0 4 (0 2 9 8 09 :52 FAX 810 988 $859

Small Angle Neutron Scattering (SANS) and Small Angle X-ray Scattering (SAXS) Analysis of

Polyaniline Salts and Blends.

Alan R. Hopkins and Paul G. Rasmussen Center for Macromolecular Science and Engineering arid Repnrm,nt of Chemisrry, The Utiiversity of Michigarr, Ann Arbor, MI 48109-10%

Rafil A. Basheer Polymers Department, Gemral Motors Research and Development Center,

Warren, M I 48090-9055

B. K. Annis and G. D. Wignall Oak Ridge Natwnal Laboratory,

Oak Ridge, TN 37831-6393

"The submitted manuscript has been authored by a eontractor of the U.S. Government under wnbact No. DEACOE960R22469. Amdiigly. the U.S. Government retains a nonexclusive.myalty-free l i m e to publish-.. orrcprcduce the p u b f i i b s d f & m ' o f t h i i ~ - ~ wntribudon, 5iiall6& o G b do"so: for S Government purposes."

[.$$J 0 fj t$@fj __ Introduction

~~- ~

Doped pol yaniline cmeraldinc salts (PANI-ES) exhibit good Pi- environmental stability with a high level of conductivity (1 to 300 Skm).

Howsver, they suffer from the fact that thcy have poor mechanical properties and must be blcnded with an insulating host polymer to be useful in industrial applications. Polyaniline blends are a new type of conducting material that typically show very low onsets of conductivity unlikc that of metal lillcd blends. This 1s primarily due to the unique cellular geometry of the PANI-ES that is formed within the insulating host polymer.' The formation of this immiscible polymer network may be due in part to structural and conformational differences in the blend components.

Electron microscopy (i.e. tunneling electron microscopy FEM))' has been extensively used to characterize the geometric derails of the fibrils that make up the wcb-like network morphology in plyaniline blends, These qualitative results define the blends as a two-phase polymer blends with semi-crystalline PAhq-ES rich fibrils intcmingled in the host polymer.

Oftea the SANS technique is a more powerful quantitative tool for evaluating domain dimensions of polymer blends in the bulk state. It takes advantage of the large difference in coherent scattering lengths between deuterium and hydrogen atoms and therefore a large contrast can be generated by dispersing a small concentration of fully deuterated polymer in a polymer matrix or vice versa2 In addition, the contrast deuteration approach does not markedly influence the thermodynamic properlies of the

polymer.2 Since previous electron micrographs suggested that blends of PAM-ES and insulating polymers form a dual phase in the host polymer. the SANS technique was used to allow further insight into the phase scparated geometry. Specifically in this study, the morphology of deuterated plyaniline emeraldine salt fully doped with camphor sulfonic acid (Le. d-

a % [

PANI-HCSA) and polycaprolactam blends were investigated. The differences in scattering length densities betwecn the labeled salt and matrix give rise to the necessary contrast to allow characterization of domain sizes. A Debye- Buechc treatment of the absolute differcntial scattering data was employed to elucidate the multi-component nature of the blend. In addition. morphology information extracted from SAXS scattering can be used to compliment SANS data.

plyanilinc sulfonate counter-anion has on the molecular structure and morphological features of the dual phase blend.

The present work is part of a larger study to understand what effect the

Experimental Procedure: Polymerized &PANI-EB was formed as follows: In an argon gas fded

glove box, approximately 5 mL of A5 miline and 50 mL of deuterium oxide were added to a h e - n e c k flask (immersed in a 35 'C oil bath) and vigorously s t i d for 10 mhuks with a mechanical stirrer. Aftcr 10 minures of stirring, 50 mL of a 2.65 M ferric chloride h D20 was drop-wise addcd oyer a So minute period Solurion was stimd for 12 hours. Thc precipitated polymer was SUb9eq~€!Dd)' washed twice with 1 .O M Ha and then with 1.0 M NH,OH. Finally, the wash was wried out wilh distilled water, plcthanol and diethyl ethcr h the order mentioned, The resulriag dark, free flowing PANI-EB powder was chien dried at 70 'C for 12 horn to remove remaining moisture and solvent. X-ray pholbelecaon spectroscopy, elemental analysis and FTIR confirmed the plyaniline in the EB oxidadon state (Le. x -0.5) and NMR ('Hand "C) confimed the PANI-EB was = 95 9bdeutemtcd

DopeddPA!!!-HCSA solutions wem prepared in HFIP by a solution dopix m c W . Au filtered, fully doped PANI-HCSA salt and blend solu&ons were sdUti0n cast onto

a Teflon coated glass substrate and covered with a glas dish to allow for a slow cvapontion at mom tcmpetaturc. All films were subscquently pcded off thc Teflon substrate and dfied in a dynamic vacuum oven at 75 'C for approximately 335 hours. From fluorine elemental d y s i s , percent residual HFIP solvent in f ibs was less than 0.5% (wtlwt). This cast proms from HFIP @wed robust. free scaadinz and solvent ftcc PANI-HCSA and PANI-ICSA / polycaprolactam films.

SANS and SAXS data acquisition. S A M data were obtained on the 30 m inslnmen? at the W.C. Koelrler facility of Oak Ridge Nationd b0ratory (OWL) via a 64 X 64 cm2 area detector with element si7s m I cm2. The Mnge of the momennun transfer or wave vector, Q, w89 0.040 - 4.0 M-' where Q = (4N'~)4n0 with h =0.475 m. the fileumn wavelength, and 6 , one half the scattering angle. Standard conectiom for detccror efficiency, backgmmd. empty cell scattering and sample transmission were applied. Free standing isotropic film samples of m 0.5 mm diameter were cut and stacked to achicvrs a sample thicbeas of a 1 m. The data were then converted to an absolute scale by comparison withsemndary standards4 and reponed as differential U-QSS sechons pcr wit volume of sample. (eldo), with Uniu of cm". This quanury expresse9 the neutron scaneMg power of a sample. Xo arrectiom were made for incoherent scattering. Typically, this is on the order of 1.

incidtntradiatiMwascopperK,(hr0,154~)fmarotatinganoderoutceequipptd with a monochromator, and a Q mgc of 0.05 - 4.3 nm" was used- For Comparison with wide angIe X-ray scattering (WAXS) (at thc same A) this comsponds IO a 29 range of 0.07 - 6.3 '. Here also standard cmections we* applicd and tbe dataconvextd to an absolute differentid scatrering cross sectiona.6

The SAXS measurpIIlcnia were conducted at both &e 10 m facility5 and O&W. The

Resultsand Discussion Scattering by salts. Reference to the SAXS data in Figure 1 shows an

I

inlense scatterjng below Q = 1 nm" for PANI-HCSA, d-PAN-HCSA and o 50 % (wtlwt) blend of the respective deuterated and hydrogenatcd salts. This

2

upswing in scattering intensity implies the presence of another component in the films such as small solvent voids. At Q = 2.5 nm", a large broad peak is seen for all three salt samples but not in the PANEEB sample which implies that thc scattering is due to the doped salt which has a pcriodicity of about 2.5 nm. For comparison with WAXS scattering, this peak corresponds to - 3.5" on a 20 scale. Interestingly, the WAXS diffracrogxam of PANI-HCSA has shown a sharp diffraction peak at - 4.5' (20 scale). Since this diffraction peak corresponds to such a large periodicity (Le. - 2.5 nm), we suggest that the charged polyaniline chains pack back-to-back with the CSA sulfonate counter-anions facing outward, as depicted in Figure 3. This type of bchavior has been proposed before by Jarvinen et. al.' and Kim el. a1.8 Both these groups report a very sharp Bragg peak at approximately 3" for plyaniline salt fully dopcd with dodecyl benzene sulfonic acid (HDBSA). Furthermore, a very similar type of WAXS diffractogram was seen by Kricheldorf e?. aI.9 using CIsH3, alkyl substituted terephthalic acid rigid rod polyesters. These reports attributed the strong, single Bragg diffraction peak at - 3" (which corresponds to a d-spacing of R 3.1 nm) to back-to-back chain packing of rigid chains with respective sulfonate countcr-anions I substituent groups radiating outward.

Scattering by d-PMI-HCSA i pdycuprolactam blends. Figure 3 shows SAXS from d-PANI-HCSA / polycaprolactam blends. At the lowest valucs of Q &e. < 0.5 nm") intense scattering again indicates the presence of some additional component (i.e. perhaps small voids) in 100 % polycaprolactam, salts and respective blends. Figure 3-6 shows that the nylon 6 matrix has a shoulder that corresponds M a spacing of about 8 rim and implies that the material is semicrystalline when cast from HFIP solvent. The same feature is found in the curves for the 5 % d-PAM-HCSA I polycaprolactam but is not apparent in the 25 % and 46 8 blends (Le. Figures 3-a, b and c respectively). This suggests that the d-PAN-HCSA salt is interacting with the polycaprolactam on a molecular scale and suppressing lamellar development at the high salt concentration. Furthermore, the structure that causes the peak at Q - 2.5 nm" remains in the blends.

polycaprolactam host polymer is seen in Hgure 4. As expected, the scattering is flal (ic. Figure 44) for the non-deuterated polycaprolactam sample and is consistent with rhc fact that there is linle scattering contrast io this host polymer. At lowest Q-values, the scattering from the 5 % d-PAM-HCSA I pdycaprolactam (Figure 4-a) is nearly an order of magnitude greater than that for thc host polymer (Figure 4-c). If the blend cornponcnts were uniformly dispersed. the difference in scattering would be much less.

The SANS cross section for the two deuterated blends, salts and

3

Consequently, most of the neutron scattering from the 5 % deuterated blend must come from a phase separated morphology.

It is surprising that in Figures 4-a and b the 25 % blend with 5 times more dcuterated material does not scatter mort than thc 5 % blend. This result may indicate a change in blend morphology. As indicated in Figure 3, the SAXS lamellar peak at Q - 0.8 nm" gradually dissipates as more d-PANI-HCSA is incorporated into the blend, indicating that the polycaprolactam lamdlae are no longer present. This indicates that the blend components might k more intermixed, explaining the lower SANS intensity seen in ]Figure 4-b.

A widely used formalism for analyzing small angle scattering data to characterize the inhomogeneties (Le. domain sizes) present in a random two-

phase material is the Debye-Bueche model" which is based on a siagle exponential correlation function. Figure 5 shows the fit of this modcl to the 5 % deutcrated blend at a low Q range (i.e. 0.5 to 1.5 nm-') by plotting (d2/dQ))*"2 vs. Q. Over this limited Q range, the correlation length. a, evaluated from the ratio of the slop to the intercept of this plot indimtes that the size of the PAXI-HCSA domain in the blend is roughly 15.3 tun. However, only a small Q range could be t i e d to the Debyc-Bueche model. The nonlinear behavior over the large Q range of Figure 5 suggesls that the d- PAM-PICSA / polycaprolactam blend cannot be treated as a simple random dual phase system.

Acknowledgement This research was in part suppomd by the Division of Materials Sciences,

Office of Basic Energy Sciencm, Department of Energy, under contract DE- AC05-%0R2W with Lockheed Martin Energy Research Corporation.

References and Notcs

(a) Reghu, M.; Yoon. C.O.; Yang, CY.; Moses, D.; Smith, P.; I3eeger.A.J. Physic. Rev. B, 1994,50, No. 19. 13931. (b) Ref. 2-a within. Wignall, G.D., Physcid Prop. of Polymers. ACS Books, 1993,313. Koehler, W.C. Physica(Amsterdam) 1986, 137B.320. Wignall, G.D.; Bates, F.S. J.Appl. Crystdlogr. 198620.28. WipnaU, G.D.; Lin, J.S.: Spooner, S. 3. Appl. C?ystdlogr. 1990, 23,241. Russell. T.P.; Lin, J.S.; Spooner, S.; WiguaU, G.D. J. Appl. Ctystdtogr. 1988.21.629. Jarvincn. H,; b, T.; Laakso, J.; Levon, K EuropeanPafenl

Kfm, M. S e i tevou, K. Amew'can Chemical Society Polymer PreprWs. 1996,37. No. 1, , 111-112. Kricbeldorf, H.R.; Wutz, C.: h b s t , N.; Domchkc. A. Gum, M. ACS Symp. Series 548, Washiaton 1 9 9 4 3 1 1-332. (a) Debye, P.; Bueche, A.M. J , ~ p p l . Pliys. 1949,20,518. (b)

# EP 0-668-594-A2,1995.

4

Debye, P.: Anderson, H. R.; Bmberger, H. J. Appl. Phys. 1958, 28, s79.

5

~- ' . . . ..

10'

l oo

i2 4% lo-l

-2 10 I I I I I I I 1

2.00 3.00 4.00 0.00 1.00

Q, nm' Flpm 1. dE(Q]/dQ vs. Q for: (a) PANI-HCSA (b) d-PANI- HCSA, (e) blcnd of 50 %I PAM-HCSA / 50 % d-PANI-HCSA and (d) PAM-EB

Back-to-back stacking or charged chalns

g?- , km&&

u 2.5 nm periodicity from SAXS data

k l p t 2 Propod &g of PN-HCSA chains from SAXS &la.

k loo e 44: -1

10

-2 l t l

SANS 3 1

10 2

ay 10

V A Q

-1 10

0.00 1.00 2.00 Q, nm'

Elgm 4. dE(Q)ldW v% Q for: (a) 5 % and (b) 15 % (WW d - PANI-HCSA I polycapmlaccam. (c) ia 100 % polymprolacw and (d) is 100 % d - PANT-HCSA.

0.00 0.50 Qa. nmQ

0.85

R p 5. Dcbyc-Bucche plot of Bcaltuing dah for 5 % (Wwt) d - PANI-HCS A I polycnpmlaMm blend.

DISCLAIMER

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof. nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsi- bility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Refer- ence herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recom- mendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

____ ___ ~-