6 (2000) 32.

1 9 Yao C X & Ke Q F, J China Text Univ, 26 (2000) 85.

20 Ke Q F, A Study 011 FUllctional Nonwoven Filters and Their Application in Leukocyte Depletion, Ph.D. thesis, Dong Hua University, Shanghai, China, 2000.

2 1 Kogelschatz U, Plasma Chem Plasma Process, 23 (2003) 1 2 .

22 Feng R, Castle G S P & Jayaram S, IEEE TrailS Ind Appl, 34 ( 1 998) 563.

23 Tang X L, Feng X P & Li Z G, Phys Experim, 24 (2004) 1 6.

24 Yang C Z, Pu X Y & Lin L Z, Spectrose Spectr Anal, 1 7 ( 1 997) 40.

25 Meiners H, J Quant Spectrosc Radiat Trallsfer, 9 ( 1 969) 1 496, 1 502.

26 Murphy P, J Opt Soc Am, 58 ( 1 968) 1 200.

27 Li T J, Allalysis of Emission Spectroscopy (Atomic energy press, Beijing, China), 1 983, 46.

28 Campbell H D, J Quallt Spectrosc Radiat Tral1.l!er, 9 ( 1 969) 461 .

29 Griem H R, Plasma Spectroscopy (McGraw-Hili Publishing Company Inc., New York, USA), 1 964, 1 48.

30 Hahn T D, Phys Rev A, 42 ( 1 990) 5747.

3 1 Wiese W L, Brault J W , Danzmann K , Helbig V & Kock M, Phys Rev A, 39 ( 1 989) 2461.